Tobacco Documents Online

1 Administrative Materials PROGRAM AGENDA FACULTY ROSTER PARTICIPANT ROSTER PROGRAM EVALUATION INSTRUMENT MAPS AND NOTES (BLUE SHEETS)

2 Arthur C. ~Jpton Chair, Institute of Environmental Medicine New York University Medical Center 550 First Avenue New York IJY 10016 212-263-5280 Peter A. Val berg Adjunct Associate Professor of Physiology Harvard School of Public Health and Senior Associate Gradient Corporation 44 Brattle Street Cambridge MA 02138 617-576-1555

9:45 - 11:15 Cancer Modeling Cohen 11:15 - 11:30 Break Applications of Expert Judyment 11:30 - 12:15 ,,c Applications of Expert Judgment Moeller 12:15- 1:15 ~ in Risk Analysis J~;S Lunch Exposure Assessment 1:15 - 2:45 ?The Respiratory System as an Entry for Exposure Valberg 2:45 •- 3:15 Refreshment Break 3:15 •- 4:45 Assessment of Exposures Ryan Friday„ September 6 Regulatory Aspects 8:30 -• 9:30 Legislative & Regulatory Aspects of Risk Brown 9:30 -• 9:45 Ref reshment Break Discussion Session - Risk Analysis for Specific Contaminants 9:45 - 11:15 1. ALAR (Daminozide) Graham 11:15 - 12:30 2. Dioxin Birnbaum 12:30 - 1:15 Lunch 1:15 - 2:00 3. Lead Hu 2:00 - 2:30 General Discussion Staff 2:30 - 2:45 Refreshment Break Course Closing 2:45 - 3:30 Risk in Perspective Wilson 3:30 - 3:45 Course Critique & Evaluation Moeller & Wilson

RISK ANALYSIS IN OCCUPATIONAL & ENVIRONMENTAL HEALTH September'4 - 6, 1991 Course Co-Leaders: Richard Wilson & Dade W. Moeller AGENDA DAY T I nzE TOP I C S PEAKER Wednesday, Septenber 4 Course Introduction 8:30 - 9:00 Welcome Moeller 9:00 -- 10:00 Introduction to Risk Analysis Wilson 10:00 -10:15 Refreshment Break 10:15 - 11:30 FDA Approach to Risk Assessments Scheuplein Introduction to Discussion Session 11:30 - 12:15 Introduction to Background Materials Moeller on: 1. ALAR (daminozide) 2. Dioxin 3. Lead 12:15 - 1:00 Lunch Tools in Risk Analysis 1:00 - 2:15 Applications of Epidemiology \ Cole 2:15 - 2:30 Refreshment Break 2:30 -;3:45 Use of Animal Data as Predictors Crouch of Human Risk 3:45 - 5:00 Endpoints Other Than Cancer Brain Thursday, September 5 Cancer & Cancer Modelintc 8:30 - 9:30 What is Cancer? 9:30 - 9:45 Refreshment Break

4 Susan R. Walker, Ph.D. Toxicologist Ecology and Environment, Inc. 1776 S. Jackson Street, Suite 200 Denver, CD 80210 303-757-4984 Thomas Wi11'ard Risk Assessment Specialist John Mathes & Associates, Inc. 701 Rodi Road, Suite 101 Pittsburgh, PA 15235 412-824-0200 Edward J. Willwerth CMC/CIH Atlantic Env & Marine Services,Inc. P0 Box 773 Plymouth, MA 02362 508-747-6944 Fred Youngs Senior Scientist National Toxics Campaign Fund 1168 Commonwealth Avenue Boston, MA 02134 617-232-0327

August 27, 1991 RISK ANALYSIS IN OCCUPATIONAl AND ENVIRONMENTAL HEALTH September 4--6, 1991 FACULTY ROSTER Course Directors Richard Wilson Mallinckrodt Professor of Physics Fellow of Adams House Physics Dept., Lyman Lab 231 Harvard University Cambridge MA 02138 617-495-3387 Dade W. Moeller Professor of Engineering Dept. Env-,ironmental Health Harvard Sc:hool of Publ i c Heal th Boston MA 02115 617-432-0793 Session Leaders Linda Birnbaum Director, Environ. Toxicology Div. Health Effects Research Laboratory U.S. Environmental Protection Agency Research Triangle Park NC 27709 919-541-2655 Joseph D. Brain Cecil K. and Philip Drinker Professor of Environmental Physiology Chair,Department of Environmental Health Harvard School of Public Health Boston MA 02115 617-432-1191 David R. Brown Chief, Environmental Epidemiology and Toxicology Hazards Section Connecticut Health Service Preventive Disease Division 150 Washington Street Hartford CT 06106 203-566-8167 Samuel M. Cohen University of Nebraska Medical Center Path./Micro. Department 600 South 42nd Street Omaha NE 68198-3135 402-559-7758 Philip Cole Department of Epidemiology University of Alabama Medical School Birmingham AL 35294 205-934-7898 Edmund A.C. Crouch Senior Scientist Cambridge Environmental, Inc. 58 Charles Street Cambridge MA 02141 617-225-0810 John D. Graham Professor of Policy and Decision Sciences, and Director, Center for Risk Analysis Harvard School of Public Health Boston MA 02115 617-432-1090 Howard Hu Assistant Professor of Occupational Medicine Harvard School of Public Health 665 Huntington Ave. Boston MA 02115 617-432-3320 Arthur Langer Environmental Sciences Laboratory Brooklyn College Brooklyn NY 11210 718-421-3851 P. Barry Ryan Associate Professor Environmental Health Harvard School of Public Health Department of Environmental Health 665 Huntington Avenue Boston MA 02115 617-432-1167 Robert Scheuplein Deputy Director Food and Drug Administration HFF101, Office of Toxicological Sciences 200 C Street, SW Washington DC 20204 202-485-0046

August 28, 1991 RISK ANALYSIS IN OCCUPATIONAL AND ENVIRONMENTAL HEALTH September 4--6, 1991 Participant Roster Paul Balserak Environmental Engineer Environmental Protection Agency 401 M Street, W.S. Washington, DC 20460 202-382-3403 ~ Mary Dymond Industrial Hygienist Barr Engineering Company 7803 Glenroy Road = Minneapolis, MN 55439 612-830-0555 Jeffrey Banikowski Pianaging Scientist O'Brien & Gere Engineers, INc. 5000 Brittonfield Parkway P.O. Box 4873 Syracuse, IV`( 13221 315-437-6100 Mark Boeni qer Industrial Hygienist NIOSH 4676 Columbia Pkwy. Cincinnati, OH 45226 513-841-4314 Marina Cofer-Wildsmith Director Environmental Lung Health American Lung Association 1118 Hampton Avenue St. Louis, MO 63139-3196 314-645-5505 Sandra A. Cotter Omaha District Corps of Engineers Attn: MROPO-T Box 1294 Dowrutown Station Omaha, NE 65,101-1294 402-221-4079 Nigel Eliam Safety Director Rohm and Haas (U.K.) Ltd. Tyneside Works, Jarrow Tyne and Wear County Durham, England NE32 3DJ 44-914-898181 , Amelia A. Gagen Acting Group Leader Lawrence Livermore National Lab P0 Box 808 Livermore, CA 94551 415-422-4278 Charles 0. Gallina Senior Nuclear Scientist Illinois Dept of Nuclear Safety 1035 Outer Park Drive Springfield, IL 62704 217-785-9936 Y. Niilagros Giles Chemist City of San Marcos 105 Meiners San Marcos, TX 78666 512-353-0426

TIPS FOR A SAFE STAY PLEASE ... Walk via main thoroughfares and well-lit areas. :C)on't wear fancy jewelry in plain sight. ~'C>on't wear your meeting badge outside the vicinity of the School. _ . }3e cognizant of and alert to your surroundings. When walking after dark, travel in groups of two or r.eore, i.e. do not travel alone. Carry your purse close to your body. Don't leave valuables in your room, use a hotel safe deposit box. Abide by common sense: if something looks suspicious, avoid it and report it .

3 Helen E. Mead Chemist USACE-MRO-I_D-EF 215N 17th Street " Omaha, NE 68102-4978 402-221-7647 Robert Mermelstein Manager, Mz~terials Safety Xerox Corporation 800 Salt Raad, Bldg. 843 Webster, NY 14580 716-422-5764 Marc Meteyer Manager Health & Environ Issues American Petroleum Institute 1220 L Street Washington, DC 20005 202-682-8209 Yusuf G. Noorani Sr. Prog/Proj Engineer E G & G, Idaho P0 Box 1625„ MS#3920 Idaho Falls„ ID 83415 208-526-6925 Thomas G. 0simitz Section Manager Product Toxicology & Env.Assessment S.C. Johnson & Son 1525 Howe Street Racine, WI 53402 414-631-2669 Sara Rasmussen Policy Analy>t Section Chief USEPA/OWS 401 M Street,. SW MC OS-311 Washington, DC 20460 202-260-3409 Bryan R. Ruble Section Manager, Analytical Lab The Drackett Company 5020 Spring Grove Avenue Cincinnati, OIH 45232 513-632-1794 i John Salladay Head, Health Physics Branch Portsmouth Naval Shipyard Portsmouth, NH 03804-5000 207-438-2588 Tom Seiter Chemist II/Haz Mat Specialist KCMO Fire Department 5130 Deramus Kansas City, MO 64120 816-483-7403 Barrie Selcoe Project Scientist O'Brien & Gere Engineers, INc. 5000 Brittonfield Parkway P.O. Box 4873 Syracuse, NY 13221 315-437-6100 Robert C. Smith V-P, Sales & Marketing The EnviroMed Companies, Inc. 414 W. California Avenue Ruston, LA 71270 318-255-0060 Richard Stark Senior Executive Consultant NUS Corporation 910 Clopper Rd. Gaithersburg, MD 20878 301-258-8599 Patricia Tilson Safety Specialist Forces Command ATTN: FCJI-SO Fort McPherson, GA 30330-6000 404-669-7481 Albert Van De Griek Deputy Director Division of Life Sciences Food & Drug Admin/CDRH/OST/DLS 12709 Twinbrook Parkway, HFZ-110 Rockville, MD 20857 301-443-2936

Wednesday, September 4 Course Introduction 8:30 - 9:00 Welcome Moeller 9:00 - 10:00 Introduction to Risk Analysis Wilson 10:00 - 10:15 Refreshment Break 10:15 - 11:30 FDA Approach to Risk Assessments Scheuplein Introduction to Discussion Session 11:30 - 12:15 Introduction to Background Materials Moeller 2:15 1:00 on: 1. ALAR (daminozide) 2. Dioxin 3. Lead Lunch Tools in Risk Analysis 1:00 -2,t15 Applications of Epidemiology Cole 2:15 - 2.,30 Refreshment Break 2:30 - 3;~45 Use of Animal Data as Predictors Crouch 3:45 -;i„00 of Human Risk Endpoints Other Than Cancer Brain

2025545687

2 Bruce G. Goodale Environmental Project Director NY State LOw Level Radioactive Waste Siting Commission 2 Third Street Troy, NY a2180 518-271-1;585 Oscar Hernandez Branch Chief US Environmental Protection Agency 11903 Antietam Road Woodbridge, VA 22192 Franca Gri1 l i Assistant Regulatory Toxicologist Hazardous Contamination Branch Ministry of' the Environment 135 St. C1' a.i r Avenue West Linda E. Jennett Director Environmental Affairs Schlumberger Env. Svces., Inc. 300 N. Main St., Suite 200 Greenville, SC 29601 803-233-0916 Suite 100 Toronto, Ontario, CAN M4V1P5 416-323-5076 Jong-Suk Kim Sidney E. Grossberg, M.D. Professor and Chairman Department of Microbiology Director-General Korean Ministry of Environment 635-298 Pongcheon 9-dong, Kwank-gu Seoul, Korea 617-432-4637 Medical College of Wisconsin 8701 Watert,Dwn Plank Rd. Milwaukee, WI 53226 414-257-8427 Betsey Kuhn Deputy Director Michael Gunn Asst. to the City manager City of Cincinnati Resources & Technology Division Economic Research Service 1301 New York Avenue, NW Washington, DC 20005 202-219-0449 801 P1 um Street, Rm. 24 Cincinnati, OH 45202 513-352-3790 Jeffrey Kutcher Aanel i a A. Hagen Acting Group Leader Lawrence Livermore National Lab Consutants in Epidemiology and Occupational Health 2428 Wisconsin Ave., NW Washington, DC 20007 202-333-2364 P0 Box 808, 7000 East Ave, L-255 Livermore, CA 94551 415-422-4278 Angela Li-Muller William C. Hayes Environmental Specialist Dow Chemical Company Senior Regulatory Toxicologist Ontario Ministry of the Environment 135 St. Clair Ave West Toronto, Ontario M4V 1P5 416-323-5114 V 2020 Dow Center Midland, MI 48674 517-636-2664 Mayada Logue O N ~11 Scientist Philip Morris v't ~ Jose E. Hernandez P0 Box 26603 Head, Indusi:ri al Hygi ene Dept. Richmond, VA 23261 ~ Navy Environrriental Health Center 804-274-3189 ~ 2510 Walmer Ave. © Norfolk, VA 23664 804-444-7575

TO: Continuing Education Participant~s ~ FROM: David A. S. Klipp, Program Coordinator RE: Telephone Locations and Protocol Here are the-telephone numbers which will serve to deliver messages to you during your stay with us at Harvard: During Class Hours: 617-432-1109 if that line is busy or there is no answer: 617-432-1171. We will take messages for you and post them either at the back of the classroom or at your place. In the event of an emergency, we will interrupt you in class. Here •is-•a- map-indicating (**)-the locations of the -four--public telephones in the School of Public Health: Lounge ** ~ ** (1st Floor) _ T I~ Elevators 11{11(stairway) -- ' \/ \) Classroom Hallway Lavatories PLEASE NOTE: ~ O ~ ~ Acknowledging the limited number of public telephones available C1Y here, we ask that you please try to consider keeping your calls 1$ brief and to a minimum. ~ ~ ~

i V A ® M1!®ENTL4L .114M l.r LucMrsllt ® SCIEMCS PARK NORTH STATtOf! MAY1iA!lRST VowooIN*hm aO'VtRM1iEHT CTRI L PARK sT/IEST A~aa~®w~ C®MMUTElI /RAEte LiNlt docrun . av.k43T~~oa ![ o'l.rf~ fn ral f~as~ fYNP110Mif 1Ml3EW/ ~E s OAK a/1c1Ft ktALDEN CENTER INELLINGTON St1LL1VAN SQUARE C09AMUfMTr CdLLSSii 0 ® =" fwxmw" FRAmam unAmm zi El REVERE BEACH dEACHMONT SUfFOLK DOWNS ORlENT HSiGNTS ® fA/W FMLL RAPID TRANSIT LINES • MO/Rk RIEIfl= COibitlt ,. yw 74. •..• ZF.glM55"11f56R-)/6

HARVARD MEDICAL AREA IKEY TO uaAPf I Rotch Bulldinp- 55 Shattuck Stre.t OTHER MEDICAL AREA FACILITIES 14 New Engiand Deaconeu Hoipital HARVAR D SCHOOL OF PUBLIC HEALTH • Francis A. Countway Librory of M.dklns 15 Beth Israel Hospital 2 Heelth Scimces :aboratovies, Dldgs, I ord 11 10 Shattuck Strett 16 Judqe Baker Guidance Center 665 Huntinpton Avenuo 9 Peter Bent Bripham HospitN 17 Boston Hospitat for Women (Lyin¢In Divkion) 10 Massochusatts MentaJ Health Canter 3 Sebastian S. Krage Educ®tiond F.cilidp Sullding 1e Madicat Ar®. Coop: 677 Huntlnptonb Avanut 11 Chlldren's Cancer Research Foundatlon 19 Jimmy Fund Auditorium 4 Henry LnSh.ttuclc International Houe. 12 Childnn'e Hospitd Medicta CentM 189-203-207 Park Drhrt 20 Harvard M.dicat School Labontory for Human 13 Shields Warren Radiation Latwr.tory Reproductlon.nd Repeoductive Biology HARVARD MEDICAL SCHOOL S Medical School Uwtdrangla 25 Shattuck ;itre.t 5A Harvard M.dica! School -Adminimadon Wpdky 3 V.nderbilt H.II -- 109 Av®nue Loui. P..tses (Medkal H.atth S.rvkatl HARVARD SCHOOL OF DENTAL MEDICINE 7 Hanard Schoot of Dantd M.edicin. 188 Lonp+vocb Av.atus Shattuck In- ~ternational House

SCIENCE AND ITS LIIVIa[TSo The Regulator's Dilemma Alvin M. Weinberg PROLOGUE: The shift in environmental concerns from visible pollution to more subtle threats, such as toxic pollutanis, presents special problems for regulators xho must fi4nction outside the limits of scientific certainly. The same handicap besets judges who tnust adjudicate disputes over claims for damages arising from new and hazardous technologies that involve adverse health effects that are latent or unpredictable. In this area of uncertainty in which accidental exposure to hazards is rare, scieni'i;,,ts resort to probabilistic risk assessment to estimate the likeli- hood and consequences of events that may carry a threat to human health. Such scienti,6c techniques for the investigation of rare events, however, often cannot provide definitive answers for regulators and judges. In this essaY phtsicist .9lvin Id'einberg suggests that instead of asking sci- entists for answers to unanswerable questions, regulators should settle for less-definitive answers and regulate on the basis of uncertainty. Technologi- cal ftxes, including greater reliance on inherent safetyfeatures that depend on the imnnutable laws of nature, can help reduce risk. But ultimately, says ti'einberg, it tna' y be necessary to establish some threshold beyond which blame for accidents and other untoward events would be unprovable and vic- tims would be compensated by a society as a whole. t Alvin A1. Id'einberg received his Ph.D. in physics from the University of Chicago in i' 939. He has been a leading figure in the development of nu- clear energy and has served as director of the Oak Ridge National Labora- ton- and as director of the Institute for Energy Analysis of the Oak Ridge Associated Universities. He is the coauthor of The Physical Theory of Neu- tron Chain Reaction (1958) and has written extensivelv on nuclear energy, nuclear proliferation, and the interaction between modern technology and s(X'iet 1'. FALL 1985 59

Despite the difficulties, scientific mechanisms have been devised for estimating, however imperfectly, the probability of rare events. For accidents the technique is probabilistic risk assessment (PRA); for low-level insults various empirical and theoretical approaches are used. Although probabilistic risk assessment had been used in the aerospace industry for a long time (for example, to predict the reliability of compo- nents), it first sprang into public prominence in 1975 with a reactor safety study directed by nuclear engineer Norman C. Rasmussen.3 The Rasmussen study, sponsored by the Atomic Energy Commission (now known as the Nuclear Regulatory Commission), was designed to estimate the public risks involved in potential accidents at commercial nuclear reactors. Probabilistic risk assessment, when applied to nuclear reactors, seeks to identify all sequences of subsystem failures that may lead to a failure of the overall system; it then tries to estimate the consequences of each subsystem failure so identified. The result is a probability distribution, P(C): that is., the probability, P, per reactor year, of a consequence having magnitude C. Consequences include both material damage and health effects. Usually, the probability of accidents having large consequences is less than the probability of accidents having small consequences. A probabilistic risk assessment for a reactor requires two separate estimates: first, an estimate of the probability of each accident sequence; second, an estimate of the consequences-particularly the damage to human health-caused by the uncontrolled radioactive effluents released in the accident. An accident sequence is a series of equipment or human malfunc- tions, such as a pump that fails to start, a valve that does not close, or an oper- ator confusing an ON with an OFF signal. We have statistical data for many of these individual events; for example, enough valves have operated for enough years so that we can, at least in principle, make pretty good estimates of the probability of failure. Uncertainties still remain, however, because we can never be certain that we have identified every relevant sequence. Proof of the adequacy of proba- bilistic risk assessment must therefore await the accumulation of operating experience. For example, the median probability of a core melt in a light water reactor, according to the 1975 Rasmussen study, was I in every 20,000 reactor-years; the core melt at Three Mile Island's number two reactor (TMI- 2) occurred after only 700 reactor-years. The number two reactor, however, differed from the reactors Rasmussen studied, and in retrospect, one could rationalize most of the discrepancy between his estimate and the seemingly premature occurrence at TMI-2. Since the core melt at Three Mile Island. the world's light water reactors have accumulated some 1,500 reactor-years of operation without a core melt. This performance places an upper limit on the a priori estimate of the core- melt probability. Thus. if this probability were as high as I in every 1,000 reactor vears. the likelihood of su5iving 1,500 reactor-years would not be more than 22 percent; put otherwise, we can say with 78 percent confidence that the core-melt probability is not as high as I in 1,000 reactor years. With 500 light water reactors on line in the world, should we survive until the year 2000 without another core melt, we could then say with 95 percent confidence 62 ISSI!ES IN SCIENCE AND TECH'.OLOG1'

I n his essay "Risk, Science, and Democracy," William D. Ruckelshaus expresses very clearly what I call the regulator's dilemma. During the past 15 years. Ruckelshaus notes, there has been a shift in public emphasis from visible and demonstrable pollution problems, such as smog resulting from automobiles and raw sewage, to potential and largely invisible problems, such as the effects of low concentrations of toxic pollutants on human health. This shift is important for two reasons. First, it has changed the way that science is applied to practical questions of public health protection and environmental regulation. Second, it has raised difficult questions about managing chronic risks within the context of free and democratic institutions.z When the environmental concern was patent and obvious-such as the problem of smog in Los Angeles-science could and did provide unequivocal answers. Smog, for example, comes from the gas emissions from burning liquid hydrocarbons, and the answer to the smog problem lies in controlling these emissions. The regulator's course was rather straightforward because the science upon which regulatory decisions are made was operating well within its power. However, when the environmental concern is subtle-for example, how much cancer is caused by an increase of 10 percent in mean background radiation-science is being asked a question that lies beyond its power; the question is trans-scientific. Yet the regulator, by law, is expected to regulate even though science can hardly help him; this is the regulator's dilemma. Although my essay is subtitled The Regulator's Dilemma, many of the same issues anse in the adjudication of disputes over who is to blame and who is to be compensated for damage allegedly caused by rare events, such as nuclear accidents. The regulator's dilemma is also faced by the judge who is presiding over a tort case involving, for example, a claim for damages blamed on a toxic waste dump. Indeed, the regulator's dilemma could equally be called the toxic tort dilemma. A lawsuit involving alleged injury from chemical pollutants is unlike the traditional liability case. If my car injures a pedestrian, I am liable to be sued. What is at issue, however, is not whether I have injured a pedestrian. Rather, it is whether I am at fault. On the other hand, if the lead from my car's exhaust is alleged to cause bodily harm, the issue is not whether my car emitted the lead but whether the lead actually caused the alleged harm. The two situations are quite diffetent.. In the first example the relation between cause and injury is not at issue. In the second it is the issue. In this essay, therefore, I try to delineate more precisely those limits to sci- ence that give rise to the regulator's dilemma. I speculate on how these intrinsic limits to science seem to have catalyzed a profound attack on science by some sociologists and public-interest activists. In addition, I offer a few ideas that may help the harried regulators finesse these trans-scientific issues. I I Science deals with regularities in our experience; art deals with singularities. It is no wonder that science tends to lose its predictive or even explanatory power when the phenomena it deals with are singular, 60 ISSUES IN SCIENCE AND TECHNOLOGY

Ii. Supreme Qturt of the United States: Industrial Union Depanment, AFl: CIO v. American Petroleum Institute. et al., argued October 10. 19'9. dee:ided July 2. 1980. No. .'8-911. Washington, DC (1980). IG. Anierican Industrial Health Council: Q mment on: A Rep ut < f the GueraKenq- Regulatory Liaix>n Group Entitled ':ticientific Ba.u~ for Ielentif}•inK Potential Carcinogens and Estimating their Ri,k." AIHC. "iar.uLle. ( btay S. 19-9 ). 17. Purchase, LF.: Inter•,pecies Compari.xxts of Carcinogenicity. Br. J. Cancer 41:454-46K (19ki0). 18. l'.S. EmOronmerttal Protection Agency: Policy and Procedures for IdentiF<ing, and A,sessinR and Regulating Airtx)rne Substances Pc .tiinR a Risk of Cancer. Pr<tpu ecl Rule Fed. Reg. a4:5£3(i42 (19t30). Received 8 28r89: review decision 12/t/89; revision 2/8190; accepted 3/9/90 APPL OCCUP. ENVIRON. HYG. 5/8) • AUGUST 1990 517

Instead of asking science for answers to unanswerable questions, regulators should be content wa th less far-reach i ng answers. negotiation between individuals in conflict because they hold different non- scientific beliefs, how can one say that this scientist's opinion is preferable to that one's? Furthermore, if the matter at issue moves across the boundary between science and trans-science, where all we can say with certainty is that uncertainties are very large, how much less able are we to distinguish between the expert and the charlatan, between the scientist who tries to adhere to the usual norms of scientific behavior and the scientist who suppresses facts that conflict with his political, social, or moral preconceptions? One way to deal with these assaults on scientists and scientific truth would be to define a new branch of science, called regulatory science, in which the norms of scientific proof are less demanding than are the norms in ordinary science. I should think that a far more honest and straightforward way of dealing with the intrinsic inability of science to predict the occurrence of rare events is to concede this limitation and not to ask of science or scientists more than they are capable of providing. Instead of asking science for answers to unanswerable questions, regulators should be content with less far-reaching answers. For example, where the ranges of uncertainty can be established, regulate on the basis of uncertainty; where the ranges of uncer- tainty are so wide as to be meaningless, recast the question so that regulation does not depend on answers to the unanswerable. Furthermore, because these same limits apply to litigation, the legal system should recognize, much more explicitly than it has, that science and scientists often have little to say, probably much less than some scientific activists would admit. The expertise of scientific adversaries is often at the heart of litigation over personal injury alleged to be caused by subtle, low-level exposures. Each side presents witnesses whose scientific credentials it regards as impeccable. Because the issues themselves tend to be trans-scientific, one can hardly decide the validity of the assertions of either side's witnesses. Under the circumstances, I suppose, one is justified in regarding a scientific witness no differently than any other witness; his credibility is judged by his past record, behavior, and general demeanor, as well as the self-consistency of his testi- mony. Such, at least, was the way in which a federal district court judge, Patrick Kelley, settled Johnston v. United States, in which the issue was the claim that exposure to radiation from reworking old aircraft instrument dials had caused injury; Kelley impugned, on grounds no different from those one would invoke in an ordinary lawsuit, the competence if not the integrity of some of the plaintiffs scientific witnesses. VII There are various ways to provide some assurance of safety despite uncertainty. Here I briefly describe two of these ways-which I call the technological fix and de minimis-without claiming that these are the most important, let alone the only, ones. Technological fix. Science cannot exactly predict the probability of a serious accident in a light water reactor or the likelihood that a radioactive waste canister in a depository will dissolve and release radioactivity to the environment. Can one design reactors or waste canisters for which the 68 ISSUES IN SCIENCE AND TECHNOLOGY

HAZARDS: SCIENCE AND ITS LIMITS boundary between science and trans-science will recede toward events of lower frequency. At any stage, however, the boundary is fuzzy, and much scientific aontroversy boils overdeciding where it lies. One need only read the violent exchange between Edward P. Radford and Harald H. Rossi over the risk of cancer from low levels of radiation to recognize that where the facts are obscure, argument--even ad hominem argument-blossoms.tz Indeed, Alice Whittemore in her "Facts and Values in Risk Analysis for Environmental Toxicants," has pointed out that facts and values are always intermingled at this "rare event" boundary between science and trans-science." A scientist who believes that nuclear energy is evil because it inevitably leads to prolifera- tion of nuclear weapons (which is a common basis for opposition to nuclear energy) is likely to judge the data on induction of leukemia from low-level exposures at Nagasaki differently than is a scientist whose whole career has been devoted to making nuclear power work. Cognitive dissonance is all but unavoidabli: when the data are ambiguous and the social and political stakes are high. VI No one -would dispute that judgments of scientific truth are much affected by the scientist's value system when the issues are at or close to the boundary lx~tween science and trans-science. On the other hand, as the matter under dispute approaches the domain of science, most would claim that the scientist's extrascientific values intrude less and less. Soviet scientists and U.S. scientists may disagree on the effectiveness of a ballistic missile defense, but they agree on the cross section of U23` or the lifetime of the pi meson. This all seems obvious, even trite. Yet in the past decade or so a school of sociology of' knowledge has sprung up in Great Britain that claims that "scientific vicws are deter7nined by social (external) conditions, rather than by the internal logic of scientific tradition and inherent characteristics of the phenomen:jl w•orld,"'4 or that "all knowledge and knowledge claims are to be treated as lxing socially constructed: genesis, acceptance, and rejection of knowledge [is] sought in the domain of the Social World rather than... the Natural Wor'l!.d."15 The attack here is not on science at the boundary with trans-science, in particular--the prediction of the frequency of rare events. At least the more extreme of the sociologists of knowledge claim that using traditional ways of establishing scientific truth-by appealing to nature in a disciplined man- ner-is not how science really works. Scientists are seen as competitors for prestige, pay, and power, and it is the interplay among these conflicting aspirations, not the working of some underlying scientific ethic, that defines scientific truth. To be sure, these attitudes toward science are not widely held by practicing scientists; however, they are taken seriously by many political activists who, though not in the mainstream of science, nevertheless exert important influence on other institutions-the press, the media, the courts- that ultimately influence public attitudes toward science and its technologies. If one takes such a caricature of science seriously, how can one trust a sci- entific expert? If scientific truth, even at the core of science, is decided by FALL 1985 67

The Delaney clause is the worst example of how a disregard of an intrinsic lir,nit of science can lead to bad policy by overenthusiastic politicians. When one compares the relative intrinsic safety of two very similar devices-such as two water-moderated reactors-probabilistic risk assess- ment is on much more solid ground. Here one is not asking for absolute estimates of risk, but rather for estimates of relative safety. If reactors A and B differ in only a few details-say reactor A has two auxiliary water feed trains whereas B has only one-the ratio of core-melt probabilities should be much more reliable than their absolute values because the ratio requires an estimate of failure of a single subsystem, in this instance the extra auxiliary water feed on reactor A. Not only can one say with reasonable assurance how much safer reactor A is than reactor B, but as a result of the detailed analysis one can identify the subsystems that contribute most to the estimated failure rate. Even if proba- bilistic risk assessment is inaccurate, it is very useful in unearthing deficien- cies; one can hardly deny that a reactor in which deficiencies revealed by probabilistic risk assessment have been corrected is safer than one in which they have not been corrected, even if one is unwilling to say how much safer. Somewhat the same considerations apply to low-level insult. An agent that does not shorten lifespan at high dose will not shorten lifespan at low dose. An agent that is a very powerful carcinogen at high dose is more likely to be a carcinogen at low dose than one that is a less powerful high-dose carcinogen. Thus, animal experiments surely are useful in deciding which agents to worry about and which not to worry about. Of course, the Ames test (which determines by a relatively simple procedure whether a substance is mutagenic) has made at least some preliminary screening of carcinogens more feasible because substances that cause mutations are considered to be poten- tial carcinogens. The difficulty today seems to be not so much identifying agents that at high dose may be carcinogens as it is prohibitin- exposures far below levels at which no effect can be, or perhaps ever will be, demonstrated. The regulator and the concerned citizen are inclined to approve the Delaney clause of the Federal Food, Drug, and Cosmetic Act, which prohibits the use of any food additive that has been shown to cause cancer in laboratory animals or humans. This clause, however, is of no help in resolving such issues as the relative risks of, say, cancer induction by nitrosamines (carcinogenic compounds that can be formed in the body from nitrites) and digestive disorders caused by meat untreated with nitrites. The Delaney clause is the worst example of how a disregard of an intrinsic limit of science can lead to bad policy by overenthusiastic politicians. Harvard physicist Harvey Brooks has oft~ en pointed out that one can never prove the impossibility of an event that is not forbidden by a law of nature. Most will agree that a perpetual motion machine is impossible because it violates the laws of thermodynamics. That one molecule of a polychlorinated biphenyl (PCB) may cause a cancer in humans is a proposition that violates no law of nature: hence many, even within the scientific community, seem willing to believe that this possibility is something to worry about. It was this error that led to the Delaney clause. When is an event so rare that the prediction of its occurrence forever lies outside the domain of science and therefore within the domain of trans- science? Clearly we cannot say, and perhaps as science progresses, this 66 ISSUES IN SCIENCE AND TECHNOLOGY

HAZARDS: SCiIENCE AND ITS LIMITS that the core-melt probability is not higher than I in 3,000 reactor-years. In the absence uf such experience, one is left with rather subjective judgments. Although Harold «' Lewis, in his critique of Rasmussen's 1975 study,' asserts that he could not place a bound on the uncertainty of probabilistic risk assessment, Rasmussen argued that his estimate of core-melt probability may be in error by about a factor of 10 either way-that is, the probability may be as high as 1 in 2,000 reactor-years or as low as I in 200,000 reactor-years. As we see, after 3,000 reactor-years of operation without a core melt, we can say with about 78 percent confidence that Rasmussen's upper limit (1 in 2,000 reactor-years) is not too optimistic. Furthermore, if we survive to the year 2000 wir.hout a core melt, the confidence level with which we can make this assertion rises to 95 percent. Our confidence in probabilistic risk assess- ment can eventually be tested against actual, observable experience. Until this experience has been accumulated, however, we must concede that any probability we predict must be highly uncertain. To this degree our science is incapable of dealing with rare accidents, but time, so to speak, annihilates uncertainty in estimates of accident probability. Unfortunately, time does not annihilate uncertainties over consequences as unequivortlly as it does uncertainties over frequency of accidents. A large reactor or chemical plant accident can canse both immediate, acute health effects and delayed, chronic effect,;. If the exposure either to radiation or to methyl isocyanate is high enough, the effect on health is quite certain. For example, asi,ngle exposure of about 400 rems will cause about half of the people exposed to die. On the other hand, in a large accident many people will also be exposed to smaller doses-indeed, to doses so low that the resulting health effects are undetectable. At Bhopal many thousands of people were exposed to rni°thyl isocyanate but they recovered. We cannot say positively whether or r,ot they will suffer some chronic disability. The very worst accident envisaged in the Rasmussen study, with a probability of I in I billion reactor-years, projected an estimated 3,300 early fatalities, 45,O{30 early illnesses, and 1,500 delayed cancers per year among 10 million exp<)&°d people. Almost all of the estimated delayed cancers are attributed to exposures ofless than 1,000 millirems per year-a level at which we are very hard put to estimate the risk of inducing cancer. Similarly, the American Physical Society's critique of the Rasmussen study attributed an additional 10,000 deaths over 30 years among 10 million people exposed to cesium-135 distributed in a very large aecident.s The average exposure in this case was assurred to be 250 millirems per year-again, a level at which our es- timates of the health effects are extremely uncertain. Has the nuclear community, particularly its regulators, figuratively shot itself in the foot by trying to estimate the number of delayed casualties as a re- sult of these low-level exposures? In retrospect, I think the Rasmussen study would have been on more solid ground had it confined its estimates to those health effects resulting from exposures at higher levels, where science makes reliable estimates. For the lower exposures the consequences could have been stated simply as the number of man-rems (the number of people multiplied by the number of rems) of exposure of individuals whose total exposure did not exceed, say, 5,000 millirems, without trying to convert this man-rems number Our confidence in probabilistic risk assessment can eventually be tested against actual, observable experience. FALL 1985 63

Introduction to Risk Analysis Wilson

H4ZARDS SCIEtiCE 4tiD ITS LIMITS irreproducible, and one of a kind-in other words, rare. Although science can often anal;~ze a rare event after the fact-for example. the extinction of ,4 ir;os:iur< during the Cretaccous-Ter;ian p,:rio.i fullo,,.ine the presumed ccllision of the earth and an asteroid-it has great difficult\ predicting when such an uncommon event will occur. I distinguish here between two sorts of rare events-accidents and low- level insults.. whose potential to cause injunt is unknown. Accidents are large- scale malfu nctions whose etiology is not in doubt, but whose likelihood is very small. The partial nuclear reactor meltdown at Three Mile Island in 1979 and the release of toxic gas from a chemical plant at Bhopal, India, in 1984 are ex- amples of accidents. The precursors to these specific events-for example, the condition of the auxiliary water feed system and other components at Three Mile Island-and the way in which the accidents unfolded are well under- stood. Estimates of the likelihood of the particular sequence of malfunctions are less firmly grounded. As the number of individual accidents increases, prediction of their probability becomes more and moi-e reliable. We can predict very well how many automobile fatalities will occur in 1986; we can hardly claim the same degree of reliability in predicting the number of serious reactor accidents in 1986. Low-level insults are rare in a rather different sense. We know that about (00 rems of'radiation will double the mutation rate in a large population of exposed mice. How many mutations will occur in a population of mice exposed to 100 millirems of radiation? In this case the mutations. if induced at all by such Iow levels of exposure. are so rare that to demonstrate an effect with 95 percent confidence would require the examination of many millions of mice. Although such an effort is not impossible in principle. it is in practice. Moreover, even if we could perform so heroic a mouse experiment, the extrapolation of these findings to humans would still be fraught with uncer- tainty. Thus. human injury or abuse from low-level exposure to radiation is a rare event whose frequency cannot be accurately predicted. III When dealing with events of this sort, science resorts to the language of probability. Instead of saying that this accident will happen on that date. or that a particular person exposed to a low-level dose of radiation will suffer a particular fate. it tries to assign probabilities for such occurrences. Of course, ~khere the number of instances is very large or the underlying mechanisms are fully under>tood, the probabilities are themselves perfectly reliable. In quan- tum mechanics there is no uncertainty as to the probability distribution of the phenomenon being described. In the class of phenomena considered here, however, even though the likelihood of an event happening or of a disease being caused by a specific exposure is given as a probability, the probability it- self is very uncertain. One can think of a somewhat fuzzy demarcation between what, I have called science and trans-science. The domain of science covers phenomena that are deterniinistic or whose probability of occurrence can itself be stated precisely; in contrast, trans-science covers those events whose probability of occurrence is itself highly uncertain. Although science can often analyze a rare event after the fact, it has great difficulty predicting when such an uncommon event will occur. F4tL P485 , 61

OSHA ha,d failed to show a significant reductior in risk going frc;m 10 to I ppm. Although OSHA did not propose a fi>rmal policy in response to the decision, the agenn, has generalhy accepted the view that quantitation of risk is required for the regulation of carcinogens, and it has in- Q-()rporated QRAs into its standard setting activitysince that time. A~iiitLst th.e controversy associated with modeling a pro- cess th: is incompletely undersaxxi scientifically and the judicial political climate which favors the use of a quan- titative procedure to help regulate carcinogens in the gen- eral environment and workplace, the EPA and other reg- ulatony agencies have opted for the use of a conservative approach in the development of risk assessment proce- dures. For example, the QRAs are usually based on t;,e most sensitive species and use of the most conservative dose-response curve, while low weight is given to neg- ative epidem iological data.~ 5) Although this procedure has been criticized by some industry representatives(16) and some academic scientists,07t it would be difficult to per- form numerous risk calculations involving all plausible options for the many judgments that must be made in the development of a QRA. For most chemicals, this would result in such a wide range of risk estimates that the anal- vsis would not be u.seful to the regulatory agency or to others formulating public policy. The CS-TLV Committee, on the other hand, provides recommendations for the use -3 -2 10 10 . rn9/m3 -1 10 10 0 of industrial hygienists rather than setting governmental standards, and the Committee bases its recommendation on the professional judgement of its members. Both the TLVs and QRAs are subject to external reviews before adoption. With this perspective in mind, the authors compared the chemical carcinogens which were quantitatively evaluated by the two procedures. The first qualitative comparison is that only 16 chemicals appear on both the CS-TLV Com- mittee list and the EPA list. However, this apparent dis- agreement is not too surprising because of the substantially different mission of these two organizations and the ap- proaches they take when classifying the hazardous "po- tency" of chemicals. TLVs are quantitative guidelines for recommended exposures in the workplace, hut there is no explicit estimate of the health risk associated with these leveLs. On the other hand, the EPA unit risk factor explicitly relates dose to cancer risk by means of a mathematical, linearized, multistage model of rarcinogenesis, but few have been translated into pc rmissible exposure levels. Op- erating as an independent organization, the IARC reviews all relevant scientific information in order to assess the evidence that an agent could alter the incidence of cancer in humans but makes no attempt to extrapolate beyond the range of the available data. Likewise, no recommen- dation is given for safe exposure levels for regulation or legislation.t t4/ 1 10 10 2 3 10 concentration (tog scate)• 10 It 10 5 FGURE 2. Comparison of TLV, EPA unit risk dose-response, and animal bioassay re.sufts for vinyl cttlotide exposure. APPL OCCUP. ENV78ON. NYG. 3/81 • AUGUST 1990 515

~a

HAZARDS: SCIENCE AND ITS LIMITS probability of such occurrences is zero-or at least, where the prevention of such mishaps relies on immutable laws of nature that can never fail rather than on the less than reliable intervention of electromechanical devices? Surprisingay, this approach to nuclear safety has come into prominence only in the past five years. Kare Hannerz in Sweden and Herbert Reutler and Gunter H, Lohnert in West Germany have proposed reactor systems whose safety does not depend on active interventions, but rather on passive, inherent characteli:,tics.16 Although one cannot say that the probability of mischance has been reduced to zero, there is little doubt that the probabilities are several, perhaps three, orders of magnitude lower than the probabilities of mischance for existing reactors. To the extent that such proposed reactors embody the principle of inherent safety, their adoption would avoid much of the dispute over reactor safety, the limits on nuclear accident liability contained in the Price-Andcrson Act, repetition of the Three Mile Island accident, and so forth. In short, such a technological fix enables one largely to ignore the uncertainties in any prediction of core-melt probabilities. The idea of incorporating inherent or passive safety into the design of chemical p,lants had been proposed by Trevor A. Kletz of the Loughborough University,of Technology in 1974, shortly after the disaster at the Flixborough cyclohexane plant, which killed 28 people." I suspect that one of the main consequences of the Bhopal disaster will be the incorporation of inherent safety features into new chemical plants; again, a way of finessing uncertainty in predicting failure probabilities. De minimis. A perfect technological fix, such as a totally safe reactor or a crash-proof car, is usually not available, at least at an affordable cost. Some low-level exposure to materials that are toxic at high levels is inevitable, even though we can never accurately establish the risk of such exposure. One way of dealing with this situation is to invoke the principle of de minimis. This principle, as Howard I. Adler and I suggested several years ago, argues that for insults that occur naturally and to which the biosphere has always been exposed and presumably to which it has adapted, one should not worry about any additional man-made exposure as long as the man-made exposure is small compared to the natural exposure.t8 The basic idea is that the natural level of a ubiquitous exposure (such as cosmic radiation), if it is deleterious, cannot have been very deleterious because in spite of its ubiquity, humans have survived. Moreover, we do not know and can never know what the residual effect of that natural exposure really is. An additional exposure that is small compared to natural background radiation should be acceptable; at the very least, its deleterious effect, if any, cannot be determined. Adler and I suggested that for radiation whose natural background is well known, one imay choose a de minimis level as the standard deviation of the natural background. This turns out to be around 20 percent of the mean background, around 20 millirems per year; this value has been used as the Environmelzial Protection Agency standard for exposure to the entire radio- chemical fuel cycle. Scientists know more about the natural incidence and biological effects of radiation than they do about any other agent. It would be natural, therefore, to use the standard established for radiation as a standard for other agents. FALL 1985 69

HAZARDS: SCIENCE AND ITS LIMITS past few years we have seen a remarkable shifft in viewpoint; whereas 15 years ago most cancer experts would have accepted a primarily environmental etiology foir cancer, today the view that natural carcinogens are far more importaM than are manmade ones has gained many converts. In his 1983 article in Science, biochemist Bruce N. Ames marshaled powerful evidence that many of our most common foods contain naturally occurring carcino- gens.9 Indeed, biochemist John R. Totter, former director of the Atomic Energy Commission's division of biology and medicine, has offered evidence for the oxygen radical theory of carcinogenesis: that we eventually get cancer because we metabolize oxygen and subsequently produce oxygen radicals that can play ihavoc with our DNA.10 As such views of the etiology of cancer acquire scientific support, I think that the trans-scientific question, as to how much cancer is caused by a tiny chemical or physical insult will be recognized as irrelevant. One does not swat gnats when pursued by elephants. Ambiguous carcinogens. To further complicate the cancer picture, there is evidence that some agents, such as dioxin, various dyes, and even moderate levels of radiation, seem to diminish the incidence of some cancers while simultaneously increasing the incidence of others. The lifespan of the animals exposed to these agents in laboratory tests on average exceeds that of animals not exposfxj.'1 A most striking example, given by biostatistician Joseph K. Haseman, is yellow dye number 14 given to leukemia-prone female rats. This dye completely suppresses leukemia, which is always fatal, but causes liver tumors, most of which are benign. I mention these two findings-or perhaps they should be considered points of view-to stress my underlying point: that when we are concerned with low-level insult to human beings, we can say very little about the cancer dose-response curve. Saying that so many cancers will be caused by so much low-level exposure to so many people, a practice that terrifies many people, goes far beyond what science actually can say. V Does the scientific community accept the notion that there are intrinsic limits to what it can say about rare events; that as events become rarer, the un- certainty in the probability of occurrence of a rare event is bound to grow? Perhaps a better way of framing this question is: Of what use can we put scientific tools of investigation of rare events, such as probabilistic risk assessment and large-scale animal experiments, if we concede that we can never get definitive answers? I believe that probabilistic risk assessment with an uncertainty factor as high as 10 is often useful, especially if one uses the technique for comparing risks. For example, the 1,500 reactor-years already experienced since the Three Mile Island accident suggest that a reactor core-melt probability is likely to be less than I in 1,000 reactor-years and may well be as low as less than I in 10,000 reactor-years. This is to be compared with dam failures whose probability, based on many hundreds of thousands of dam-years (where time has annihilated uncertainty), is around I in 10,000 dam-years. Even with an uncertainty factor of 10, we can judge how safe reactors are compared to dams. Some agents, such as dioxin, various dyes, and even moderate levels of radiation, seem to diminish the incidence of some cancers while simultaneously increasing the incidence of others. FALL 1985 65

One can argue that an accident whose occurrence requires an exceedingly unlikely sequence of untoward ~e»ents may also be regarded as an act of G-i9d. This approach has been used by chemist T. Westermark of the Royal Institute of Technology in Stockholm. He has suggested that for naturally occurring carcinogens such as arsenic, chromium, and beryllium, one may choose a de minimis level to be, say, 10 percent of the natural background.19 Clearly, a dee minimis level will always be somewhat arbitrary. Neverthe- less, it seems to me that unless such a level is established, we shall forever be involved in fruitless arguments, the only beneficiaries of which will be the toxic tort lawyers. Could the principle of de minimis be applied in litigation in much the same way it may be applied to regulation-that is, if the exposure is below de minimis, then the blame is intrinsically unprovable and cannot be litigated? I would imagine that the legal de minimis may be set higher than the regulatory de minimis; for example, the legal de minimis for radiation could be the background (after all, the BEIR-III committee concedes there is no way of knowing whether or not such levels are deleterious). The regulatory de minimis could justifiably be lower, simply on grounds of erring on the side of safety. One approach may bc to concede that there is some level of exposure that is beyond demonstrable effect. This defines a trans-scientific threshold. A de minimis level could then be established at some fraction, say one-tenth, of this beyond-demonstrable-effect level. For example, if we take 100 millirems per year of radiation as the beyond-demonstrable-effect level for general somatic effects (damaging somatic cells as opposed to germline cells), which is the value according to the BEIR-III committee, a de minimis level could be set at 10 millirems per year. Of course, such a procedure would evoke much controversy as to what is the beyond-demonstrable-effect level or whether 10 is an ample safety factor. This example demonstrates, however, that at least in the case of low-level radiation, a scientific committee has been able to agree on a beyond-demonstrable-effect level. As for the safety factor of 10, this cannot be adjudicated on scientific grounds. The most one can say is that tradition of- ten supports a safety factor of 10-forexample, the old standard for public ex- posure (500 millirems per year) was set at one-tenth of the tolerance level for workers (5,000 millirems per year). Can the principle of de minimis be applied to accidents? What I have in mind is the notion that accidents that are sufficiently rare may be regarded somehow in the same category as acts of God and be compensated accord- ingly. We already recognize that natural disasters should be compensated by the society as a whole. One can argue that an accident whose occurrence requires an exceedingly unlikely sequence of untoward events may also be regarded as an act of God. Thus, the Price-Anderson Act could be modified so that, quite explicitly, accidents whose consequences exceeded a certain level, and whose probability as estimated by probabilistic risk assessment would be less than, say, I in I billion per year, would be treated as acts of God. Compensation in excess of the amount stipulated in the revised act would be the responsibility of Congress. The cutoff for either compensation or for probabilities would be negotiable, and perhaps it would be revised every 10 years or so. One not entirely fanciful suggestion may be to set any probability of the order of I in 10 million to I in 100 million per year to be a de minimis cutoff, this being the frequency at which the earth may have been visited by 70 1SSUES IN SCIENCE AND TECHNOLOGY

into numbers of latent cancers. Thus, health consequence would be reported in two categories: (1) for highly exposed individuals, the number of health effects; and (2) for slightly exposed individuals, the total man-rems or even the distribution of exposures accrued by the large number of individuals so exposed. Perhaps such a scheme could be adopted in reporting the results of future probabilistic risk assessments; at least it has the virtue of being more faithful than the present convention to the state of scientific knowledge IV In both of my examples of accidents (Bhopal and nuclear accidents), many people are exposed to low-level insult. The uncertainties inherent in estimating the effects of such low-level exposure are heaped on top of the uncertainties in estimating the probability of the accident that may lead to exposure in the first place. Science has exerted great effort to ascertain the shape of the dose- response curve at low dose-but very little, if anything, can be said with certainty about the low-dose response. Thus, to quote the report of the National Research Council, The Effects on Populations of Exposure to Low Levels of Ionizing Radiation: 1980 (also known as BEIR-III, for the commit- tee that prepared it, the Committee on the Biological Effects of Ionizing Radiation). "The Committee does not know whether dose rates of gamma or x-rays of about 100 mrads/yr are detrimental to man.... It is unlikely that carcinogenic and teratogenic effects of doses of low-LET radiation adminis- tered at this dose rate will be demonstrable in the foreseeable future."6 This prompted Philip Handler, then president of the National Academy of Sci- ences, to comment in his letter of transmittal to the Environmental Protection Agency, which had requested the study, "It is not unusual for scientists to disagree ...(and) ... the sparser and less reliable the data base, the more opportunity for disagreement.... The report has been delayed ... to permit time ... to display all of the valid opinions rather than distribute a report that might create the false impression of a clear consensus where none exists."' This forthright admission that science can say little about low-level insults I find admirable. It represents an improvement over the unjustified assertion in the BEIR-II report of 1972 that 170 millirems per year over 30 years, if imposed on the entire U.S. population, would cause between 3,000 and 15,000 cancer deaths per year.8 I do not quarrel with the estimated upper limit-which amounts to I cancer per 2.500 man-rems, but I regard placing the lower limit at 3,000 rather than at zero as unjustified. Moreover, I think it has caused great harm. The proper statement should have been that at 170 millirems per year, we estimate the upper limit for the number of cancers to be 15.000 per year; the lower limit may be zero. Since the appearance of the BEIR reports, two other developments have added to the burden of those who must judge the carcinogenic hazard of low- level insults: an awareness and study of (1) natural carcinogens, and (2) ambiguous carcinogens. Natural carcinogens. Is cancer environmental in the sense of being caused by technology's effluents. or is it a natural consequence of aging? In the 64 . ISSC'.ES IN SC'IEtiCE AND TECHNOLOGY

COMMENTARY Interview wath a Risk Expert* Daniel E. K:~shland, Jr. Science: "Dr. Noitall, you are the ultimate authority on all types of risks, a revered figure who has just appeared on national television." Noitall: "A vast understatement of my true value." Science: "You must have a large laboratory to uncover so many facLs not available to the regulatory agencies." Noitall: "Facts are no longer created in laboratories; they are created in c:he media. Any pronouncement of mine repeated in three periodicals, four newspapers, or one television pro- gram is considered a fact My appearance on three talk shows is enough to qualify me as an expert It is no longer necessary to have a laboratory in my profession." Science: "Could you give an example of how to avoid risks?" Noitall: "Stay out of the home. More than 3 million people in the United States were injured in 1987 in home accidents; 90 percent of all automobile accidents occur within 10 miles of home. It is imperative that you stay away from home." Science: "But I've heard that many accidents occur on high- ways." Noitall: "That is true. There is one fatality for every 10 minutes of driving on the highways in the United States. I have developed a rigorous formula that shows that the more time spent on the highway, the greater the chance of an accident Therefore, I recommend driving 80 miles per hour as a way of reducing the time spent on highways and thus reducing your chance of accident Science: "If one stays away from home, is there not an in- creased chance of infectious diseases?" Noitall: "One has to give up sexual intercourse entirely. The danger of disease from that source is far greater than from eating an apple, and it should be avoided at all costs." Science: "Are there other dangers about which the Environ- mental Protetfion Agency has failed to advise us?" Noitall: "Breathing. All breathing generates oxygen radicals, which are the main sources of mutations in DNA, leading to cancer, birtl'i defects, and very peculiarly shaped mole- cules in the urine. Breathing has been observed 3 minutes before death irt 100 percent of all fatalities. We urge every- one to stop breathing until the proper research has been carried out The EPA has been told about this relation and has failed to act on it, a scandalous display of irresponsi- bility." Science: "What about hazards from crime?" Noitall: "A third of all homicides are committed on inti- mates, about a third on acquaintances, and about a third on strangers. Hence, it is imperative to avoid intimates, acquaintances, and strangers in order to reduce your risk of homicide significantly." .ScieTIC2: "Can one ever completely eliminate a given risk?" *Reprinted with permission from Science, Vol. 244, June 30, 1989, p. 1529. Copyright AAAS, 1989. Noitall: "One can reduce a risk to essentially zero by adopt- ing what I call 'the riskier altemative strategy.' For example, one could take up hang gliding, as it has been conclusively demonstrated that fewer hang gliders die of passive cigarette smoke than those who never participate in the sport. Peo- ple who bicycle without a helmet need not worry about a little nuclear reactor nearby. People who have a cocktail before dinner or wine with a meal need never worry about a little trichloroethylene in their drinking water. By the proper choice of alternative strategies, it is possible to reduce one's chance of dying of any particular disorder to any de- sired level. It has relieved many people of risk anxiety syn- drome." Science: "This seems so sensible; I am surprised people don't follow your advice." Noitall: "Most ignoramuses are in fact following my formula without knowing it Millions of people commute 20 miles to work, take airplanes, and choose hopelessly short-lived grandparents and still worry about clean drinking water. These people are secret admirers of peptic ulcers." Science: "We can't thank you enough for the time you are spending with us, but I have one last question. Do you practice what you preach?" Noitall: "Sadly, the answer is no. My family on the pater- nal side has a hereditary weakness whose clinical manifesta- tion is the 'eat, drink, and be merry' psychosis. As a result, all my ancestors on that side of the family have died pre- maturely, in their early nineties. I doubt whether I will escape the family curse." HPS Nevvsletter, September 1989 . 9

"ANY POLITICIAN WOULD PREFER A DEAD BODY TO A FRIGHTENED VOTER" -- JOHN DUNSTER, U. K. HEALTH AND SAFETY INSPECTOR

12 RISK/BENEFIT ANALYSIS THE MEANING OF RISK 13 Figure 2-1. Accidental Deaths per Million Tons of Coal Mined - . .._- ~ in the United States. _ FiW_lre 2-2. .,wwrai n_:'_._cn.. Deaths per Thoucanri Cea! ~^ --- ~... .. ~a•...., u~~ Niuyce5 in the United States. 2.50 N ~ U O ~ 2.25 c ~ 1 0-I v T O F-O . d W 2.00 a L n O v ° 1.75 ~ L h 0.5 ~ m v C ~ L n R 1.50 V Q c V a J Q 1.25 0 1950 1955 1960 1965 1970 0 00 Year . 1950 What measures of risk are appropriate for a particular risk assess- ment depend on the specific details of the question the assessment is designed to illuminate. Presumably they will be the measures corre- sponding asinearly as possible to the way in which the risks are per- ceived. In what follows we will usually be limiting consideration to risks of death (measured by probabilities of dying or expected excess numbers of deaths) resulting from various actions, although other risks will occasionally be mentioned. Although we shall not concern ourselves much with it, the dis- tinction between risks and measures of risk is not totally academic. A simple example is the American coal industry, taken as a whole, between 1950 and 1970. Figure 2-1 is a plot of one measure of risk in this industry-the number of accidental deaths per million tons of coal mined. Clearly this measure steadily declined during this period, so that, if we follow the industry through successive years, it appears to be getting safer. Looking at Figure 2-2, which shows the behavior of another measure of risk-the number of accidental deaths per 1955 1960 1965 1970 Year thousand persons employed-one might naively assert that the indus- try is getting more dangerous, not safer. Evidently the two measures illustrated might be used to support opposing views on the safety of coal mining. Neither measure taken alone is right or wrong, nor are they even contradictory even though they may be so perceived. Any risk assessment supposed to be com- plete would have to draw attention to the two aspects of the risk of coal mining gauged by the two different measures and would have to take both into account, depending exactly on the purpose of the risk assessment. From a national point of view, given that a certain amount of coal has to be obtained, deaths per million tons of coal is the more appropriate measure of risk, whereas from a labor leader's point of view, deaths per thousand persons employed may be more relevant. What steps to take to reduce the risk will depend on which of the two measures is used. Doubling the number of miners, each working

4. U.S. Nuclear Regulatory Commission, Risk .4ssessmenr RevieK• Group Report to the L'.S. Nuclear Regulatort• Commrss on (NUREG/CR-0400) (Washington, D.C.. September 1978), vi. 5. "Repon to the American Physical Society by the Study Group on Light Water Reactor Safety," Reciexs oJAfodern Physics 47 (Supplement 1) (Summer 1975). 6. National Research Council, The Effects on Populations of Exposure to Low Levels of loni=ing Radiation: 1980 (BEIR-IlI), (Washington, D.C.: National Academy Press, 1980), 2. 7. National Research Council, The Effects on Populations ojExposure to Low Levels of lonizing Radiation: 1980 (BEIR-1I1), iii. 8. National Research Council, The Effects on Populations of Exposure to Low Levels of loni:ing Radiation (BEIR-II), (Washington, D.C.: National Academy Press, 1972), 2. 9. Bruce N. Ames, "Dietary Carcinogens and Anticarcinogens," Science 221 (Sept. 23, 1983): 1249, 1256-64. 10. John R. Totter, "Spontaneous Cancer and ]ts Possible Relationship to Oxygen Metabolism," Proceedings ojthe National Academy oJSciences 77 (April 1980): 1763-67. 11. Alvin M. Weinberg and John B. Storer, "On 'Ambiguous' Carcinogens and Their Regulation," Risk Analysis 5 (June 1985): 151-55. 12. National Research Council, The Effects on Populations of Exposure to Low Levels oj lonizing Radiation.• 1980 (BE1R-tIl), 287-321. 13. Alice Whittemore, "Facts and Values in Risk Analysis for Environmental Toxicants," Risk Analysis 3 (March 1983): 23-33. 14. John Ben-David, "Emergence of National Traditions in the Sociology of Science: The United States and Great Britain," Sociological Inquiry 48, nos. 3 and 4(1978): 209. 15. Trevor J. Pinch and Wiebe E. Bijker, "The Social Construction of Facts and Artefacts: Or How the Sociology of Science and the Sociology of Technology Might Benefit Each Other," Social Studies of Science 14 (1984): 401. 16. KAre Hannerz, Towards Intrinsicalfr Sa,fe Light li•ater Reactors (ORAU/IEA-83-2(M) Rev.) (Oak Ridge, Tenn.: Oak Ridge Associated Universities. Institute for Energy Analysis, June 1983): Herben Reutler and Gtinter H. Lohnert, "The Modular High Temperature Reactor," Nuclear Technologr 62 (July 1983): 22-30. 17. Trevor A. kletz. Cheaper, Safer Plants or I{ •ealth and Sqfett• at Gi'ork: Notes on Inherentlr Safer and Simpler Plants (Rugby. England: The Institution of Chemical Engineers, 1984). 18. Howard 1. Adler and Alvin M. Weinberg. "An Approach to Setting Radiation Standards," Health Ph, vstcs 34 (June 1978): 719-20. 19. T. Westermark, Persistent Genoroxic I;'¢ues.• .9n .9ttempt at a Risk Assessment (Stockholm: Royal Institute of Technology, 1980). 20. William C. Clark, it'itches. Floods, and Lf onder Drugs: Historical Perspectives on Risk Management (RR-81-3) (Laxenburg, Austria: International Institute for Applied Systems Analysis, March 1981). 72 ISSUES IN SCIENCE AND TECHNOLOGY

~ Table 7-•2'. Some Commonplace Risks of Death in the United States, Based on Estimated U.S. Resident v Population (Source 1). a Risk Annual per Capita Risk a Annual Trendb Uariability, Percentc Based On z Source x Motor vehicle accident Total .4 x 10-4 .. 0 950-78 m z m 1 ~ -~ Collision wirh pedestrian 4.2 x 10-s -3.9 x 10' 10 1950-78 2 a Home accidentsd Falls 1.1 X 10-4 6.2 x 10-s -2.9 x 10 6 -3.0 x 10-6 5 6 1950-78 1963-77 2 z a 2 r -~ -s ~ Drowning 3.6 x 10 ... 7 1963-77 2 N Fires 2.8 x 10-5 -1.0 x 10-6 5 1963-77 2 Inhalation and ingestion of objects 1.5 x 10-s ... 10 1968-77 2 Firearms 1.0 x 10-s -2.4 x 10-' 8 1968-77 2 Accidental poisoning Gases and vapors 7.7 x 10-6 ... 5 1963-77 2 Solids and liquids (Not drugs or medicaments) 6.0 x 10-6 ... 10 1971-77 2 Electrocution 5.3 x 10-6 ... 5 1971-77 2 Tornadoes 6 x 10-' ... 100 1950-77 1 Floods 6 x 10' ... 100 1950-77 1 Lightning 5 x 70-' ... 18 1971-77 2 Tropical cyclones and h urricanes 3 x 10 ' ... 160 1952-77 '. Bites and stings by venomous animals and insects 2.4 x 10-' .. 13 1971-77 2 Air Pollution 2.4 x 1074 ... -see text- a. Average over indicated years, if no trend is shown. The value of trend line in last year of indicated years is used if a trend is shown. b. Average annual change of annual per capita risk during years shown. Least squares straight line fit of annual risk versus time. A trend is shown if the estimated trend was significant at the 5 percent level (two-tailed). c. Estimated standard deviation of annual per capita risk about the trend line (trend) or of the mean value (no trend). d. Home accidents inciude,s some proportion of some of the following seven risks. Sources: 1. U.S. Bureau of the Census (1975, Annual). 2. National Safety Council (Annual). ~ ~ ~

RISK = PROBABILITY OF OCCURRENCE x SEVERITY OF EVENT RISK UNITS: DEATHS/YEAR DEATHS/TON OF COAL (BTU, khw) DEATHS/WORKING MAN LOSS OF LIFE EXPECTANCY WORKING DAYS LOST (WDL) PUBLIC DAYS LOST (PDL) RISK <----> UNCERTAINTY If I am certain I will die of cancer, I do not describe it by the word RISK. Risk changes as information improves either as events develop by study Use of the word risk implies that there is uncertainty. When there is uncertainty there remains a risk.

(A LEAST) TWO TYPES OF UNCERTAINTY 1. DOES THE PROCESS OCCUR? (Do cars kill people?) (Does benzene cause cancer) 2. STOCBASTIC Will this particular car kill me? or pass me by? Will this particular carcinogen give me cancer? (Onemthird of cigarette smokers die of their habit we do not know, and probably never will know, which) DO SOME CANCERS FOLLOW THIS LINE? 4.

Death Rates at Five Year Intcrvafs from 1900 to 1975 for Various Age Groups: United States. 4 a> a ?t)ct 2OU .b 30-a 1,0 Reduction principally 0•J due to reduction in 0.7 communicable disease CI.i a.5 0.4-1 1900 " 1920 65-74 55 -64 45-54 ..,-r-Car accidents 25 -34 15 -24 ~ 1 -4 5-14 19,10 1960 1980 Date Age Grocip UVf R 85 . Hard to reduce further

~~- RESEARCH i RISK ASSESSMENT i RISK MANAC E n.~e ~ ~ 1nV~ ~ Laboratory and field observations of adverse health effects and ex- posures to particular agents Hazard Identification (Does the agent cause the adverse effect?) I Development of regulatory options Information on extrapolation methods for high to low dose and animal to human Field measurements, estimated exposures, characterization of populations Dose-Response Assessment (What is the relationship between dose and inci- dence in humans?) Exposure Assessment (What exposures are currently experienced or anticipated under different conditions?) I Evaluation of public health, economic, social, political consequences of regulatory options Risk Characterization (What is the estimated incidence of the ad- verse effect in a 7) I v n L gi e popu aton. ~ I i Agency decisions and actions Elements of risk assessment and risk management. s~.'4StTSSZOz

Table 7-3. Some Occupational Risks of Death. ~ ~ ~ Occupation or In,dastry Annual Riska Annua! Trendb Variability, Percent` Based On z Source ~ ~ Manufacturing 8.2 x 10-5 -1.6 x 10-6 8 1955-78 m m 1 z -5 -6 ~ Trade 5.3 x 10 -2.3 x 10 15 1955-78 1 ~ "4 -6 -~ Service and government 1.0 x 10 -2.0 x 10 8 1955-78 1 a Transport and public utilities 3.7 x 10'4 ... 16 1955-78 1 z Agricultured 0 x 10-4 6 9 1955-78 a 1 r . ... < Construction 6.1 x 10-4 -7.0 x 10 6 6 1955-78 1 ~ ~ Mining and Quarrying 9.5 x 10-4 ... 22 1955-78 1 Farminge 3.6 x 10-4 -5.0 x 10-6 7 1964-77 1, 2 Tractor fatalities per tractor 8.8 x 10-$ -1.0 x 10-5 22 1969-77 1 Metal mining and milling 9.4 x 10-4 ... 15 1959-71 3 Nonmetal mining ,and milling 7.1 x 10-4 +2.3 x 10-5 15 1959-71 3 Stone quarries and rnills 5.9 x 10-4 ... 20 1959-71 3 Coal mining (accidents) 6.3 x 10-4 -1.0 x 10-4 46f 1963-77 4 Police officers killed in line of duty Total 2.2 x 10-4 ... 19 1975-78 4 By felons 1.3 x 10 ° -2.1 x 10-5 8 1975-78 4 Railroad employees 2.4 x 10-4 -6.0 x 10 6 7 1963-77 1,4 Steel worker (accident only) 2.8 x 10 4 ... ? 1969-72 5 Fire fighter 8.0 x 10 4 ? 1971-72 5 a. Per person at risk. Average over indicated years, if no trend is shown. The value of trend line in last year of indicated years is used if a trend is shown. b. Average annual change of annual risk during indicated years. Least squares straight line fit of annual risk versus time. A trend is shown if the estimated trend was siE;nificant at the 5 percent level. Note that the error estimates for these trends are generally large. c. Estimated standard deviation of annual risk about the trend line (trend) or of the mean value (no trend). Expressed as a percentage of the risk shown in the first column. d. Not strictly compara.ble with farming category, includes transport accidents and all agriculture. e. Not strictly comparable with agriculture category, refers to nontransport deaths occurring on farms, the population at risk being assumed to be all employed workers, unpaid family members working more than fifteen hours per week and operators working more than one hour per week. f. The large variability is due to the bad choice of model (straight line fit) and the large changes occurring in the years indicated. Sources: i. National :iat'ety Council (Annual). 2. U.S. Department of Agriculture (1979). 3. U.S. Bureau of Mines (Annual). 4. U.S. Bureau of the Census (Annual). 5. Baldewicz et al. (1974).,

PROPORTION OF CANCER 'DEATHS ATTRIBUTABLE TO DIFFERENT FACTORS Per cent of all cancer deaths Factor or class of fac:tors Best estimate Range of acceptable estimates Cigarett:e smoking 25 20 to 30 Alcohol 3 2 to 4 Diet 35 10 to 70 Food additives >1 -5* to 2 Reproductive and sexual behavior 7 1 to 13 Occupation 5 s2 to 10 Pollution 1 >1 to 3 Industrial products >1 >1 to 2 Medicines and medical procedures 2 1 to 3 Geophysical factors+ 3 2 to 3 Infection. 3 1 to ? Unknown ? ? Source: R. Doll and R. Peto, J. Natl. Cancer Inst. 66, 1191 (1984).

T0,000c 1900 1920 1940 1960 1980 r I I 1 I II 1 ( I Tl.gure 1. Death rates by age in the United States, for' se-lected years, .1900-1977 (from the U.S.. Department ® of: Health, Education and Welf are statistics). N . C ~ cZk . w . r~

REASONS FOR CHOOSING RATS AND MICE FOR EXPERIMENTS They are easy to breed. They are maramais and have some similarity in metabolic processes to man. They live only two years so that an experiment can be completed within a human lifetime.

CERTAINTY OF INFORMATION WITH MAGNITIIDE OF RISK E.g. Vinyl Chloride o Human carcinogen o Mutagen (with activation) o Animal carcinogen May be initiator, i.e. linear dose-response BUT only 130 angiosarcomas from 30 years exposure worldwide; 70 more from past exposures 200 other cancers from past exposures 130 from 30 years exposure worldwide 400 TOTAL over 30 years worldwide MOREOVER, EXPOSIIRES NOW DOWN 1000-FOLD SO P'E EXPECT < 1/60 YR IN THE FUTURE RISK OF A NEW CHEMICAL AF2DENT ENVIRONMENTALIST --> 1 (because we don't Jmow that it is safe) MYOPIC INDUSTRIALIST --> 0 (because we don't }aow.that it is dangerous) *************,r*************************te*********** * * * The job of a RISK ASSESSOR is to find a number * * * * in between with its uncertainty * * * * 1 > R > 0 * * * ************************************,r************** perhaps

' AQP 1tr11rt11rP nf Ct-itii. D......L.r:.-.... 1A-7A A_ __ _fr- _ r. ...o_ -.. ~~... .,. J14liV 1 vNula~lvll- I 7~~r r1~e- r j peci~ic Death tcates. ~-- Approximation c ~ ~ ~ ~ N 0.6-i C 0.5-{ O U rro ~ U- 0.4-I 6 0.3~ 0.2-1 0.1 -1 10 20 vv4'svsszoz 40 50 60 Age 70 80

THE MEANING ®E RISK l I RISK(SENEFIT ANALYSIS 4 Figure 1-1. Death Rates at Five Year Intervals from 1900 to 1975 for Various Age Groups: United States. Date s T4s1Vsszoz Age Group ovrk 85 75 -84 65 -7 4 UNDER I 55 -64 45-54 35-44 25-34 15-24 1-4 5-14 1980 mum from some point of view. One attempt at reducing such possi- bilities is the objective analysis of risk, which is pursued throughout this book. To make any start cr; objcciive assessment it is neces;a~ °-- y to realize what is being measured. Death is one clear objective measure, The total annual risk of death at any age is just the probability of dying within one year. In the absence of any extra causes, population aver- ages for this measure are obtained from national mortality tables (see Chapter 7). But in risk assessments we are interested in additional risks of death or components of the total risk of death due to some specific actions undertaken either voluntarily or involuntarily. More often, we are interested in how much of an action to undertake, so that we wish to evaluate measures such as extra probability of death per unit of action (per cigarette smoked, or pet ton of coal mined, for example). Death is not the only measure of risk of interest, for, although it is probably the most objective one and for this reason often used, it may not capture large components of what are perceived as risks. In balanced decisions it may become vital to consider other measures. A few possible such measures are: by age Deaths by cause Injuries Illness Man-days lost by cause by type by severity index by cause by type by severity index I by cause Days of impaired health Days of pain Loss of life expectancy I Total numbers (whole task) or probabilities (for individuals) per unit operation size per event per unit dose (per cigarette, per ton produced, per unit output, etc.)

Risk Management Commentary for Dr. D. Allan Bromley Assistant to the President for Science and Technology I. I'residential Obj ectives: 1. The very best science should determine the allocation of resources. Public fears drive much of regulatory and Congressional actions. These public concerns are often inconsistent with science-based selection of those factors most effective in improving public health. 2. Coherent policies, methodologies, and procedures for establishing regulatory objectives of the federal agencies on a scientific basis. These should effectively incorporate current science, peer review, and public participation. 3. Maximize the effectiveness of our huge national expenditures arising from regulations for improving health and safety by a balanced allocation of priorities and resources. 4. Minimize congressional prescriptions on details of risk management to give the agencies more flexibility to meet regulatory objectives with minimal impact on economic activities. II. Recommendations: 1. Appoint a deputy for a full-time focus on Risk Assessment and Management. Select someone who understands the complexities of the issues. 2. Establish a Council (or PCAST Subcommittee) on Risk Assessment and Management for review and guidance of agency criteria, methodologies and procedures. This would provide scientific oversight on proposed agency activities and regulations. It would also coordinate the large number of existing government committees and programs already engaged on specific agency tasks.. 3. Coordinate with OMB on the process for economic evaluations and agency implementation of OMB procedural recommendations.

RISK/BENEFIT ANALYSIS EDMUND A.C. CROUCH RICHARD WILSON 3 The importance of perceptions of risks is illustrated by Table 1-1, which summarizes results of a public opinion survey. Most people seem to believe that life is becoming more dangerous, even rhoug!; most objective ;reu,suri,S show the contrary to be true. The expecta- tio;; of life, for example, an inverse measure of the probability of dying, has steadily increased, from perhaps twenty-eight years fifteen centuries ago, to fifty years one century ago, to about seventy-two years currently, although the rate of increase has been decreasing. The increase has been brought about by the elimination of many large risks to life, among them many infectious and contagious dis- eases, poor working conditions, and inadequate nutrition. Figure 1-1 shows the reduction in death rates in this century by age group. De- tailed examination shows that the increase since 1960 in the 15 to 24 age group is due to automobile accidents. Doll (1979) has also shown how health, as measured by most medical indicators, is improving. It is now necessary to concentrate on the many smaller risks, often poorly understood, in order to further reduce total risks. Perhaps it is Table 1-1. Public Opinion Survey Comparing Risk Today to Risk of Twenty Years Ago. BALLINGER PUBLISHING COMPANY Cambridge, Massachusetts A Subsidiary of Harper & Row, Publishers, Inc. PERSPECTIVE ON RISK Q: Thinking about the actual amount of risk facing our society, would you say that people are subject to more risk today than they were twenty years ago, less risk today, or about the same amount of risk today as twenty years ago? Q: I'd like to start by asking you a few questions about the amount of risk we face in our day-to-day living. Thinking about the actual amount of risk facing our society, would you say people are subject to more risk today than th( ~ were twenty years ago, less risk today, or about the same amount of risk today as twenty years ago? Top Corporate Investors, Federal Executives (N = 401) Lenders (N = 104) Congress (N = 47) Regulators (N = 47) Publlc (N = 7,488) More risk 38 60 (percent) 55 43 78 Less risk 36 13 26 13 6 Same amount 24 26 19 40 14 Not sure 1 1 ... 4 2 Source: Marsh & McLennan Companies (1980).

W t (,S G-i,~ FIGURE 2 $WRVIY FoR FiAASH -MLLElVdtlON Question: Thinking about the octuol amrrodn/ oJrisk facing our society, would you say that people are subject to more risk today than they were 20 years ato. less risk today. or about the same amount of risk today as 20 years ago? Percent of responses Top corporate tnvestors/ Federal executives Lenders Congress regulators Public Lcvel of risk (401) (104) (47) (47) (1,488) P',)re risk 38 60 55 43 78 Less r isk 36 13 26 13 6 Same amount 24 26 19 40 14 Not sure I I - 4 2 Question: Would you say that people are subject to more risk today than they were 20 years ago, less risk today, or about the same amount of risk today as 20 years a=o? Percent of responses Demographic Number of More Less Same Not characteristics respondents risk risk amount sure Sex Mate 702 74 10 15 2 Female 766. 82 2 14 2 Age 55 and over 179 86 2 10 2 Education Not high school graduate 269, 85 3 9 3 Marital status Separated, divorced, widowed 226 82 2 13 3 Household income S7,5f)U or Iess 211 83 4 10 3 Race White 1,258 77 6 15 2 i3lack 155 Y8 3 6 2 Spanish-American 37 94 2 7 4 Total 1.488 79 6 14 2 'Sourcc: Marsh & McLennon Co.. Inc. Risk in a cos+tpkx society: A public opinion survey. "Numbers in parentheses indicate number of rexpoodewts in each category. ~

.CARCINOGEN RISK ASSESSMENT ~~ FIRST RULE: WHENEVER THERE IS DATA ON HUMANS -- USE THEM * ~~~r****~****~**~****************~r***************************** CIGARETTES STUPIDITY NAPTHALYAMINE OCCUPATIONAL BENZIDINE OCCUPATIONAL ARSENIC -- Inhalation OCCUPATIONAL -- Ingestion OCCUPATIONAL BENZENE OCCUPATIONAL AFLATOXIN Bl FOOD RADIATION WAR (Monson will discuss how epidemiology works) ---------------------------------------------------------------------------- THERE IS NEGATIVE EPIDEMIOLOGY PEOPLE HAVE BEEN EXPOSED BUT THERE IS NO SIGN OF AN EFFECT E.G. There is no firm evidence that unleaded gasoline in a refinery causes cancer. ThereforEa normal users of gasoline, with lower exposure, have less risk. ------------------------------------------------------------------------------ The: two assumptions made (in using human epidemiology) are: 1. EXPOSURE in the situation being assessed is ~ O similar to that of the epidemiological study j~ ~ 2. The DOSE-RESPONSE RELATION must be assumed, ~ ~ ~ since we need risk at low doses ~ ~. ~

Ytr4strsszoz

CERTAINTY OF INFORMATION WITH MAGNITUDE OF RISK E.g. Vinyl Chloride o Human carcinogen o Mutagen (with activation) o Animal carcinogen May be initiator, i.e. linear dose-response BUT only 130 angiosarcomas from 30 years exposure worldwide; perhaps 70 more from past exposures 200 other cancers from past exposures 130 from 30 years exposure worldwide 400 TOTAL over 30 years worldwide MOREOVER, EXPOSURES NOW DOWN 1000-FOLD SO PIE EXPECT < 1/60 YR IN THE FUTURE

PETO ARGUMENT IF ALL TISSUES ARE EQUAL WHETHER ATTACHED TO MOUSE OR MAN, WE EXPECT BILLIONS MORE CANCERS IN MAN THAN MOUSE l. CANCER INCREASES WITH WEIGHT (M) 2. CANCER IS KNOWN TO VARY AS A HIGH POWER OF AGE (T4) .. At the end of life the ratio of cancer in man to cancer in mouse for equal daily intake (roughly proportional to mass x time) 4 4 f Mman Tman Tman 70 kg f 70 x = X Mmouse Tmouse Tmouse 30 kg 2 SEVERAL BILLION! BUT WE KNOW IT IS CLOSER TO 1 BECAUSE (i) of background (ii) of experiment SO THE METABOLISM MUST ACCOUNT FOR A FACTOR OF A BILLION (Calabrese will discuss)

HAZARCS: SCIENCE AND ITS LIMITS the comel:arv asteroids that may have caused the extinction of species in past geologic eras. As in most such questions, identifying and characterizing the problem is easier tha n solving it. That the dilemma of the regulator and the toxic tort judge is rooted in science's inability to predict rare events cannot be denied. Getting lthe regulator and the toxic tort judge off the horns of the dilemma is far from easy, and my two suggestions-the technological fix and de minimis--are offered tentatively and with diffidence. Equaily obvious is the intrinsic social dimension of the issue. In an open, litigious democracy such as ours, any regulation and any judicial decision can be appealed, and if the courts offer no redress, Congress, in principle, can do so. These legal mechanisms are ponderous, however. The result seems to me to be a gradual slowing of our technological-social engine as we become more and mom enmeshed in fruitless argument over unresolvable questions. Wes',~ern society was debilitated once before by such fruitless tilting with Don Quixotian windmills. I refer of course to the devastating campaign against witches from the fourteenth century to the early seventeenth century. As ecolog~s,t William Clark has put it so vividly, society took it for granted dur- ing that period that death, disease, and crop failure could be caused by witches.x' To avoid such catastrophes, one had to burn the witches responsible for them--and consequently some million innocent people were burned. Finally, in 1610, the Spanish inquisitor Alonzo Salazar y Frias realized there was no demonstrated connection between catastrophe and witches. Although he did not prohibit the burning of witches, he did prohibit use of torture to ex- tract confessions. The burning of witches, and witch hunting generally, declined precipitously. I have recounted this story many times by now. Yet it still seems to me to capture the essence of our dilemma: the connection between low-level insult and bodily harm is probably as difficult to prove as the connection between witches and failed crops. I regard it as an aberration that our society has allowed this issue to emerge as a serious social concern, which in the modern context is hardly less fatuous than were the witch hunts of the past. That dark phase in western society died out only aft' er several centuries. I hope our open, democratic society can regain its sense of proportion far sooner and can get on with managing the many real problems we always will face rather than waste its energies on essentially insoluble, and by comparison, intrinsically un- important, problems. ® NOTES: I. This article was adapted from a paper delivered at a June 3-4, 1985. National Academy of Engineering symposium on "Hazards: Technology and Fairness." A report on that symposium will be published in book form by the National Academy Press. 2. William 1). Ruckelshaus. "Risk, Science, and Democracy," Issues in Science and Technology I (Spring 1985): 19-38. 3. U.S. Nuclear Regulatory Commission, Reactor SafetyStudt•: An Assessment ofAccident Risk in U.S. Commercrat Nuclear Plants (WASH-1400. NUREG 75/014) (Washington, D.C., 1975). FALL 1985 71

4. Ini.t.iate a public education program on the realities of risks and choices (both governmental and individual), comparative risks, benefit-risk balance, and the acceptability of low-level exposures in daily life. The Society for Risk Analysis may be a vehicle for such a program. Modest financial support would be needed for workshops, bulletins, teaching material, etc. III. Background Issues: The cumulative national direct cost burden of proposed reductions of environmental risks is likely to exceed $100 billions per year (e.g. clean air, toxic, CFCs, (:02, electricity fields, house radon, hazardous waste, etc.). Indirect costs will increase this burden. These costs will become publicly apparent through their large inflationary effect on costs. They cannot be ignored. Unfortunately, there is no upper bound to such expenditures as long as "zero" risk is the political target, except a national resource limitation. Recognizing that poverty is the greatest social pollutant, the effect on the productivity of the economy must be considered. Congre~,s has implied in several acts that the cost of meeting health and safety objectives should not be considered in setting regulatory objectives. However, this does not preclude the establishment of regulatory targets consistent with the optimal distribution of our resources to improve health and safety. This is a central issue, as it is the operational intersection of risk assessment with risk management. Even a crude disclosure of the relative importance of risks might diminish the prescriptive tendencies of Congress, and shift the decision initiatives to the agencies on the allocation of funds and attention. Public confidence in the wisdom, objectivity, credibility, and feasibility of goverru-n.ent regulatory actions is an essential objective. This requires that the suggested OSTP Council be broadly based with expertise from all stakeholders (academia, government, industry, public groups, etc.) and is constituted to consider risk assessment, management, and communication. Obviously, such a Council would need staff support for organizing briefings, discussions, and fundings. CS/dt/Papers/Bromley/Version 2/Revised 2

FDA Approach to Risk Assessments Scheuplein

are usually justified on the basis of individual risk reductions, without re,Fe,rence to the comparative cost-effectiveness of other risk reduction measures. 3. The responsibility for the implementation of risk management is shared in practice between government, industry, and the public. This is a very com.plicated interaction and can only be successful if all sectors agree on the obj1ectives. In this respect, the Nuclear Regulatory Commission has pioneered in a sharing of responsibilities with the industry. It should be studied as a developing risk management system. 4. The individual exposure to risks arises from food, water, air, industry, life- style systems, work, etc. Many federal agencies are involved (FDA, EPA, NR.C:, HHS, OSHA, DOE, DOD, DOT,etc). The need for an Executive Office guidance and overview has been growing during this past decade. 5. The financial consequences of risk management regulations may be one of our largest national cost factors, indirectly comparable to health services and defense. It requires a scientific basis for rational decision making. It deserves OST]P attention at the highest level. 6. The'Senate version of the Clean Air Act, S.1630, contains a long section on the establishment of a 10 member Risk Assessment and Management Commission. As presently proposed, this Commission reports to the President and Congress. It is intended to have a 4 year life. If this legislation is passed, its relationship to PCAST and the OSTP should be considered. This proposed Commission is focussed on air pollution., The OSTP interest shou]',d encompass the wider spectrum of risks covered by other federal agencies. Presented by: August, 1990 R. C. R. Hart, Starr, Wilson and CS/dt/PapersBromley/Version 2/Revised 4

Table 7-4. Annual Fatality Risks in Sports a ~ 0 port Average Annual Riskb A verage Annual Deaths Estimated Population at Risk r Years of Coverage z Source ~ X d h ~ Aerial acrobatics (professional) , <2X10-3 ~ 0.22 360 1970-78 1 z Air show/air racing and acrobatics 5 x 10-3 4.9 1,050c 1971-77 1 m Flying amateur/home built aircraft 3x 10-3 25 8,000` 1970-77 1 Bicycle racing (registered) < 9 x 10-s d,e 0.33 9,800 1970-78 1 a z Boating 5 x 10 s 1,300 27 x 106 1972-78i 2, 3 a r Bobsledding < 7 X lC-4d,f 0 450 1970-78 -G 1 ~ Footbai l Sandlot 2 x 10-6 1.7 106 1970-78 V) 1 Professional and Semiprofessional <4x10-4 d, 9 ~ 0.11 1,500 1970- 78 1 High school 1 x 10-51 13 106 1970-78 1 College 3 x 10"sl 1.2 40,000 1970-78 1 Glider flying 4 x 10-4 7 18,000` 1970-77 1 Hang gliding ~8x 10-4 31 20,000-60,000 1974-78 1 Hunting 3 x 10 5 600-800 22 x 106 1972 2,3 Ice yachting < 1 x 10-4d,h ~ 0.22 4,500-6,500 1970-78 1 Lighter-than-air flying 9x 10-4 2.6 3,000c 1970-77 1 Mountaineering 6 x 10-" 34 60,000 1970-78 1 Mountaineeringk 7 x 10-4 12 19,000 1951-60 4 Power boat racing 8 X 10 4 5.2 6,500 1970-78 1 Professional stunting <_ 1 x 10'd°r 1 200 1975-78 1 Rodeo < 3 x 10-s d, e 0.33 34,000 1970-78 1 Scuba diving 4 x 10-4 126 300,000 1970=76 1 Ski racing 2 x 10-$ 2 81,000 1970-78 1 Spelunking < 1 X 10 4 d' i 0.44 10,000 1970-78 1 Sport parachuting 2 x 10 3 41 25,000 1970-78 1 Thoroughbred horseracing 1 x 10-3 2.6 1,800 1970-78 1 Swimming 3 x 10-5 2,600 82 x 106 1972-78j 2,3 a. No error estimates are: given. The reason is that, although we could give statistical sampling errors on the risks shown, the population size is so uncertain in most cases (by a factor of 2 to 3) that this uncertainty dominates. b. Per person at risk. See preceding note on error estimates. c. This population corresponds only to pilots certified by the Federal Aviation Administration. d. The value shown is statistical 95 percent confidence upper bound, assuming risk proportional to person-years of exposure and a Poisson dis- tribution of deaths. See also note a on error estimates. e. Three deaths observed in time indicated. f. No deaths observed in time indicated. g. One death observed in time indicated. h. Two deaths observed in time indicated. i. Four deaths observed in time indicated. j. Population figures from 1972, deaths from 1978. We have assumed a similar population went swimming or boating in 1978. k. Not strictly comparable with the preceding entry, also labeled Mountaineering. The figure in the population column is total man-mountain- ays, and the risk is per man-mountain-day. This agrees with the previous figure for annual risk if an average of - 0.9 days per year is spent moun- aineering, but note that the year; of coverage differ also. I. If participation has remaiined constant, as we assume, there are possibly decreasing trends in these risks. Sources: 1. Metropolitan Life Insurance Company (1979). 2. U.S. Bureau of the Census (Annual). 3. National Safety Council (Annual). 4. Fer- s(1963). (The article also discusses some of the problems of interpretation of risks such as those shown in this table). ~ tb ~

IV. Risk Assessment: The quantitative establishment of the relationship between public exposure to a hazard and its health and safety consequences suffers from several handicaps. 1. Analytic uncertainties become larger as the exposure levels per person becomes smaller because the data base becomes vague and finally nonexistent. At the same time, the number of people involved tends to increase, so the cumulative public risk becomes indeterminate. As a substitute for scientific information, agencies use simplified extrapolations of high level da.ta and "worst case" projections for regulatory purposes. 2. The agencies treat each hazard as an independent source of risk, and use fixed criteria for setting targets (e.g. EPA's 1 in a million lifetime risk). Comparative risk analysis is generally absent, although implicit balancing of the :public perceptions of the relative importance of risks is subconsciously involved in agency attention. 3. The "worst case" syndrome pervades agency decisions. Given the uncertainties of the data, it is easier to publicly defend a "worst case" choice. However, the economic consequences may be extremely large, particularly when orders of magnitude are involved. 4. It is :impossible to consistently use "worst case" analysis. Agencies tend to apply "worst case" projections to selected situations they choose to analyze, and to ignore the actual risks of unanalyzed hazards. This tends to make regulation arbitrary, capricious, and independent of the actual risk. V. Risk Management: 1. There are no generally accepted measures as yet developed to compare the unlike consequences of a variety of risks (e.g. life threatening vs physical impairment, human vs ecologic, physical vs psychological, short term vs long term, etc). Nevertheless, such implicit evaluations are being made, and are shaped by the cultural biases of the decision makers. 2. There are no guidelines for comparative cost-effectiveness of the remedial measures imposed by regulations to achieve improvement in health or ecology, or of the reduction of involuntary exposures to risks. Regulations CS/dt/Papers/Bromley/Version 2/Revised 3

CONTRAST WITH SACCHARIN Wide use; if dose-response is LINEAR, FDA calculate 500 cancers/year (April 15, 1987 Federal Register) (x 10f uncertainty) ******************************** * * * RISK IS BIGGER FOR SACCHARIN * * * ~e9t*~k****~h~k****~k~k~h*9e*~kok**~F****~E*~F But dose-response relation may be non-linear Not found in humans (sensitivity too small) Not a mutagen *****~~****:r******************************************************* * * * CERTAINTY OF (DEDUCTION FROM) DATA IS BIGGER FOR VINYL CHLORIDE * * * *~eok~k~kk,k*4e**~e~h~F***ok4e*~k~F*~es@**skoY*9i*****~e*~F~k~k~e~k~k**~k~kotek~k~F~k*~k~k*~k*~e**** ***

174 RISK/BENEFIT ANALYSIS EVERYDAY LIFE: A CATALOGUE OF RISKS 175 Table 7-1. Some One in a Million Risks. Living in the United States: Time to accumulate a one in a million risk of death from the cause indicated. Motor vehicle accident 1.5 days Falls 6 days Drowning 10 days Fires 13 days Firearms 36 days Electrocution 2 months Tornadoes 20 months Floods 20 months Lightning 2 years Animal bite or sting 4 years Occupational Risks. Time to accumulate a one in a million risk of death in the occupation indicated. General Manufacturing 4.5 days Trade 7 days Service and Government 3.5 days Transport and Public Utilities 1 day Agriculture 15 hours Construction 14 hours Mining and Quarrying 9 hours Specific Coal Mining (accidents) 14 hours Police duty 1.5 days Railroad Employment 1.5 days Fire Fighting 11 hours Other Risks. Cosmic Rays. One transcontinental round trip by air. Living 1.5 months in Colorado compared to New York. Camping at 15,000 feet for 6 days compared to sea level. Other Radiation 20 days of sea level natural background radiation. 2.5 months in masonry rather than wood building. 1/7 of a chest X-ray using modern equipment. ~~~St7sizQZ Table 7-1. continued Eating and Drinking. 40 diet sodas (saccharin) 6 pounds of peanut butter (aflatoxin). 180 pints of milk (aflatoxin). 200 gallons of drinking water from Miami or New Orleans. 90 pounds of broiled steak (cancer risk only). Smoking 2 cigarettes Source: Tables 7-2 to 7-5. clearly into categories within which intercomparison is more easily justified and probably more accurate. Table 7-2 is a list of various commonplace risks of death, most of which would be considered involuntary. Notice that there may be some overlapping between categories (home accidents, for example, includes falls within the home). Table 7-3 shows some occupational risks, mostly risks of fatal accidents. Again, most such risks would be considered involun- tary by those exposed. Table 7-4, in contrast, shows a set of volun- tary risks of death, those incurred in sporting activities. Table 7-5 is a further set of everyday risks, but now specialized to cancer risks, selected because such risks arouse particularly strong emotions. Before discussing these risks in more detail and indicating how they are all estimated, we would like to give another example that may help place these risks in perspective. Four tablespoons of pea- nut butter per day is shown as giving a risk of liver cancer of 8 X 10-6 per year, or a lifetime risk of 6 X 10-4. But four tablespoons of peanut butter corresponds to 400 kilocalories (Kcal), so if one were to eat only peanut butter, daily energy requirements would be sup- plied by 26 tablespoons per day, giving a lifetime liver cancer risk of 4 X 10-3, or 0.004. This should be cor -oared with a lifetime proba- bility of any kind of cancer of about 0.25, even in the absence of peanut butter. r

ILLUSTRATIVE CATEGORIZATION OF EVIDENCE BASED ON ANIMAL AND HUMAN DATA1 Human Evidence Sufficient Limited Animal Evidence Inadequate No Data No Evidence Sufficient A A A A A Limited B1 B1 B1 B1 Bl Inadequate. B2 C D D . D No data B3 C D D E No evidence B4 C D D E Source: Fed. Reg. 51, No. 185, Wednesday, Sept. 24, 1986. 1The above assignments are presented for illustrative purposes. There may be nuances in the classification of both animal and human data indicating that different categorizations than those given in the table should be assigned. Furthermore, these assignments are tentative and may be modified by ancillary evidence. In this regard all relevant information should be evaluated to determine if the designation of the overall weight of evidence needs to be modified. Fa:levant factors to be included along with the tumor data from human and animal studies include structure-activity relationships, short-term test findings, results of appropriate physiological, biochemical and toxicological observations, and comparative metabolism and pharmacokinetic studies. The nature of these findings may cause an adjustment of the overall characterization of the weight of evidence.

CRITICAL COMMENTS BY DISTINGUISHED PEOPLE ON RISK ASSESSMENTS FOR CARCINOGENS -------------------®--------------------------------------------------------- '~RICHwRD PETO, Reader in Epidemiology, University of Oxford: All important carcinogens found by epidemiology; none by animal tests *~* All that can be done is form a priority list Some index of nastiness N Some index of exposure E Prioritize by N x E Use more than one index to be sure that nothing is missed *in Assessment of Risk from Low-Level Exposure to Radiation and Chemicals eds., A.D. Woodhead, et al., (Plenum, New York), 1985. --------------------®-------------------------------------------------------- **AMES (Professor of Biochemistry, U.C. Berkeley) AND OTHERS Prepare a priority list from animal tests **Science, ;?36, 271 (1987). ------------------------------------------------------------------------------- DOUGLAS FOY, (lawyer for Conservation Law Foundation) Although risk numbers are not believeable, they are useful as a priority list ------------------------------------------------------------------------------- AHMED (Environmental Defense Fund) Statement similar to Foy's ------------------------------------------------------------------------------ CONCLUSION: A CALCULATED RISK OF A SINGLE CHEMICAL BY ITSELF IS NOT USEFUL ONE NEEDS A LIST OF SEVERAL

Idealiseci Scherne for Risk Analysis. .b a Scientific Data Economic and Engineerin6 Data Interested Parties \ ASSUMPTIONS "Dc Minimis" Risk. STOP 10'5/yr. Occupational 10"6/yr. General Total Societal Impact 10/yr. ~-~ Nurnbers t Unct:rtainty Risk Asscssrnent ~ Cost Assessment Benefit Assessment Valuc f udt;emcnts Numbers ± Uncertainty Numbers ± Uncertainty . Risk/(3enefit Risk/Risk Risk/Cost Comparisons is-- DECISIONS Results of decision Kniglrthoods, I nsuI ls, Anonyrnous letters etc.

. s RISK ASSESSMENT OF CHEMICAL CARCINOGENS: IS IT TIME FOR A CHANGE? By Robert J. Scheuplein, Ph.D. Director, Office of Toxicological Sciences Center for Food Safety and Applied Nutrition Food and Drug Administration Washington, D.C. 20204 Thank you, Carol, for the introduction and to ILSI for inviting me to, speak to you today. My subject is the Quantitative Risk Assessment of Carcinogens: Is it timE: for a change? Before I speculate about the direction in which risk assE>.s,sment might be headed or ought to be turned, I would like to give you some idea of where I think its come from, what science it rests on, what: it is and what it is not. This will involve a brief scientific history of cancer risk assessment with a few detours to places where it touches social policy or has been influ,enced by the Congress and the regulatory agencies. This interaction between science and regulation is important to understand, because a part of my thesis will be that despite well over 500 papers on cancer risk assessment, on the bioassay, on cancer thresholds and numerous related subjects, since 1961 (the date of Mantel and Bryan's paper) -- cancer risk assessment is still more of a regulatory tool than a scientific discipline and rests more on regulatory need than scientific plausibility. Presented at Brookings Institution, Washington, D.C.~June 17, 1991 ~

N carcinogenic chemicals were contributing to the world-wide diversity in cancer incidence. Percival Pott had established the connection between soot and scrotal cancer in chimney sweepers several generations earlier. By the 20's it was clear that polycyclic hydrocarbons were the carcino- genic ingredients in soot, tar and oil. Some cancers on the abdomen could be attributed to carrying a basket of live coals beneath the clothes to keep warm in winter; some cancers in the buccal cavity could be attributed to chewing various mixtures of betel, tobacco and lime and some on the palate to smoking cigars. As Richard Doll (1977) has pointed out„ oncologists who worked chiefly in Europe and North America tended to regard these incidents as oddities and irrelevant to the production of ord:Cnary cancers. So it took a while for people to associate the major- ity of cancers with environmental factors, but soon the association became obvious. These concerns in the 30's and 40's motivated an effort to bring the kno,~nT occupational hazards under control, either by banning their produc- tion or controlling the manufacturing process to reduce exposure to employees. (Slide-4) - Occupational Cancers- Doll (1977). But these occupational hazards could not be responsible for the large observed incidence of cancer. Whole populations, however, had been exposed to lower levels of these same agents. These included polycyclic hydrocarbons, produced by the combustion of coal, wood and oil. It was known, for example, that residents of large towns in the U.K. may have been exposed - mainly through the combustion of domestic coal - to . something like 1/100 the amount of benzo(a)pyrene regularly inhaled by men working in the manufacture of coal gas and these men experienced only

r, -D- misc:oding sufficient to produce eventually a malignant cell. The cell then can reproduce itself in an irreversible and unregulated manner to yield a malignant tumor. But this is theory! What does the experimental evidence show? So far, for any carcinogenic or mutagenic response in any given situation, be it man, mouse, isolated organ or a Salmonella plate assay, there is a demo°:nstrable threshold or "no effect level." In thousands of studies with hundreds of thousands of animals, not a single carcinogen has been found that has not exhibited an experimental threshold. However, animal studies are insensitive and thresholds will vary from individual to individual. It is completely impractical to determine the level at which the most susceptible individual in the whole population might fail to respond. And worse yet, if a gargantuan animal study were done, assuming all the experimental difficulties involved in such a study could be overcome, people would point out that experimental animals are more inbred than people and the result would probably be discounted. The o,ne-molecule theory has to be argued at the theoretical level, so let's look at it. Nothing (in the one-molecule theory) is mentioned about the relationship between the intake dose and the final concen- tratiori of the chemical carcinogen in the nucleus of the cell where it interacts with DNA. Substances that are ingested have to be absorbed, distributed and metabolized usually before they can reach critical organs in chemically activated form. Then the activated molecules have to run a ; gauntlet of sequestration by other uninvolved macromolecules and overcome

TABLE 111. Estimated Lifetime Cancef R6sh from Ocmpafionaf ExpoSVre to the TLV IARC TLV Adjusted Daify Expsure (µg/m') Assaciated with Risk of Estirrtated Life6me Carrcer Risk from Substance Class µg/V Unit Risk' 1/10' 1/101 ExQostrre to TLV Acry)arnide 28 30 2.4 x 10-' 4.2 x 10-; 4.2 0.0072 Acrylonitrile 2A 4500 1.3 x 10-5 7.7 x 10-z 7.7 x 10' 0.057 Benzene 1 30000 1.5 x 10-6 6.7 x 10-' 6.7 x 102 0.044 Beryllium 2A 2 4.5 x 10-' 2.2 x 10-' 2.2 0.0009 1,3-Butadiene 28 22000 5.2 x 10-5 1.9 x 10-2 1.9 x 10' 0.68 Cadmium 2A 10 3.3 x 10-' 3.0 x 10-3 3.0 0.0033 Carbon tetrachloride 2B 30000 2.8 x 10-6 3.6 x 10-' 3.6 x 102 0.081 Chloroform 213 50000 4.3 x 10-b 2.3x10-' 23x102 0.19 bis(Chlorometltyl)e~ 1 5 1.2 x 10-2 8.3 x 10-5 8.3 x 10-z 0.058 Chromium (VI) 1 50 2.2 x 10-3 4.5 x 10-* 4.5 x 10-' 0.10 Dichloromethane 2B 175000 7.6 x 10-' 1.3 1.3 x 103 0.12 (M4etiiylene chloride) Ethylane oxide 2A 2000 2.0 x 10-5 5.0x10-z 5.0x10' 0.039 Famnalde?* 2A 1500 2.4 x 10-6 4.2 x 10-' 4.2 x 102 0.0036 Hexachlorobuhadierre 3 240 4.1 x 10-6 2.4 x 10-' 2.4 x 102 0.00098 fdickel refinery dusf 1 1000 4.5 x 10-5 22x10-2 22x10' 0.044 Vinyl chloride 1 10000 1.3 x 10-6 7.7 x 10-' 7.7 x 102 0.013 'From Table If adoed fa ocapatiatal aqwstxe. Estirnaled risic b hanm kan epOSUe b a time-aeigtled ayeape d 1 µphrr' br a noma! 8-hau wakday. 40-1m workweelc 4anar career (see ie4. contains the Iinternational Agency for Research on Cancer's (IARC) classification of each of these chemicals.(1`') This classification scheme evaluates the likelihood that these chemicals are human carcinogens but makes no attempt to quantify their potential risk or to set "safe" exposure levels. Hexachlorobutadiene is classified by IARC in cate- gory 3, "the agent is not classifiable as to its carcinogenicity to humans:'t14> Seven other chemicals: acrylamide (2B), acrylonitrile (2A), beryllium (2A),1,3-butadiene (2B), cad- mium (2A), carbon tetrachloride (2B), chloroform (2B), methylene chloride (dichloromethane) (2B), ethylene ox- ide (2A), and formaldehyde (2A) are in IARC category 2, "the agent is probably (2A) or possibly (2B) carcinogenic to humans." The remaining five chemicals, benzene, bis(chloromethyl) ether, chromium VI, nickel refinery dust (nickel comlxiund5), and vinyl chloride are in IARC cate- gory i "human carcinogens." Using vinyl ~chloride as an example, Figure 2 illustrates the typical relationship found between the dose-response curve resulting from a QRA of the type performed by the EPA, the empirical data on wh;.:i the modeling is per- formed, and the TLV established by the ACGII 1. The slope of the dase-response curve shown here (i.e., 0.0013) is derived from ihe EPA unit risk factor for vinyl chloride adjusted to reflect the exposure situation of the occupa- tional environment. Discussion In this set of 16 chemicals, both the EPA and the ACGIH approaches r,tnk them in approximately the same order of carcinogenic risl:. However, the EPA is far more con- servative, refle(aing the agency's objective to piotect all members of the community, not just healthy adults. The authors could ;not definitively comment on the relative accuracy of the two approaches because our theoreticall understanding of the dose-response relationship for oc- cupational carcinogens is still elementary, and we are, therefore, limited in our ability to discriminate between the accuracy of the'ILV and QRA approaches. One is further hampered by the fact that the empirical data available to assess the carcinogenicity of specific chemicals are usually the result of animal experiments at high doses, together with a battery of short-term tests which are sometimes augmented by epidemiology studies that usually have scanty exposure information. The available occupational cohort studies have not followed workers for their entire lifetime and, thus, do not give complete inforntation on agents which cause cancer many years after exposure. Conse- quently, no one at the present time can speak with scientific certainty about "safe" levels of exposure to carcinogens. Although decisions on the perrnissible exposure to car- cinogens are fraught with difficulty, we believe that rec- ommending maximum levels of occupational exposure should be guided by three principles: 1. Scientifically, one should seek the most appropriate data and methnds for predicting the effect of human exposure to carcinogens based on our latest theo- retical understaaditt¢; of the prooess of carcinogenesis. 2. As a public health issue, one should admit the im- precision of our knowledge and compensate for our uncertainty by building into the system a margin of safety. 3. As public policy, one should explicitly document the methodology. The increasing motivation to use QRA as a tool to es- tablish occupational health standards dates from the 1980 decision by the Supreme Court to overturn the Occupa- tional Safety and Health Administratfon's (OSHA) newly proposed benzene standard.0 5) The court maintained that 514 APPt- oCCUP. eW1NaVt PlYS. 3/t! • A06tJSt 1990

addi.tives as methods improved to the point of being capable of detecting trace carcinogenic contaminants. This was the scientific and regulatory setting when in 1969 the FDA gathered together a group of scientists to consider how food additives and pesticides should be tested and evaluated for possible carcino- genicity. The Report of that FDA Advisory Committee on Protocols for Safety Evaluation and in particular the Panel on Carcinogenesis was publashed in 1971. This report stated among other things that: 1) Testing should be done at doses and under experimental conditions likely to yield maximum tumor incidence; 2) And that at least two species should be used for all carcinogenicity tests, and that, 3) For compounds judged carcinogenic at test levels, a virtually safe dose could, in principle, be estimated by downward extrapolation using some arbitrarily selected but conservative dose-response ;.urve. . These recommendations initiated the regulatory use of the MTD bioassay and IjLiiear risk assessment. Details on both the bioassay and the method 0 of ext:rapolation would evolve, but this 1971 Panel Report gave their

r, -6- diffusional barriers before they can first enter a cell, and then later enter the nucleus of a cell. When the carcinogen is in the nucleus and poised to react with DNA, nothing is mentioned (in the one-molecule theory) about the constraints imposed on chemical reactions by the requirements of mass action or the need i:o acquire a transition state configuration or an activation energy prior to reaction, or even that the final adduct be somewhat stable so that 1Ct can last long enough for replication. And :ii` a reaction with DNA does occur, nothing is mentioned about the intron content of the DNA, that the amount of DNA in the nucleus that is not overtly expressed as protein may amount up to 90% of the total. These regions of DNA, which correspond mostly to the centromeric parts of the chromosomes almost certainly instruct no other process than their own repl:[c:ation. And finally nothing is said of the ability of the organism to accommodate to adverse effects - in this case by DNA repair mechanisms and enzyme induction. An example on this last:point was published in 1977 by Tony Pegg of the Hershey Medical Center. Dimei:hylnitrosamine (DMN) is a potent carcinogen in many species. It is well established that DitN exerts its carcinogenic effect after its metabclic conversion into a reactive methylating agent. The electrophile then methylates DNA nucleosides which are likely to miscode. One partic- ular adduct - 06 - methylguanine appears to be promutagenic or tumori- genic: in several studies and it has been identified as probably the adduc:t responsible for tumor induction in animals.

that a substance is a carcinogen but are no* usually helpfu! in setting a TLV. A safety factor is often applied to estabiisE; a TLV for carcinogens, by taking the lowest level known to induct cancer (or the no-effect level) and then dividing that hv in arbitrarv factor, such as 10 or 100. The CS-TLV C;ommittee. realizing the imprecision of setting TLV,, for carcinogtn.ti. recommends that. for all carcinogens having a TlY. "worker exposure by all routes should he carefully controlled to levels as low as reasonably achievable (A1ARA) below the TLV:"') In the earlv 1970s, the U.S. Environmental Frotection Agency (EPA) developed an approach that wa5 different from that of the CS-TLV Committee. Early decisions by the EPA conveyed the idea that the only acceptable degree of regulation of carcinogens would be a total ban on expo- sures.(3.") However, the impracticality of achieving zero risk on a broad scale for a large number of economically important chemicals became increasingly apparent to many, including die U.S. Congress. As a result, the EPA in 1976 became the first federal agency to adopt formal guidelines embracing a two-step process of risk assessment. The first step is a determination of whether a particular substance constitutes a cancer risk, i.e., hazard identification. The second step includes a quantitative risk assessment (QRA) as a key component of determining the degree of regu- latory action needed to protect the publict5~ - As part of the QRA process, the EPA computes dose- response curves, makes low-dose extrapolations, and es- timates the size and degree of exposure of the exposed populations in order to estimate the number of excess cancers exp~cted in the total U.S. population. The rationale and procedures for the EPA approach are used to guide regulatory actions which are meant to protect each mem- ber of the general public over a lifetime against exposure via inhalation or ingestion.(6) Regulatory action is taken only after the results of the QRA are integrated with en- gineering data and with social, economic, and political concerns.C> Confusion-has arisen from the different approaches used by the CS-TI V Committee and the EPA in estimating risk Although tihe TLVs continue to be used widely by profes- sional industrial hygienists around the world to evaluate the safety of workplace exposures, the QRA approach is viewed by some as riore objective. Recently, criticism of TLVs has focused attention on the objectivity and scientific standards of the CS-TLV Committee.ts> Several examples were given of chemical substances for which unpublished data (primarily from the files of industrial companies) were important in setting the recommended TLV. Since, in many instances, the TLV is the only number available to industrial hygienists, it is important that the CS-TLV Committee's pol- icies and procedures regarding carcinogens be reviewed to assess the results of the current TLV approach. We be- lieve a quani:itative comparison of the EPA and the TLV approaches may provide some important information re- garding thi.; assessment. Reflecting on some of these issues, Andersent9> pre- sented a critical review of quantitative risk assessment in occupational health in the 1988 Herbert E. Stokinger Lec- ture, concluding that, "Quantitative Risk ASSessment is not just coming to the occupational environment. It is here now and is an issue to be reckoned with by everyone of us in the industrial hygiene profession"") in his review, Andersen suggetits that QRA during the past 13 years has been "damned" hv its misapplication. Overly conservative quantitative approaches to predicting risk would lead to risk estimates that "greatly restrict commercial operations, decrease our ability to compete in world markets, and lead to large expenditures to change work practices with no concomitant increase in health protection:' He went on to suggest that the problems faced by the use of overly con- servative techniques can be overcome in part by the use of recent cancer models that have greater biological rel- evance, e.g., the physiologically based pharmacokinetics models (PB-PK)t10> and the Moolgavkar, Venzon, Knudson (MVK) models.t t t> Although the theoretical appeal of these cancer models is clear, the bulk of the QRAs developed and published since 1976 have come from regulatory agen- cies which have not used these new techniques. We cannot compare current TLVs to the results of risk assessments using the MVK or PB-PK approaches; however, comparing established TLVs for carcinogens with the results of the EPA QRAs may help determine whether, and under what circumstances, the CS-TLV Committee may consider using QRAs as part of its decision-making process. This article presents a comparison between the ACGIH TLVs and the EPA QRAs for the 16 chemical carcinogens that have been evaluated by both groups. These QRAs were chosen for comparison since they are the largest available collection of risk as.5essment5 developed by a standard methodologic approach. f ARedods 8etd Reaults The comparison reported herz is derived from the ACGIH 1988-1989 list of'II.Vst12) and an EPA list of carcinogens taken from the Integrated Risk Information System.t13)'Itie ACGIH list contains over 700 agents of which 55 are clas- sified by the Committee as carcinogens in the adopted list plus 3 in the Notice of Intended Chznges List. These 55 substances are listed along with their TLVs, where available, in Table I. The EPA list, in Table II, oontains 54 agents, including a substantial number of pesticides and nitrosa- mines for which a unit risk factor fbr inhalation exposure is available. The EPA's unit risk factor is a conservatively estimated risk to humans from constant lifetime exposure of breathing contaminated air at a level of 1 µggm3. This risk estimate is derived from the availabL- results of animal bioassays, biochemical studies, and epidemiologic studies. To assure safety, conservative assumpaons are used to sup- plement missing or unknown information (e.g-, using re- sults from the most sensitive animal species and the lin- earized multistage dose-response model and extrapolaring using the upper 95 percent confidence limit of the ex- perimental evidence). The ACGIH TLVs are compared with the EPA QRAs APPL OCCUP. ENb7RON. 1fYG. 50 • AUGUST 1990 511

r, What Pegg did was to give DMN to rats and then analyze the livers for the 06-~adduct after 4 and 24 hours. His reE;ults are shown in the slide. (Slide 5). 06 _Methylguanine Levels in Rat Liver DNA At first glance, there appears to be a linear relationship between the administered dose and t:he formed adduct. But if the data are extended to lower doses, this is shown not to be true. At very low doses, 06-adducts levels are many times less than expected on the basis of linearity. Later studies showed that: a saturable enzymic repair system was responsible for the removal of the 06-adduct and that the repair system operated in both liver and kidney cells. These studies don't necessarily imply an absolute threshold for tumor induction for DMN because it was possible to detect some 06 - adduct in DNA 24 hours after a single dose and we don't know what level of 06- methy1guanine may be necessary to initiate tumor induction. But they do show t:hat liver and kidney can protect itself against a low-dose, car- cinogenic stimulus and that linear extrapolation is probably unjustified at low doses. These studies by Pegg are pertinent because they provide information well below the observable tumorigenicity range - over a 20,000 fold dose range - and measure the concentration of the specific adduc.t identified to be the one likely to initiate tumors. In summary, there are many reasons to contest the "one-molecule theory" and to anticipate, in conformity with the animal evidence, that signifi- cant concentrations of a carcinogen might be required to elicit cancer. But nevertheless, the "one-molecule" concept or its equivalent that exceed:Cngly small levels of carcinogens ingested daily for a lifetime !

-13- rfant:el and Bryan had already introduced the concept of using this upper limit of the negative study to base the downward extrapolation on, using a I probit per 10 fold dilution slope. And so, in principle, quantita- tive extrapolation as a regulatory method (QRA) was approved. But it was never assumed that cancer necessarily occurred at low doses, but oinly that if it did, it would be safely bounded by the extrapolation procedure. Now extrapolating a positive response instead of an upper limit to a negative response makes no essential difference as Starr and othe•ts> have shown unless we know what the dose response curve is. Since our e);trapolation models are quite simplistic and without an adequate biological basis - there need not be any cancer at all at low doses - and if there is, we have no idea about its actual dose response. The notion that one can calculate expected values of an actual risk from such an analysis is really quite bizarre. Extrapolation using bioassay data 3-4 orders of magnitude removed is not a procedure for estimating risk - it's regulatory standard setting - It's not science, it's Po__~licX. And it's conservative or, if you will, prudent policy - and I'm just describing it, not advocating it. The scientists who wrote the 1971 Protocols Report had to face two problems. The QRA procedure they were recommending was extremely conservative, it would ban any carcinogenic food additive, because the amount of additive is usually substantial. Remember the additive has to be used in amounts sufficient to accomplish its intended effect. But this didn't really bother anyone at the time - carcinogenic additives had no place in food anyway. And they pointed out that this application of

I -15- time though with a legal loophole called the DES Proviso. It said in effect that you could use effective animal drugs even if they were carcinogenic so long as none remained in the edible tissue of the animal after slaughter - no residue would be permitted and the FDA was given the task of approving the analytical, method to assure it. This ushered in r the era that some industry groups characterized as "chasing zero." The ot:her related difficulty occurred in food and color additives. Analytical methods were becoming so sensitive that traces of carcinogenic contaminants were being found, particularly in colors. It was hard not to firid carcinogenic derivatives of aniline, a carcinogen, in aniline- based colors. And so the question was how do you regulate a substance which does not test out as a carcinogen itself, but which contains a chemical at low but detectable levels which is a known carcinogen? Both of these problems would challenge FDA for many years, the first culminat- ing in the final SOM document in 1985, the other in the constituents policy in 1983. The need for both a procedure for risk assessment and a level of acceptable risk were common to both issues and of course they are interrelated. If you use the Probit model and a IO-8 acceptable risk level, you come out about the same place as if you used a linear model and a 10-6 acceptable risk level. Extrapolation Models and Background Additivity In the late 70's and early 80's, there was a good deal of debate over the best form of the extrapolation model. The original Probit model of 0 Mantel and Bryan was considered too arbitrary and not conservative enough - with the developing trend toward a 10-6 or one in a million

11 -2- Tt nonetheless involves several scientific matters and any good history should include them: (Slide 1) - Major Scientific_Issues in Cancer Risk Assessment. These issues are all involved in QRA and the final result - the deter- minai:ion of a safe level - depends on how these issues are resolved. And I`ll touch briefly on most of them. Righ- after World War I, scientists began to experiment with large colonies of rodents. Shortly thereafter, Yamagiwa and Ichikawa dis- covered that stdn cancers could be induced in mice by the repeated application of gas tars (ref.). The number of chemical agents tested grew steadily from year to year and it became difficult to analyze the available experimental results because of the variety of different methods adopted. By 1930, efforts began at standardizing these methods. (Slide 2) - Paper_TitlebyTwort and Twort. By 1939, methods were sufficiently well developed that lists of substances with relative potencies were published based on the ability of a c.ompound°s capacity to produce tumors in the shortest possible time. (Slide 3) - List-of Carcinogenic Comgounds Arranged in Descending Order of_Fotency - John_Iball_(1939). In this list by John Iball in 1939, the index of potency is the percentage of tumours A divided by the mean latent period B, recorded in the last column. . There were good social reasons for these academic efforts. It was becoming clear that environmental and occupational exposure to

-17- yielded greater conservatism and had the right name, but what really clinched linear risk assessment was an idea published in 2aor 3 papers in 1977 and 1978 by Crump, Hoel, Peto and their co-workers. These papers contained a notion which today is unfortunately part of quantitative risk assessment mythology, namely that there are sound biological reasons for belj:eving that every carcinogen response curve is linear at low dose rates, as far as humans are concerned. This proposition rests on the presence of background carcinogens and the way they interact with the carcinogen in question. The reasoning is part biological and part mat:hematical. The_biological_)art Approximately one of five Americans develops a cancer, and every person is exposed to thousands of carcinogens in food, in the environment and even endogenously. This 20% background rate, from these many different chemj'.cals must surely provide some significant mechanisms that are shared by t:he carcinogen in question. In other words, the carcinogen being adde>.d' and some of the background carcinogens must share a common pathway to carcinogenesis - and thus produce cancer through an identical mechanism. Their effects are functionally indistinguishable. Now the mathematical part As the slide shows, (Slide 9) - Background Additivity, the cancer inci- dence I(d) will be a linear function of the dose rate A at low dose rates provided that the slope (F1 (Do))is positive. They defend the assump- tion that the slope is positive by arguing that even if there were a threshold, it would be a threshold for each cell; there would be a

70 RISK/BENEFIT ANALYSIS This form of analysis constitutes an all-at-once technique, but one that is less useful than the preceding example (chemical hazards) be- cause it aIves no clue as to how to reduce the risks. The limitatin n ca, ~e Overcome if it is possible to analyze each event leading to risks as a sequence of well-understood events forming an event tree. A set of event trees would cover all possible cases leading to the final risky event. The most well-known use of event tree analysis is, perhaps, the analysis of nuclear reactor accidents in the so-called Rasmussen report (Nuclear Regulatory Commission 1975). The procedure will be briefly outlined here with a highly simplified example from the report. The first step is identification of all possible sequences of events that may lead to serious consequences to public health and safety, followed by the separation of these sequences into segments that are approximately independent of each other. By analyzing each event in each sequence separately, using theory or past experience or both, the overall probability of occurrence of the whole sequence can be evaluated. Thus the most probable (highly simplified) se- quence for catastrophic failure in a PWR is shown in Figure 3-6. The overall accident probability (with assumptions to be mentioned) is then equal to Probability of a pipe break (from theory and past data on other pipes) x probability of failure of emergency core cooling system (from a fault tree analysis) X probability of containment violation (more fault trees) x probability of unfavorable weather (past data on wind patterns, rain- fall, and so on) ~ PI P2 P3 Pa • The accident described by this event tree is initiated by the break of N a water pipe in the cooling system causing loss of coolant and result- ing, ing, if the emergency core cooling system subsequently fails, in the ~ meltdown of the reactor core and the release of the fission products ~ therein. This may cause a violation of the concrete containment ves- ~ sel, so that if the wind direction is right the released fission products ~ may be blown over population centers, possibly causing radiation .~ overdoses to a large segment of the population. In each case the probability of the event and also its severity must be evaluated. As ~ indicated, the probability (pl ) of a pipe break may be estimated from historical experience with pipes first in other industries, sec- 104 RISK/BENEFIT ANALYSIS cases are examined in detail major flaws ~^u wedK '- u.. ncsses can be dis- covered_ We are l:eartc;,cd, iiowever, by the realization that many of the flaws can easily be remedied and that in at least one case (saccha- rin) common sense filled in the gaps. Before beginning a detailed discussion of individual studies, it is useful to picture an idealized scheme (Figure 6-? ) of the complete decision process, and the place of risk assessment within it. Informa- tion is passed from scientist, engineer, and economist to risk assessor and to cost and benefit assessors. The results of these assessments and comparisons between them are made available to the decision- Figure 6-1. Idealised Scheme for Risk Analysis. I ASSUM PTIONS " " De Minimis Risk. STO P 10's/yr. Occupational 10'6/yr. General Total Societal Impact 10/ yr. Scientific R isk C Data Asses sment --~~ Numbers ± ~ Uncertainty 1 Economic and ~ C ost I Engineering Data I Asses sment Numbers ± ( \ Uncertainty Risk/Benefit Risk/Risk ~ ~ Risk/Cost ~ ( Numb Comparisons Ben efit Uncertaint ~ Asses 9 sment y ~ I ~ i Interested ~ Va lue Parties ludge ments 1- DECISIONS ( L --l Results of decision Questions Alternative f tw 1 P i l K or Decision . oss e b Decisions nighthoods, Insults, Anonymous letters etc.

a -26- iii. mechanism studies in vivo, e.g., foci development iv. secondary mechanism for non-mutagenic carcinogens

-10- The Secretary also said, "'6,ihenever a sound scientific basis is developed for the establish- inent of tolerances for carcinogens, we will request the Congress to give us that authority:" and a].so, "'So long as the outstanding experts in the National Cancer Institute and the Food and Drug Administration tell us that they do not know how to establish with any assurance at all a safe dose in man's food for a cancer-producing substance, the principal in the anticancer clause is sound." The Congressional response was, of course, predictable and the new Color Additives Amendments of 1960 contained its own Delaney anticancer clause. Congress concluded that there was'too much uncertainty and it would require FDA to regulate on the side of caution by banning all animal carcinogens from the food supply. Of course, it was only 1960 - analyt- ical methods were typically capable of detecting a few parts per million at best:. There were relatively few carcinogens known and it was not apprer_j'.ated that traditional food and spices and ordinary cooking prac- tices would eventually be found to account for many if not most of them. The widespread contamination of food by low levels of environmental contaminants like dioxins, PCBs and aflatoxin had not yet been discovered. Nor was the fact that the failure to specify "safe" levels would assure the triggering of the Delaney Clause on food and color

-21- o - The correlation in the carcinogenic potency estimates in both rats and mice are determined nearly entirely by the magnitude of the MTD used and only minimally by the extent of the carcinogenic response. c - Based on upper limits, inferred potencies from some substances giving no response in the MTD-bioassay appear to pose a possi- ble carcinogen risk as high as 10,000 times greater than other demonstrated carcinogens. o - There is no reason to believe that the inverse MTD or its equivalent I/TD50 should be regarded as a valid indicator of the low dose risk either to animals or to humans. (Slide-12) - Starr's-Comparison of I/MTD with Potency Estimates-of 83 Rat Carcinogens There is a plausible explanation for the strong 1/MTD vs. potency corre].ation despite the fact that it is hard to prove. Many believe, as I do„ that the high doses used in the bioassays are capable of producing carcinogenic responses not necessarily present at lower doses of the same chemical. Depending on the chemical the mechanisms will vary, e.g., altered metabolic pathways, ala Perry Gehring; altered physiology, e.g., d-limonene, NTA, saccharin, enhanced cell proliferation, ala Bruce Ames; altered endocrine or hormonal status, e.g., mammary and thyroid cancers v and many others. If this is true, and more evidence is accumulating that it is, then high-to-low-dose extrapolation of carcinogenic effects, on

-9- virus assay work. From this experience they felt that a probit slope of one per ten-fold dose dilution would likely be quite conservative for carcinogens. This would mean in the example above that the safe dose corrE:aponding to a risk of 1/100 million would be 1/8,300th of that which produced no tumors in the actual study. This 8,300 corresponds to a safety factor based on an upper confidence limit to a negative study. Later Carrol Weil would propose a factor of 5000 for known carcinogens, using a factor of 10 for animal. variation, a factor of 10 to translate animal results to human, a factor of 10 for cancer on the theory that it is less reversible than other toxic effects and a factor of 5 if the data used were a minimum effect level. (Weil 1972). The Delaney Clause had been passed by Congress a few years earlier as part of the Food Additives Amendment in 1958. In 1960,,there were hearings in the House to deal with some unfinished business regarding Color Additives. Cancer experts testified to the Committee,that there was a great deal of uncertainty about cancer induction at low doses. A report prepared by G. Burroughs Mider, then Associate Director for Research at NCI, was quoted by the Secretary of HEW at the hearings. It had an impact on the Committee. It stated: "No one at this time can tell how much or how little of a carcinogen would be required to produce cancer, or how long it would take the 0 c:a,ncer to develop."

-12- deveP.opment official sanction. The Panel's recommendations were moti- vated essentially by two concerns - 1) the fears over persistent low doses of carcinogens in food the limitations of negative studies on small numbers of animals It was clear that no unqualified negative answer is ever possible. That all a negative study can do is to supply an upper limit to the possible carcinogenicity. It was pointed out that these upper limits are uncom- fortably large. Even with as many as 1000 test animals and using only 90% confidence limits, the upper limit yielded by a negative experiment is 2.3 cancers per 1000 test animals. The report contains the following statement: "'No one would wish to introduce an agent into a human population for which no more could be said that it would probably produce no more than 2 tumors per 1000." So how does one increase the sensitivity of the bioassay with a limited number of animals? The answer - increase the dose well beyond the anticipated use level and extrapolate the results down to these low doses. If the study is positive, the fidelity of the extrapolation depends on the dose-response curve - a small part of which is accessible . in the experimental range. And if the study is negative, it has a theoretical positive upper bound.

chloroform while the EPA reverseti thiy order In addition Tl. ,~ Order EPA Order ti>rmaldehycie is substantially higher on the ordering by ACGIII than by EPA. In establishing TLVs for vinyl chloride and chloroform, the C5-TLV Committee probably weig'.ited liearihy the positive epidemiologic evidence for vinyl chlo- ride, in deciding to establish a relatively more protective value for vinyl chloride than for chloroform. The TLV for formaldehyde is based primarily on prevention of eye, nc>se. and tl• roat irritation. These acute effects have been ohserved in humans at levels below the lowest effect seen for carcinogenicity in rodents. The discrepancy for 1,3- buradiene rin be explained, in part, by the CS-TLV Com- mittee minimizing the relevance of an animal bioassay which induced angiosarcomas of the heart, a rare tumor in humans. ' Table III gives the TLVs and adjusted unit risk for these agents along with the EPA's estimate of daily occupational exposure h---,els corresponding to lifetime cancer risks of one in a million and one in a thousand. This table also gives an e;itimate of the lifetime risk from occupational exposure to a daily level at the TLV. Eight of these 16 estimated lifi>time cancer risks from occupational expo- sure to the TI1,V lie between 1 and 10 percent with the two highest estimates being chloroform at 19 percent and 1,3- hutadiene at 158 percent while the two lowest estimates are he.Y:.tchlorobutadiene at 0.1 percent and beryllium at 0.09 percent. On the average, the TLVs for these 16 agents are over 25 tiree,, greater than the EPA estimated daily expo- Beryllium B is(chloromethyl )ether Cadmium Acrylamide Chromium VI Hexachlorobutadiene Nickel Refinery Dust Ethylene Oxide Acrylonitrile Vinyl Chloride 1,3 Butadiene Benzene Carbon Tetrachloride Chloroform V Bis(chloromethyl)ether Chromium VI Beryllium Cadmium Acrylamide 1,3 Butadiene Nickel Refinery Dust Ethylene Oxide Acrylotritrile Chloroform Hexachlorobutadiene Carbon Tetrachloride Formaldehyde Benzene Vut--yI Moride etT-h`ylene i^ on e RO1RE 1. Ordering of chemicals by estimated risk by the Cheatlical Sub- sure level a.s;;ocia¢ed with a risk of 1/1000. Table III also stances TLV Committee and the U.S. Envirorurwtal Protntim Agency. TABLE ll. C#ttmicals Catdnwgens for Mfttickt QuaciftOM RiSIcs Fare Been Group Compour>ds _ Unii Risit Factors' Acetaldehyde Acrylamide Acrylonitrile Aldrin Arsenic Asbestos Azobenzene Benzene Beuidene Beryllium 1,3-Butadierte Cadmium Carbon tetrachloride Chlordane bis(2-cttloroeih!(t)~tter Chloroform bis(chlorometh}l)Eltter Chromium VI DDT 1,2-Dibromoettlan, Dibutylnitrosamir>E, 1,2-Dichloroeft>itrH: 1,1-Dichloroett>ulene (Vinylidene chloride) DichlorometharNS (Methylerre chloride) Dieldrin Diethylnibosamine Dimethylnitrosarnine 22x10-6 1.3x10-3 6.8 x 1U-5 4.9x10-3 4.3 x 10-3 2-3 x 10-t 3.1 x 1o--5 8.3 x 10-6 6.7 x 10-z 2.4 x 10-3 28 x 10-4 1.8 x 10-3 1.5 x 10-5 3.7x10-5 3.3x10-' 2.3 x 10-5 62x10-2 12 x 1D-z 9.7 x 10-s 22 x 10-t 1.6 x 10-3 26 x 10-s 5.0x10-5 4.1 x 10-6 4.6 x 10-3 4.3 x i0-2 1.4 x 1U-2 0 for httalabon 6tpo:are by U.S al1's CsrraaoyAn AnewrorE CUMMEXIS Uait Risit FsdoPS` 12-Dipt>erryttt}rdrarirte 4.5 x 10-' Epidtlorotrydrin 12 x 10-6 Ethylene oxide Fomlaldefrytie 1.8 x 10-2 1.1 x iU-' Heptactda 1.3x10-3 H"chlor epo" 26x10-3 Nexadtlorobutadiene Headtbrocycolmrte 22 x 10-5 21 x tUs Whnical Wade 2.0 x 10-' alpha ison>et 1.8 x 1EF3 befa isomw 5.3x1Q-' Hexischlotod+bqtzoftirt 1.3 x 104 tiytbatirtPhtyd~aatte srtffale 4.9 x 10-3 Nidtei ref'urery dusi 24x10-4 Nir;kei Wsufflde 4.8x10-4 NiVoso-dime~tirle 1.4x10-2 Di-bt4yiamirte 1.6x10-' Dewftsamine 43x10-z N-rtitrosopysrolidMe 6.1 x 10-4 2-Tehactdaoetttane 1,1,1 7.4 x 10; , 22-Tetrachloroefharre 1 1 5.8x10-s , , Toxaphene 32x1o-4 1,1,2-Tricttloroefirte 1.6 x 10-5 Tridtioroethyfene 13 x 104 2,4,6-Trichloropt>enoi 5.7x10-6 Vinyl chloride 7.1 x 1Q-i `Eslimafed risk !o hunws from consW lifetime expme ol txeatMrg cordamirgted air at a IeveM ot 1 µryrrt3. APPL OCCUP. ENVIRON. HYG. 50 • AUGUST 19911 513

-8- could be harmful, prevailed - in the 40's and 50's, and I suspect prevail today. Now the question of the existence of a threshold in an individual is a problem in biology - the question of determining the range of possible susceptibility in a large population that is assumed to have individuals capable of responding is a problem in the statistics of sampling to which we now turn. We have to skip over (in the interests of time) a rich part of the biological science that was developed in the 40's and 50's to arrive at our principal focus, the idea of using risk assessment for setting safe population exposure levels for carcinogens, This idea was first pub- lishe:d, so far as I am aware, by Mantel and Bryan in 1961. They showed that a negative animal study - particularly a small one with 100 animals or so, does not necessarily demonstrate that the treatment was safe. Studies of feasible size can be used to establish directly only risks of the order of 1/100 or higher. When a study with 100 animals is negative, we can only claim that we are reasonably sure (assurance levell of 99%) that the true risk is no greater than 4.5 percent. (Slide-6) - Interpretation of Negative Studies. Mantel and Brya:n then proposed to rely on the dose-response principle and extrapolate from this upper limit conservatively to human exposure levels. They examined the dose-responses of many chemicals and concluded that most chemical responses involving lethality decreased very rapidly, with dose-response slopes steeper than.10 or more probits per ten-fold . dose dilution. They noted probit slopes for the therapeutic effects of antibiotics of 2-3, and still lower probit slopes, the order of 2, in

an B0Z excess risk of lung cancer (Doll, et al, 1972). It was easy enough to dismiss the corresponding risks on the grounds that the doses were minute, but one did not then (or now) assume for cancer the exist:ence of a threshold. So some form of quantitative relationship between the dose and the resulting incidence was needed. But in the absence of such a relationship, decisions had to be made and FDA banned carcinogens from food during the years well prior to the enactment of the Delaney Clause. Arnold Lehman, the chief toxicologist at FDA in the 40's, stated in an article in 1949 that: "a finding that a substance caused cancer in animals was regarded as so °alarming' as to exclude it from consideration." In 1945, the FDA banned Butter Yellow; in 1950 Dulcin and P-4000, two artificial sweeteners; in 1950 also tonka beans and coumarin; and in 1959, aminotriazol on cranberries, all on the grounds they were carcinogens and had no place in foods. The reasons that scientists were unwilling to assume the existence of thresholds for carcinogens are interesting - primarily because they were large7'.y theoretical. went like this: suggesting that mutation. This mutat:i on and on 0 was argued that Essentially the argument recast in modern terms Cancer is caused by agents known to be mutagenic at least one crucial, rate limiting step is a somatic focused attention on the nature of the genes that undergo the amount of chemical needed to affect that change. It only one molecule was necessary to produce a mutation in the DN'A within the nucleus of a cell. This in turn could lead to a

-19- these mutations are expected to be produced by "background carcinogens," not ,just aflatoxin. MTD The use of MTDs (Maximum Tolerated Doses) had been challenged throughout the period. Perry Gehring and Phil Watanabe had shown in 1976 that large doses could exceed metabolic and physiological thresholds, leading to prolonged retention in the body, formation of different metabolites and in some cases disproportionate increases in reactions between reactive electrophilic metabolites and macromolecules. They concluded that dose-dependent alterations in the fate of chemicals must be considered or at high doses you risk the likelihood of disproportionate increases in toxicity including carcinogenesis. They reported evidence of possible dose.-dependent effects-in styrene, ethylene glycol, aniline, carbon disu:li`ide, 2-naphthylamine, benzopyrene, bis-hydroxycoumarin, salicyl- amide, amphetamine and sulfobromophthalin (Gehring, et al) (1976). By the: late 1970's, enough bioassay data had accumulated largely owing to NCi and later NTP studies, to provide a sufficient basis to examine the results of the studies for correlations between the responses in rats and mice. In 1979, Crouch and Wilson examined the carcinogenic potencies for 70 chemicals in the two species. They demonstrated empirically that good correlations existed for.the potencies between the different species. This was an important finding, because if there were good interspecies correlations between potency estimates for rats and mice, then it was reasonable to believe that humans and animals might also be similar in their carcinogenic responses. But in 1985, Berstein, et al, reported

THE AMERICAN JOURNAL OF CANCER A Coatinastioa of The Jw=aal of Caacer Raaw& VG~L,UXE XVII FEBRUARYf 1fl33 NUMBER 2 SUGGESTED METHODS FOR THE STANDARDISATION OF THE CARCINOGENIC ACTItTITY OF DIFFERENT AGENTS FOR THE SKIN OF AtICE C. C. TWORT AxD J. M. TWORT (Frons the Laboratoriea of the 3Ia»cheatev Committee on Cancer) s ~

. o - Try to discourage media hype. Incessant coverage of the risk of real or suspect carcinogens - buoyed up by the exaggerated claims of QRA determined risks - makes it unnecessarily diffi- cult to get the public to appreciate the overwhelming impor- tance of smoking and the diet to cancer causation. o - Try to discourage the use of health warnings on trivial risks. It was absolutely appalling that for many years the health warning on saccharin in the U.S. was at least as strong as that on cigarettes. THAT'S NOT RISK COMMUNICATION! Cigarettes probably contribute some 150,000 deaths from cancer each year. Saccharin was banned by the FDA in 1977 on the grounds that it wasn't shown to be safe and on the Delaney Clause - not because it was known to produce cancer in humans. o - Finally, do some good research mechanistic work on cancer. . There are, I think, three areas to focus on: 1) Theoretical work 2) Work at the cellular level, biochemistry, oncogenes 3) Work in whole animals - (not MTD Testing) i. dosing regimens, effects of diet ii. effective dose studies and pkBP

F -14- QRA would be consistent with the Delaney Clause. They didn't think back then r:n terms of impurities and low level environmental contaminants - and if they had, I don't think they would have recommended high to low dose extrapolation. I believe this is true because of how they handled the second problem. If you tested a food additive in a carcinogen bioassay and the result was negative, the logical result of their analysis would require a downward extrapolation from the upper 90% confidence limit ala Mantel and Bryan. But again, because of the substantial amount of food additive required for a functional effect - typically at least several ppm, this extrapo- lation would ordinarily result in the ban of the additive at effective and useful doses. What did they do? They ignored their discussion on QRA, they ignored Mantel and Bryan and said that the sensible thing to do was to use a 100-fold safety factor! Their statement was that for agents not judged carcinogenic the use of QRA to estimate a safe dose would be logical, but would give a level so low as virtually to exclude from use agents for which there was no ~ positive evidence of carcinogenicity. And they wouldn't do it. This c:cmmonsense approach to the cancer problem was soon to be challenged by two related difficulties that defied easy solution. The first was how to deal with animal drug residues - those in food-producing animals as the result of ingestion of added drugs for prophylaxis, for the treatment of disease or for growth promotion. In 1962, Congress-had put yet another Delaney Clause in the Act with the Animal Drug Amendments - this

distribution of these and at least one of which would be below the critical value. Since cancer is believed to be of single cell origin, this one activated cell would initiate the cancer, Fi(Do) would be positive and the probability of response would be linear at low doses. They conclude "... in environments already containing appreciable amounts of carcino- genic processes, the effects of any slight addition to these processes will be proportional to the amount added. ... its implications are that much previous investigation of the form of the dose-response relationship at infinitesimal doses is irrelevant to the interpretation of animal studies for the formulation of social policy." It's hard to know what to say in the face of such confidence - for which there is no experimental data at all. These excathedra pronouncements are not believed by everyone but they continue to haunt some people including some in the regulatory agencies. The implications are, if this is trite, that dose-response curves become approximately linear just below the observable range - so long as they are roughly linear in the observ- able range. Not everyone believes this - I certainly don't. Alice Whittemore, Mel Andersen and other pharmacokineticists still believe that tumor probabilities are proportional to effective doses and these generate very non-linear dose-response curves. And I suppose that the folks in Kurt Harris's lab at NCI still feel they have accomplished somethang by finding mutations in p53, a putative tumor suppressor gene d in hua:~n hepatocellular carcinomas in China, despite the theory that says

-20- that the MTD's used in 186 NCI experiments were also highly correlated witpa potency. (Sl:tde 10) - MTD - Potency Correlation. Correlations between MTDs in rats and mice are not surprising because both species could respond similarly to high doses of different chemi- cals. However, the strong correlation between these MTDs and the derived carcinogenic potencies is startling. The correlation is surprising because MTDs are determined in a 90 day study and this time period has been regarded as too small a f raction of a rodent's lifetime to reflect the presence of a carcinogenic process - much less predict the strength of the carcinogen. Berstein and co-workers showed (Slide-11) that potency estimates from NCI Bioassays were restricted to an approximately 30-fold range surrounding ln(2). The TDsn is TD50 virtually the same as the MTD. They used a one-hit model to show this and an idealized 2-Group experimental design, but they and others have shown that this high correlation is not sensitive to "reasonable" depar- tures from either the experimental design or the extrapolation model used. Riethh and Starr (1989), and others since, have investigated these corre- lations in detail. It's clear now that: . o - The correlations between the MTDs and the estimated potencies are real. They do not depend on a "selected" data base. I

, 151 .3 -J P(D) .2- t ! ( l r i r ( . ~ 01 . ~ 0 ~ i P(U) 10) I Subst.mce No.10: SODIUM SACCHARIN Source: Taylor, ct al, Tozic. Appl. Pharmacul., 29, 154, Abstr. 200, 1974 Fig. 2a (fit'•Hil Armitdl;t'-Ih lf \Vt•ibull - Fig. 2b 10''-~ r T 10'3 f--T l(Yr 1 1)

n the life span, can be considered as the maximum tolerable dose for human beings. This, however, only in such cases, in which a com- plete exclusion from [the] human is not feasible." Some Conclusions I don't have the answers to these current scientific issues in risk assessment, but I do have some suggestions as to how we should behave about them. CANCER IS A VITAL HEALTH ISSUE -(Slide 15) - AND WE SHOULD TREAT IT SERIOUSLY AND DETERMINE WHERE THE REAL RISKS ARE. o -• Face up to the fact that, as we use Quantitative Risk Assess- ment today, it is justified almost entirely as a very prudent regulatory standard - if that's what we really want. It does not estimate risk and we will have to eapect that it won't for decades. o - Stop the codification of risk assessment "acceptable levels" and risk assessment methodology,in Federal Statutes. We are just creating other kinds of Delaney Clauses. o •- Try harder to examine some of the cancer mythology that under- lies our beliefs concerning thresholds, additivity and standard testing procedures for carcinogens.

-22- the 'basis of bioassay results only, is not credible. This is not to say that all high dose carcinogens are not carcinogenic at low doses. But the current NTP cancer bioassay data base with approximately 50% of the compounds testing positive contains many compounds that are probably not carcinogenic at low doses mixed in with many that are. The point is the test does not discriminate between them. Bac'k to Druckrey While Secretary Fleming was reading the MiDER Report to Congress in 1960, Hans Druckrey in Germany was preparing to publish a review of his life- Ionl; work on chemical carcinogenesis. He published it in 1966 in a rev:Cew article entitled: "Quantitative Aspects in Chemical Carcinogenesis." He ar.d his colleagues were very well recognized - Druckrey, Preusman, Schmahl, Nakayima and others were major contributors to the field of chemical carcinogenesis. Their work spanned 25 years and included the testing of over 100 different chemical compounds in some 10,000 rats. An example of his work on diethylnitrosamine is shown on the next slide. He administered daily doses in the drinking water until 50% of the animals in each group had liver tumors. The slide (Slide 13) shows that at the lowest daily dose rate (0.075 mg/kg/d) the cumulative dose re- quired to produce cancer in 50% of the animals was only 1/15 of that required at a daily dose of 14.2 mg/kg/d. 0

-23- This fact, that lower doses applied for a longer period could be more potent than larger doses of less duration was avidly learned and made a permacient feature of the lore of chemical carcinogenesis. But part of Druckrey's work seems to have been ignored. First, the price that is paid for a more efficient dose response is a longer induction time. And second, the cumulative doses involved are large, comparable to ingesting 1/10 of the MTD daily for a lifetime. This nitrosamine work was pub- lished in 1963 and, before that in 1961, Shubik had shown that not only were t:umors generally slower developing at low doses, they also were more benign. The Druckrey data are plotted in the next slide (Slide 14). You can see the :Lncrease in the induction time with'decreasing dosage. The data show very clearly that 0.075 mg/kg/d is close to a practical threshold based on the fact that the induction time required for the development of the tumors approaches the lifetime of the animals. Druckrey's rats didn't live much longer than 2z years or .t: 900 days -- and these days they don't live nearly as long. And he concluded that "With very low dosage the induction time can be longer than the life expectancy and that this is apparently a limiting factor in car- cinogenesis." He even had some regulatory advice for us: "'As a basis for future discussions it is proposed, that 1 per cent of the lowest dosage, which, given daily over the whole life span to ,ausceptible experimental animals, produces cancer only at the end of

-16- acceptable risk level. The One-hit model came from the one molecule theory and was easier to explain, but it wasn't a good fit for most data. The,re were other models. You may recall the Logit, the Weibull, the Mult.-ihit and the Gamma Multi Hit - all of them competing with the Mu1t:i-S_tage. None of these were based on biology. The critical steps and mechanisms in the development of tumours were and are still unknown. But: the multistage model had the best biological credentials, having first been used to explain the steep increase in the age adjusted rates of some cancers in humans by Armitage and Doll in 1961. And most people believed cancer was a multistage process, so there was a simulacrum of a biological basis. Since the early 80's, the strongly curvilinear models have virtually been abandoned. What was that? Well, first the low level risks that emerged from these different models were embarrassingly divergent. When the various models were applied to the risk of saccharin by t:b,e NAS in 1978, the risk estimates ranged over 5 million. (Slides 7 and 8)- Saccharin Risk Estimates (1978). If OM:B had been paying attention back then, they would have been exultant - this risk assessment certainly made the uncertainties in modelling crystal clear! Today our risk assessments don't differ very much. EPA ordinarily uses the MultistaRe with an algorithm that constrains it to be linear and FDA uses the Gaylor_Kode11 procedure for most carcinogens, which is designed to be linear. The other models could not be easily linearized and were abandoned. Since then our risk assessments have been 0 more nearly in agreement, more uniformly conservative and much less revealing of still unresolved uncertainties. The linear multistage

Table 2 Occupational Cancers "55~o"'A' '45b:4L rl977J UiP/9/Ns' BF/y!/rf9,4"/ efAw&:~-'e Co4o -~;,Q/svs fJA.~Bo~ l/oL ~A .4gent Occupation Site of cancer Ionizing radiations radon certain underground miners bronchus (uranium, fluorspar, hematite) X rays, xadium radiologists, radiographers skin radium luminous dial painters bone Ultraviolet light farmers, sailors skin Polycyclic hydrocarbons in chimney sweepers scrotum soot, tar„ oil manufacturers of coal gas skin many other groups of ex- bronchus posed industrial workers 2-Naphthyl amine; 1-naph- chemical workers; rubber bladder thyJam:tine workers; manufacturers Benzidine; 4-aminobiphenyl Asbestos Arsenic t Bis (chloromethyl) ether Benzene Mustard gas Vinyl chloride (Chrome ores) (Nickel ore) (Isopropyl oil) * * * Specific agent not identified. of coal gas ' chemical workers asbestos workers; shipyard and insulation workers sheep dip manufacturers; gold miners; some vine- yard workers and ore smelters makers of ion-exchange resins 1 workers with glues, varnishes, etc.'• poison gas makers PVC manufacturers chromate manufacturers nickel refiners isopropylene manufacturers hardwood furniture makers ~ leather workers bladder bronchus pleura and peritoneum skin and bronchus bronchus marrow (leukemia) bronchus; larynx; nasal sinuses liver (angiosarcoma) bronchus bronchus; nasal sinuses nasal sinuses nasal sinuses nasal sinuses `T

~TER /P fIDiYI/,u/STiPfJT/oN U~ S/.1/~SLE 1~OfGS df D,fTjf/ ZAo Z0 bv, ,w 2ov -5 / Ao /., ~-0 o,y,v ~M~y,~y1 1•S ~5-

Table '7••5. Everyday Cancer Risks a Annual Average Riskc Estimated Uncertaintyd Source Current Cancer Rai`esb r All Cancers 2.8 x 10-3 Buccal cavity, pharynx, respiratory 7.2 x 10-4 Digestive organs and peritoneum 7.5 x 10-4 Bone, connective tissue, skin, breast 3.1 x 10-` ^' 20% 1 Genital organs 3.2 x 10-4 Urinary tract 1.2 x 10-a Other 2.7 x 10-4 Leukemia and other blood and lymph 2.6 x 10-4 Cosmic Ray Risks° Airline pilot (50 hours per month at 12 kilometers altitude) 4 x 10-$ 2 One transcontinental round trip by air per year 10-6 2 Frequent airline passenger (4 hours per week flying) 10-$ i 2 Living in Colorado compared to New York Camping at 15,000 feet for 4 months per year 8 x 10-6 2 x 10-s ~ Factor of 3f 3 3 Other Radiation Risks Natural background radiation (sea level) 2.0 x 10-s / 3 Average diagnostic medical X-rays in the United States Living in masonry building rather than wood 2.0 x 10-5 5.0 x 10-6 {` 4 5 Eating and Drinking One 12h ounce diet drink per day 10-$ Average saccharin consumption in the United States 2.0 x 1Q 6 Four tablespoons peanut butter per day' One pint milk per dayi Miami or New Orleans drinking water charcoal broiled steak per week 1/2 lb 8.0 X 10-6 2.0 x lv 6 10-6 Factor of order 10 See text. . (cancer risk only; heart attack and other risks additional) 3.0 x 10-' ,,Icohol, averager< over smokers and nonsmokersg 5.0 x 10-s ~ Factor of See text. Alcohol, light drinker (one beer per day) g 2.0 x 10-5 order 10 Tobaccoh Smoker, cancer only 1.2 x 10-3 Factor Smoker, all effects (including heart disease) 3.0 x 10"3 of 3 See text. Person sharing room with smoker 10-s ~ Factor of 10 Air Pollution Polycyclic organics, all effects 1.5 x 10_s See text. See text. a. These are risks of death, the difference between incidence and mortality being well within the uncertainties shown, (except for the Current Cancer Rates category. b. Included to give some perspective. The figures given correspond approximately to the lifetime risk divided by the lifetime. The lifetime risk is estimated by the fraction of those dying who die of the given cancer, average lifetime is estimated as seventy years. Since cancer rates increase rapidly with age and the population age structure is changing, these figures are only approximate. Data from Vital Statistics of the United States, 1975. c. Averaged over the whole population of the United States. d. Even the uncertainties in these estimates can be very large. The uncertainties are mostly estimated subjectively and are conditional on the models used for extrapolation being approximately correct. e. Averaged over males and females. The risk is approximately double for females only. f. We assume a linear model with a total of 1 cancer per 5,000 man-rem, corresponding to BElR"1972. More recent estimates of the BEIR committee (1980) would give >lightly lower estimates. g. Cirrhosis of the liver. N'ot a cancer, but included here since the methods used are similar. It is possible that in this case there is a threshold effect for damage. In addition there is some evidence that moderate alcohol consumption is associated with lower death rates from other diseases. h. Based on human data. Based on human data for aflatoxin carcinogenicity. Note that we assume that the measured aflatoxins are aflatoxin B, the most potent. If .e corresponds to other atlatoxins, these estimated risks should be reduced. Sources• (The following references are the sources of data used in the models. We have estimated the risks). 1. U.S. Department of Health, Education and Welfare (1975). 2. United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) (1962). 3. Oakley (1972). 4.. United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) (1977). We have used the bone marrow dose here. 5. Moeller and Underhill (1976). ~ 00 tJ

Risk Assessment and Comparisons: An Introduction RICHARD WILSON AND E. A. C. CROUCH Risk assessment is presented as a way of examining risks so that they rnay be better avoided, reduced, or otherwise managed. Risk implies uncertainty, so that risk assess- ment is larg(.l;v concerned with uncertainty and hence with a concept of probability that is hard to grasp. The results of even the simplest risk assessments need to be compared with similar assessments of commonplace situ- ations to give them some meaning. We compare and contrast some risk estimates to display their similarities and differences,, VERY DAY INE TAKE RISKS AND AVOID OTHERS. IT STARTS AS soon as we wake up. One of us lives in an old house that had old wiring. Each time he turned.on the light, there was a small risk of electrocution. Every year about 200 people are electrocuted in tlie United States in accidents involving home wiring or appliances, repn:senting a risk of death of about 10-6 per year, or 7 x 10-' per iuf-time. To reduce this risk, he got the wiring replaced. When we-wallc downstairs, we recall that 7000 people die each vear in falls in U.S. homes. But most are over 65, so we pav little attention to c`Lis risk since both of us are younger than that. How should we go to work? Walking is probably safer than using a bicvcle, but would take five times as long and provide less healthful exercise. A car or, better, public transport would be both safer and faster. Expediencv wins out, and the car comes out of the garage. Fortunatelv, the c.hoice nowadays is not between horse or canoe- both of which are much more dangerous. The day has just begun, and already we are aware of several risks, and have made decisions about them. Most of us act semi-automatically to minimize our risks. We also expect society to rniaimize the risks suffered by its members, subject to overriding moral, economic, or other constraints. In some cases these constraints will dominate, in others there will be trade-offs between the values assigned to risks and the constraints. Risk assessments, except in the simplest of circumstances, are not de- signed for making judgments, but to illuminate them (1). To effectivelv illuminate, and then to minimize, risks requires knowing what they are and how big they are. This knowledge usually is gained through experience, and the essence of risk assessment is the application of this knowledge of past mistakes (and deliberate actions) in an attempt to prevent new mistakes in new siruadons. The results of risk a.ssessments will necessarily be in the form of an estimate of probabilities for various events, usually injurious. The goal in performing a risk assessment is to obtain such estimates, although we consider the major value in performing a risk assess- The authors are in the DcPartmcnt of Phvsics and the Energy and Environmental Poliey Center, Harvard Universm, Cambridge, lAA 02138. ment is the exercise itself, in which (ideally) all aspects of some action are explored. The results, goals, and values of performing the risk assessment must be sharply contrasted with the cultural values assigned to the results. Such cultural alues will presumably be factors influencing societal decisions and mav differ even for risk estimates that are identical in probabilitv. Risk and Uncertainty The concept of risk and the notion of uncertainty are closely related. We may say that the lifetime risk of cancer is 25%, meaning that approximately 25% of all people develop cancer in their lifetimes. Once an individual develops cancer, we can no longer talk about the risk of cancer, for it is a certaintv. Similarlv if a man lies dying after a car accident, the risk of his dving of cancer drops to near zero. Thus estimates of risks, insofar as thev are expressions of uncertainty, will change as knowledge improves. Different uncertainties appear in risk estimation in different ways (2). There is clearly a risk that an individual will be killed by a car if that person walks blindfolded across a crowded street. One part of this risk is stochastic; it depends on whether the individual steps off the curb at the precise moment that a car arrives. Another part of the risk might be systematic; it will depend on the nature of the fenders and other features of the car. Similarly, if two people are both heavy cigarette smokers, one may die of cancer and the other not; we cannot tell in advance. However there is a systematic difference in this respect between being, for instance, a heavy smoker and a gluttonous eater of peanut butter, which contains aflatoxin. Al- though aflatoxin is known to cause cancer (quite likely even in humans), the risk of cancer from eating peanut butter is much lower than that from smoking cigarettes. Exactly how much lower is uncertain, but it is possible to make estimates of how much lower and also to make estimates of how uncertain we are about the difference. Some estimates of uncertainties are subjective, with differences of opinion arising because there is a disagreement among those assessing the risks. Suppose one wishes to assess the risk (to humans) of some new chemical being introduced into the environ- ment, or of a new technology. Without any further information, all we can say about any measure of the risk is that it lies between zero and unity. Extreme opinions might be voiced; one person might say that we should initially assume a risk of unity, because we do not know that the chemical or technology is safe; another might take the opposite extreme, and argue that we should initially assume that there is zero risk, because'nothing has been proven dangerous. Here and elsewhere, we argue that it is the task of the risk assessor to use whatever information is available to obtain a number between zero and one for a risk estimate, with as much precision as possible, together with an estimate of the impre-ision. In this context, the statement "I do not know" can be viewed only as procrastination 17 APRIL 1987 ARTICLES 267

70 I 6 ,g ~; - *' RIETH AND STARR T i i 4 INVERSE MAXIMUM DOSE ---- MAXIMUM LIKELIHOOD ••••••• UPPER-BOUND 2 v Cl Y v - 6 f i 83 RAT CARCINOGENS N FIG. 4. Comparison of the inverse of the maximum dose tested with maximum likelihood and upper- ~OV bound potency estima;es of rat carcinogens.

-2- (7) Hearings before the Committee on Interstate and Foreign Commerce, House of Representatives, 86th Cong., 2d Sess. (1960). (8) Gehring, P.J., Blau, G.E., Wantanabel, P.G., Pharmacokinetic studies in evaluation of the toxicological and environmental hazard of chemicals. In Adv. in Modern Toxicology - New Concepts in Safety Evaluation, Hemisphere Publishing, Wash., D.C. (1976).

Application of Epidemiology Cole

C,~lD~GN, r.U/LSe~ 44 4 -r- I / / / / / + / ./ / , At3RERVATIC}NS, BUT p NOT MEABURABLE 2 3 -.. -S ^o4 •.-3 •g •J . 0 1 L09sc(s/yaxfmum Applted 8o.a) FIGURE 1. Lo;garithm of carcinogenic pp(pnq vorsNi laaatirithm of the Inverse of the~: appfied dose for female B6C3F1 mlco ff1 (ht NCI/NTP spriai of bloassays, Each point repewar. separate experiment. The dotted IinG Ii a ImWqUarQi fft to these points. . /b

~SlG ~/G/vT/F/e --CSS%~S l/t/ ~,aNG,El~ ~sk ~ss~ss,~l~.vT ~ -- Sl ~E oF flD~ilA.v G'A~t.GE/2 ~t'lS.~ e~R®~j Lvl.v ws,E ~x/oasv~Q,~s wse• ,Qr- ~o,vsC Fo~ C9~eiir~o6Evs ® - ~i~AD,~d~ E ~ovT~.r~Li-r fo,P ~ R ~i.Yo66.vs ® /NS~it~S'lT/v/7`/ BF r4s/1W1x71- 5'T ~ ~,r/STf1.~GE AN,o .tilEi9s~.~/3/~lTf/ e~ T~it'l'TfldGG~

Introduction to Background Materials Moeller

REFERENCES (1) Doll, R., Introduction in Origins of Human Cancer, Book A, Incidence of Cancer in Humans. Cold Spring Harbor Conferences on Cell Prolif- eration, Vol. 4 (1977) Edited by Hiatt, H.H., Watson, J.D. and Winsten, J.A. (2) Doll, R., Vessey, et al. The Mortality of gas-workers. Final report of a prospective study. Br.J.Ind.Med. 29, 394 (1972). (3) l?egg, Anthony E., Alkylation of Rat Liver DNA by Dimethyl- nitrosamine: Effect of Dosage on 06-Methylguanine, J. Nat'l Cancer Inst., 58, No. 3 (1977) 681-7. (4) Pegg, A.E., and Georgiani Hui, Formation and Subsequent Removal of 06-methylguanine from DNA in Rat Liver and kidney after small doses of DMN. Biochem. J. 173, 739-748 (1978). (5) Mantel, N. and Bryan, W.R., "Safety" Testing of Carcinogenic Agents, J. Nat'l Can. Inst. 27, No. 2, 455-49 (1961). (6) Weil, C. Statistics vs. Safety Factors and Scientific Judgment in the Evaluation of Safety for Man, Toxicol. and Applied Pharmacol. ~ Q 2I 454-463 (1972) . ~ , ~01

between this and providing a numerical estimate is important, but is one of presentation rather than substance. If there are no animal data, or if in an animal experiment there is no statistically siiplificant effect, it does not necessarily mean that there is no risk. If the experimenters have been diligent, the risk is probably small, ai:though never zero, even though that may be the best estimate. Various attempts are made to use data even less direct than the animal bioassavs to estimate risks in such cases. These include simple analogies based on chemical similarity (10), and comparison witC1 outcomes other than cancer-for example, muta- genesis (11) and acute toxicity (12, 13). Not surprisingly, these more indirect procedures arouse even more controversy than the animal bioassavs. There have been few attempts to perform risk assessments for biological end points other than cancer. However, it is known that the pollutants in cigarette smoke cause at least as many deaths through heart problems as bv cancer (14), and we should not be surprised if other carcinogens were to produce chronic effects other than cancer. For now, the cancer risk assessment has to act as surrogate for these other risks also. Risk Value Versus Certainty of Information After risks of a number of situations have been assessed, we often want to order them in order to decide which should command our attention. It is not alwavs the order of increasing risk that is used for such purposes. There have been proposals to order potential carcinogens on other factors (8, 15), such as the certainty of information. Vinyl chloride gas has been found to cause angiosarcomas both in people and in rats. Since an angiosarcoma is a rare tumor, the risk ratio (the ratio oEthe observed number of cancers in those exposed to the number expected by chance) is of order 100 or more in some cases. If an angiosarcoma is seen in a vinyl chloride worker, the attribution to vinvl chloride exposure is almost certain. On the other hand, the number af persons who have been heavily exposed to vinyl chloride is small, so that only about 125 angiosarcomas have been seen among vinyl chloride workers worldwide in the last 20 years. Now that exposure; in the workplace have been greatly reduced, no angiosarcomas attributable to recent occupational exposure have been seen. We do not know the dose-response relation, but it is generally believed that the response falls at least linearly as the exposure is reduced, so that no more than one cancer is expected in several years. We can compare this with the possible cancer incidence that was predicted by the Food and Drug Administration (FDA) in 1977 from use of saccharin (16). This was based on experiments with rats, leading to an additional uncertainty. More people ate saccharin than were exposed to vim11 chloride, and nearly 500 cancers per year were estimated for the United States alone. For vinyl chloride we therefore have the s:ituarion that the individual risk is now low, yet there is considerable certainty that there is a risk. For saccharin the risk is higher, but there is more uncertainty about the value of the risk. Some persons, in some situations, may demand that more attention be given. to the risk from vinyl chloride than to the risk from saccharin; for other persons or situations the reverse may be the case. Comparison of Risks The purpose of risk assessment is to be useful in making decisions about the hazards causing risks, and so it is important to gain some I'7 APRIL 1987 perspective about the meaning of the magnitude of the risk. Comparisons can be useful. We are not born with an instinctive feeling for what a risk of one in a million per lifetime means, although we do learn that some risks are small and others large. It is particularly helpful to compare risks that are calculated in a similar way. For example, the risk of traveling by automobile can be compared to that of traveling by horse with the use of historical data. Another common procedure is to compare exposures only. Table 1 shows a list of radiation exposures in typical situations (17). The dose-response relation for radiations with similar energy deposition per unit track length will be similar, although there may be some correction required for dose-rate effects, so that ordering by expo- sure should be similar to ordering by risk. In estimating the number of lethal cancers on a linear hypothesis, we have here assumed approximately 8000 man-rems per cancer (at low doses), in itself uncertain by 30% or more. , An example of comparison of risks that are similarly calculated is the'comparison of risks of various chlorinated hvdrocarbons in drinking water. The risks to humans are estimated from carcinogen bioassavs in rodents (rats and mice). Since these are similar materi- als, we might expect that the dose-response relationships have the same shape. Chloroform, which is produced by interaction of chlorine with organic matter during the chlorination of surface waters to kill bacteria, produces cancer in animals 20 times as readilv as does trichloroethvlene, an industrial solvent that is occasionallv found in well waters as a result of accidental pollution. Although neither is known to cause cancer in people, we might expect that chloroform would do so about 20 times as readily. Table 2 shows a variety of risks calculated in various ways and our estimate of the uncertainty. They are deliberately jumbled to provoke thought by juxtaposition. [Risk estimates quoted by the Environmental Protection Agency (EPA) for carcinogens tend to be greater than those shown in- Table 2 by a factor approximately equal to the uncertaintv factor-this is not accidental (5, 18).l Contrasting Risks Objections have been raised to risk comparisons on the ground that they are misleading. This would be true if all risks of the same numerical magnitude were treated in the same way. But they are not. In some cases it is useful to contrast risks to indicate the different ways in which they are treated in society. In Table 3 we give an example by comparing and contrasting the carcinogenic effects of aflatoxin B 1 and dioxin, both among the most carcinogenic chemi- cals known. The difference in treatment of these two materials is perhaps a reflection of different values assigned to various aspects of the problems caused by their presence. Aflatoxin and dioxin have similar toxicities and carcinogenic potency (perhaps within a factor of 10, although both measures for both chemicals van• substantially with species tested). The certainty of information for aflatoxin is great There is less information about carcinogenicitv of dioxin. Dioxin may be a promoter and pose a minuscule risk at low doses, whereas aflatoxin is almost certainly an initiator also. Nonetheless such standards as there are appear to be more stringent for dioxin, possibly because dioxin is an artificial chemical and possibly because it was a trace component of a chemical mixture (Agent Orange) that was used in warfare. The small risk of a large accident in a nuclear power plant can also be contrasted with the more numerous small accidents or events that occur every day in the mining, transport, and burning of coal. One feature that is brought out dearly here is that we do not always compare the risk averaged over time, but worry more about risks that are sharply peaked in time. ARTICLES 269 ,

'5~& ic~;- AL - - . . - • -- A -' ~ A, Th'~ 22-7 o 2 " .~1-12> TDo ~A~G.o /o 5 X 0 ~ ~S } ~ Z ~oLO ~.liEE ~`

/N c91~ ~ /~~~s ~ 'e'll i913 3 / 1Y d 'A~~Izz 4~v"onra ~®/ld Z 5~25' v~`~ ~ l 37 ~~- ,~.y 36~36 Z8'S ~3~ Z?.d 14 4/`!~ 2~ 3 3rS ~,,ls a~/3o ~T/ 609 tV o• 07S 5'7 ~ ~ ~ OD

- 5 - C. Issues - _A'lodel CissLCmes - - linear dose response - non-threshold - other 4 September 1991 Philip Cole, M.D. Austin H, Delzell E, Cole P: Benzene and leukemia: A review of thE: literature and a risk assessment. Am J Epidemiol 127: 419-439, 1988.

TABLE 4 ESTIMATED VIRTUAL SAFE DOSE (VSD) FOR FOUR MODELS FOR FOURTEEN SUBSTANCES SAGC9lRR~~ Estimated VSD at Risk Level 10-6 One-Hit Armitage-Doll Weibull Multi-Hit 2.0 x 10-s 1.9 x 10-4 .52 .80 3.4 x 10-5 7.9 x 10-4 4.0 x 10-2 .28 4.5 x 10-5 .35 .59 2.3 5.2 x 10-6 1.6x 10-3 1.7 x 10-3 3.8 x 10-3 3.2x10-s 1.9x10-2 1.9x10-z 7.7 x 10-2 2.0 x 10-2 2.0 x 10-2 2.1 x'10-9 3.9 x 10-10 2.1 x 10-4 2.2 x 10-4 2.6 x 10-4 2.6 x 10-4 8.4x10-$ 4.2x10-3 4.3x10-3 1•3x10-2 1.6x10-4 4.0 x 10-4 3.1x10-2 3.7 x 10-2 ---m 1 4.3 x 10-s .33 .53 1 1 I 5.5 x 10-4 4.5 6.0 . 33.5 5.7 x 10-6 2.2 x 10-5 1.2 x 10-3 6.7 x 10-3 2.8 x 10-4 6.4 x 10-4 1.7 x 10-2 4.9 x 10-2 3.7 x 10-s 5.7 x 10-5 1.1 x 10-3 3.8 x 10-3 /3a.4 ~.lf~1`i ~od.vcr~ a~,~~,oe.Py ~/v .~ /q8o~~o /y9

rors- aF ~NG.,-R ~w >f~E U .S: ~I98'7) ~17SD~ o -- c9sos 9d~, o" O ~®SP ~vDvGT/Y/`/ 91 4f ~oRTi9l,/Ty ~~ x 72 bl:wD,v /la 8~ rv2/o /6'

TABLE L_CPtemicaf Substances Classified as Carciragens by I4C6M1 witth Tfieir Respective TtVs (1988-1989 Adopted YaNye.s) Substance TLV I Subsfanca TLV Acrylamide-Skin" 0.03 rnqhrr' Ethylene dibromide-Skin - Acrylonitrile--Skin^ 4.5 mg/m3 Ethytene oxide" 1.8 mglm3 4-Aminodiphynyl-Skin e Formaldehyde" 1.5 mg m' Antimony triouide production Hexachforobutadiene-Skin" 0.21 mghrrj Arsenic trioxice production Hexamethyl phosphoramide-Skin - Asbestos Hydrazine-Skin 0.13 mg/m' Amosite 0.5 fiber/cc 4,4'-Methylene bis(2-chloroaniline)-Skin 0.22 mg/m3 Chrysotile 2 fibers/cc Methylene chloride (Dichloromethane)" 175 mg/rn3 CrocidoliCe 0.2 fiber/cc 4,4'-Methylene dianiline 0.81 nxyrrt3 Other fonns 2 fiberslcc Methyf hydrdzine---Skin 0.35 nxynr3 Benzene' 32 mg/m3 Methyl iodide-Skin 12 mg/n~ Benzidine-Sk:n B P-Naphthylarnine B Benzo(a)pyrene - Nicked sulfide roasting, furne & dust 1 mWm', as Ni BeryfliurM 0.002 mgm3 4-Nitrodiphenyt ® 1,3-Butadierk'A 22 rig(mrl 2-Nitropropane 35 mg(m3 Carbon tetrachloride-Skin" 31 mg/rn3 N-Nitrosodimethybmine--Skin Chloroform" 49 mcym3 N-Pherryl ~pt~r}2mir~e bis-(ChtoronMethyf)eW 0.005 mghrr3 Ptery4trydrazine-Skin 22 mgrm' Chlomiethyf r*fryf ethes - PrOpane sJlforle - Chromates of Is3d, as Cr 0.05 rtxyrtrj P-PrombcWt 1.5 mg+m' Chromite ore pnxessin0 (chromate) 0.05 m{ynP, as Cr Propylene imine-51dn 4.7 mgW Chromium (Y), certain water insoluble compouro 0.05 nxym;, as Cr o-Totidine-Sfdn - Chrysene - o-Toluidine--Skin 9 MgVm' Coal tar pitch velaCGfes 02 rrKym3, as bentene p-Toluidins---Skin 9 RWTP solubles Vinyl bromide 22 rnPP 3,3'-Dichforotenadine--Skin Ynyf dtrotide 13 mom' Dimethyl carlkunlyl chloride Vrryi *Wmerie 6WAe-_%n 57 mg/rn3 1,1-Dirnethyftdrazine--Skin 12 mg/rr-P Zinc dvoirraft 0.01mgim3,asCr Dimethyf sulfale--Skin 0.5 mg(nr3 _ Ploticce o( kftXW OWN)es ('Oor 1985-1999) Cadmium and canpocmds" 0.1 mgirna Etlry1 wY& 20 mglm3 Xytidine (nroW isorr~rs}- Skin 25 mglrts' "(3t~emiqls corKainEd on both Ux TLV and EPA rardnogen IisL BSubstance desigried by CS-TLV Cartmittee as a confimred txman caraaogen wftait a nV. Wa4ds oTosed bfia arbslarxe should be `popey equiav~d n vfrYaMy d;ma;* ap oPOMue •~ in two ways: 1) do the ACGIH and EPA place these chemicals in the same order of toxicity? and 2) what level of risk do tf, e EPA unit risk factors imply from exposure to the ACGIH's TLVs? The EPA dose-response assessment commonly begins with the multistage model, P(d) = I -- exp(qid + q2d2 + . .. + qkdk), puts an upjD<!r 95 percent confidence limit on the linear term of the dase-response (qIs) based on a sratistical eeal- uacion of anitnal bioassay data (with consideration of spe- cies, route of administration, duration of exposure and followup, and other experimental design criteria deemed most relevantt to human risk assessment), and then uses the linearized multistage model (only the linear term is included) to estimate the risk of lifetime exposure to low doses. Bera,uLq.e the linearized multistage model used by the EPA for its unit risk Pactor is equivalent to the single hit model, our estimate of lifetime risk of developing can- cer from occupational exposure is based on the model, Prob (d) = 1 - exp(-ad), where Prob(d) is the lifetime probability of developing cancer from exposure to a daily level of d µg/m3 during a working lifetime of a 40-hour workweek/168-hour week, a 50-week/workyear, a 40-year career, and an average life span of 74 years. The slope of this dose-response curve (a) directly indicates cancer risk, thus a larger slope im- plies a larger risk at the same dose. The slope is derived from the EPA unit risk factor and is adjusted as follows to reflect different exposure situations. To adjust for different exposure durations, we use the simple assumption ti= the dose-response slope for oc- cupational exposure is (40/168) x(50/52) x(40/ 74) = 0.124 of the complete lifetime exposure slope. The EPA assumes a normal respit:gion rate of 20 cubic meters in a 24-hour period white sm assume a r2te of 10 cubic meters in an 8-l~our avurking day. Ttyerefcxe, to adjust for different breathing rates for working and nonworking per- sons, we assume the occupational exposure slope is (10/8)/(20r24) = 1.5 times the EPAA slope. Figure 1 displays the comparison of the ACGIH and EPA arrangements of the 16 common agents in decreasing or- der of risk (increasing TLV level and decreasing unit risk order). The spearman rank eorrelation coefficient for these two orderings is r = 0.78 implying substantial, yet im- perfect, agreement. The major disagreements are the or- derings of hexachlorobutadiene,l,3-butadiene, vinyl chlo- ride, formaldehyde, 'and chloroform. The ACGIH has hexachlorobutadiene with a greater carcinogenic risk than 1,3-butadiene, and vinyl chloride with a greater risk than 512 APPL OCCt1P. &VNROAb M'6. SfN • AUBUSi i!!0

~ _ ~it/G/DZVGE wfAT0 OU,~ T"O B.9G~6~40G~r~ GiLP~/NOGE/ys iiYAes"O"Dr1yT o10 PsPi~r•9i~y c~Acr•v~'E~ 7`o S~IGrG,p~G~v C~,Q~.voGE,ys 2 . ~"y.G r~1GT w~Trs' ~E~`i~NiS/~s ~iv G®,r~~e N urifH T/~,Gr 1<d ) - ~; ~ 1'z d) 2: ° ~, " Fla'J f tgw- yy y.~R/fo `C FDR ~Dw a®SES ® ~ ~ ~ ~ rob

o_ ~~li g-,'Z elV'-, lf,1sk- w17~4 j~ o - _ ~305 o - ~~1fDY1 ~'L /LI!5-14 G

1168 ALICE S. WHITTEMORE 37. Ostennan-Golkar S. Ehrenberg L: Dosimetr~ of electophilic compounds by means of hemoglobin alkylation. Ann Res Public Health 4: 397-402, 1983 38. Holrr.quist ND, Detectton of unnary cancer with urinahsi: sediment. J Urol 123 1RR-189, 1980 39. Hollstein M. McCann J. Angelosanto FA, Nichols WW: short term tests for carcinogens and mutagens Siutat Res 65. 133-226. 1979 40. Doughert) RC. Whitaker M1, Tang S-Y et ol: Sperm dcnsit} and toxic suhstanetr a piuential key to environmental health hazards. In Emironmental Health Chemistr%. hlcKtnne} JD (Ed.). Ann Arbor. Mtch. Ann Arbor Science Publishers. 1981. pp 263- 278

Use of Animal & Other Data As Predictors of Human Risk Crouch

115~~~z C1r9391 RELATIVE POTENCY OF CARCINOGENIC COMPOUNDS TAnt.t: 1: C itrrinoRr)ric Cor))pnrurris :1 rrrtngtd i)r I)rscruding Order of Pore»ry 189 'nttttxiuncl \i. af tnic C :Ji.•e u•hcn tirst tuiuour :,piKar, \n. nf tu- m"ur, t'cr- CCnt- agr uf tu- mnur? (:\ ) Yapil- Iouta E pi- t hi•li- ttrna A.•cr- aQr latrnt Ix•ritK! (li) ndcz (.\ l H X 100) t I. 9: 10-Uiruet hy1-1 : 2-henzartt hracene 20 13 65 6 r 43 151 2. \leth)•tcholanthrenc (a) 18 18 100 1 17 99 101 3. \lethyicht,l:)nthrcnc (b) 8 S 62.5 () S lit 41 4. Meth.•icht)lanthrene (a antl h ncltletl tu- );ether) 26 23 :+:1.5 1 2-7 109 Xt) 5. 3 : -1-Iienzp)•rene (from pitch) 1t) 10 1(H1 2 x 127 I 79 6. 3 : -1-tienzln•renc (s)')tt het ic) 9 i ii•i 2 5 109 72 7. 3:-1-I1enzpyrene (5 antl 6 added tt)- t;ethcr) i 19 17 h`.5 4 13 119 f I ia 8. C'ht)l;tnthrcne 49 28 23 112 51 9. S: 6-rvr-Ird'ente nu-1 : 2-benr.anthracene 14 I 1.3 93 12 194 .}K 10. 2-Met h}•1-3 : 4-hcnzphenanthrene 16 12 7 5 S 7 155 48 11. 1 Q•AIet hyl-I : 2-benrant hracene 18 12 663 ? 10 147 .)S 12. 5 : 6-I)intethy1-1 : 2-bcnzanthrncenc 19 16 8-1 () 16 220 38 13. 6-isol'rc)tty1-1 : 2-I>enz.) nt hracene 15 11 73.5 1 10 201 36 14. 3 : -1 : S : 6-1)ibcn•r.carbar.nle 19 9 -1 7.5 4 5 143 33 15. 3 : -1 : 8 : 9-I )ihenzp)•rene 17 10 5n a 10 205 29 16. 5-Met hti•1-1 : 2-hen•r.anthrncene 8 7 :ii.5 ? 5 317 29 17. 5-1:thyI-1 : 2-I>enzstnthracene 9 7 77.5 ) 5 285 27 18. 1 : 2: J: 6-1 ))1cr17.ant hracene 65 41 63 33 239 26 19. 3 : -d-iier)it)henanthrene 18 12 67 5 7 387 17 20. 1 :2 : 5: 6- I) i 1)enzca r ba •r.t)1 e 9 4 -14.3 i 3 263 17 21. 5-)r- I'rt)t))•1-1 : 2-henzant hra cene 20 6 30 3 3 192 16 3: 4 : S : t5-1)il)enz:u•ririine 22 28 11 3 > 9 ` 35'r 11 . 23. 3'-\iethyt-1 : 2 : 5: 6-tlil>enza)tthracene 25 ~ 7 ? ~. 6 I 325 I t) 1 : 2 5: b-I )ihenr:)cricfinc 24 2, 6 24 4 ~ 35 U ~ 7 . . ' I _ TtT7 .V.s i - f ~ .in5 ~ 60 2-} 5 i ) t is still the possibility of an error due to the fact that a number of aninials maN (lie soon ttftt,r the tirst tun1()ttrs are st'rn and before the majority have appeared 0 'I'heoreticallv it would be an advantage in ubtainin; a (luantitutivc' COill- {)ari"on of the p()tency of several compounds if all the experiments were carrit•d Out on I)urv-line mice under as nectrly u; possible the same conditions. Thi~ «•t)tt1(I°re(luct• the v<lriatit)n between the batches of animals used for different CUnI()t)ttn(IS, but the results of the comparison wt)u1d then apply UnI1' tt) th;lt

2 C. Relative incidence rate (relative risk, RR) R?q= the incidence rate in an exposed group divided by that in a non-exposed group example: the RI of leukemia among rubber workers is 4.5 (base = 1.0) D. Standardized mortality ratio SMR = the number of deaths observed (usually in an occupational group) divided by the number of deaths expected example: among pliofilm workers the SMR is 337 (base = 100) V. Study designs - general A. Descriptive studies aF, `1~ »cl~vrcluc~ ~'~0c`~ ~ejn-c~ u s~ud~~ C~rrela E~,~Q 5~~d~es o_qroucto 1.3 B. Follow-up (cohort) studies a. prospective is b. retrospective + C. Case-control- lcss «,j2,,4,,+ ~~- R A- s elec~n- °-6 \DXI,tr) w;-~h PeePz, wO"rt ar'd wi1'hoLd' drstLLse 14 de1ermrne exFb2U'u_ D. Proportional mortality ratio (PMR) VI. Study designs - specific Limitations: A. The retrospective follow-up design Example: 1165 rubber hydrochloride (pliofilm) workers followed-up from 1950-81 experi- enced 9 deaths from leukemia with 2.7 expected, an SMR of 337 Advantages: fast, inexpensive (- -0 exposure based profile of effects (all causes of death) relatively free of bias ( s'{6ic,omct~l~ cc-rcr~ inadequate exposure-possible (jn~a inadequate exposure documentation - usual ~ prone to chance•- 0 prone to confounding N C11

J CZUa+ CHs Vo!, 39, No 12. pp 1157-1168, 1986 0021 -9681,86 53.00 + 0.00 Printed in Great Sntatn. All rights reserved _ Copyrtght f 1986 Pergamon Journals Ltd EPIDEMIOLOGY IN RISK ASSESSMENT FOR REGULATORY POLICY ALICE S. WHITTEMORE Department of Family. Community and Preventive Medicine. Stanford University School of Medicine, Stanford, CA 94205, U.S.A. "The uncharted galaxies of epidemiology are numerous" Lilienfeld et al. [11] I. INTRODUCTION THE Tw,- rrTIETH century has seen the rapid evolution of many new fields concerned~with protecting public health. Epidemiology and risk assessment have several of the features common,-to these. new fields, and important differences. Both are needed to make the difficult decisions required in setting standards for levels of toxic agents in the workplace and environment. They differ in their aims, orientation, and time scale. According to Lilienfeld and Lilienfeld [2], epidemiology is "the study of the distribution of a disease or a physiological condition in human populations and of the factors tha.t influena: this distribution" (italics added). By contrast, health risk assessment denotes research and evaluation to characterize the probability of physical harm to humans attributable to a particular agent or group oj agents. While the distribution of disease provide;, the focus for epidemiologic research, concern for adverse effects of specific toxicants drives risk assessment. Moreover, while epidemiologic studies proceed at the, glacier-liike pace needed to mobilize large staffs of support personnel and to monitor large populations over long periods of time, risk assessment activities acquire the urgency felt by regulators, who must make decisions (including decisions to postpone decisions) today. Most important; while epidemiology is a scientific field that draws upon medicine, demography, and statistics, risk assessment is a hybrid of science and policy that draws not only upon fields such as epidemiology, toxicology, chemistry and engineering, but also upon psy<:hology, politics, economics, law and social justice. These i nherent differences in emphasis, timing, and nature complicate the role played by epidemiology in risk'assessment for regulatory policy. In 1985, this role is still largely one of epidemiology's uncharted galaxies. In the sections below, I review the role's history, and the reasons why it will continue to play an essential part in regulatory decision- making. The role has placed epidemiologic findings and epidemiologists at the center of political controversies, and I discuss. the positive and negative side effects of this new visibility. :Finally, I explore ways to prevent the negative side effects and ways to increase the utiSity oT epidemiologic data for regulatory risk assessment. II. THE ROLE OF EPIDEMIOLOGY IN RISK ASSESSMENT Concern about industrially related contaminants in our air, water, and food began gathering momentum shortly before World War II, and accelerated with the publication in 1963 of Rachel Carson's book, The Silent Spring. The spectre she painted of man's i 1157

Risk Analysis in Environmental and Occupational Health Use of animal and other data as predictors of human risk Edmund Crouch 1 Background Information ............................................. 2 2 Known Human Carcinogens .......................................... 3 3 4 5 Target Risks. The Necessity of Extrapolation .............................. The Nature of Carcinogenesis ......................................... TGiEa Standard Anima! Test................. ........................... 5 6 9 6 Raw Results - and what to do with them . ................................ 10 7 The Two Major Extrapolations ........................................ 14 8 Interspecies Comparison - Constant Relative Potency ....................... 17 9 Interspecies comparisons - practical and theoretical ......................... 18 An example - 1,2 Dibromoethane ................................. ....... 19 10

Site or Site or Mateiial/Action type of Material/Action type of tumor tumor 4-Aminobiphenyl Arsenic (compounds) Aurarnine manufacture Asbestos BenzidIne BCME Chiormaphazine Bladder CCME Lung Cyclophosphamide Chromium (VI compounds) 2-Naphthylamine Mustard gas Nickel refining Arsenic Benzene PUVA Skin Myleran Leukemia Soots„ Tars, Minera9 oils Chiormabucil Meiphalan DES (In utero) Vagina Vinyl Chloride Liver The "naf;ural" rates for these cancers, expressed in terms of lifetime risk and annual average risk, are shown in the following table. S6te or type of tumor Lifetime Risk Annual Average (In ABSENCE of exposure) Bladder 5 x 10-1 7 x 10-5 Lung (Pop". ave.) 4 x 10-2 6 x 10-4 Skin (deaths) 3 x 10-' 4 x 10"5 Liver 1 x 10-3 2 x 10*5 Vagina 7 x 10-3 9 x 10-5 Leukemia 8 x 10"3 1 x 10-4 4

1164 ALICE S. WHITTEMORE year, and seeks to determine the major causes of those actual deaths. Such a perspective is much more likely to overlook a large number of small effects of various chemicals than laboratory science might be, but it is much less likel~ to overlook the chief determinants of current mortality rates-2nd trends, especially if these are not simple direct effects of individual chemicals on molecular DNA". Quality control and data analysis also are complicated by the political climate surrounding many studies of environmentally induced disease. The possibility of subjective reporting bias is increased, causing greater need for exposure and outcome validation (20]. For subjective disease assessments such as miscarriages and asthma attacks, there is need for difficult and expensive validation of negative outcomes among both exposed and unexposed populations. Political pressures have their largest impact on the interpretation of epidemiologic data. Pressure to provide "bottom lines" produces quantitative risk estimates with spurious precision, numbers that, out of context, take on a life of their own. Such numbers are overinterpreted by laymen who expect a study to produce unequivocal answers, and when it does not, who criticize epidemiology for failing to achieve aims that go beyond available resources or methodologic capabilities. Perhaps the most troubling impact of risk assessment activities concerns their side• effects for the epidemiologist. He has joined the ranks of psychiatrists, statisticians, and clinicians who take the stand as expert witnesseses in multimillion dollar lawsuits. While this activity •helps keep bread on the table, one worries about the conflict between the one-sidedness of such an advoc4cy position and all of one's training to strive for a balanced perspective in weighing the strengths and limitations of a data set and placing it in the broader context of other data. Apart from the monetary inducements to take a unilateral view, there can also be pressures from peers and employers. Espousal of unpopular views may cost an epidemiologist invitations to conferences, permission by an employer to attend conferences [21], favorable reviews of papers, or even a job [22]. These hazards of course are not quite unique to the epidemiologist, but are shared by all those in the environmental health sciences whose work impinges on risk assessment for regulatory policy. Equally worrisome is the tendency for political and philosophical differences to masquerade as scientific disputes. By now we have become inured to the familiar spectacle of government and industry epidemiologists aligning themselves in predictable camps in hassles over such issues as the incidence of brain tumors in the petrochemical industry [23], the fraction of U.S. cancer deaths attributable to occupational exposures [24], and the toxic importance of lead in automobile exhaust relative to that of lead in paint [251. A second manifestation of this masquerade is the overkill in critiques of completed studies whose results have undesirable implications for the interests of one or another faction in a regulatory issue. While constructive peer review is a useful process, critiques that exaggerate a study's flaws and overlook its strengths for the purpose of discrediting its conclusions are counterproductive and a poor use of resources [26]. One can look back in history for more subtle and therefore perhaps more disturbing examples of how values influence scientific conclusions. Samuel George Morton was a 19th century self-styled "objective empiricist" who used his extensive collection of human skulls to study racial differences in cranial capacity, a putative marker for intelligence. His findings supported contemporary caucasian beliefs: whites above indians, blacks at the bottom. Stephen Jay Gould reanalyzed Morton's meticulously recorded raw data, and found a fabric of apparently unconscious manipulations in the form of errors, mis- calculations and omissions, all in favor of white supremacy [27]. Gould notes that unconscious or dimly perceived finagling is probably endemic in science, since scientists are human beings rooted in political and culturar contexts. This example serves as a sobering reminder that reporting and interpreting one's data can require soul-searching, ruthless honesty, and courage.

- 4 - C. Specific individual case - ~or CC.-ses relevant exposure absence of alternative cause IX. Benzene and leukemia - A model risk assessment A. Basic science not genotoxic n 'c- damages chromosomes -~nv'- ciec- hcd ~-nec i anlsy~ B. Animal studies carcinogenic leukemogenicity problematic 14 p~ C. Epidemiologic studies l'7 s1;eg ~~~\,generally positive for AML -r\ecc-~,Ue~ S~~zs ~G; ~poor quantification of exposure some potential confounding - o{-t,,r so1,;{rrt D. Epidemiologic data* - observed deaths : CL 19 - expected deaths: 9.6 - total deaths: 1273 - mean cum. exposure: 42 ppm-yrs E. Risk assessment - excess deaths: 19-9.6 = 9.4 - excess deaths/1000: 9. 4/1 .273 = 7. 4/1000 -~~1'~'~~ - baseline risk: 7/1000 - doubling dose: - ~ (14/14.7)(42 ppm-yrs) = 40 ppm-yrs ~'rn c~es ~ ,~eed ~ be cxposed 4-0 do«b!e AhL P ts iC X. The OSHA standard ~~G Pp-ft.U A. For many years = 10 ppm 8 hr TWA 30 yrs x 10 ppm = 300 ppm-yrs ti 7 doublings = 800/1000 = unacceptable 1.7 additions ti 56 deaths/1000 B. Currently =IL ppm 8 hr TWA '3 ` 30 ppm yrs ti 1.75 baseline ti 5 excess deaths/1000 exposed

ANIMAL CARCINOGEN => HUMAN CARCINOGEN and to work from here. This assumption is not unreasonable, in view of what is known about carcina,;Ienesis - although it is something which can be argued about in specific cases. It is also well to be aware of the phrase emphasized - "under suitable conditions". While it may be true that animal carcinogens are indeed human carcinogens, the conditions of exposure of humans may typically be very different from the conditions under which the material is carcina,ienic to animals. It may be that under the conditions of human exposure, the material is not carcinogenic in animals or humans. 4 The Nature of Carcinogenesis. In what follows, it is useful to keep in mind some information about the process of carcinocenesis. This information has been derived from studies of humans and animals, and from experiments performed in vivo or in vitro. It is based partly on experimental studies, and partly on 1:heoretical ideas suggested by those studies. • (;ancers arise from one (or more) individual cell(s) which have gone "out of control" in some way - the cell becomes immortal, with no limit on the number of cell divisions, and the usual constraints on cell division no longer apply. A cell may pass through several stages before reaching this state. The underlying cause of such behavior is probably some effect(s) on the genetic materiai o# the cell, but the exact mechanism(s) is (are) unknown. • The occurrence of such events appears to be a random process at some level. One cannot tell which individual cell or animal or person will be affected. Hence we talk about the PROBABILITIES of cancer - the chance that some event will occur. • When we feed materials to experimental animals, the probability for cancer depend on various factors which can be manipulated. For example, the probability varies with: The total AMOUNT of material (the total dose) The AGE at which dosing takes place The RATE OF APPLICATION, or the time over which dosing continues OTHER FACTORS (some known - stress, dietary factors, ..., others unknown) We therefore expect, and in practice observe, DOSE-RESPONSE curves. Such dose-response curves are fundamental in extrapolating risks to humans. I like to draw an analogy to the similar problem of extrapolation which arises for acute toxicity - in both 6

The simplest sort of analysis can be performed if all the animals survived for the whole length of the expariment - and in practice the same sort of analysis is performed provided a reasonable fraction survived that long and provided there were not too many early deaths. In that case, we can simpVy list the dose groups and the numbers of animals with tumors compared with the total number of animals examined; for example: Dose Number with Tumor Number Examined 0 (control) 10 50 0.5 x MTD 25 50 MTD 30 50 However, things are not usually this simple. Similar results are available for Many different sites a Many different tumor types Combinations of these as will be seen in the examples to follow. To determine whether the rate of cancer has been increased involves comparing the proportion with tumor in the control group with the proportion with tumor in the dosed groups, and deciding whether there is a significant increase in any dosed group(s). The choice of which sites and/or types of tumors to combine before performing such statistical tests can be difficult. Generally, various grades of tumors (nodules, adenomas, carcinomcas) may be combined for any given site. In addition to the simple numbers of animals with tumor, there is additional information available which mai y be used in more complicated cases. The date of death of each animal is recorded, and may be taken into account in time-adjusted analyses of tumor incidence and in the life-table tests meftfioned on the appended material. . IU O N Cst ~11 ~ ~ ~ 11 ~

are here averaging over a lifetime - the figure is not meant to imply that the risk is the same in each year of life - we have just seen that it varies drastically with age. When discussing the risks of carcinogens, the same caveats have to be borne in mind. We usually attempt to estimate a lifetime risk but may express this, for comparison purposes, as an annual average risk. For an individual exposed continuously to a carcinogen, we would expect that the risk of cancer increases with age in a fashion similar to the risk of other (naturally occurring) cancers. There is another reason also for quoting an annual average risk obtained by averaging over a lifetime. When estimating risks of carcinogens, one is often interested in the response of a population to exposure to the carcinogen. In this case, one should strictly (if it were possible) estimate what the effects at all future times would be on individuals of different ages at the times of exposure. The effects at all future times on the whole population would then be an average over the effects on all the individuals in the population (who were of different ages at the times of exposure. Thus, to obtain an estimate of the effects on a population, one implicitly performs an average over the age groups present in the population. If the population were stationary (and if certain other conditions were fulfilled) this average would be the same as an average over a lifetime. This explains the usefulness of a lifetime average, since one may argue that the differences between population and lifetime averages are small compared with other uncertainties inherent in all the procedures we will describe later. The preceding discussion must be considered only a heuristic argument for accepting a lifetime average as being useful. In practice, people will be exposed at different ages, and for varying periods, ito different amounts of carcinogens. AII these differences (and many more besides) will affect the probability of carcinogenesis for each of them. 2 KnOwn Human Carcinogens There is now good evidence that human exposure to certain materials can, under certain conditions), increase the rate of human cancer. The evidence comes from various types of epidemiological investigation (discussed in other talks in this course). In all cases, exposures to these materials has been high, compared with population exposures, and the population exposed has been small compared with the total U.S. population. The resultant risks to those exposed has been substantial. The following table indicates a few of these materials, and the types of cancer which have been caused in humans by exposure to them. 3

1162 ALICE S. WHITTEMORE SKIN ORAL LUNG B FEMALES PANCREAS COLON & RECTUM PROSTATE BREAST URINARY LEUKEMIA & LYMPHOMA OVARY UTERUS ALL OTHER 30 20 10 PERCENT 0 I t t t 10 20 30 PERCENT FtG. 6. Estimated percentage of all incident cancers occurring by site of origin in United States males or females in 1985, excluding nonmelanoma skin cancer and carcinoma in siru. (Source l13].) private health resources so as to avoid spending disproportionate sums of money on minor hazards, while neglec'ting major ones. Figure 6 shows estimates of the percentage of all cancers diagnosed in the U.S. in 1985 occurring among the major sites, for men and women separately. Among men, cancers of the lung, large intestine and prostate account for about 56% of all new cancers (and 57% of all cancer deaths). Among women, cancers of the lung, large intestine and breast comprise 52% of all new cancers (and 51 % of all cancer deaths). Table 7 shows that occupational and environmental factors do not play an appreciable role in the etiology of these major causes of morbidity and mortality, except for lung cancer. Moreover, the contribution to lung cancer is dwarfed by that of tobacco, which has been estimated to account for 91 and 79% of lung cancer deaths among U.S. men and women, respectively [14]. (The sum of the percentages for males exceeds 100% because of the multifactorial etiology of lung cancer.) To date, we have made slow progress in preventing cancers of the breast, prostate, and large intestine, which are more likely to TAllLE 7. ESTIMATED PERCE!vTAGES OF TfiE 1WOR S7TE-SPECIFIC CANCERS ATTRIBL7ABLE TO OCCt:- MTIONAL AND ENV7RONMENTAL FACTORS' Mates Females Lung 15 5 Colon and rectum 2 t Prostate <t Breast 0 •Source [14). TABLE 8. BIOLOGICAL MARKERS FOR ENVtROKNEN-TAL E%POSLItF_S Marker Specfinen Methodology Ref's Chromosome aberrations Blood lymphocytes, Autoradiograph): (breaks, n-arrangements, erythrocytes in bone phytohemagglutinin sister chromatid exchanges) marrow stimulation o! 1}mphocytes (33.341 Micronuclei Erythrocytes in bone Microscopic eiamination [35) marrow Covalent binding to DNA Blood lymphocytes, tissue Radioactive labefing; explants immunoassavs; indirect immunofluoreicenR microscopy [361 Co%alent binding to cellular Hemoglobin Chromatograph. (3'] proteins Cellular atypia Sputum. cervical Microscopic ezam nation 1381 epithelium Mutagens Urine, feces, cervical Ames salmonella test (391 secretions, breast flui3f' Sperm abnormalities Semen 140]

Typically, in epidemiological studies, a relative risk of more than 2 is required in order to detect any effecf:. Thus the (epidemiologically) discoverable population average human risks are > 10-5 per year, or 10-3 per lifetime, and probably much larger. For the small subgroups of the population usually available for study, the observable risks are generally much larger. For example, in the groups of workers exposed to vinyl chloride, the relative risk for angiosarcoma of the liver was huge, mainly because angiosarcoma of the liver is such a rare disease. Had vinyl chloride caused a more common tumor of the liver, it is quite likely that the association with vinyl chloride exposure would have been missed. In animals, vinyl chloride induces other tumors at a greater rate than angiosarcomas (although it also induces them), and current quantitative risk assessments are based on these other tumor types. 3 Target Risks. The Necessity of Extrapolation. When considering the size of acceptable risks to the public at large, the usual targets are much smaller than the discoverable risks discussed above. Typically they will be less than 10 per year. Note that the EPA and the FDA set targets of order 10 to 10-4 per lifetime, that is, of order 10 to 10' per year. It must also be borne in mind that there are a large number of materials which are of potential interest. The Chemical Abstracts Service (CAS) has now given names to well over six million distinct chemicals which have been mentioned in scientific literature, and there have been various estimates of the number (around 50,000) of chemicals in general commercial use. With such numbers, it should be immediately apparent that there are just too many time, money and logistical constraints to directly detecting any adverse effects from such a plethora of materials to which humans may be exposed. Notice that a risk of 10-' per lifetime corresponds to a rate of aloout 3 per year in the whole U.S. population. Thus, even if the whole U.S. population were exposed to some material causing a risk of death of 10-' per lifetime, the resulting deaths would be statistically indistinguishable in the usual two million deaths per year (unless there were something extremely unusual about the deaths). Extrapolation is therefore essential in order to estimate the sizes of risks, and hence be in a position to demand that risks be reduced to the levels mentioned. The fundamental observation on which such extrapolation is based is that: HUMAN CARCINOGEN =#, ANIMAL CARCINOGEN In other words, every known material which has been shown to be a human carcinogen is also known to cause tumors in animals under suitable conditions. This observation is not very useful in itself, bu1, what is done in order to allow risk assessments is to assume its converse: 5

TOOLS OF RISK ANALYSIS Applications of Epidemiology I. Overview II. Risk assessment epidemiology A. Definition: a description of the change in the incidence rate of a disease due to a known change in the level of exposure to a cause B. Purposes: - guide public health policies - guide the regulatory process __~ assist in tort resolution C. Foundations • - basic science - ~Q~e ~ ~~~e~ ~ - animal studies `dcseribe eney D. Growing importance of epidemiology : - 0-5u-3~3 tkie- ~-'advances in methodology a tj~-F IcLat. !~ reduced reliance on animal research CSF~_c~ts oaU~~~ bases in law ~'nlmciCs hcwe c~ stnc~ l~ expo:~~ o rx~ ~hC_ destre_ ~~rc&~ e. o rz oS{'ec"E- III. Epidemiology - general A. Definitions: the study of the distribution and determinants of disease in man an observational science dealing with the environmental causes of diseases of human beings B. Strengths - human beings - human lifestyles C. Limitations - non-experimental - often qualitative IV. Selected measures A. Incidence rate I = new cases/(population x time) ~ example: the incidence rate of leukemia is ~ i k 10.1 cases per 100,000 person-years N ~ B. R s R = new cases/population C11 L example: the lifetime risk of developing leukemia is 700 per 100,000 * lk C11 ~ persons, or 0.7% ~ ;A

- 3 - B. The case-control design 0~'4cr) ct6cd -A- 1?14 ~ o~ ~f s r~ ereal -a) 0 J- (J ~i cJ }~reer s~ Example: 138 adults with leukemia, resident in Olmsted County M, were compared with 276 adults without leukemia. Informa- tion on benzene exposure was abstracted from medical records. Among persons with benzene exposure, the RI of leuke- mia was 3.3 compared to persons without exposure. Advantages : fast 6 s profile of exposures control confounding precise (not prone to chance) suitable for rare disease Limitations: single disease only relative measures cp juests prone to bias •- c/rffrcc,~ -1-o fi)+-ove-. 7"h-sa.m e 2s e, e,~,zcs VI I. Interpretations I0-Au e no e, oczf-ccrn e- A. Chance B. Bias co/f"h fooss161c qn -phc~~e i u C. Confounding ~ qoJ-her- D. Valid d 15 eQ.,SeS causal null Comment: not mutually exclusive not permanent VIII. Causality A. Individual study strength internal consistency biological credibility B. Abstract, general case external consistency response to manipulation

5 The Standartl Animal Test The requirements for a"standard" animal test are quite severe. The animals involved have to be as similar to humans as possible - in metabolism, in being omnivorous, in their sensitivity to chemicals, for example - yet as different as possible in their life span and cost of upkeep (so that we can get resulis in a reasonable time at a reasonable cost). In practice, there is little option but to use standard laboratory animals. The usual choices are rodents - rats and mice; with oG,&sional tests being performed on golden hamsters or guinea pigs. Other animals (e.g. gerbils) have been proposed, but for now the experience built up in handling laboratory rodents is a strong I`.ncentive for continuing their use despite certain known disadvantages. Any change would no'N have to be done gradually, and with much cross checking with previous results. It is now standard to require tests to be performed in at least two species (practically always rats and mice,) and on both sexes, in case one or the other species or sex is peculiarly resistant to the material under test. A compromise has to be made over the number of animals to test. It would be desirable to have as many as logistically possible, to increase the statistical sensitivity of the experiment; but as few as possible to minimize the costs of testing (since there is always another material to test). The current recommendation is for at least 50 per group of similarly treated animals. There is a similar trade-off between costs and the number of dose levels to test in a given experiment. The current recommendation is to have at least three, preferably four or more, dose groups -- an undosed group (the conf•-` group), a group tested at the maximum tolerated dose (MTD) of lfte material under test, and third group tested at some intermediate dose (usually 1/4 to 1/2 of the MTD). The MTD of a material is roughly defined to be as much as possible, but not enough to kill off the animals e-arly or to cause too large other overt effects (like loss of weight). The reason for using it in these. experiments is to increase the sensitivity, on the basis that giving more of something is more likely to produce a response if any response if going to happen at all. The sensitivity has to be as high as possible, since the observable responses are of the order 10-' (10%) while the risks of interest are of order 10 (100,000 times smaller). The alternative way of increasing sensitivity is to increase the number of animals tested (within reason), but this only increases sensitivity in proportion to the square root of the numbers tested, while increasing the dose gives an increase in sensitivity roughly proportional to the dose. Clearly the latter is most cost effective. Even with such a minimum design, there are: 3 dose groups x 2 sexes x 2 species x 50 animals per group 9

For risk assessment purposes, it is necessary to make various assumptions about the behavior of animals in experiments like these. For example, it is assumed that: • Animals are affected independently (a tumor in one animal has no effect on any other animal). • Animals are equally likely to be affected • Each animal receives the same dose and so forth. It is assumed that cage effects, littermate effects, the effects of heating, lighting, stress etc. are either not present, or are randomized among all the animals in such a way that there will be no effect on 1he final analysis. With such assumptions, the probability of an animal having a tumor is related to the dose by some sort of dose-response relationship, so that at any given dose this probability can be computed.. The observed results, a number of animals with tumor out of a larger number examined, is then a binomial sample with this probability. In practice, we don't know what the dose-response relationship is - we wish to estimate it from the results. But we assume that we know thO ,3HAPE of the dose-response relationship (specified by a mathematical formula), so that all that is required is to estimate some PARAMETERS in the mathematical formula. For example, the E.P.A. uses a dose-response relationship of the form: p=1-exp{-(qo+q,d+q2d2+...+qk-,dk-1 )I when there are k doses in an experiment, where p is the lifetime probability of tumor at dose d. It is usual -to use a maximum likelihood technique to estimate the various parameters q0, q1, q2, ... qk_,, given 'the observed numbers of animals with tumors and the numbers of animals examined at each dose. In cases where there is appreciable early mortality in the experiment, so that the observed numbers of animals with tumors are likely to be underestimates of what would have been observed at the end of a perfect experiment, one can make modifications to the dose response relationship, just as one can make life-table adjustments to standard statistical tests. One technique used is to modify the dose response curve to explicitly include length of life, using the idea that probability of tumor is likely to increase with a power of age (see page 2): p = 1 -exp{-(qo+q,d+q2d2+...+qk-,dk-') (t/L)"~ where t is t'he age at death, and L is a standard lifetime. The parameter n can either be fixed at some reasonable value (in the range 2 to 11), or estimated from the experimental results. This technique suffers from the same limitations as the usual modifications to the standard statistical 12

Expression of Risks Jusr as a comparison of risks is an aid in understanding them, so is a careful selection of the methods of expression. It is hard to comprehend the statistical (stochastic) nature of risk. There are ways to mitigate this difficulty in comprehension. We are almost all used to one such statistical concept the expectation of life. When we talk about the expectar.on of life being 79 years (for a nonsmoking male in the United Stares) we all know that some die young and that many live to be over 80. Thus the expression of a risk as the reduction of life expectancy caused by the risky action conveys some of the statistical concept essential to its understanding. One particu- lar calculation of th is type can be used as an anchor for many people, because it is easy to remember. The reduction of life expectancy by smoking cigarettes can be calculated from the risk, one in 2 million, of smoking one cigarette, multiplied by the difference of the average life-span of a nonsrnoker and a lung cancer victim. This turns out to be 5 minutes, or the time it takes to smoke the one cigarette. It is important ro realize that risks appear to be very different when expressed in different ways (19). One example of this can be seen if we consider the cancer risk to those persons exposed to radionuclides after the Chernobvl disaster. According to the Soviets (20), the 24,000 persons between 3 and 15 kilometers from the plant, but excluding the town of Pripyat, received and are expected to receive 1.05 million man-rems total integrated dose, or about 44 Table 3. Comparison of two very touc chemicals, afiatoxin Bl (22) and dioxin (23); CDC, Centers for Disease Control. Measure Aflatoxin B1 Dioxin Acute toxicity High Equal Carcinogenic potency to people -500 Unknown [(kg • dav)/mg] Carcinogenic potency to rats -5000 -5000 [(kg • dav)/mg] Mutagenic Yes No Certaintv of information on human High Low carcinogenicitv Activin, (initiator or promoter) Initiator Promoter (?) Possibilitt• of threshold dose response Low High Source Natural Artificial Common knowledge Little known Agent Orange FDA action level in peanuts (ppb) 20 CDC level of concem in soil (ppb) 1 on waste disposal. Economists and others often argue that efficiency depends on adjusting society until the amounts spent to save lives in different situations are equalized. It seems to us that society does not work that way. People are aware of the order of magnitude of these differences, and approve of them. Nonetheless, we believe that providing this information to a decision-maker is essential for an informed decision. rems average. Even if we assume a linear dose-response relation, with 8000 man-rems per cancer, the risk may be expressed in different wavs. Dividing 1.05 million man-rems by 8000 gives 131 cancers expected in the lifetimes of that population. This is larger than, and for some people more alarming than, the 31 people within the power plant itself who died within 60 days of acute radiation sickness combined with burns. Dividing the 131 again by the approximately 5000 cancer deaths expected from other causes, the accident caused "ordy" a 2.6% increase in cancer. This seems small compared to the 30% of cancers attributable to cigarette smoking. The difference is even more striking if we consider the 75 million people in Bvelorussia and the Ukraine who received, and will receive, 29 million. man-rems over their lifetimes. On the linear dose- response relation this leads to 3500 "extra cancers," surelv a large number for one accident. But dividing by the 15 million cancers expected in this population leads to an"insignificant" increase of 0.0 a 3%. Of course:, none of the methods of expressing the risk can be considered "right" in an absolute sense. Indeed, it is our belief that a full undersrnding of the risk involves expressing it in as many diffierent ways as possible. Cost of Reducing a Risk Another interesiarlg and instructive way of comparing risks is by comparing the amount people have paid in the past to reduce them. It might be thought that people would try to adjust their activities until the amount spent is roughly the same. Cohen (21) has shown that the amounts sp ent vary by a factor of more than a million. He shows that it would be possible even for an American to save lives in Indonesia by aiding in immunization at $100 per life saved. Society is willing to spend :more on environmental protection to prevent cancer (over $1 m:ill:ion per life) than on cures (about $50,000 per life with the high value of $200,000 for kidney dialysis raising some objections). This ratio is in rough accord with the maxim "an ounce of protection is bexer than a pound of cure." Z'eople are willing to spend still more on radiation protection at nuclear power plants and REFERENCES AND NOTES 1. L. B. Lave, Sci'ence 236. 291 (1987). 2. R. Wilson, E. A. C. Crouch, L. Zeise, in Risk Quanritation and Requlatorv Policy (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY,1985), Banbury Repon, vol. 19, pp. 133-147. 3. N. C. Rasmussen er al., "Reactor safetv studv-an assessment of accident risks in U.S. commercial nuclear power plants" (WASH 1400. \'UREG 75/014, U.S. Nuclear Re¢ulatorv Commission, Washington, DC, 1975). See also D. Okrent. Science 236,296 (1987). 4. R. Doll and R. Peto, J. Nat1. Cancer Imt. 66, 1191 (1984). 5. E. L. Anderson et al., Rirk Arurl. 3, 277 (1983). 6. E. A. C. Crouch and R. Wilson, J Taxical. Enriron. Health 5, 1095 (1979). 7. E. J. Calabrese, Principles ofAnimal Extrapolation (Wilev, New York, 1983). 8. R. Peto, inAssesanent ofRlsk from Low-Lcrrl Erporure to ftadiation and Chemualt, A. D. Woodhead, C. J. Shellabarger, V. Pond, A. Hollaender, Eds. (Plenum, New York, 1985), pp. 3-16. 9. B. N. rlmes, R. Magaw•, L. S. Gold, Scicnce 236, 271 (1987). 10. "Control of trihalomethanes in drinking water." proposed rule. Fed. Reqirr. 43. 5756 (1968). See also the advanced notice [ibid. 41. 28991 (1976)] and the final rule [ibid. 44, 68624 (1979)]. 11. M. MeseLson and K. Russell- in rns of Human Canca. H. H. Hiatt, J. D. Watson, J. A. Winstcn. Eds. (Col Spring Harbor Laboraton', Cold Spring Harbor, ; 7Y, 1977) p. 1473. 12. S. Parodi, M. Tamngher, P. Boero, L. Santi, Mutat. Res. 93, 1(1982). 13. L. Zeise, R. Wilson, E. A. C. Crouch, RirkAnal. 4, 187 (1984). 14. Smoking and Health, a Report oftheSurgeon General (PHS79-50066, Public Health Service, Washington, DC, 1979). 15. R A. Squire, Scrcnu 214, 877 (1981). 16. "Sacchann and its salts," proposed rule and hearing, Fed. Regict. 42, 19996 (1977). 17. R. Wilson and W. J. Jones, E~scr,~y Ecok~tv and theEnrironmenr (Academic Press, New York, 1974), table 9-6. Other entnes mav be readilv calculated from data in the reports of the United Nations scientific committec on the effects of atomic radiation ["Sources and effects of ionizing radiation" (United Nations, New York, 1977)] and the report of the Committee on the Biological Effects of Ionizing Radiations ["The effects on populations of exposure to low levels of ionizing radiations" (National Acadcmy Press, Washington. DC, 1980)]. 18. M. Russcll and M. Gruber, Scicncc 236, 286 (1987). 19. A. Tverskv and D. Kahneman, ibid. 211, 453 (1981). See also P. Slovic, ibid. 236, 280 (1987). 20. L'.S.S.R State Committee for the Utilization of Atomic Energy, `°Che accident at the Chernobvl Nuclear Power Plant and its consequences," working document for the Post Accident Review Meeting, 25-29 August 1986, International Atomic Encrgq Agency, Vienna. 21. B. L. Cohen,'Healrh P{m. 38, 33 (1980). 22. H. R. Roberts, "The regulatory outlook for nut produczs," paper presented at the Annual Convention of the Peanut Butter Manufacturers and Nut Salters Associa- don, West Palm Beach. FL, November 1977. 23. R. D. Kimbrough, H. Falk, P. Stehr, G. Fries, J. Taritol. Environ. Health 14, 47 (1984). 24. Our work on risk assessment has been supported by donations from Clairol, Inc., the Dow Chemical Company, the Cabot Corporation, the General Electric Foundation, and the Monsanto Corporation. . 270 SCIENCE, VOL. 236

Epidemiology irt Risk Assessment for Regulatory Policy 1165 IV. THE FUTURE It seents likely that public concern for environmental issues will not abate within this century, that public and corporate funds will continue to support research to monitor and evaluate environmental and occupational hazards to health, and that epidemiology will continue,to play a critical role in this endeavor. It is therefore worthwhile to ask how regulators and epidemiologists can counteract the negative impacts of the political pressures endemic to regulation, and how epidemiologists and epidemiologic studies can provide guidance and support for the overall thrust of regulatory policy, as well as for the difficult decisions faced by regulators. One antidote for the negative side effects of politicization on epidemiologic research is awarenes s of the hybrid nature of risk assessment activities. We need to recognize that a neat separation of regulatory policy into matters of fact and value is illusionary, and to sensitize ourselves to value judgements when they occur. They will and must occur, because setting standards for hazards at work and in the environment is a social and political process. It is possible to abate political pressures by allocating sufficient funds, time and qualified personnel to the careful conduct of well designed studies, and by incorporating into the studies th e,advice of experts chosen to represent the concerns of all sides in sensitive issues. Recent invostigations of pregnancy outcomes among women whose drinking water had been con3aminated by a chemical leak from an underground tank at an electronics company provide a model for achieving such abatement. These investigations were conducted by the California Department of Health Services with the cooperation of the Santa Clara County Department of Health [32). Before beginning the studies, the principal investigators formed an advisory committee of epidemiologists representing the inte.r.ests.ef industry and of the citizens. The committee had a voice in the design, the data collection., the analyses and the interpretation of findings. The resulting consensus report provided a voice of reason that cooled many tempers in the heated political dispute surrounding, the issues. Epidemiologists can make their data more useful to regulators in several ways. A first step is good documentation. Clear, thorough and complete recording of the details and data that led to a study's conclusions are needed by regulatory scientists who must use the conclusions to formulate policy statements for public approval. The completeness of recording is important. Serfling [28] has decried the filtering of data and relevant research results that seem to contradict strongly held views about exposure effect, citing some occupational studies as examples. In 1981 the Interagency Regulatory Liaison Group published guidelines for documentation of epidemiologic studies [29, 30). There now seems to be a consensus that these guidelines have been helpful in improving the clarity and completeness of'study reports, and that they have not been the unwelcome intrusion of government agencies into epidemiologic turf feared by some. Apart from the regulatory scientists' need for documentation of technical details and raw data, there is -the layman's need for clear, nontechnical documentation of a study's conclusions„ with particular emphasis on the degree of precision and sources of uncertainty associated with the conclusions. The policy decisions for which epidemiologic evidence is needed concern the public, and the public must make those decisions. Informed decisions by laymen require 'exposition of the major findings of a study. as well as the sources and nature of uncertainty about the findings in clear English without the use of esoteric jargon. A_ , s_econdstep to enhance the utility of epidemiologic data involves more even- handedness among epidemiologists about the strengths and weaknesses of a study, and less dredging for flaws with intent to discredit. It is imperative that scientists attempt to form a consensus about the interpretation of data, so that the courts are not forced to resolve technical scientific issues they are ill-equipped to handle. David L. Bazelon, Senior Court Judge of the U.S. Court of Appeals for the District of Columbia Court, complained that scientists cannot agree about the reliability of data, and that ". .. they disagree even more about the inferences to be drawn from the facts. Often, they can tell us only of 'the risk

EpidemioloEy in Risk Assessment for Regulatory Policy 1163 kill us than are the pesticides we use to attack the insects in our homes. Such a perspective could help to assuage fear of cancer from environmental toxicants, and to direct the expenditure of public funds toward more cost-effective priorities. III. THE IMPACT OF RISK ASSESSMENT ON EPIDEMIOLOGY Clearly, epidemiologic observations continue to play an indispensable role in risk assessment for regulatory policy, and conversely, increasingly many epidemiologic studies are devoted to occupationally and environmentally induced disease. Increased public awareness of environmental issues and the need for risk assessment has brought epi- demiology into courts, into homes on the evening news, and into leisure reading in the Sunday newspaper supplement. Thanks to such publicity, epidemiology is no longer an arcane word for an esoteric specialty. The need for epidemiology in risk assessment has brought cmployment opportunities and interesting scientific problems to epidemiologists. But it has also produced negative effects. Problems arise because risk assessment is not a science, but rather a complex and often subtle ,fusion of facts and values. The problems are aggravated by the prevailing misconc:eption that risk assessment for toxic substances is (or should be) entirely objective and scientific. This misconception is illustrated by the statement of the Office of Science and Technology Policy, Executive Office of the President, that toxic substance regulation consists of two stages: Stage I (risk assessment) uses empirical data and scientific judgeme,nt to characterize human exposure and risk; Stage II (policy) uses social and political action to decide regulatory action [5]. This separatist view is echoed by the National Academy of Sciences Committee on the Institutional Means for Assessment of RiskT ic_,1Public Health, which reported: I "1Ne recommend that regulatory agencies take steps to establish and mzintain a clear conceptual distinction between assessment of risks and consideration of the risk management alternatives; that is, the scientific findings and policy judgements embodied in risk assessments should be explicitly distinguished from the political, economic, and technical considerations that inPruenee the design and choice of regulatory strategies" [16]. While it is useful to call attention to the desirability of such a distinction, I believe that in practice it is an unrealistic and unattainable goa1. Values enter toxic risk assessment in many covert ways. They determine the quantity and quality of information obtained about a chemical, influence explicit and implicit assumptions used to analyze data, affect the way data are interpretect, and influence the weights used to combine disparate sets of data (see Ref. [17] for examples). This mix of science and policy can have undesirable effects on the quality cf epidemiologic research by compromising the design, conduct, analysis, and interpretat!ion of sludies. Adverse effects on the design and conduct of studies can occur in several ways. Political pressures to find quick answers to difficult questions have prompted poorly designed and hastily conducted investigations of possible danger from air pollutants and toxic wastes (e.g. Ref. ['d8]). The findings of such studies have been heavily criticized and the resulting controversies do not help the image of the field. Sometimes political pressures completely prevent a s1 udy. For example, an attempted county-wide investigation of the reproductive effects of ai:rial malathion spraying for the Mediterranean fruit fly was aborted because -the- fiospitafi*i1h the largest proportion of births declined to participate, due to the inflammatory political climate at the time [19]. Conversely. political pressures have initiated unwarranted studies virtually doomed ta be inconclusive because of low, poorly documented exposures and lack of focus on specific disease entities. In the words of Doll [7]: ."Art epidemiological perspective starts not with the 10,000 chemicals that pollute a particular area, but with the 10,000 deaths that occur in that area each C D )4 1:-O

tests -- one has to introduce additional assumptions in order to apply it. In this case, one has to decide whether the tumors were a cause of death, or simply incidental. An alternative technique used when there is early mortality is to estimate the age dependence directly from the data, using a (so-called) non-parametric technique. This approach has been used to assemble a large database of comparable analyses of animal bioassays. This msV hodofogy has taken the raw results of the animal experiment, and summarized them in the form of a dose-response curve with known parameters. It is also possible to estimate how uncertain one is about a given parameter, using the same maximum likelihood techniques used to obtain point estimates of them - indeed, one can plot the uncertainty distribution for any of the parameters. For example, for the parameter q, (which will turn out to be the one of interest), we can plot the probability that q, lies below any given value: ..... . . .. .~ . __..... ....::- - 1; i i ... . . ,.... . . ... f. ...... '--. ..:.... .., x R In particular, we can find that value q,* such that there is 95% probability that q, < q,*. However, it is important to note that the uncertainty distribution so plotted contains only the uncertain#y due to the numerical size of the experiment - the uncertainty that arises because we used a small number of animals, instead of an infinite number. It does not include the uncertainties which must be present because of the shakiness of all our assumptions -- i.e. the major urn;Ortainties. I ._ _..._, 13

1160 c E z ~ ALICE S. WHITTEMORE aizo3so 84o tsoo 3720 WORKING LEVEL MONTHS FtG. 4. Standardized mortality ratios (SMRs) for lung cancer by working-level-months (WLM) of cumu(ative exposure in United States uranium miners. SMRs were normalized to 100% for exposure category 0-21 WLM. (Source [4)). 1300 1100 7oo 5o0 100 0 , Monitoring populations for disease'.is time-consuming, expensive, and vulnerable to serious bias. One must worry that comparisons between exposed and unexposed popu- ,lations are not confoutfded by differences in smoking and other determinants of health, nor biased by differences in subjective assessments of disease. Such worries are aggravated in studies of environmentally induced disease, because the effects are likely to be small and the danger of reporting bias great. These obstacles do not vitiate the strengths of epidemiology in risk assessment for regulatory policy. As noted by Doll in the context of policy-setting for the prevention of cancer [7J, human observations continue to make several essential contributions to risk assessment. In the paragraphs below I list some of the reasons why human data are needed for regulatory decisions. First, they are needed to detect unsuspected hazards that have not emerged from laboratory tests. Animal experiments are still imperfect tools for detecting human cancer, largely because of the great variability across species in response to chemicals, and our lack of .understanding about the causes of this variability. The International Agency for Research on Cancer has determined that there is sufficient evidence from human observations, but limited, inadequate, or nonexistent evidence from animal experiments, to classify as carcinogens the chemicals or chemical processes listed in Table 4. The fact that most of these chemicals have tested positive in one or more of the short-term in vitro or in vivo tests now in use, reflects not the sensitivity of the test battery but rather the intense scrutiny the chemicals have received, relative to those for which no human data are available. Moreover, the tests are not specific; one or more of them have been positive for a vast number of chemicals occurring naturally in the foods we eat and the products we use. Thus laboratory tests do not yet provide a reliable screen for human carcinogens, and they are of limited or no utility for many other diseases or conditions associated with environmental exposures. Human data will continue to be needed, despite the obvious desirability of discovering health hazards before human exposure to them. Second, human data are needed to estimate exposure levels producing the highest additional risk that is socially acceptable. Just as laboratory tests provide imperfect screens TAOLF 4. CCHEMICALS 0!t tNDL3TRtAL PROCESYS WITH SUFFlCiETiT* EITDENCE PrMt CARCI\tX'iEVtCITY IN HCMANS DLT VoT IN EXlERfMENTAL ANI4ALSt Arsenic and certain arsenic compounds Manufacture of auramine Benzene N,N-bis(2tholorethyl)-2-naphthylamine (chtornaphazine) Undergroud mining of hematite Manufacture of isopropyl alcohol (strong acid process) Mustard ps Nickel refining +. •As defined by the International Agency for Research on Cancer (8). tsoura: (8). •

Risk Assessment for Carcinogens: A Cornparison of Approaches of the ACGIH and the EPA Michael C.R. Alavanja,A Charles Brown,8 Robert Spirtas,1-D and Manuel Gomez" AEpiderniology and Biostatistics Program, Division of Cancer Etiology, National Cancer Institute, Executive Plaza North ® uilding, Room 543, Bethesda, Maryland 20892; BBiostatistical Methodology and Cancer Control Epidemiology Section, Division of Cancer Prevention and Control, National Cancer Institute, Executive Plaza North Building, Room 344, Bethesda, Maryland 20892; cNational Institute of Child Health and Human DeveH)pment, Executive Plaza North Building, Room f07, Bethesda, Maryland 20892; °Mernber of the Americzu1 Conference of Governmental Industrial Hygienists' Chemical Substances Threshold Lin'it Values Committee The rel ative carcinogenic potency of 16 chemicals evaluated by both the U.S. Environmental Protection Agency (EPA) and the Chemdcal Substances Threshold Limit Values (CS-TLV) Com- mittee of the American Conference of Governmental Industrial Hygienista, (ACGIH) were compared. The estimated cancer risk resulting from occupational exposure to the threshold limit values (TLVs) were also computed using dose-response curves devel- oped as a part of EPA quantitative risk assessments. Subst:untial agreement between the EPA and the CS-TLV Com- mittee was found when the relative potency of these carcinogens was compared. Use of EPA's risk model to estimate lifetime cancer risk from occupationall exposure at the TLV levels often resulted in high cancer risk estimates. The approaches used to assess cancer risk by both groups is described and a suggestion is made i~oir incorporaring existing quantitative risk assessments into the IjLV evaluation procedure. AlStla* M.C.R.; BroMfft, C.; Spirtas, R.; hmez, lbl.: Risit Ameststent for Carctrogem A foffoarison of Approa%es of 4he ACG}9 and tbe F.PII. Appi. Occup. Erwrwt. Ftyg. 5:510-51 7; 1990. Introducrtiion The Chemical Substances Threshold Limit Values (CS- TLV) Committee of the American Conference of Govern- mental Industrial Hygienists (ACGIH) has been reviewing its policie.s and procedures regarding carcinogens. Spirtas et al. (1985) described the current process the CS-TLV Committee uses to make the qualitative decision to des- ignate a chemical as a workplace carcinogen and the quan- titative decision to recommend levels of exposure for the guidance of industrial hygienists.0) Threshold limit values (TLVs) (for carcinogens as well as other toxic agents) are time-weighted averages (TWAs) for a normal 8-hour work- day, 40-hour workweek. The TLV is set for inhalation ex- posure, with special notifications for agents where ab- sorption from skin exposure is important. The TLV is assumed to be protective for "nearly all workers" assuming the workers to be healthy adults <') TLVs are guidelines for good work practices to be used only by professional industrial hygienists. For substances which cause chronic diseases such as cancer, however, there may not be a sharp cutoff point (threshold) between effect and no effect; it is, therefore, important that professional judgment be used in monitoring and protecting workers exposed to such substances. When deciding on guidelines for carcinogens, the CS- TLV Committee gives greatest weight: to epidemiologic studies having data onn quantitative exposure levels.<' ) Such substances receive an Al categorir.aion and are calied "Confirmed Human Cardnogens:'IVext in importance, and more typically available, are mamn>alnn toxicologic stud- ies having whole-bocty bioasstys. Such %Aam= are given an A2 designation and are called "Suspected Human Car- cinogens:" In reviewing the key experimental toxicology studies, the Committee considers route of entry (greatest weight given to inhalation studies), dose-response gra- dient, potency, mechanism of action, <2ncer site, time-to- tumor, length of exposure, and underiyi.ng incidence rate for the type of cancer and species under study. Replication of results is important, especialty if eomparable in different species. Other types of studies are useful in confirming Thiti article representti the views of the authors aad not tfiose of the Americui Ccxiference of Governmental InKlustrial Hygienists or its Chetn- ical tiutwances Threshold Limit Values Committee, or those of ttie U.S. De}rartment of Health and Human Services, the Natiotnl ir>stitutes of Health, and the National Carxer Institute 510 APPL OCCUP. pYYHiONI. NY& 50 • AU6lSSi 1le0

were observed in the testes and ovaries at concentrations below 300 ppm, and lesions were observed in the thymus, bone marrow, lymph nodes, spleen, ovaries, and testes in mice inhaling 300 ppm. The alterations were more severe in the males rhan in the females. In rats, the only exposure- related pathology was a slight reduction in femoral marrow cellularit,v at 300 ppm ~3> Studies to identify the target cells for benzene hema- topoietic toxicity indicated that benzene exposure dam- aged mouse pluripotent stem cells, the colony-forming cell units in the spleen, and the progenitor cells for granulo- cytes and mr.crophages.4-6> Hematopoietic depression in rodents was observed at benzene concentrations as low as 103 ppm after a 5-day exposure.(7) Cronkite et a1.(g,9) reported a series of studies where CBA/CA mice were ex- posed to benzene at 10, 25, 100, 300, or 400 ppm, 6 hours per day, 5 days per week for 2, 4, 8, and 16 weeks. Exposure to 100 ppm or greater for two weeks reduced bone mar- row cellularat}:(8) When C57BU6J mice inhaled 300 ppm benzene 6 haurs per day, 5 days per week for a total of 115 exposures, the numbers of B-lymphocytes in bone marrow and spleen and the numbers of T-lymphocytes in thymus and spleen were reduced00> When BALB/C mice were exposed at 50 or 200 ppm benzene 6 hours per day for 7 or 14 da.ys, the ratios and the absolute numbers of T- and B-lymphocytes in blood and spleen were de- pressed!lt> Depression of B-lymphocytes was dose- dependent, and it was more severe than that of the T-cells." t> When male C`.i7BL mice inhaled 10 ppm, 6 hours per day for 6 days, a significant depression in colony-forming units in B-Ivmphocy:e,s was observed; similar inhalation of 31 ppm resulted in depressed blastogenesis of T-lymphocytes.(12) Chronic/Carcinogenicity When groups of 40 CD-i mice were exposed to benzene in air at 100 or 300 ppm, 6 hours per day, 5 days per week for life, two mice in the high dose group developed mye- logenous leukemia. No leukemia was observed in the 100-ppm dose group.73> Snyder et a1.04> found that after groups of 40 C57BL mice inhaled 300 ppm benzene for 6 hours per day, 5 days per week for 2 years, eight cases of lymphoreticular neoplasia (six thymic lymphocytic lym- phomas, one plasmocytoma, and one hemocytoblastic leu- kemia) occurred; two mice in the control group developed lymphocytic lymphomas. The incidence of tumors in the benzene-treated mice was significantly greater (p = 0.005) than that in the control.('') In a lifetime carcinogenicity bioa.ssav in which oral doses of benzene were adminis- tered at 50 and 250 mg/kg-day, 4-5 days per week for 52 weeks, there was a dose-dependent increase in total can- cers.(15> The most prominent rat tumors observed were Zymbal gland carcinomas, mammary carcinomas, and leu- kemia. When Wistar rats and Swiss mice were given ben- zene at 500 mg/kg-day, 4 or 5 days per week for 104 or 78 weeks, respectively, the numbers of Z}Trtbal gland car- cinomas, hem.:olymphoreticular neoplasias, and total ma- lignant tumors were increased in the rats; increases in mouse Zvmbal gland d};splasia and carcinomas, mammarv carcinomas, pulmonary tumors, and total malignant tumors were observed."6' In the National Toxicology Program lifetime bioa5say,' t'1 50 F344/N rar,s of each sex per dose group were treated with benzene by oral gavage at doses of 50, 100, or 200 mg/kg-day for the males and at 25, 50, or 100 mg/kg-day for the females for two years. Similar groups of B6C3F1 mice of both sexes were treated with 25, 50, or 100 mg/kg- day. For the male and female rats, increases of Zymbal gland carcinoma, squamous cell papilloma, and squamous cell carcinoma of the mouth were observed. In the male rats, squamous papilloma and squamous cell carcinoma of the skin were also increased. For male mice, increased numbers of animals with Zymbal gland carcinoma, malig- nant lymphoma, alveolar/bronchiolar carcinoma, and al- veolarbronchiolar adenoma or carcinoma (combined), Harderian gland adenoma, and squamous carcinoma of the preputial gland were observed. For female mice, in- creased numbers of animals compared to the control were afflicted with malignant lymphoma, ovarian granular cell carcinoma, carcinosarcoma of the mammary gland, alveo- lar/bronchiolar adenoma, and alveolar/bronchiolar carci- noma were reported.ti7> Cronkite(IR) conducted a carcinogenicity bioassay wherein male and female C57BL/6 and CBA/Ca mice inhaled 100- 300 ppm benzene, 6 hours per day, 5 days per week for 16 weeks and found benzene-induced leukemia in the males. When mice inhaled 25 ppm benzene for as few as ten such exposures, lymphopenia resulted.08> Reprod uctive/Devetopmental Studies on the potential developmental toxicity of ben- zene administered by subcutaneous injections, ingestion, or inhalation have generally failed to show significant ad- verse effects in mice, rats, or rabbits (for review, see Schwetz(14)). Adverse developmental effects have been de- scribed in an unpublished rat bioassay performed by Litton Bionetics(20) wherein Sprague-Dawley rats inhaled 10-40 ppm benzene, 6 hours per day on days 6-15 of gestation. Embrvonic death increased from the control (6.2%) to 8.1 and 9.5 percent for rats exposed to 10 and 40 ppm ben- zene, respectively. However, the Litton study(20) was con- founded by the high ambient temperature in one of the exposure chambers during the study; maternal hyper- thermia is a known rodent teratogen. Kuna and Kapp(=t) conducted an inhalation study in which pregnant Sprague-Dawley rats were exposed to benzene at 10, 50, or 500 ppm 7 hours per day on days 6-15 of gestation. Significant reductions in mean maternal body weight gain occurred. Mean fetal body weight was reduced. Fetal crown-rump distance was decreased significantly at 500 ppm, and developmental delay was evident upon ex- amination of the fetal skeletons. Benzene was judged by these authors('1) to be fetotoxic in rats at 50 and 500 ppm and to manifest teratogenicity at 500 ppm. Coate et a1.~") found that when pregnant Sprague-Dawley rats inhaled 1, 10, 40, or 100 ppm benzene 6 hours per day on days 6-15 of gestation, no maternal toxicity was noted; however, APPL OCCUP. ENVIRON. HYG. 5(7) •-JULY 1990

7 The Two Major Extrapolations The assumptions made so far have allowed us to parametrize an animal dose-response relationship, obtaining values for the parameters which are presumably reasonably appropriate for high doses. Strictly speaking, this parametrization of the dose-response curve only enables us to estimate'the results we would expect to see at high doses in animals -the dose-response relationsh;ip can only be relied on to interpolate between high doses and perhaps to extrapolate a short distance outside the experimental range of doses. The problem now is to perform two extrapolaiaons - from animals to humans, and from high dose to low dose: Animal High Dose Low Dose Human Observed ~ ~ ~ ~ Required LOGICALL Y there are two distinct routes to follow in this extrapolation, since there are logically two distinct dose-response curves involved (see below). One can extrapolate from high dose to low dose using the ANIMAL dose-response curve, and then extrapolate to humans (dashed lines), or extrapolate to humans at high doses and then use a HUMAN dose-response curve to extrapolatE~ to low doses. We have seen how to estimate the parameters of the (high dose region of) the animal dose-respcnse curve. In practice, the same curve (with the same parameters) is used to extrapolate to low doses, by building into the mathematical structure of the dose-tesponse curve all our as3umptions about low dose behavior. How is this relevant for estimating human risk? Consider a generalized situation in which we wish to estimate the response (R) of humans to some dose (D) of material, when there is a response (r) in some experimental system at dose (d). Notice that nothing implies that r, R measure tha same sort of response - they could be completely different (r could be acute toxicity to the lung of a mouse, R could be skin rashes in humans). Similarly, the dose measures d, D may be completely different. In the case immediately at hand, r is the lifetime probability of tumor in animals, iand d is a dose as measured in the animal experiment. There are other cases of practical importance however - r might be some measure of response (such as number of 14

and not responsive to the request for a risk estimate (although this should not be re:ad as condemning procrastination in all circum- stances). The second extreme mentioned, the assumption of zero risk, can arise because people and government agencies have a propensity to ignore anything that is not a proven hazard. We argue that this attitude is incon>istent if the objective is to improve the public health, may also f.ead to economic inefficiencies, and often leads to unnecessary contention between experts who disagree strongly. Fortunately, if risk assessors have been diligent in searching out hazards to assess, Few hazards posing large risks will be missed in this wav, so that there may be minor direct danger to human health from a continuation of the attitude. Risk Estunxtion Based on Historical Data The way in which risks are perceived is strongly correlated with the way in which they are calculated. Risks based on historical data are particularly casv to understand and are often perceived reliably. It is therefore easv to illustrate a risk calculated from historical data to understand some characteristics of risk estimation. There are plenty of data on automobile accidents (although never enough to make risk assessors happy). One thing that these data can tell us is the frequency oE such accidents in the past and their trend through time. To make p;redictions, however, we must use a model. The simplest model is that there will be as many accidents next year as last, to within a statistical error of the square root of the number. A slightly more complicated, but perhaps more accurate, model might be to fit a mathernatical funcrion to numbers from previous years and to argue that next year's accidents will follow the trend given by this function. A possibly better and possibly more accurate model still might use all available information that might influence accident trends. For exaniple, an oil embargo with a concomitant rise in oil price and reduction in automobile travel would be likely to reduce the risk of accid:nt. In anv event, it becomes clear that it is impossible to calculate anv risk without a model of some sort, even the simple one that tomorrow will be like today. Risks of New Technologies We can only usc: the historical approach to estimating risks when the hazard (for example, technology, chemical, or simply some action) has been present for some time and the risk is large enough to be directly measured (although when it is not large enough to be Table 1. Comparison of several common radiation risks. Cancers if all Dose S population U ttion A (mrem/ . . . year) exposed (assuming linearity) Medical x-rays 40 1100 Radon gas (1.5 pCiLliter, equivalent dose)* 500 13,500 Potassium in own bodv 30 1000 Cosmic radiation at sea level 40 1100 Cosmic radiation at Denver 65 1800 Dose to average resident near C.hemobvl first year 5000 Not relevant One traziscontinental round trip by air 5 135 Average within 20 miles of nuclear plant 0.02 > 1 *The radon exposure is to the lungs and cannot be directiv compared to whole body external exposure. The comparison here is on the basis of thc same magnitude of risk. The uncertamtv of the radon number is at least a factor of 3. 268 Table 2. Some commonplace risks (mean values with uncertainty). Action Annual risk Uncertainn, Motor vehicle accident (total) 2.4 x 10-4 10% Motor vehicle accident (pedestrian only) 4.2 x 10-5 10% Home accidents 1.1 x 10-4 5% Electrocution 5.3 x 10-6 5% Air pollution, eastern United States 2 x 10-4 Factor of 20 downward onlv Cigarette smoking, one pack per day 3.6 x 10' Factor of 3 Sea-level background radiation (except radon) 2 x 10-5 Factor of 3 All cancers 2.8 x 10-i 10% Four tablespoons peanut butter per day 8 x 10-6 Factor of 3 Drinking water with EPA limit of chloroform 6 x 10' Factor of 10 Drinking water with EPA limit of trichloroethylene 2 x 10-9 Factor of 10 Alcohol, light drinker 2 x 10-5 Factor of 10 Police killed in line of dutv (total) 2.2 x 10-4 20% Police killed in line of dutny (by felons) 1.3 x 10-4 10% Frequent fiying professor 5 x 10-5 50% Mountaineering (mountaineers) 6 x 10-x 50% measured, an upper limit may be calculated, if one assumes some sort of model). If there is no historical database for the hazard (a new power plant or industrial facilitv, for instance), one approach is to consider it in separate parts, calculating the risks from each part and adding them together to estimate a risk for the whole. For example, all possible chains of events from an initiator to a final accident are followed in an "event tree," with the probabilities of each event in the tree being estimated from historical data in different situations. A particularly well-known example is the calculation of the probabilitv of a severe accident at a nuclear power plant (3). That this procedure has at least a partial validity is due to the fact that the design of nuclear power plants proceeded in approximately this factorable way; attempts were made to imagine all major accident possibilities, "maximum credible accidents" or "design basis acci- dents," and then to add an independent device to prevent this accident from having severe consequences. To the extent that the added safetv device is independent, the failure probability is inde- pendent, and the small overall accident probability is the product of individual failure probabilities which are larger. Risks by Analogy: Carcinogenic Risks Some carcinogenic risks may be estimated from historical data. But this is complicated by the time delay between the insult and the final cancer, one reason why causality is hard to prove if the risk is small. This is the difficult field of epidemiology. Although some of the largest cancer risks have been identified through the use of epidemiology (4), preventive public health suggests that we endeavor to estimate risks even where no historical data exist and the risk is small. This is often done by analogy with the cancer risks to animals, usually rodents, which are deliberately exposed to large enough quantities of pollutant so that an effect is observed. To use these data to estimate the risk at low doses in people involves (to oversimplify matters) two difficult steps: the comparison of carcinogenic potency in animal and man (5-7) and the extrapolation from a high dose to a low dose. Because both steps require a certain amount of theory, they are controversial. Indeed, there are those who regard the uncertainty as so great that the} prefer not to provide numerical estimates of risk (8, 9), although they may order materials in carcinogenic.potenc,v. The difference SCIENCE, VOL. 236

r c4ises, we have measurement difficulties at low doses, and in both cases there is some sort of dose-response relationship (which I deliberately leave vague for now): 1 t ; I . / C ` .~csc tD~~ Dt~SC • Evidently there will be some AGE STRUCTURE to the probabilities of cancer. As nientioned, for many cancers in humans the death rate from cancers increases with a power of age. In experimental studies involving long term feeding of rodents, the same soil of age structure is found for the incidence of tumors. A"LIFETIME" probability thus depends on when you measure it - the usual practice is to assume a"standard" lifetime of -70 years for humans and -2 years for rodents. 0 At high enough doses (i.e. at high RESPONSES) one sees interactions between different materials in both animal experiments and in human data (e.g. smoking and alcohol consumption, smoking and radon exposure, smoking and asbestos exposure). The effect o1f such interactions is to make the effect of two or more materials different from the sum o1` the effects of the materials individually (at the same doses). • It is not possible to make direct measurements of what happens at low doses (i.e. at LOW RESPONSES). In this context, low dose means a dose at which the response probability is < 0.1 usually, and < 0.01 certainly. Any attempt at studying lower doses runs up against problems of logistics, cost and the background cancer rate. • The shape of dose-response curves assumed for the low dose regions are thus based on: Theoretical ideas Prejudice Guesswork 7

1 Erackground Information It is useful to bear in mind a few sobering facts about total populations at risk, and the normal total risk of death and of dying of cancer. For the U.S., the total population is about 240 million, while the annual number of deaths is about 2 million per year and the annual number of cancer deaths is about 400 thousand. These figures imply an annual average total risk of death of about 10-2 (1 percent per year), and a lifetime risk of cancer of about 0.2 (20 percent, or 200,000 x 10), estimates you can obtain simply by dividing one figure by another. Of course, simply dividing one by another is not a particularly accurate way of computing such estimatOs - one should do the correct thing and take the age structure of the population into account, and the variation of risks with age, and so on. But even when you do precisely that, the average lifetime risk of cancer comes out to be about 20 to 25 percent. We can expect this figure to get higher as the expectation of life increases, and as other causes of death are eliminated (assuming - pessimistically - that most cancers cannot be eliminated). It is mainly the increase in expectation of life which has made cancer such a prominent cause of death in the (historically) recent past, because cancers tend to be diseases of old age. For many cancers it is found that the death rate varies as a power of age:- rate - age" where the exponent n is in the range 4 to 11. For such cancers, this pattern seems to hold over the age range from about 30 to 65. At lower ages the rates tend to be very small but almost independent of age (and the cancers may be completely different diseases in youngsters), while at higher ages the reported death rates are lower than would be predicted by this sort of formula - and in sc me cases the reported death rates are actually lower for old enough groups. It is unclear whether these reductions in death rates in the elderly are real, or are simply due to a difference in the accuracy of diagnosis and reporting. It is also possible that the reduction in reported death rates is real, but is due to the winnowing out of the population of those who are susceptible to these particular cancers, leaving a core of more resistant individuals. The major exceptions to the power law variation of death rate with age are the cancers which are known to be hormonaliy dependent (e.g. breast cancer), or are highly curable (skin cancers), or in which the natural progression is altered by intervention (e.g. a high proportion of women have had hysterectomies by age 65, so that they cannot be at risk of uterine cancers thereafter). With this age variation of risk of cancer understood, we can now oversimplify again and quote a lifetime average annual risk for cancer, obtained simply by dividing the lifetime risk by an average lifetime of about 70 years. This give an average annual risk of about 2-3 x 10". Notice that we 2

Epidemiology in Risk Assessment for Regulatory Policy 1159 discussed. 1t is unthinkable today that a U.S. President would undergo furtive oral cancer surgery on a yacht in New York's East River to keep it from his constituents. as did Grover Clevetand in 1893. The environmental movement of the 1970's has had a direct impact on the substance of epiderriiologic studies. Figure 2 shows the increase with time in the proportion of those articles in t~e American Journal of Epidemiology that are devoted to the adverse effects of physical and chemical agents in the workplace and environment. Although a sizable part of this new research has examined acute and chronic respiratory disorders and re- productive disorders, the largest portion has dealt with environmentally and occu- pationally induced cancer. That cancer should monopolize a disproportionate share of the research reflects patterns of research funding, which in turn reflect priority patterns of public fear. Many of the examples and much of the discussion in this paper concern the relationship between epidemiology and risk assessment for cancer, although the problems and future prospects apply to other diseases as well. Estimating risks to health from environmental agents using human data must proceed in the face of formidable obstacles. Most toxic exposures occur chronically at levels that are low, variable, and measured with substantial error. Epidemiologic studies are likely to overbaok a large number of small effects associated with such exposures. Data from those occupational studies dealing with high exposures and large effects typically provide limited guidance about risks at low environmental levels, as can be seen by comparing the very high lung cancer death rates of U.S. uranium miners with those of smoking and nonsmo!king U.S. veterans, shown in Fig: 3. An individual living in the U.S. today inhales y occurring radon gas and its radioactive decay products at an average rate of naturall, roughly two-tenths of a WLM per year [6]. (A WLM, the acronym for "working-level- month", is_a unit qf cumulative exposure to a-radiation.) By age 70 he will have inhaled a to~Taf about 14 WLM, a small amount in comparison with totals in excess of 3000 WLM inhaled by U.S. uranium miners before the establishment of a federal standard in 1970. The startling excess of lung cancer among these miners relative to that of other U.S. whiti; males illustrates the difficulty in at(empting to use these data to estimate risks from low levels of radiation. The difFiculty is also evident upon examination of the standardized mortality ratios (SMR's) shown in Fig. 4. The SMR's were computed using the entire cohort as the standard, and were normalized so that miners in the lowest exposure category of 0-21 WLD+i form the referent group. Interest centers on risk among miners in the 22-119 WLM range, because these exposures approximate those experienced by indi= viduals living in areas with very high background radon levels. However, the evidence is equivoczJ: although the, death rate for the 22-119 WLM group is almost twice that of the referent group, the increase is not statistically significant at the 5% level. J 60 65 70 75 BO AGE (ytwr:) . FtG. 3. Age-specific lung cancer mortality rates in U.S. uranium miners (p-pi, smoking U.S. veterans (O-Q) and non-smoking U.S. veterans (®-t•). (Source (4.5).)

Epidemiology in Risk Assessment for Regulatory Policy 1161 TAaLE S. FSTiL(ATFD HL'1tAN aLADDl7t CANCEa RtSK3 (CANCERS; IO` PO/t;LAT)Oti) POR L1FETnMI SACCHAa1N n:GFSSION OF 0.I20/DAY* Low dose extnpolation method Interspecies extrapolation method Sinjk-hit Multisuje Multihit Probit Body surface area 1200 5 0.001 450 .mg.'kg,day - 210 0.001 21 mg1kg.tlifetitne 5200 0.001 4200 •Eatrapolated from Rat Bladder Tumor Data. Source (9). TABt.E 6. RISK OF BLADDER CANCER AMONG USEBS OF AaTiFTCIAL SwEFiENE7t5 AELATTVE TO RLSK AMONG NONtAE7t5 (EST7MATED FR01t CASE-CONTROL 6fUDtES) Authors Males Females Hoover et al. (10) 0.99 1.00 Kessler and Clark [i I) 0.97 1.01 for por,ential toxicants, so also are they extremely limited tools for obtaining quantitative estimates of risk. Table 5 shows that estimates of human bladder cancer risk associated with saccharin, derived from a single positive experiment in laboratory rats, can differ by as much as six orders of magnitude, depending on the assumptions used to extrapolate across species and dose level. By contrast, the consistent lack of association found in six case-control studies of bladder cancer (see Table 6 for a sample) provide an upper bound on the slctual level of human risk. Of course, neither human nor laboratory data can prove that a substance is harmless, but consistent negative findings in humans pfovide reassurance about the probable magnitude of the hazard. Third, human data are needed to check inferences about a putative cause for a disease by-mtsnitoring- the 'effect of its removal. Such checks require time, due to the long lag between exposure onset (or termination) and disease occurrence that is characteristic of many chronic diseases. For example, we can only now begin to monitor the effects on U.S. lung cancer rates of reductions in tar and nicotine content of cigarettes and in cigarette use since the 1950's. Figure 5 shows a modest but clear downward trend with year of birth in age-specific lung cancer death rates among young U.S. white males. Each successive birth cohort contains fewer men who started smoking, and among those who did, a higher proportion who smoked low tar cigarettes. Finally, human data are needed to provide a sense of perspective about the magnitude of various hazards to health, in order to set priorities for the "penditure of public and 50 z ~ ¢ W 6 O 30 Q U 3D-g4 35-39 40-44 45-49 AGE (Y•an) Fua. 5. Age-specific lung cancer mortality rates in United States white male cohorts. (Source [12).)

The difference between these two logically distinct routes of extrapolation might be important in some circumstances. For cancer risk assessment based on animal carcinogenesis bioassays, however, the distinction is glossed over (one might even say, ignored), by the practice of assuming the same (or very similar) mathematical form for the dose-response curve in both humans End animals (or more generally, in all species), and interpreting the parameters in the same way for both compared species. In the general case, however, what is required is some sort of relationship between the parameters of the dose-response curves: Animal Human r = f(d; a,b,c...t) R = F(D; A,B,C...T) We need to be able to derive the parameters A,B,C... from the values a,b,c which can be estimated from experiments, and then use the human dose-response curve to extrapolate to low doses. The prac.tin,al approach is to seek parametrizations of the dose-response curve which result in the derivation of A,B,C... being simple given a,b,c... Consider the case of acute toxicity, for example,. It is found that the shape of the dose-response curve for acute toxicity, in which the response~ is death, is very similar for a large number of toxins and for many different species. There is, in this case, a threshold-type dose-response curve which can be nicely parametrized by two values: the dose at which 50% of the animals tested can be expected to die (under suitable conditions), and the slope of the dose-response curve at this dose. The first parameter is known as the LDS, (the second has no special name). Why is this parametrization useful? If the LDs of various materials in one species are plotted against the LD,,s of the same materials in another species, one finds approximate proportionality between them (the plot is a straight line). This can be expressed as, for example, LD,(rabbit) is proportional to LD50(mouse). Even more remarkable, it turns out (at least, it did for a particular group of chemicals) that if the dose is measured in a suitable way, as (amount)/(surface area of animal), then approximately we have numericai equality in the values of LD.: LD,(rabbit) = LD50(mouse) = LD50(other species) It is this approximate equality which explains the utility of the LD50. The other parameter used in defining the dose-response curve, the slope of the curve at the LD50, is not involved in this 16

techol and hydroduinone were potent SCE inducers at 4.4 p,g/mL"O Glutathione (GSH) inhibited benzene-induced SCE formation, and it was hypothesized that GSH conju- gation to benzene metabolites prevented DNA damage.«'> Benzene and its metabolites were reported to decrease mitotic index, to inhibit cell cycle transverse, and to in- crease ',CE frequency in cultured human T-lymphocytes. The relative pote,zc'y of benzene metabolites for SCE in- duction were catec:hol > 1,4-benzoquinone > hydroqui- none > 1,2,4-benzenetriol > phenol > benzene.t4$> Tice et a1.t49> found a concentration-dependent increase in DBA/2 mouse bone marrow lymphocytes after a single, 4-hour inhalation s udy of benzene at 28-3000 ppm; an increase in SCE was detected at 28 ppm. This response was strain-dependent as DBA/2 mice were more sensitive than C57B1:/6 mice, young DBA/2 mice (three months) were more sensitive than older mice (10 months), and male mice were more sensitive than female mice. Following intraperitoneal injection, a linear dose-dependent increase in SCE was observed in DBA/2 mice.t49> DNA Damage Benzene failed repeatedly to exhibit genotoxicity in tests for unscheduled DNA systhesis (UDS) in cultured primary rat hepatocytes. Benzene is consistently negative in HeLa cells with or without metabolic activation. Glauert et at.t5o> published the single positive report for increased UDS in cultured primary rat hepatocytes associated with benzene exposure. In a study of in t4tro DNA damage, mouse L5178 YS lymphoma cells failed to show single strand breaks after exposure to 1.0 mM benzene, phenol, or catechol or to 0.1 mM hydroquinone; however, a dose-dependent in- crease in DNA dama;e was observed after treatment with para-henzoquinone or 1,2,4-benzenetrioLtSt> Para-benzo- quinone at 6 µM induced 70 percent single strand DNA breaks within 3 minutes of exposure; the same damage was achieved by benzenetriol within 60 minutes.(51> A concentration-dependent increase in mouse periph- eral hlcxod micronuclei was observed after C57BU6 mice inhaled 10, 25, 100, or 400 ppm, 6 hours per day, for 9 days.1 s') When C57F3V6 mice inhaled 300 ppm benzene for 16 weeks under a similar protocol and the patterns for micronucleus induction monitored, the initial increase was followed by a gradual decrease.r53> When the peripheral blood of B6C3F1 miice given oral benzenet t'~ was studied, a dose-dependent increase in the numbers of circulating ervthrocyte micronuclei occurred. A significant increase was observed in male mice given a dose as low as 25 mg/kg- day for 120 davs.t'4> Pretreatment of male and female CD-i mice with metabolic enzyme-inducing agents (phenobar- bital, SKF-525A, Arcchlor 1254) failed to protect against the clastogenic effect of benzene exposure, but pretreat- ment with 3-methylcholanthrene potentiated benzene mvelcx-latitogenicia:",S1 Male mice were more sensitive than female mice and chromosomal damage was greater after oral than after intraperitoneal administration."s> Chromosome aherrations were induced in Wistar rats after inhalation of 100 or 1000 ppm benzene.l'1'> When male DBA2 mice inhaled benzene at 0, 10, 100, or 1000 ppm or male Sprague-Dawley rats inhaled benzene at 0, 0.1, 0.3, 1, 3, 10, or 30ppm for 6 l-rours, significant (dose- dependent) increases in SCE and micronuclei were ob- served in mice at = 10 ppm, and increased SCE and mi- cronuclei were observed in rats inhaling ; 3 ppm and at I ppm, respectively! 5'> The Erexson data(57> are the lowest concentrations of inhaled benzene that have been reported to induce genotoxicity. Neoplastic Transformation Using morphologically transformed colonies as a marker, benzene was considered mutagenic in Syrian hamster em- bryo (SHE) cells, but it was not considered mutagenic in cultured Balb/C 3T3 mouse fibroblasts, in Simian adeno- virus-transformed SHE cells, and in Chinese hamster ovary (CHO) cells.(58) Benzene, hydroquinone, and para-ben- zoquinone were reported to alter gene expression in cul- tured Swiss mouse spleen lymphocytes, where hydroqui- none and para-benzoquinone at 10-20 µM inhibited RNA synthesis 50 percent.t59> Inhibition of T-cell proliferation and reduced production of interleukin-2 (a T-cell growth factor) by 5 µM para-benzoquinone was suggested to ac- count, in part, for benzene-induced aplastic anemia.(s9) Human Cytogenicity Forni etal.t60> found a significant increase in lymphocyte chromosome aberrations in two groups of workers with overt benzene intoxication as compared to age-matched controls. One group consisted of 25 individuals recovered from benzene hemopathy 1-18 years previous along with 4 additional workers currently suffering from acute ben- zene poisoning. The second group consisted of 34 workers in a rotogravure plant exposed at 125-532 ppm benzene in air from 1952 to 1953. Tough et a1.t61,6z> found an in- creased incidence of chromosome aberrations in 38 work- ers inhaling 25-150 ppm benzene for 1-25 years com- pared to the incidence in the general human population. These individuals had been exposed to benzene until two to four years prior to the study.(6',62) Watanabe et a1.t63> found an increase in the frequency of SCE among nine females at six onths after cessation of benzene exposure at 1-9 ppm for 1-20 years and among seven females ex- posed to benzene at 3-50 ppm for 2-12 years. Killian and Danielt64> found a significant increase in chromosomal aberrations among workers exposed to average benzene levels below 10 ppm. Workers exposed to benzene (av- erage, 56.6 months) had a doubling of chromosomal breaks and a threefold increase in rings and dicentric chromo- somes. Almost twice as many benzene-exposed workers as controls exhibited both chromosome breaks and rings and dicentric chromosomes.",'> Piccianotci5-'> examined the Killian and Danielt64> data and reported that 38 (73.1 %) of 52 workers exposed to mean ambient benzene at less than 10 ppm had chromo- sonie hreaks as compared with 18 (40.9%) of 44 matched (unexposed) controls. When individuals with both chro- 456 APPL OCCUP. ENVIRON. HYG. 5/7) • JULY 1990

1158 ALIC'£ S. WHITTEMORE TABLE I. PARTIAL LIST Ot FEDLRAL LEGtSLATtO~ RFGCLATING TOXIC Sl'BSTAVCFS Legislation Year passed Delaney Clause of Food. Drug and Cosmetic Act 1959 Federal Hazardous Substances Act 1960 Clean Air Act 1970 Occupattonel Safct} and Health Act 1970 Consumer Product S,tfety Act 1972 Federal Envtronmental Pesticide Control Act 1972 Federal Insecticide. Fungicide and Rodentitide Act 1972 Safe Dnnking Water Act 1974 Resource Conservation and Recovery Act 1976 Toxic Substjntes Control Act 1976 Clean Water Act 1977 TABLE 2. FEDERAL AGENC'fES REGULATIVG TOXiC 5L3STAKEs Agency Year established Food and Drug Administration 1929 Environmental Protection Agency 1970 Occupational Safety and Health Administration 1970 Consumer Product Safety Commission 1972 TADLF. 3. FEDERAL RESEaRCH ORGANlZATiO?IS rM1'YFSTIGATtttG TOXIC SL'BSTAVCES Organization Year established National Cancer tnstitute 1937 National Institute for Environmental Health Sciencts 1969 Nationat Institute for Occupational Safety and Health 1971 National Tozicology Program ` 1978 destruction of the earth with industrial emissions fueled public pressure for a rash of environmental legislation, some of which is listed in Table 1. Tables 2 and 3 show the parallel evolution of federal agencies created by Congress to regulate and control toxic emissions, and of federal research institutes to provide the scientific basis for such regulation. These developments have led many to regard the 1970's as "the decade of the environment". Although motivation for the environmental movement included concern about the adverse effects of contaminants on respiratory function, reproductive outcomes and genetic mutations, the most compelling constituent was public fear that the global destruction predicted by Carson would include an epidemic of chemically induced cancers. Figure 1 shows temporal trends in the estimated probability that a white male baby born in the U.S. will either develop cancer or die from it. The temporal increase does not reflect the feared epidemic. Rather it reflects the greater proportion of men who will survive to old age when cancer risks are highest, as well as the more accurate diagnoses among the elderly; and the effects of tobacco. Apart from this real increase in cancer incidence and mortality, there is a perceived one due to the openness with which the disease is now 20 16 10 _~--------d o...------ ! t t z 12 W U G e a o ~-s- 1970 Y " ~ t < < 1972 1974 1976 1978 t t 1980 1982 1975 1980 1985 YEAR YEAR FtG. 2. Proportion of articles published in American FtG. t. Trend of lifetime probability for developing Journal of Epidemiology concerned with adverse (-) or dying (---) of cancer, white male born in health ettects of physical and chemical agents in the the United States. (Source (3n. salorkplace or general environment.

relationship. Had we chosen some other method of parametrization, it is quite possible the required interspecies relationship between parameters would be much more complicated. 8 Ini:erspecies Comparison - Constant Relative Potency What is sought is a simple relationship between the parameters of dose-response relationships in different species. When it is assumed that the dose-response relationship includes a term linear in dose, there is a simple measure of the strength of a carcinogen - the carcinogenic potency (the slope of the dose-response curve at low dose). The simplest hypothesis is that for different species, the ratio of carcinogenic potencies is constant for different materials, so that if material A is twice as potent a carcinogen as material B in species 1, it will also be twice as potent as material A in species 2. This is the idea of constant relative potency, as applied to carcinogenesis, and it underlies the standard approaches to estimating human risks from animals. There is even some data which supports this idea! There have been several hundred bioassays performed simultaneously on rats and mice, and when the results of these are parametrized using a close-response relationship which includes a linear term, we can estimate the potency in two species for each material tested. Plotting the potency measured in rats versus the potency measured in mice for each material then gives the figure shown (page 24). Notice that each measurement is uncertain to greater or lesser degree, due to the relatively small numbers of animals tested. If the idea of constant relative potency were exactly correct, these points would all lie on a straight line on the figure - or at least, a!l would lie sufficiently close to such a!ine that the measurement uncertainty bars on each point would encompass the line. From the figures, one can see that: (1) (2) On average, potency in one species is proportional to potency in the other species. There is a large scatter of the points around the lines of exact proportionality - a scatter bigger than would be expected from the measurement errors alone. A similar ~comparison can be attempted between the potencies measured in animal experiments, and those observed in humans (page 24). These cases have arisen in the past where humans have been exposed to materials before they were known to be carcinogenic. We can make use of other's misfortune to estimate how potent each such material is in humans, and compare with estimates obtained for mice and rats in laboratory experiments. In this case, the uncertainties are so large that little can be quantitatively stated, although qualitatively the idea of constant relative potency does not seem to be disproved. A more recent and much more thorough study of comparisons between humans and animals has been carried out for the E.P.A. by Allen, Crump &S'hipp and the qualitative results are similar (page 25) - although Allen et a!. do not quantitativefy evaluate the correlation. 17

POTENCY IN HUMAN (nq'k9 d) 0 M T N W a , '-i POTENCY IN HUMAN (mgf'ky d) a, a~ _ s aN u - 4 T I 9v &svssZoz T I LoBjp): F344 (p= 0.025) 1 ci~ i, w ~ cn \ \ T T 11 r 1 N t I i , \ \ 0

10y ~ 1'03 0 s+ p 102 C O E ~ = j 10 ~ o v t- o r .~ • a 1,0 o~ E M / W 10-1 n N O 0-Z A 6-.a i 0,.3 10-41 L 10-5 I 10-4 10-3 10-2 10-1 1.0 101 102 103 5 -4 '^'---~ 104 o TOY5 Eatimotea from Animal Data (^'9/k9/daY) I'ig'• 2. Human TDI estimates versus animal TDu estimates obtained from base case (analysis 0); log-log plot. N O ~ . ~ ~ ~ ~ G~D 24 ,~ Ec e I

9 lnterspecies comparisons - practical and theoretical The measure of carcinogenic potency introduced above was roughly defined as the ratio of (excess tumor probability)/(dose), at low enough dose. For the E.P.A. model usually used in risk assessments: p =1 -exP{-(qo+q,d+q2d2+...+qk-,dk-')! the corresponding measure is q1. When this dose-response relationship is used with real data, it is usual 1!o use an "upper 95% confidence limit" estimate q,* of q, as the measure of potency, since such an estimate is always non-zero (while, for example, the maximum likelihood estimate is often z,ero). The "upper 95% confidence limit" is with respect to the numerical uncertainties of the experiment only, and so this estimate of potency is in no sense an upper limit with respect to all the other uncertainties involved. To compare humans with animals, the approach taken is to postulate a similar dose-response relation ship in both cases: Animal Human p=1 -exp{-(qo+q,d+...)} p = 1 -exp{-(Qo+Q,D+..)} and then ifhe constant relative potency hypothesis suggests that Q, is proportional to q1, and so one hop es to say that: Q, = constant x q,` or at least Q, < constant x q,* where the constant depends only on which animals species is used. We expect the constant to be different for different animal species - it will presumably depend on how we measure dose, on the relative lifespans of animal and human, on relative metabolic rates, and a whole host of other factors. With enough experiments, we could measure the constant in this relationship - at least in comparing animal with animal, rather than human with animal - and (in theory) empirically determine how it varies with these factors. The figures mentioned above suggest that the constant is not completely constant, but that there is some sort of random uncertainty built in (or at least, an uncertainty that we can treat as random), amounting to an average factor of about 5. If we are very lucky, it may be possible to find some way of measuring dose so that the constant in the above relationship is numerically equal to 1, so that the potency is equal in different species (up to the uncertainties) - just as it was possible to find such a measure in the case of the LD,50. 18

For performing risk assessments for human safety purposes, there is naturally a prejudice to be conservative. It is generally agreed that assuming LINEARITY between dose and response (for our discussion, this means the lifetime probability of a cancer) at low enough doses is CONSERVATIVE. This assumption is made in a theoretical way - it is assumed that the true relationship between dose and response lies, at low enough doses, entirely below (or at worst on) a linear curve joining the responsa. at zero dose (background) with the response at some higher (but still low) dose. I'1-'Tf LiC l~ci, ~c%l- ~- ( 0 A Typically,l,he background rate is of order 10-' to 10-', and we are interested in excesses over the background of order 10 to 10-4, so this diagram is not to scale. It is useful to define the POTENCY of a carcinogen as the ratio of excess lifetime probability of cancer to the dose causing that excess (at low enough doses). On the diagram, this is the ratio i/d. The potency is thus the slope of the dose-response curve at low enough dose, and we have the basic equation: EXCESS RtSK = POTENCY x DOSE There is reasonable evidence that some mechanisms of carcinogenesis result in a THRESHOLD - i.e. thai, there is some (threshold) dose below which the excess incidence of cancer is much lower thu would be predicted by a linear extrapolation from doses above the threshold, and possibly that the excess incidence of cancer is literally zero below such a threshold (excess, here, means excess over the background occurrence of cancer). Some of the evidence for such mechanisms comes from observation of the dose-response curves in experimental situations - the experiments on saccharin provide a good example. However, there is stilf the possibility that a linear mechanism may still operate at low enough doses, and so any human risk assessment has to take that possibility into account. 8

1166 ALICE S. WHtTTEMORE of risk'. . . . Courts must not be expected to resolve such questions. What judge knows enough to understand issues on the frontiers of nuclear physics, toxicology and other specialities informing health and safety regulations?" [31). While it may be naive to think that epidemiologists can reach a consensus about uncertain data when millions of dollars and lives are at stake, there is no feasible alternative but to try to do so. The reproductive studies in Santa Clara County, and others like them. provide a paradigm for achieving such a consensus. A third step to increase the utility of epidemiologic observations for risk assessment is aggressive monitoring of occupationally exposed popiilations. This is largely a job for industrial epidemiologists and occupational physicians, who should keep computerized. annually updated and linkable medical, job and smoking histories for all current (and to the extent feasible, former) employees. As noted by Doll [7), this monitoring makes sense from the industrial point of view, since most such studies would reveal no excess risk, and the accumulated negative human evidence, coupled with estimates of exposure levels for various agents, would be useful in resisting overzealous regulation. The monitoring also makes sense from the worker's point of view, because real hazar¢s would be detected earlier than they otherwise might be..Finally, it makes sense for the public who would learn that prolonged exposure to quantified levels of many of the agents feared harmful have not produced observable human hazards. The most promising developments in the monitoring of exposed populations involve the use of biological exposure markers in blood, tissue, urine, feces, hair or nail samples. Table 8 lists several of the markers detectable and quantifiable in human specimens. Such markers have the potential to document exposure levels, identify and quantify unusual susceptibility to environmental toxicants, detect precursors of injury or organ dysfunction, an;d provide etiologically supportive biological links between exposure and disease. Epidemiologic studies are needed to determine how well they correlate with exposure, and with preclinical or clinical manifestations of disease. They are also needed to detertpine the marker's reproducibility and persistence over time. Industrially exposed cohorts *and cohorts of patients undergoing chemotherapy are ideal populations for such'studies. CONCLUSIONS Epidemiology continues to play an indispensable role in risk assessment for regulatory purposes. Human data are needed to detect hazards missed by laboratory experiments, estimate exposure levels producing the highest socially acceptable risks, monitor changes in disease rates after the removal of putative causal agents, and provide a perspective for cost-effective allocation of public health resources. Epidemiologists can make their data even more useful for risk assessment by providing clear and complete documentation for other scientists, and jargon-free documentation for those not versed in epidemiologic methods. Equally important is data interpretation with more balance and less factiousness. All of these objectives would be facilitated by the dialogue resulting from symposia, and from postdoctoral fellowships and visiting ap- pointments allowing academic, regulatory and industrial epidemiologists to visit one another's worksites. Occupationally exposed populations should be monitored for exposure levels, morbidity and mortality. Biological markers in human specimens promise to afford useful indices for exposures and for unusual susceptibility to exposures. There is need for work to correlate these markers with exposure history and with disease, and to establish their reproducibility, variability and persistence over time. Risk assessment is both a political and a scientific process, and politicization will continue to complicate the conduct of epidemiologic research on the effects of environ- mental toxicants. Some restraint of political pressures can be achieved by allocating the funds and time for studies of high quality, with-®ngoing input from epidemiologists representing all interested parties.

It has now become standard practice for risk assessments to assume that the constant is exactly unity if the dose is measured as a (daily average amount)/(surface area of animal), by analogy with the t.D50 case. The graphs shown on page 24 actually suggest that it would be better to assume an average factor of unity, with an uncertainty factor of about 5 to 7, when the dose is measured as a (daily average amount)/(bodyweight of animal). This assumption will probably change some time in the future when better information is available, or when an alternative theoretical framework suggests a better idea. 10 An example - 1,2 Dibromoethane As an example of the procedures usually adopted, let us look at the case of 1,2-Dibromoethane. What fol€ows is by now means complete, but it indicates the sort of analysis which has to be performed. This example is confined to analyzing just one result out of many, in a single bioassay (of about 5). In practice, it is essential to look at all the results. The bioas>ay I have chosen was an inhalation bioassay in the National Toxicology Program series. A summary of the study design for rats (the design for mice is very similar) is: Initial Concentration Time on study (weeks) number of animals ppm (6 hr/d, 5 d/wk) Exposed Observed C Male Rats _ Control 50 0 0 104-106 Low dose 50 10 103 1 High close 50 40 88 0-1 ~_ Female Rats Control 50 0 1 104-106 Low dose 50 10 103 1 High dose 50 40 91 0-1 We will look, only at the results in female rats. First, their survival was not as good as might be desired (see graph below) in such an experiment, but the early mortality was probably largely due to the cancers appearing in the study, so it is acceptable - we can use (at least initially) the 19

pared to 27 expected (SN1R = 33-), and four cases of multipl,-:~ myeloma were observed compared to one ex- pected (tiMR =-i09) f all cases statisticalh• signiticant]. Rin- sky et al.'`'-) determined that cumulative exposure to ben- zene (measured as ppm-years) was the most reliable predictor of death from henzene-induced leukemia. In- creases in cumulative exposure were associated with marked progressk,e increases in the SMR for leukemia: among workers with less than 40 ppm-years cumulative exposure, the StiiR = 109; with 40 to 199.99 ppm-years cumulative exposure, the SMR = 322; with 200 to 399.99 ppm-years cumulative exposure, the SMR = 1186; and with 400 or more ppm-years, the SMR = 6637. (The ppm-years were calculated a5 40 years at 10 ppm average exposure/year = -i00 ppm,years.) Seven of the nine leukemia deaths with multiple -nyeloma had less than 40 ppm-years of benzene exposure. Rinsky et aP 9?> concluded that protection from benzene-induced leukemia increased exponentially with reductions in exposure time. Yin et al.t9"> conducted a retrospective cohort study of 28,460 workers exposed to 3-308 ppm benzene (with the majority exposed to 15-150 ppm) compared to a control cohort oF 28,257 workers not known to be exposed to henzene. Thim, cases of leukemia were found in the ex- posed population compared to four such cases in the con- trol. The benzene cohort experienced a leukemia mortality rate of 14 per 100,000 person-years, and the control pop- ulation experienced a leukemia mortality rate of 2 per 100,000 person years (SMR = 5.74). In an additional study authored by Yin and as.sociates; 99) ambient benzene con- centrations for 508,818 workers averaged 5.6 ppm with 65 percent of the workplaces having less than 12 ppm and 1.3 percent having benzene levels greater than 308 ppm. Aplastic anemia occurred at 12.1 per 100,000 persons in this cohort and represented a 5.8-fold increase over that of the general population. Ott er al.1 too~ carried out a mortality study of 594 white male workers exposed to benzene from 1940-1970. The Occupational Safety and Health Administration (OSHA)t1o1) concluded that the Ott cohort was exposed to an average of 5 ppm: for an average of nine years. Three cases of myelocyt,ic leukemia (2 classified as acute) were found compared to 0.8 cases expected (p < 0.047). Bond et al.(102) extended the cohort definition for the Ott study to include those employees who worked for at least one month (1938- 1978) ~ind increased the observation follow-up to 1982, bringing the total persons studied to 956. Four deaths due to myelogenous leukemia were observed with 0.9 ex- pected (5MR = 444). Decouile et al.(103) found a fourfold excess risk for lym- phatic and hematopoietic cancers among oil refinery and chemical plant workers exposed to benzene. The expo- sures were very poorly documented, but they resulted primarily from plant fugitive emissions and perhaps ac- companied by gross exposures from cleaning tools, hands, and clothing with liquid benzene. The historical cohort mortialiity study of 259 male employees found four deaths from lymphoretic6lar cancers compared to 1.1 expected (SN1R = 364), and three deaths due to leukemia where 0.4 were expected. The multiple myelomas observed here, taken together with previous reports of benzene-asscxiated myeloma. prompted the suggestion that the pathogenesis of human multiple myeloma and chronic lymphatic leu- kemia may arise from damage to B-cell lineage."Dj' Wong' 1"4.iu'' divided the benzene exposure for 4602 work- ers (minimum time of 6 months) into four categories: < 1 ppm: 1-10 ppm; 11-50 ppm; and 50 ppm, with peak exposures of < 25 ppm, 25-100 ppm, and > 100 ppm. He compared their mortality with that of 3074 employees from the same or similar plants who had no known occupational benzene exposure. When all lymphatic and hemotopoietic cancers were considered, there was a significantly elevated risk (p = 0.03) for benzene-exposed white males when compared to unexposed workers. There was a significant concentration-dependent increase for all lymphohemato- poietic cancers (p = 0.02), for leukemia (p = 0.01), with borderline significance (p = 0.057) for non-Hodgkin's lymphopoietic cancers. Prolonged cumulative exposures were judged more important for human benzene carcin- ogenicity than maximum peak exposures, and the au- thors~ 10-',1051 concluded that there was a significant asso- ciation between occupational benzene exposure and the occurrence of leukemia, all lymphopoietic cancers, and non-Hodgkin's lymphopoietic cancers. A number of epidemiologic studiesc106,1171 have consid- ered the mortality and cancer incidence among petroleum and rubber workers. Most of these studies, however, failed to quantify the benzene exposures adequately, failed to determine whether the toxicity reported was indeed as- sociated with benzene exposures, and were confounded by difficulties in confirming the validity of the diagnoses upon which the SMR and other risk estimates were made. The latency period for benzene induction of human leukemia varies from 2 to 50 years. Aksoy etal.(87-91) found that the induction period ranged from 6 to 14 years (me- dian, 11 years). Vigliani(92) reported an induction period of 3 to 23 years (median, 9 years), and Rinsky(96) indicated a median latency of 12 years (2-22 years). The Shell Oil study4 113) indicated a latency of 17-54 years between the date of hire and date of death from leukemia. Ynt98> es- timated the average latency time for benzene-induced leu- kemia as 11.4 years. The 1985 OSHA report(to1) concluded that 11 years was a reasonable estimate for the average duration of leukemia induction associated with occupa- tional benzene exposure. Basis of the TLV Although benzene has long been recognized as a mye- lotoxicant (e.g., more than 140 fatalities due to benzene poisoning were recorded in the open scientific literature prior to 1959), the carcinogenic activity of chronic expo- sure to relatively low ambient concentrations of benzene in workplace air was not recognized until the last ten years. Benzene is a human and rodent clastogen and carcinogen. Adverse health effects in animals exposed to benzene mir- APPL OCCUF: ENVIRON. HYG. 5l71 • JULY 1990 459

Hematcpoietic System: Monocytic leukemia 6/50 5/50 1/50 Circulatory System: Hemangiosarcoma 0/50 0/50 5/50 Circulatory System: Hemangiosarcoma or Hemangiosarcoma, invasive 0/50 0/50 5/50 Liver: iNeoplastic nodule 2/50 0/49 3/48 Liver: C-iapatocellular carcinoma 0/50 1/49 3/48 Liver: Neopiastic nodule or Hepatocellular carcinoma 2/50 1/49 5/48 Pituitaiy: Adenoma, NOS 1/50 18/49 4/45 Pituitaiy: Chromophobe adenoma 20/50 0/49 0/45 Adrenal: Pheochromocytoma 3/50 1/49 0/47 Thyroid: C-cell Carcinoma 1/49 3/48 1/45 Mammaiy Gland: Adenocarcinoma,NOS 1/50 0/50 4/50 Mammaiy Gland: Fibroadenoma 4/50 29/50 24/50 e Notice e,>pecially the various groupings which are employed - this is a matter of judgement. It is clear that the major effect is in the nasal cavity, but observe also the effect on fibroadenomas in the mammary gland, and the negative trend seen in the pituitary. Such negative trends are generally ignored. Further analysis, taking account of the age at death, might show such a negative trend is an artifact caused by the early deaths in the dosed groups, but here the result in the low dose group suggests that the effect is real. Using the combined results in the nasal cavity, we fit the E.P.A. multistage model and find best estimates of: qa = 2.699 x 10'2; q, = 6.876 x 10'2; q2 = 0; and obtain an upper confidence limit for q, of q,' = 8.6 x 10"2, in each case using as doses the values 0, 10 and 40 ppm from the experimental design. In fact, the earlier figure of a distribution of values for q, is taken from this example - you can read the probability of q, being less than any given value from that figure. What this means is that the linear term in the relation between risk and dose is probably less than 8.6 x 10-2 per ppm (under the conditions of the experiment). Now what do we do with this estimate? That depends on the application, but we will assume that we wish to make a "UNIT RISK" estimate for humans from it - that is, estimate an upper bound Iifetime risk to a human exposed to 1 µg/m3 of dibromoethane for life. 21

revertants per culture dish) in a mutagenesis bioassay, with d the dose applied to each culture dish. l Sysl,em -> Arbitrary Animal bioassay Human example Re>ponse r p (lifetime probY. of R tumor Dose measure d d (as used in expt.) D Dose-response r = f(d; a,b,c,..,t) p=1-exp{-(qo+q,d+..)} R = F(D; A,B,C,..,t) curve What is required is some connection between the parameters a,b,c,... of the dose-response relationship in the experimental system and the parameters A,B,C,... of the human dose-response relationship. These parameters presumably include those mentioned in Section 6, and I have explicitly included age amongst them. Given such a connection, the extrapolation to humans of the results in the animal studies is perfectly straightforward. The problem lies in finding the connection. Once such a conn-', ;;ion is found (by whatever means) we have the methodology for the two extrapolations required. Notice the difference between what is done in the two distinct pathways of extrapolation mentioned above: In the first, the shape of the dose-response curves are examined, and it is decided how they may be (separately) extrapolated to low doses. Then some relationship is postulated between the parameters of the dose-response curves at low doses (it has to be postulated, since nothing can be measured at such low doses). One potential advantage of this approach is that the animal dose-respanse curve could be measured, in principle and by heroic experimentation, down to lower response rates than usual (and this has been done in some cases) - allowing greater confidence in this extrapolation to low dose. - In the second, some relation between the parameters of the dose-response curves is obtained at high doses (and this may be done experimentally, in principle, since at high doses the responses are measurable). Then it is decided how the human dose-response curve should be extrapolated to low doses. The advantage here is the possibility of direct comparison between species, albeit at high dose. 15

Also shown are the "natural" rates for such cancers, expressed in terms of lifetime risk and annual average risk. Material/Action Site Lifetime Annual or Risk Average Type Risk (In absence of exposures) ------------------------------------------------------------------------ 4-Ajninobiphenyl Auramine manufacture Benzidine Chlornaphazine Cyclophosphamide 2-Naphthylamine Bladder 5 x 10-3 7 x 10-5 Arsenic (compounds) Asbea tos BCME CCME ung x 10-2 x 10-4 Chromium (VI compounds) (Pop'. ave.) Mustard gas Nic:kel refining Arsenic PUVA Skin 3 x 10-3 4 x 10-5 Socts, Tars, (Deaths!) Mineral oils Vinyl chloride Liver 1 x 10-3 2 x 10-5 DES (In Utero) Vagina 7 x 10-3 9 x 10-5 Benzene Myleran Chlo:rambucil Leukemia 8 x 10-3 1 x 10-4 Melphalan Typically, in epidemiological studies, a relative risk of >2 is required in order to detect any effect. Thus the (epidemiologically) discoverable population average human risks are > 10-5 per year, or 10-•3 per lifetime, and probably much larger. For the small subgroups of the population usually available for study, the observable risks are generally much larger. 3. Target Risks. The Necessity of Extrapolation. When considering the size of acceptable risks to the public at large, the usual targets are much smaller than the discoverable risks discussed above. Typically they will be of order < 10-6 per year. Note that the EPA and the FDA set targets of order 10-6 per lifetime, that is, of order 10-8 per year. It must also be borne in mind that there are a large number of materials which are of potential interest. The Chemical Abstracts -3-

mosome breaks and chromosome markers ( ringti. dicen- tric chr(rme>somes) were compared, less rhan 3 percent of the nonexposed group showed genetic damage where 2' percent of the exposed workers were afflicted with chro- mosemie aberrations (p < 0.001). A number of reports suggests that benzene-incluced hu- man chromosome damage is site-specific. Ding et al.'O-' reported a c~•togenetic studv of 21 patients (8 male and 13 female) with chronic benzene poisoning who had been exposed to unspecified benzene concentrations for 1-28 years (average, 6 years). At the time of cytogenetic analyses, all individuals had not been exposed for 5-20 years (av- erage, ae years), and all but one had recovered from clin ical signs of benzene poisoning. Hvpodiploid and hyper- diploid cells were increased significantly in the benzene- exposed patients, and chromosome deletions in the hy- podiploid cells involved groups C, E, and G chromosomes and chromosome gains in the hyperdiploid cells involved groups C and E. Similar findings were also reported by Sasiadek and Jagielskil68> where chromosomal aberrations were detected more frequently in chromosomes 2 (Group A), 4 (Group B), and 6 and 9 (both are Group C). Sarto and associates(G9) found an increase in chromosome ab- errations among 22 workers inhaling 0.2-12.4 ppm ben- zene for 11.4 ± 7.0 years; a control population was matched for sex, age, smoking habits, and site of residence. Pharmacokinetic/Metabolism Studies Rusch et al.(') concluded that humans absorb approxi- mately 46 percent of the benzene that is inhaled. Assuming a respiratory rate of 16 per.minute and a tidal volume of 0.5 liters, approximately 7.5 µL benzene can be expected to be absorbed each hour through the lungs of a person inhaling, air containing 10 ppm benzene.(7) Benzene dermal absorption was 0.05 percent when neat liquid benzene was applied directly to a human forearm Glucuronid• Sullat• at 0.0022 mg/cm2 and allowed to dR,1'0' and the flux of henzene through cultured human abdominal skin from air saturated with benzene at 31°C was 1.0 µLcm-'~hr-1.1'1> Susten et al.l''1 found that after dermal application of S µLl-'C-labeled benzene to intact skin of hairless mice, maximal skin radioactivirv occurred at 1.5 min, and it re- mained "essentially unchanged for at least 2.5 hr:' Perme- abilin• is, however, dependent upon presence of solvents. Blank and McAuliffel'1 ) found the constants to be 111, 3.73, 2.4, and 1.4 X 10-j µL cm-z-hr-1, respectively, for water, hexadecane, isooctane, hexane, and gasoline. Based on in vitro percutaneous absorption and in vino inhalation data, one example of calculated total benzene exposure used an adult working in ambient air containing 10 ppm ben- zene with 100 cm'- skin surface in direct contact with gas- oline containing 5 percent benzene. It was estimated that if the worker's entire skin surface was in contact with ambient air, the individual would absorb 7.5 µL benzene via inhalation in one hour, 7.0 µL from direct dermal con- tact with g soline, and 1.5 µL from body surface exposure to ambient air.(71) Sabourin et al.(73) investigated the absorption and elim- ination of benzene in F344/N rats, Sprague-Dawley rats, and B6C3F1 mice after an oral or intraperitoneal dose of 0.5-150 mg/kg. They reported that gastrointestinal absorp- tion was essentially complete. The toxicity of benzene has been attributed to its me- tabolites.('4) A major metabolite is phenol (Figure 1), gen- erated by oxidation of benzene by the liver cytochrome microsomal system(75) via the reactive epoxide interme- diate, benzene oxide. Results of physiologically based pharmacokinetic modeling of benzene metabolism found that mice metabolized a greater proportion of absorbed benzene to the hydroquinone conjugates and muconic acid than did rats.t76> Rats metabolized benzene primarily to the phenyl conjugates and to the phenyl mercapturic B.nz.na ~~ _n1 O \\ ~ Prt-Ph.nyl Phonol Mucona{dehyds 0 Marcaptu ri,-: Acid OH 0A~ I S-N-Acatyl-Cys V 1 Phanyl M.rcapturic Acid ir{ S-N-Acatyl-Cys FlGURE 1. Major pathways of benzene metabolism. (Reproduced with permission from reference 76.) H APPL OCCUP. ENVIRON. HYG. 5p) - JULY 1990

giving a minimum of 600 animals per experiment. AII the animals have to be carefully housed (under standard conditions), cared for, and individually tracked throughout their two year lifetime. They are then sacrificed and a large number of their tissues examined individually. None of this comes ch eap - the cost of such an experiment is unlikely to be less than $200,000, and may run above $1,000,000. It should be noted that the type of experiment detailed here is the minimum considered necessary to answer a YES/NO question: Is this material carcinogenic under the conditions of this standard bioassay? The experimental design and analyses performed are designed to be unlikely to answer YES if there is no carcinogenic action present (so that the experiments have low alpha error), but they can easily answer NO even in the presence of carcinogenic action. This sori of test is exactly what is required, of course, if one is interested in identifying materials which are surely carcinogens; in order to study their mechanism of action for example - one doesn't want to accidentally end up with a material with no carcinogenic action. I would submit, however, that for the purposes of protection of public health, the questions asked of the tests are entirely the wrong way round. For protecting public health, one should surely ask not whether this material is almost surely a carcinogen, but how strong a carcinogen it could be, given the results of the experiment. The fact that the same sort of analysis is applied now as in the past is perhaps a combination of accident and inertia, but one has to admit that, for the most part, the methodology has been largely successful so far. 6 Raw Results - and what to do with them. Having spent 2 years performing the experiment described above, what output do we get? When the animals are sacrificed, they are dissected and a whole list of tissues examined, both macroscopically and microscopically. All lesions, whether related to cancer or not, are noted down and iJsually (nowadays) recorded in some sort of computer database. The pathologists performing the examinations usually use some sort of standardized nomenclature for what they observe -- for example, the National Toxicology Program uses a modified version of the Systematized Nomenclature for Pathology (SNOP). Other information about individual animals is also recorded - such information as where they came from, which cages they were kept in, when they died (e.g. if they died naturally, or were sacrificed at the end of the experiment, or sacrificed earlier because they clearly would not survive), and so forth. The outcome is that for each animal, we have a list of the lesions affecting them when they died. An examp,l0 of a condensed listing of just the cancer-related lesions is appended. From such listings, wie can perform various analyses and statistical tests to see whether the rate of cancer was increalsed at any site or for any type of cancer. 10

Service (CAS) has now given names to well over six million distinct chemicals which have been mentioned in scientific literature, and there have been various estimates of the number (around 50,000) of chemicals in general commercial use. With such numbers, it should be immediately apparent that there are jusi: too many time, money and logistical constraints to directly detecting any adverse effects from such a plethora of materials to which humans may be exposed. Notice that a risk of 10-6'per lifetime corresponds to a rate of about 3 per year in the whole U.S. population. Thus, even if the whole U.S. population were exposed to some material causing a risk of death of 10-6 per lifetime, the resulting deaths would be statistically indistinguishable in the usual two million deaths per year (unless there were something extremely unusual about the deaths). Extrapolation is therefore essential in order to estimate the sizes of risks, and hence be in a position to demand that risks be reduced to the levels mentioned. The fundamental observation on which such extrapolation is based is that: HUMAN CARCINOGEN -->implies--> ANIMAL CARCINOGEN In other words, every known material which has been shown to be a human carcinogen is also known to cause tumors in animals under suitable conditions. The only current possible exception to this is arsenic, but it is quite plausible that this is simply because it has not been tested adequately. This observation is not very useful in itself, but what is done in order to allow risk assessments is to assume its converse: ANIMAL CARCINOGEN -->implies--> HUMAN CARCINOGEN and to work from here. This assumption is not unreasonable, in view of what is known about carcinogenesis - although it is something which can be argued about in specific cases. 4. The Nature of Carcinogenesis. In what follows, it is useful to keep in mind some information about the process of carcinogenesis. This information has been derived from studies of humans and animals, and from experiments performed in vivo or in vitro. It is based partly on experimental studies, and partly on theoretical ideas suggested by those studies. (a.) Cancers arise from one (or more) individual cell(s) which have gone "out of control" in some way - the cell becomes immortal, with no limit on the number of cell divisions, and the usual constraints on cell division no longer apply. A cell may pass through several stages before reaching this state. (b.) The underlying cause of such behavior is probably some effect(s) on the genetic material of the cell, but the exact mechanism(s) is (are) unknown. -4-

(c.) The occurrence of such events appears to be a random process at some level. One cannot tell which individual cell or animal or person will be affected. Hence we talk about the PROBABILITIES of cancer - the chance that some event will occur. (d.) When we feed materials to experimental animals, these prob<<bilities depend on various factors which can be manipulated. For example, they vary with: The total AMOUNT of material (the total dose) The AGE at which dosing takes place The RATE OF APPLICATION, or the time over which dosing continues OTHER FACTORS (some known -- stress, dietary factors, others unknown) We therefore expect, and in practice observe, DOSE-RESPONSE curves. Such dose-response curves are fundamental in extrapolating risks to humans. I like to draw an analogy to the similar problem of extrapolation which arises for acute toxicity -- in both cases, we have measurement difficulties at low doses, and in both cases there is some sort of dose-response relationship (which I deliberately leave vague for now): i LIFE'T-IME PlZo g r OF Ti4w1oR cffACIrtcSt;-~rtEC iTy PRoe oF pEATN 0 (e,) Evidently there will be some AGE STRUCTURE to the probabilities of cancer. As mentioned, for many cancers in humans the death rate from cancers increases with a power of age. In experimental studies involving long term feeding of rodents, the same sort of age structure is found for the incidence of tumors. A "LIFETIME" probability thus depends on how youe measure it - the usual practice is to assume a "standard" lifeti.me of -70 years for humans and -2 years for rodents. (f.) At high enough doses ( i.e.. at high RESPONSES one sees interactions between different materials in both animal experiments and in human data ( e.g. smoking and alcohol consumption, smoking and radon exposure, smoking and asbestos exposure). The effect of such interactions is to make the effect of two or more materials different from the sum of the effects of the materials individually (at the same -5-

Or\J3O At r -a~ ( E (0IcL) Male Rats Control Low High Lung: alveolar/bronchiolar adenoma or carcinoma 1 /20 3/50 0/50 Pituitary: chromophobe adenoma 0/18 2/43 2/49 Adrenal: pheochromocytoma or pheochromocytoma, malignant 2/20 2/50 2/50 Thyroid: C-cel{ carcinoma 0/16 0/38 2/43 Thyroid: C-cell adenoma or carcinoma 2/16 3/38 2/43 Preputial gland: adenoma or carcinoma, NOS 2/20 1/50 4/50 Tests: interstitial-cell tumor 13/20 46/50 47/50 Female Rats Lung: aivEIclar/bronchiolar adenoma 0/20 0/50 4/48 Hematopoietic System: leukemia 0/20 4/50 1/50 Pituitary: chromophobe adenoma 0/19 3/45 0/43 Pituitary: chromophobe adenoma or carcinoma 3/19 10/45 8/43 Thyroid: C-cell carcinoma 2/15 1/38 0/44 Thyroid: C-cell adenoma or carcinoma 4/15 3/38 2/44 Mammary Caland: fibroadenoma 3/20 9/50 2/50 Uterus:leiomyosarcoma 0/19 1/50 3/50 Uterus: endometrial stromai polyp 0/19 6/50 4/50 Uterus/Enclometrium: adenocarcinoma, NOS 0/19 5/50 3/50 Mesentery°, Iipoma 0/20 0/50 2/50 Male mice Lung: alveolar/bronchiolar carcinoma 2/14 9/50 12/46 Lung: aiveoi,ar/bronchiolar adenoma or carcinoma 4/14 15/50 18/46 Hematopoietic System: lymphoma or leukemia 2/14 6/50 11/46 Liver: hepa,tocelluiar carcinoma 0/14 7/50 13/46 Liver: hepatccellular adenoma or carcinoma 1/14 9/50 14/46 Female mice Lung: alveolar/bronchiolar carcinoma 1/20 4/39 2/48 Lung: a(veolar/bronchiolar adenoma or carcinoma 1/20 8/39 10/48 Hematopoietic system: all neoplasms 5/20 16/41 14/48 Hematopoietiw system: malignant lymphoma, lymphocytic leukemia, or leukemia, NOS 5/20 14/41 13/48 ~ All sites: hemangioma 0/20 4/41 0/48 O Liver: hepatweilular carcinoma 0/20 3/40 0/48 N ~ Liver: hepata,ellular adenoma or carcinoma 1/20 4/40 0/48 Pituitary: chromophobe adenoma 2/14 1/19 0/14 ~ Uterus: endornetriai stromal polyp 0/20 3/37 0/45 CA Peritoneum:lipoma 2/20 0/41 0/48 m ~ Mesentery: lipoma 2/20 1/41 0/48 ~

dosEs). (g.) It is not possible to make direct measurements of what happens at :Low doses ( i.e. at LOW RESPONSES ). In this context, low dose means a dose at which the response probability is < 0.1 usually, and 0.071 certainly. Any attempt at studying lower doses runs up against problems of logistics, cost and the background cancer rate. (h.) The shape of dose-response curves assumed for the low dose regions are thus based on: Theoretical ideas Prejudice Guesswork For performing risk assessments for human safety purposes, naturally a prejudice to be conservative. there is < It is generally agreed that assuming LINEARITY between dose and response (for our discussion, this means the lifetime probability of a cance:_) at low enough doses is CONSERVATIVE. This assumption is made in a theoretical way -- it is assumed that the true relationship between dose and response lies, at low enough doses, entirely below (or at worst on) a linear curve joining the response at zero dose (background) with the response at some higher (but still low) dose. VCTiME P2oBy ~ ~ -r n ~ i hc~~rcun ~ ; '~'t,Lm o l' I (c~~ 0 0 G~. Typically, the background rate is of order 10"4 to 10-1, and we are interested in excesses over the background of order 10-6 to 10-4, so this df.agram is not to scale. It is useful to define the POTENCY of a carcinogen as the ratio of excess lifetime probability of cancer to the dose causing that excess (at low enough doses). On the diagram, this is the rat:io i/d. The potency is thus the slope of the dose-response curve at low enough dose, and we have the basic equation: EXCESS RISK - POTENCY x DOSE There is reasonable evidence that some mechanisms of carcinogenesis result: in a THRESHOLD -- i.e. that there is some (threshold) dose below which the excess incidence of cancer is much lower than would be -6-

the dose gives an increase in sensitivity roughly proportional to the dose. Clearly the latter is most cost effective. Even with such a minimum design, there are: 3 dose groups x 2 sexes x 2 species x 50 animals per group ' giving a minimum of 600 animals per experiment. All the animals have to be carefully housed (under standard conditions), cared for, and in(E _vidually tracked throughout their two year lifetime. They are then sacrificed and a large number of their tissues examined individually. None~ of this comes cheap -- the cost of such an experiment is unlikely to be less than $200,000, and may run above $1,000,000. It should be noted that the type of experiment detailed here is the minimum considered necessary to answer a YES/NO question: Is this material carcinogenic under the conditions of this standard bioassay? The. experimental design and analyses performed are designed to be unlikely to answer YES if there is no carcinogenic action present (so tha.t the experiments have low alpha error), but they can easily answer NO even in the presenci--,,of carcinogenic action. This sort of test is exactly what is required, of course, if one is interested in identifying materials which are surely carcinogens; in order to study their mechanism of action for example -- one doesn't want to accidentally end up with a material with no carcinogenic action. I would submit, however, that for the purposes of protection of public health, the questions asked of the tests are entirely the wrong way round. For protecting public health, one should surely ask not whether this material is almost surely a carcinogen, but how strong a carcl!nogen it could be, given the results of the experiment. The fact that the same sort of analysis is applied now as in the past is perhaps a combination of accident and inertia, but one has to admit that, for the most part, the methodology has been largely successful so far. 6. Raw Results - and what to do with them. Having spent 2 years performing the experiment described above, what output do we get? When the animals are sacrificed, they are dissected and a whole list of tissues examined, both macroscopically and microscopically. All lesions, whether related to cancer or not, are noted down and usually (nowadays) recorded in some sort of computer database. The pathologists performing the examinations usually use some sort of standardized nomenclature for what they observe -- for example, the National Toxicology Program uses a modified version of the Systematized Nomenclature for Pathology (SNOP). Other information about individual animals is also recorded -- such information as where they came from, which cages,they were kept in, when they died ( e.g. if they died naturally, or were sacrificed at the end of the experiment, or sacrificed earlier because they clearly would not survive), and so fortli„ The outcome is that for each animal, we have a list of the lesions affect:ing them when they died. An example of a condensed listing of just -8-

is then a binomial sample with this probability. In practice, we don't know what the dose-response relationship is - we wish to estimate it from the results. But we assume that we know the SHAPE of the dose-response relationship (specified by a mathematical formula), so that all that is required is to estimate some PARAMETERS in the mathematical formula. For example, the E.P.A. uses a dose-response relationship of the fo.rrx : p == 1 - exp( - (q0 + q1.d + q2.d2 + .... + qk-1•dk-1)) when there are k doses in an experiment, where p is the lifetime probability of tumor at dose d. It is usual to use a maximum likelihood tec:h.nique to estimate the various parameters q0, ql, q2, ••• qk-1> given the observed numbers of animals with tumors and the numbers of animals examined at each dose. In cases where there is appreciable early mortality in the experiment, so that the observed numbers of animals with tumors are likely to be underestimates of what would have been observed at the end of a perfect experiment, one can make modifications to the dose response relationship, just as one can make life-table adjustments to standard statistical tests. One'technique used is to modify the dose response curve to explicitly include length of life, using the idea that probability of tumor is likely to increase with a power of age (see page 1) . p = 1 - exp{ - (q0 + q1.d + q2.d2 + .... + qk-1•dk-1) (t/j,)n} where t is the age at death, and L is a standard lifetime. The parameter n can either be fixed at some reasonable value (in the range 2 to 11), or estimated from the experimental results. This technique suffers from the same limitations as the usual modifications to the standard statistical tests -- one has to introduce additional assumptions in order to apply it. In this case, one has to decide whether the tumors were a cause of death, or simply incidental. An alternative technique used when there is early mortality is to estimate the age dependence directly from the data, using a (so-called) non-parametric technique. This approach has been used to assemble a large database of comparable analyses of animal bioassays. T',7is methodology has taken the raw results of the animal experiment, and summarized them in the form of a dose-response curve with known parameters. It is also possible to estimate how uncertain one is about a given parameter, using the same maximum likelihood techniques used to obtain point estimates of them - indeed, one can plot the uncertainty distri'_bution for any of the parameters. For example, for the parameter ql (which will turn out to be the one of interest), we can plot the probability that ql lies below any given value:

Risk Analysis in Environmental and Occupational Health September 1, 1987 Uncertainties in Predicting Human Risks Edmund Crouch 1. Background Information It is useful to bear in mind a few sobering facts about total populations at risk, and the normal total risk of death and of dying of cancer. For the U.S., the total population is about 240 million, while the annual number of deaths is about 2 million per year and the annual number of cancer deaths is about 400 thousand. These figures imply an annual average total risk of death of about 10-2 (1 percent per year), and a lifetime risk of cancer of about 0.2 (20 percent, or 200,000 x 10-6), estimates you can obtain simply by dividing one figure by another. Of course, simply dividing one by another is not a particularly accurate way of computing such estimates -- one should do the correct thing and take the age structure of the population into account, and the var:Cation of risks with age, and so on. But even when you do precisely that, the average lifetime risk of cancer comes out to be about 20 to 25 percent. We can expect this figure to get higher as the expectation of life increases, and as other causes of death are eliminated (assuming -- pessimistically --that most cancers cannot be eliminated). It is mainly the increase in expectation of life which has made cancer such a prominent cause of death in the (historically) recent past, because cancers tend to be diseases of old age. For many cancers it is found that the death rate varies as a power of age:- rate - agen where the exponent n is in the range 4 to 11. For such cancers, this pattern seems to hold over the age range from about 30 to 65. At lower ages the rates tend to,be very small but almost independent of age (and the cancers may be completely different diseases in youngsters), while at higher ages the reported death rates are lower than would be predicted by this sort of formula - and in some cases the reported death rates are actually lower for old enough groups. It is unclear whether these reductions in death rates in the elderly are real, or are simply due to a difference in the accuracy of diagnosis and reporting. It is also possible that the reduction in reported death rates is real, but is due to the winnowing out of the population of those who are susceptible to these particular cancers, leaving a core of more resistant individuals. The major exceptions to the power law variation of death rate with age are the cancers which are known to be hormonally dependent ( e.g. breast: cancer), or are highly curable ( skin cancers ), or in which the natural progression is altered by interventicn ( e.g. a high proportion of women have had hysterectomies by age 65, so that they cannot be at

Epidemiology ia Risk Assessment for Regulatory Policy 1167 ArknoKlydgrrrrents-Supported by NIH Grant CA 23214 and by grants to the Society for industrial and Applied Mathematics tnstitute for Mathematics and Society, from the Sloan Foundation, the Environmental Protection Agency, and the National Science Foundation. The author is grateful to Patricia Ford and Joseph B. Keller for advice and assistance in the preparation of the manuscript. M REFERENCES 1. Lili enfeld DE, Lilienfeld AM: Epidemiology 101: The new frontier. Int J Epidemiol 7: 377-380, 1978 2. Lilicnfeld AM. Lilienfeld DE: Foundations of Epidemiology, 2nd edn. New York: Oxford University Press. 1980 3. Stidman H, Mushinski MH. Gelb SK, Silverburg E: Probability of eventually developing or dying of cancer-United States,1985. CA-A 35: 36-56, 1985 4. Halpern J, Whittemore AS: Issues in analyzing cohort data: application to lung cancer mortality in uranium mirners. J Chron Dis In press 5. Kahn HA: The Dorn study of smoking and mortality among US veterans: report on 81 years of obsrrvations. In Epidemiologic Approaches to the Study of Cancer and Other Chronic Diseases, Haenszel W(',Ed.). NCI Monogr No. 19, 1966. pp. 1-126 6. G:orge AC, Breslin AJ: The distribution of ambient radon and radon daughters in residential buildings in the New Jersey-New York area. In The Natural Radiation Environment III, Gessel TF. Lowder WM (Eds). Washington. DC: Technical Information Center-U.S. Department of Energy. 1980 7. Doll R: Relevance of epidemiology to policies for the prevention of cancer. J Occup !11ed 23: 601-609. 1981 8. Clremicals attd.Induserial Processes Associated Mith Cancer in Humans. IARC Monogr. Vols 1-20. Lyon, Frar ce 9. Federal Register 45(15), 22 January. 1980. p. 5200 10. Hoover RN, Strasser PH: Anificial sweetness and human bladder cancer-preliminary• results. Lancet 8173: 873-840, 1980 11. Kessler 11, Clark JP: Saccharin, cyclamate and human bladder. JAMA 240: 349-355. 1978 12. ViP,al Sta6sft of the US 1960-1979, Vol. I1-9tortality, Part A. Hyattsville. Md: National Center for Health Statistics 13. Silverberg E: Cancer statistics. CA-A 35: I9-35, 1985 14. Dull,R, Peto R: The causes of cancer: quantitative estimates of avoidable risks of cancer in the Unites States - toda;y• J Natl Cancer Itut 66: 1191-1308, 1981 IS. Calkins DR, Dixon RL, Gerber CR, Zarin D, Ornenn GS: Identification, characterization and control of po.lential human carcinogens: A framework for federal decision-making. J Nati Cancer Irsst 64: 169-176, 19130 ~ gr 16. Committee on the Instttutional Means for Assessment of Risks to Public Health. Risk Assesstnent in the Federal Govettune®t: ManaeinY the Process. Washington D.C.: National Academy Press. 1983 17. Whittemore AS: Facts and values in nsk analysis for environmental toxicants. Risk Anal 3: 23-34, 1983 18. Kolata GB: Love Canal: False alarm caused by botched study. Science 208: 1239-1282. 1980 19. Petitr,i D: Studying potential reproductive hazards. In En.irortmental EpideYniology: Risk Assessment, Prentice RL, Whittemore AS (Eds). Philadelphia. Penn. SIAM Publications. 1982. pp. 49-62 20. Roht LH, Vernon SW, Weir FW, Pier SM. Sullivan P. Reed U: Community exposure to hazardous waste, disposal sites: Assessing reporting bias. Am J Epidetnfol 122: 418433, 1985 21. Sun M: EPA said to bar official from Meeting. Science 214: 629, 1981 22. Sun 14: A firing over formaldehyde. Science 213: 630-b3t, 1991 ' 23. Lev~in R: Government;industry brain tumor risk. Science 210: 996-997. 1980 24. Occupational Safety and Health Administration: Estimates of the Fraction of Cancer in the United States Related to Occupational Factors, filed in Congressional Record (September 15. 1978) 25. Marshall E: The politics of lead. Science 216: 496. 1982 26. Soskolne CL: Epidemiologic research, interest groups and the reviev% process. J Pubt Health Policy 6: 173-184. 1985 ` 27. Gould SJ: Morton's ranking of races by cranial capacity. Science 200: 503-509, 1978 28. Serllittg TD: Filtering information about occupation, smoking and disease. J Chron Dis 37: 227-230, 1984 29. Verhalen RD: Guidelines for documentation of epidemiologic studies. Am J Epidemiol 114: 609-613. 1981 30. Labarthe DR: Commentary: The Interagency Regulatory Liaison Group "Guidelines for Documentation of Epidemiological Studies". Am J Epidemiot 114: 614418. 1981 31. Bazelon DL: Science. technology and the court Science 208: 661. 1980 - 32. Prel,n,ancy Outcomes in Santa Clara County 1980-1982. Reports of two epidemiologic studies. Berkeley. CA: _Califomia State Department of Health Sen•ices, 1985 33. Raposa T: Sister chromatid exchange studies for monitoring DNA damage and repair capacity after cytostatics in t•itro and in lymphocytes of leukaemic patients under cytostatic therapy. Mutat Res 57: 241-251, 1978 34. Evans HJ, O'Riordan ML: Human peripheral blood lymphocytes for the analysis of chromosome aberrations in mutagen tests. Mutat Res 31: 135-148. 1975 35. Hogstedt B. Gullberg B. Mark-Vandel E. Mitelman F. Skerf%ing S: Micronuclei and chromosome aberrations in bone marrow cells and lymphocytes of humans exposed to petroleum vapors. Heriditas 94: 179-187, 1981 36. Perera FP, Weinstein 113. Molecular epidemiology and carcinogen-DNA adduct detection: new approaches to studies of human cancer causation. 3 Chron Dis 35: 581-600. 1982

APPLIED OCCUPATIONAL AND ENVIRONMENTAL HYGIENE, Vol. 5, No. 7. pages 453 - 463 (July 1990). Notice of Intended Changes®Benzene Edttors Note: In anticipation of significant interest and to ensure reader awareness of the proposed revision of the Threshold Limit Value (TLV) for benzene, publication of this revised documentation is issued at this time and in advance of the publication of the Threshold Limit Values and Biological Exposure Indices for 1990-1991 booklet. The recommendation of the Chemical Substances TLV Committee received approval from the ACGIH Board of Directors and members in attendance at the annual ACGIH business meeting on May 16, 1990. The recommendation is that benzene be listed on the Chemical Substances TLV Notice of Intended Changes for 1990-91 at 0.1 ppm as a time-weighted average (TWA) with a Skin notation and designation as an Al carcinogen (confirmed human carcinogen). This proposed reduction for the adopted benzene TLV-1'ViA of 10 ppm and A2 carcinogen designation (sus- pected human carcinogen) will undoubtedly prompt spec- ulation as to the basis for the proposed revision. An esti- mated exposure to benzene of 238,000 U.S. workers and usage of more than 11 billion gallons of benzene per year are added incentives to publish the revised documentation at this time. The proposed revision will remain on the Notice of Intended Changes for a period of at least two years during which comment and substantive evidence for or aga;nst the appropriateness of the revised TLV is solic- ited by the TLV Committee. This publication of the documentation in Applied pro- vides an additional opportunity for comment. Benzene CAS: ?1-43-2 Benzol; phenyl hydride; cyclohexatriene; coal naptha C6H6 Skin TLV-1VJA, 0.1 ppm (0.3 mg/m3) Al-Confirmed Human Carcinogen TLV-TYP,, 100 ppm, 1946 TLV-TW4, 50 ppm, 1947 TLV-TWA, 35 ppm, 1948-1956 TLV-TNiA, 25 ppm, 1957-1962 TLV-Ceiling, 25 ppm, Skin, 1963-1976 TLV-TWA, 10 ppm,A2, Skin, 1977-presenh Skin notation deleted 1978 TLV-STEL, 25 ppm, A2, 1980-1987 TLV-TWA, ().1 ppm, Al, Skin: proposed 1990 DocumentaCion revised, 1990 Chemic:al and Physical Properties Benzene is a colorless, highly flammable, nonpolar liq- uid with an odor that is characteristic of aromatic hydro- carbons. Benzene can be sgpplied as industrial grade, ni- tration grade, or refined. Phvsicochemicat properties of reagent grade benzene include: Molecu lar weight: 78.11 Specific gravity: 0.87865 at 20°C Melting point: 5.5°C Boiling point: 80.1°C Vapor pressure: 75 torr at 20°C Closed cup flash point: -11.1°C Autoignition temperature: 562°C Flammability limit in air: 1.5-8.0 vol% Odor threshold: 12 ppm Saturated air at 25°C contains 120,000 ppm Solubility: 0.180 g/100 ml water at 25°C; miscible in all proportions with carbon tetrachloride, ethanol, chlo- roform, diethyl ether, carbon disulfide, acetone, gla- cial acetic acid, and oils. Major Uses and Sources of ®ccupational Exposure At one time, benzene was an important solvent. espe- cially for inks, rubber, lacquers, and paint removers. At present, such uses are minimal; most benzene is consumed in the chemical industry as a raw material for numerous organic chemicals and in plastics manufacture. It is found in gasoline from trace amounts to as much as 30 percent in some countries (U.S. average, 1-3%). Total benzene usage exceeds 11 billion gallons per year; i> and it is es- timated that 238,000 employees in U.S. petrochemical plants, petroleum refineries, coke and coal operations, tire man- ufacturers, bulk terminals and plants, and in truck transport are exposed to benzene.(2) Benzene is a myelotoxicant, known to suppress bone marrow cell proliferation and to'induee hematologic dis- orders in humans and in animals. Signs of benzene- induced aplastic anemia include suppression of leukocytes (leukopenia), red cells (anemia), platelets (thrombocyto- penia), or all three cell types (pancytopenia). Classic symp- toms include weakness, purpura, hemorrhage, pancyto- penia, and aplastic anemia. Animal Studies Subchronic When Sprague-Dawley rats and CD-1 mice of either sex were exposed by inhalation to benzene at 1, 10, 30, or 300 ppm, 6 hours per day, 5 days per week for 13 weeks, treatment-related pathology was observed in the high dose (300 ppm) groups of both species.(3) In mice, hematologic changes included decreased hematocrit, total hemoglobin, erythrocyte/leukocyte count, platelet count, and mye- loid:erythroid ratio. In rats, decreased lymphocyte count and a relative increase in neutrophil count were the only exposure-related dinical change. Histopathological changes APPL OCCUP. ENVIRON. HY6. 50 • JULY 1990 1047-3Y2)V9U/D607-4535215/11 © 1990 AIH 453

risk of uterine cancers thereafter). With this age variation of risk of cancer understood, we can now oversimplify again and quote a lifetime average annual risk for cancer, obtained simply by dividing the lifetime risk by an average lifetime of about 70 years. This give an average annual risk of about 2 - 3 x 10-3. Notice that we are here averaging over a lifetime -- the figure is not meant to imply that the risk is the same in each year of life -- we have just seen that it varies drastically with age. When discussing the risks of carcinogens, the same caveats have to be borne in mind. We usually attempt to estimate a lifetime risk, but may express this, for comparison purposes, as an annual average risk. For an individual exposed continuously to a carcinogen, we would expect that: the risk of cancer increases with age in a fashion similar to the risl: of other (naturally occurring) cancers. There is another reason also for quoting an annual average risk obtained by averaging over a lifetime. When estimating risks of carcinogens, one is often interested in the response of a population to exposure to the carcinogen. In this case, one should strictly (if it were possible) estimate what the effects at all future times would be on individuals of different ages at the times of exposure. The effects at all. future times on the whole population would then be an average over the effects on all the.individuals in the population (who were of different ages at the times of exposure. Thus, to obtain an estimate of the effects on a population, one implicitly performs an average over the age groups present in the population. If the population were stationary (and if certain other conditions were fulfilled) this average would be the same as an average over a lifetime. This explains the usefulness of a lifetime average, since one may argue that the differences between population and lifetime averages are small compared with other uncertainties inherent in all the procedures we will describe later. The preceding discussion must be considered only a heuristic argument for accepting a lifetime average as being useful. In practice, people will be exposed at different ages, and for varying periods, to different amounts of carcinogens. All these differences (and many more besides) will affect the probability of carcinogenesis for each of them. 2. Knokm Htuman Carcinogens There is now good evidence that human exposure to certain materials can,, under certain conditions, increase the rate of human cancer. The evidence comes from various types of epidemiological investigation (discussed in other talks in this course). In all cases, exposures to these materials has been high, compared with population exposures, and the population exposed has been small compared with the total U.S. population. The resultant risks to those exposed has been substantial. The following table indicates a few of these materials, and the types of cancer which have been caused in humans by exposure to them. -2-

predicted by a linear extrapolation from doses above the threshold, and possibly that the excess incidence of cancer is literally zero below such a threshold ( excess, here, means excess over the background occurrence of cancer). Some of the evidence for such mechanisms comes from observation of the dose-response curves in experimental situations -- the experiments on saccharin provide a good example. However, there is still the possibility that a linear mechanism may still oper-4te at low enough doses, and so any human risk assessment has to take that possibility into account. 5. The Standard Animal Test. The requirements for a "standard" animal test are quite severe. The animals involved have to be as similar to humans as possible -- in metabolism, in being omnivorous, in their sensitivity to chemicals, for example -- yet as different as possible in their life span and cost of upkeap (so that we can get results in a reasonable time at a reasonable cost). In practice, there is little option but to use standard laboratory animals. The usual choices are rodents -- rats and mice; with occasional tests being performed on golden hamsters or guinea pigs. Other animals (e.g. gerbils) have been proposed, but for now the experience built up in handling laboratory rodents is a strong incentive for continuing their use despite certain known disadvantages. Any change would now have to be done gradually, and with much cross checking with previous results. ^:t is now standard to require tests to be performed in at least two specLes (practically always rats and mice) and on both sexes, in case one or the other species or sex is peculiarly resistant to the material under test. A compromise has to be made over the number of animals to test:. It would be desirable to have as many as logistically possible, to increase the statistical sensitivity of the experiment; but as few as possible to minimize the costs of testing (since there is always another material to test). The current recommendation is for at least 50 per group, of similarly treated animals. There is a similar trade-off between costs and the number of dose levels to test in a given experiment. The current recommendation is to havee at least three dose groups -- an undosed group (the control group), a group tested at the maximum tolerated dose (MTD) of the material under test, and the third group tested at some intermediate dose (usually 1/4 to 1/2 of the MTD). The MTD of a material is roughly defined to be as much as possible, but not enough to kill off the animals early or to cause too large other overt effects (like loss of weight). The reason for using it in these exper:iments is to increase the sensitivity, on the basis that giving more of something is more likely to produce a response if any response if going to happen at all. The sensitivity has to be as high as possible, since the observable responses are of the order 10-1 (10%) while the risks of interest are of order 10-6 (100,000 times smaller). The alternative way of increasing sensitivity is to increase the number of animals tested (within reason), but this only increases sensitivity in proportion to the square root of the numbers tested, while increasing -7-

simplest analysis based on "end-of-fife" data, without having to worry too much about the age dependence (this should always be backed up by further analysis, of course). TIME ON STUDY (WEEKSI 0 ~.~.. ................ . ------- + or .......... ....... ^ -------'t•---~'tr --------- ~a oso --~ --- oao 4 I o- 0 70 i ~ ~ . o.w ~ i I osa ' i ' I o .o l . f----- --------ffii' I FEMALE RATS l o.m p Lwrn i EATED CONTRO 005E 0 lOW f ' . 0 NIGHDOSE ~ ~.oo Is so s w n 90 tos tm TIME ON STUDY (WEEKS) Figure 2. Survival Curves for Rats Exposed to Air Containin®1, 2-Dibromoethana Tumors were found in many tissues. A summary of those tissues where more than 5% of the animals in any group were found with tumors is (for female rats): Control Low High Subcutaneous tissue: fibroma 0/50 0/50 3/50 Subcutaneous tissue: fibroma or fibrosarcoma 0/50 0/50 4/50 Nasal 0avity: Carcinoma, NOS 0/50 0/50 25/50 Nasal Cavity: Squamous cell carcinoma 1/50 1/50 5/50 Nasal C Davity: Adenoma, NOS 0/50 11/50 3/50 Nasal Cavity: Adenocarcinoma, NOS 0/50 20/50 29/50 Nasal E~avity: Adenomatous Polyp, NOS 0/50 5/50 5/50 Nasal Cavity: Papillary Adenoma 0/50 3/50 0/50 Nasal Cavity: Adenoma,NOS; Carcinoma, NOS; Adenocarcinoma,NOS; Papillary Adenoma 1/50 34/50 43/50 Adenomcatous polyp,NOS; and Squamous cell Carcincma Lung: Alveolar/Bronchiolar Carcinoma 0/50 0/48 4/47 Lung: Alveolar/Bronchiolar Carcinoma or Adenoma 0/50 0/48 5/47 Hematopoietic System: All leukemias 6/50 7/50 1/50 20

'fABLE E3. ANALYSIS OF PRIMARY TUMORS IN MALE MICE (Continued) Vehicle Control 500 mg/kg 1,000 mg/kg Circulatory System: Hemangiosarcoma Overall Rates (a) 4/50 (8%) 3/49 (6 r) 1/50(2~,i) Adjusted Rates (b) 10.1% 8.8(~i 2.6(/r Terminal Rates (c) 3/38 (8 c) 2/33 W/'i) 1/39 (3 r) Life Table Tests (d) P=0.1301 P=0.559\ P=0.169 N Incidental Tumor Tests (d) P=0.097\ P=0.408 \ P=0.176N Cochran-Armitage Trend Test (d) P=0.134N F';sher Exact Tests P=0.512 \ 18 11\ P=O Circulatory System: Hemangioma or Hemangiosarcoma . Overall Rates (a) 4/50 (8 c) 4/49 WID 1;50 (2c;(') Adjusted Rates (b) 10.1% 11.Kr 2.6 ii Terminal Rates (r) 3/38 (8Si) 3/33 (9 r) 1139 (3Si) Life Table Tests (d) P=0.142N P=0.579 P=O. 169N Incidental Tumor Tests (d) P=0.1 I01 P=0.573 \ P=0.I76\ Cochran-Armitage Trend Test (d) P=0.1471\ Fisher Exact Tests P=0.631 P=0.181 N Liver: Adenoma Overall Rates (a) 0/50 (0 c) 5-49 (10(/i) 13, 50 (26r/i) Adjusted Rates (h) 0.0cli I3.0ii 33.3cii Te:rminal Rates (c•) 0/38 (09j) 31'33 (9i0 131'39 (33(ii) Life Table Tests (cl) P<0.001 P=0.030 P<0.001 incidental Tumor Tests (d) P<0.00I P=0.023 P<0.001 Cochran-Armitage Trend Test (cQ P<0.00I 1=isher Exact Tests P=0.027 P<0.001 Li,ver: Carcinoma Overall Rates (a) 10, 50 (20~-j) 14, 49 (29(,*i ) 12 50 (24 ~,(') Adjusted Rates (h) • 24.3~~ 35.9c;(' 25.lic.i Terminal Rates (c•) 7138 (18Sc) 9 33 (27(,~(') 5 39 (13c;(') Life Table Tests (cl) P=0.427 P=0.183 P=0.463 Incidental Tumor Tests (cl) P=0.536 P=0.379 P=0.548\ Ccchran-Armitage Trend Test (d) P=0.363 Fisher Exact Tests P=0.224 P=0.405 Lirer: Adenoma or Carcinoma Overall Rates (a) 10 50 (20~-j) 18 49 (37;(-) 23 50 (46(;i) Adjusted Rates (h) 24.3cii 45.1 !'( ' 49.8c(' Tcrminal Ratcs (c) 7 38 ( lBSr) 12 33 (361.'1) 16 39 (41c;~) Life Table Tests (cl) P=0.013 P=0.042 P=0.0)4 Incidental Tumor Tests (d) P=0.009 P=0.098 P=0.0 ) 9 C'o,.hran-Armitage Trend Test (d) P=0.004 Fisher Exact Tests P=0.052 P=0.005 Forestomach: Squamous Cell Papilloma Overall Rates (a) 3 49 (64'i) 3 48 (6~'(') 9 49 (18Si ) Adjusted Rates (h) 7.9~'j 9.1 c'i 23.1"'1 Terminal Rates (r) 3 38 Wi) 3 33 (9~i) 9 39 (23~(') ~ Life Table Tests (cl) P=0.038. P=0.597 I'=0.065 ~ Incidental Tumor Tests (cl) P=0.038 P=0.597 P=0.065 ~ Cochran-Armitage Trend Test (cl) P=0.034 ~ Fisher Exact Tests P=0.65 I P=0.060 ~ ~ ~ ~ qw Benzyl Acetate ~ ~ I ~

to 10 3 2 ~-t LisM~C u d ~ ~ ~ 10 1 -3 10 . 104 tci~~ 162 io1 1 101 402 103 POTENCY IN RAT (mq'Kg d) ®enn0~n~ + / 2 f 1 ~ Lr~ ` _.L_ > > ' 10 '101 Y ~ -3 10 -2 1 -5 -4 -3 -2 -t 0 Log„(o) B6C3F1 / / (p= 0 025) / I + . , . + J_ + ~ T r _1_ f f t f ~ 10-3 10 Z 10 1 1 101 10 2 10 3 POTENCY IN MOUSE (mg" kg d) 5 -4 -3 -2 -1 0 1 2 3 og„(g): B6C3F1 (p= 0.025) Iq

Now what do we do with this estimate? That depends on the app:Lication, but we will assume that we wish to make a"tTNIT RISK" est::mate for humans from it - that is, estimate the lifetime risk to a hwnan exposed to 1 microgram/m3 of dibromoethane for life. There are several extrapolations required. First, the animals were dosed for a lifetime, but not continuously. Correcting for continuous exposure introduces a factor of 7/5 x 24/6 (for days/week and hours/day) - but notice the subtle assumptions being made here, that it is average exposure that matters (and not peak exposure, for example). Now we estimate that a female rat will suffer and increased lifetime risk.of about 0.48 per ppm in the air (we assume that we are talking about such low doses that the excess risk is small). 1 ppm for 1,2-dibromoethane corresponds to about 7.6 mg/m3 (one would estimate a lit:tle higher from the perfect gas laws), or 7600 microgram/m3, so that the increased lifetime risk to a female rat exposed continuously to 1 migrogram/m3 is about 6.3 x 10-5 What about humans? We saw before that the assumption made was that humans are just as sensitive as animals - i.e. they suffer equal lifetime risks - if exposed at doses which are equal on an (am.ount)/(surface area) basis. Now it turns out that, approximately, equal concentrations in air lead to exposures which are equivalent on this basis, provided the species under consideration absorb about the same amount from the air they breathe. Thus the extrapolation to humans is simple in this case- one simply takes the same value for humans - a "UNIT RISK" of about 6.3 x 10-5 (i.e. that is the lifetime risk from cont:inuous exposure to 1 migrogram/m3 of dibromoethane in the air). It may be desired to estimate from this the effect on humans of ingestion of dibromoethane. In this case there are actually other bioa:>says in which dibromoethane was fed to animals under various condb:tions, but suppose that we have to make some estimate from the inhalation data. The "standard" human inhales, on average, about 20 m3 of air per day, and so inhales about 20 microgram/day of contaminant from air contaminated with 1 microgram/m3. If we assume that 100% of this contaminant is absorbed, the human's daily dose is 20 micrograms/day, or about 20/70 microgram/kg-day (as a fraction of bod,,N,eight), or 2.9 x 10-4 mg/kg-day in the conventional units used. This results in a risk of about 6.3 x 10-5, as detailed above, so that the potency is just the ratio of these -- 0.22 (mg/kg-day)-l. These short outline calculations have made several assumptions which require examination in any particular case. We have not looked at all the bioassay results, so one cannot expect that the numbers obtained here will correspond with what anybody e1se,.who has done a more thorough job, will obtain -- they are placed here in order to show in outline what is done. In practice, one has to decide that the tumor site and type combinations are appropriate for combination in the animal species. That these tumors are relevant end points for estimating the probable effects on humans. That the route of administration, and method of adninistration are reasonable to produce results that may be extra:polated to humans. And a myriad of other details which have only been Lightly touched upon, or completely omitted, in this sketch. s -18-

the. cancer-related lesions is appended. From such listings, we can perform various analyses and statistical tests to see whether the rate off cancer was increased at any site or for any type of cancer. The simplest sort of analysis can be performed if all the animals survived for the whole length of the experiment -- and in practice the same sort of analysis is performed provided a reasonable fraction survived that long and provided there were not too many early deaths. In that case, we can simply list the dose groups and the numbers of animals with tumors compared with the total number of animals examined; Dose Number Number with Examined Tumor ---------------------------------- 0 (control) 10 50 0.5 MTD 25 50 for example MTD 30 50 However, things are not this simple. Similar results are o Many different sites ( See below ) o Many different tumor types ( for examples ) o Combinations of these available for To dletermine whether the rate of cancer has been increased involves comparing the proportion with tumor in the control group with the proportion with tumor in the dosed groups, and deciding whether there is a significant increase in any dosed group(s). The choice of which sites and/or types of tumors to combine before performing such statistical tests can be difficult. Generally, various grades of tumors (nodules, adenomas, carcinomas) may be combined for any given site. Table 2 gives an example of the sort of combination and testing which is performed. In addition to the simple numbers of animals with tumor, there is additional information available which may be used in more complicated cases. The date of death of each animal is recorded, and may be taken into account in time-adjusted analyses of tumor incidence and in the life-table tests mentioned on the appended material. For risk assessment purposes, it is necessary to make various assi.unptions about the behavior of animals in experiments like these. For example, it is assumed that: o Animals are affected independently (a tumor in one animal has no effect on any other animal). o Animals are equally likely to be affected a Each animal receives the same dose .................. It is assumed that cage effects, littermate effects, the effects of heating, lighting, stress etc. are either not present, or are randomized among all the animals in such a way that there will be no effect on the final analysis. With such assumptions, the probability of an animal having a tumor is related to the dose'by some sort of dose-response relationship, so that at any given dose this probability can be computed. The observed results, a number of animals with tumor out of a larger number examined, -9-

There are several extrapolations required. First, the animals were dosed for a lifetime, but not continuously. Correcting for continuous exposure introduces a factor of 7/5 x 24/6 (for days/wOek and hours/day) - but notice the subtle assumptions being made here, that it is averagse exposure that matters (and not peak exposure, for example). Now we estimate that a female rat will suffer and increased lifetime risk of less than 7/5 x 24/6 x 8.6 x 10,2 = 0.48 per ppm in the air (we assume that we are talking about such low doses tha,t the excess risk is small). 1 ppm for 1,2-dibromoethane corresponds to about 7.6 mg/m3 (one would estimate a little higher from the perfect gas laws), or 7600 µg/m3, so that the increased lifetime risk to a female rat exposed continuously to 1 µg/m3 is less than about 0.48/760G = 6.3 x 10-5. What about humans? We saw before that the assumption made was that humans are just as sensitive~ as animals - i.e. they suffer equal lifetime risks - if exposed at doses which are equal on an (arnount)/(surface area) basis. Now it turns out that, approximately, equal concentrations in air lead to exposures which are equivalent on this basis, provided the species under consideration absorb about the same amount from the air they breathe. Thus the extrapolation to humans is simple in this case -one simply takes the same value for humans - a "UNIT RISK" of less than about 6.3 x 10" (i.e. this is our overestimate for the lifetime risk from continuous exposurO to 1 µg/m3 of dibromoethane in the air). It may be desired to estimate from this the effect on humans of ingestion of dibromoethane. In this case there are actually other bioassays in which dibromoethane was fed to animals under various conditions, but suppose that we have to make some estimate from the inhalation data. The "standard" human inhales, on average, about 20 m3 of air per day, and so inhales about 20 µglday of contaminant from air contaminated with 1 µg/m3. If we assume that 100% of this contaminant is absorbed, the human's daily dose is 20 µg/day, or about 20/70 µg/kg-day (as a fraction of bodyweight), or 2.9 x 10'4mg/kg-day in the conventional units used. This results in a risk of about 6.3 x 10-5, as detailed above, so that the potency is just the ratio of these - 0.22 (mg/kg-day)-, These short outline calculations have made several assumptions which require examination in any particular case. We have not looked at all the bioassay results, so one cannot expect that the numbars obtained here will correspond with what anybody else, who has done a more thorough Job, will obtain - they are placed here in order to show in outline what is done. In practice, one has to decide that the tumor site and type combinations are appropriate for combination in the animal species. That these tumors are relevant end points for estimating the probable effects on humans. That the route of administration, and method of administration are reasonable to produce results that may be extrapolated to humans. And a myriad of other details which have only been lightly touched upon, or completely omitted, in this sketch. 22

~ Cambridge Environmental Inc 58 Charles Street Cambridge, Massachusetts 02141 617•225•0810 617•225•0813 FAX Edmund A.C. Crouch, Ph.D. Senior Scientist

curves involved (see below). One can extrapolate from high dose to low dose using the ANIMAL dose-response curve, and then extrapolate to humans (dashed lines), o,r extrapolate to humans at high doses and then use a HUMAN dose-response curve to extrapolate to low doses. We have seen how to estimate the parameters of the (high dose region of) the animal dose-response curve. In practice, the same curve (with the same parameters) is is used to extrapolate to low doses, by building ir.to the mathematical structure of the dose-response curve all our assumptions about low dose behavior. How is this relevant for estimating human risk? Consider a generalized situation in which we wish to estimate the response (R) of humans to some dose (D) of material, when there is a response (r) in some animal at dose (d). Notice that nothing implies that r, R measure the same sort of response - they could be completely different (r could be acute toxicity to the lung of a mouse, R could be skin rashes in hwnans). Similarly, the dose measures d, D may be completely different. In the case immediately at hand, r is the lifetime probability of tumor in animals, and d is a dose as measured in the animal experiment. There are other cases of practical importance however - r might be some measure of response (such as number of revertants per culture dish) in a mutagenesis bioassay, with d the dose applied to each culture dish. Animal Human Response: r R (lifetime probability of tumor, p) Dose measure: d D (as used in experiments) Dose-response curve: r= f(d; a,b,c,...t ) R= F(D; A,B,C,....T ) [p = 1 - exp{-(q0+q1.d +...)} ) What is required is some connection between the parameters a,b,c,... of the animal dose-response relationship and the parameters A,B,C,... of the human dose-response relationship. These parameters presumably include those mentioned in section 6, and I have explicitly included age amongst them. Given such a connection, the extrapolation to humans of the results in the animal studies is perfectly straightforward. The probLem lies in finding the connection. Once such a connection is found (by whatever means) we have the methodology for the two extrapolations required. Notice the difference between what is done in the two distinct pathways of extrapolation ment:ioned above: In the first, the shape of the dose-response curves are examined, and it is decided how they may be (separately) extrapolated to low doses. Then some relationship is postulated between the parameters of the dose-response curves at low doses (it has to be postulated, since nothing can be measured at such low doses). One potential advantage of this approach is that the animal dose-response curve could be measured,, in principle and by heroic experimentation, down to lower response rates than usual (and this has been done in some cases) - allowing greater confidence in this extrapolation to low dose. -12-

Had we chosen some other method of parametrization, it is quite possible thO required interspecies relationship between parameters would be much mor'e complicated. 8. Interspecies Comparison - Constant Relative Potency What is sought is a simple relationship between the parameters of dose-response relationships in different species. When it is assumed that the dose-response relationship includes a term linear in dose, there is a simple measure of the strength of a carcinogen - the carcinogenic potency (the slope of the dose-response curve at low dose). The simplest hypothesis is that for different species, the ratio of carcinogenic potencies is constant for different materials, so that if material A is twice as potent a carcinogen as material B in species 1, it will also be twice as potent as material A in species 2. This is the idea of constant relative potency, as applied to carcinogenesis, and it underlies the standard approaches to estimating human risks from animals. There is even some data which supports this idea! There have been several hundred bioassays performed simultaneously on rats and mice, and when the results of these are parametrized using a dose-response relationship which includes a linear term, we can estimate the potency in t:wo species for each material tested. Plotting the potency measured in rats versus the potency measured in mice for each material then gives the figure shown. Notice that each measurement is uncertain to greater or lesser degree, due to the relatively small numbers of animals tested. If the idea of constant relative potency were exactly correct, these points would all lie on a straight line on the figure - or at least, all would lie sufficiently close to such a line that the measurement uncertainty bars on each point would encompass the line. From the figures, one can see that: (1) On average, potency in one species is proportional to potency in the other species. (2) There is a large scatter of the points around the lines of exact prop,ortionality - a scatter bigger than would be expected from the measurement errors alone. A similar comparison can be attempted between the potencies measured in animal experiments, and those observed in humans. These cases have arisen in the past where humans have been exposed to materials before they were known to be carcinogenic. We can make use of other's misfortune to estimate how potent each such material is in humans, and compare with estimates obtained for mice and rats'~in laboratory experiments. In this case, the uncertainties are so large that little can be quantitatively states, although qualitatively the idea of const:ant relative potency does not seem to be disproved. A more recent and much more thorough study of comparisons between humans and animals has b~een carried out for the E.P.A. by Dr. Kenny Crump, and we can expect that to be published soon - I understand that conclusions are qualitatively similar.

In the second, some relation between the parameters of the dose-response curves is obtained at high doses (and this may be done experimentally, in principle, since at high doses the responses are measurable). Then it is decided how the human dose-response curve should be extrapolated to low doses. The advantage here is the possibility of direct comparison between species, albeit at high dose. The difference between these two logically distinct routes of extrapolation might be important in some circumstances. For cancer risk assessment based on animal carcinogenesis bioassays, however, the distinction is glossed over (one might even say, ignored), by the practice of assuming the same mathematical form for the dose-response curve in both humans and animals (or more generally, in all species), and interpreting the parameters in the same way for both compared species. In the general case, however, what is required is some sort of relationship between the parameters of the dose-response curves: Animal Human r= f(d; a,b,c...t) R- F(D; A,B,C...T) We need to be able to derive the parameters A,B,C... from the values a,b,c which can be estimated from experiments, and then use the human dose~-response curve to extrapolate to low doses. The practical approach is to seek parametrizations of the dose-response curve which result in the derivation of A,B,C... being simple given'a,b,c... Consider the case of acute toxicity, for example. It is found that the shape of the dose-response curve for acute toxicity, in which the response is death, is very similar for a large number of toxins and for many different species. There is, in this case, a t:hreshold-type dose-response curve which can be nicely parametrized by two values: the dose at which 50% of the animals tested can be expected to die (under suitable conditions), and the slope of the dose-response curve at this dose. The first parameter is known as the LDSO (the second has no special name). 1'7hy is this parametrization useful? If the LD50s of various mate:rials in one species are plotted against the LD50s of the same materials in another species, one finds approximate proportionality between them (the plot is a straight line). This can be expressed as, for example, LD50(rabbit) is proportial to LD50(mouse). Even more remarkable, it turns out that if the dose is measured in a suitable way, as (amount)/(surface area of animal), then approximately we actually have LD50(rabbit) - LD50(mouse) - LD50(other species) and it is this approximate equality which explains the utility of the LD5C1. The other parameter used in defining the dose-response curve, the slope of the curve at the LD50, is not involved in this relationship. -13-

The principal reason for this wide disparity between the EPA and the CS-TLV Committee may be explained primarily bv the underlying philosophical principles governing the two organizations rather than the technical differences be- tween the two methods. The CS-TLV Committee is gov- erned hy the principle that "Threshold limit values refer te~ airh<;rne concentrations of substances and represent cUnditiUns under which it is believed that nearly all work- ers ma%- he repeatedh• exposed day after day without ad- verse effect. Because of wide variation in individual sus- ceptibiliiv, however, a small percentage of workers may experience discomfort from some substances at concen- trations at or below the threshold limit; a smaller per- centage miy be affected more seriously by aggravation of a pre-exi.sting condition or by development of an occu- pational il;lness"11z> Use of the TLV for other purposes, such as community air standards, is specifically discour- aged by the Committee. Thus, the CS-TLV Committee rec- ommendations imply that there is a small degree of risk of occupat.ional illness to some workers who are more susceptible than others. The Clean Air Act, which in part governs EPA's approach to performing QRAs, is more philosophically conservative. The Act states that Primary Air Standards must protect the public heal¢h with an adequate margin of safety based on a review of air quality criteria which reflects the latest state of scientific knowledge about the pollutant. The require- ment for an "adequate margin of safety" is intended both to address inconclusive scientific and technical information and to provide a reasonable degree of protection against hazards that: research has not ,vet identified Recognizing that imposing zero emission for some substances would impose too heavy an economic burden on society, EPA has addressed tlle problem by proposing that the Best Avail- able Technology (BAT) be used to control carcinogens. If BAT controls leave an unreasonable residual risk, further controls will be considered ! 17) When making a quantitative comparison between the ACGIH and the EPA approaches, substantial agreement is found when classifying the relative potencies of these car- cinogens, but substantial disparity in the actual levels pro- posed or rec.omtnended Estimating lifetime cancer risks from occupational exposure at the ACGIH's 'ILV levels by using the EM's QRA model sometimes resulted in extraor- dinarily high risk estimates, 68 percent from exposure to 1,3-butadiene and 19 percent from exposure to chloro- form, which may reflect either limitations in the QRA mod- eling approach or the TLV safety factor approach. A safety /actor approach, such as that used by the CS- TLV Committee, is theoretically no more or less conserv- ative than a QRA approach which is linear at low doses and assumes no threshold. In practice, however, use of a safety factor of 5-10 or even 100-1000 is markedly less conservative than the QRA approach which determines an exposure level associated with a very small risk level such as 1/1oc'. This p:)int is illustrated for vinyl chloride in Figure 2 which compares the EPA unit risk factor for the upper confidence limit on the estimated human dose-response with the TLV and the results of animal bioas.tiays. With no attempt made to acknowledge the inconsistency produced by these differing methcxis, confusion and skepticism have resulted. Although there are strengths and weaknesses as- sociated with the approach of each group, it would seem that the CS-TL%' Committee could make a major contri- bution to fostering control of carcinogens in the workplace bk• reviewing any available EPA QRA, or comparable mod- eling data when it updates or establishes a new TLV for a conlirmed or suspected human carcinogen. When possi- ble, the CS-TLV Committee should also consider the results of studies that use more refined modeLs for QRA Being les,s constrained by the judicial-political climate than the regulatory agencies, the CS-TLV Cotnmittee should be bet- ter able to promptly adopt the most scientifically defensible extrapolation procedures available when a particular chemical is being studied in terms of recommended oc- cupational exposure values. Although many scientists remain skeptical about the pos- sibility of extrapolating the effects of carcinogens to low doses, a systematic evaluation of the results of these esti- mates in future editions of the TLV Documentation volume would help alleviate the confusion that now exists. Referenees 1. Spirtas, R.; Steinberg, M.; Wands, RC.; Weisburger, EK: Identification and Clmsification of Carcinogens Procedures of the Chemical Sub- uancesThreshold timit Value Committee, ACGQ-I. Am.J. PubLc Health 76(10):1232-1235 (1985). 2. American Conference of Governmental ItAtstrial Hygienists: Thresh- ' old Limit Values for Chemical Substances Biotogical Exposure Indices for 1989-1990. ACGIH, Cmcinnui, OH (1989). 3. U.S. Environmental Protection Agesx.y: Respondans brief in support of propevied findings, catcfusions and order at 63-64, in re-. Stevens tndtn•try, Inc (Consotidued DJT Hearings), Agril 5, 1972. 4. U.S. Environmental Pnoteaiat Agency: Respondents motion to de- termine wltether or not the negistratiotis of mhrx should be canceled cx amenafed; Attachment A, September 5, 1975. 5. U.S. Environmental Protection Agesxp: lnoerim Procedures and Guidelines for lealth Risk and Economic lmpoct Asewtents of Sas- pected Cardnogem Fed- Reg. 41:21402-21405 (1976). 6. Office of Science and Tecimoiogy: Policy on: Qtanial Carcittogenr A Review ofthe Scieneeand itsAssociMedPrindpals. Fed. Reg 50:10372- 10442(Febrtnty 19B5k 7. xatkx>at Research Courtcit: Risk Aneswmt in the Federal Govertr metu: iNaaagirg the Process. NatiotW t.ratlemy Ptess, Rashington DC (1983). 8. Cauletrnn, B.L; Zietn, G.E.: Corpotate lulhxtaoe on Threshold limit Vafte-s. Am. J. Ind. Med. 13:531-559 (1968)- 9. Andersen, M.E: Qu7¢vtiu&ve Risk AssewweM mtd Occvpational Car- cino~ Appl. ind Hyg. 3(l0}267-272 (1964 10. Comlie4d, J_ Carcinogenic Risc Assessz~ Science 19&693--b49 (1977). 11. M<x>Jgavkar, S.FL; Knudson, Jr., AG.: lAtxuion and Cancer: A Model for Hurmn Cucinoget>ests. J. Natl. Cx-scer Itffit. 66:1037-1052 (1961). 12_ American Conference of C'iowertutient IndssttW f•iygicnist: 74trestroid Limit Values and Biological Exposure Indices for 1967-1988. ACGIH, Cincinnati, OH (1967). 13. U S. Environmental Protection Agetxy: Integnted Risk Infam=ion System. EPN600V8-86/0321 Office of Health and Envim.urterual As- s~xr~t Us Environrnerual Protet.~tion Agency, Washagton, DC (1987). 14. tnternaticxtal Agency for Research on Csneer: Overall Evsluxions of Careinogenicity, Suppi. F. An Updatiag of IARC!<4onographs Volumes 1 to 42. JARC, Lyon, France (Mardi 1987). 516 APPL OCCUP. EiWIROM NYG. 50 . AUGUST 1990

Tumors were found in many tissues. A summary of those tissues where more: than 5% of the animals in any group were found with tumors is (for female rats): Control Low High Subcutaneous tissue: fibroma 0/50 0/50 3/50 Subcutaneous tissue: fibroma or fibrosarcoma 0/50 0/50 4/50 Nasal Cavity: Carcinoma, NOS 0/50 0/50 25/50 Nasal Cavity: Sqamous cell carcinoma 1/50 1/50 5/50 Nasal Caavity: Adenoma, NOS 0/50 11/50 3/50 Nasal Cavity: Adenocarcinoma, NOS 0/50 20/50 29/50 Nasal Cavity: Adenomatous Polyp, NOS 0/50 5/50 5/50 Nasal Cavity: Papillary Adenoma 0/50 3/50 0/50 Nasal Cavity: Adenoma,NOS; Carcinoma, NOS; Adenocarcinoma,NOS; Papillary Adenoma; Adenomatou s polyp,NOS; and Sqamous cell Carcinoma 1/50 34/50 43/50 Lung: Alveolar/Bronchiolar Carcinoma 0/50 0/48 4/47 Lung: Alveolar/Bronchiolar Carcinoma or Adenoma 0/50 0/48 5/47 Hematopoietic System: All leukemias 6/50 7/50 1/50 Hematopoietic System: Monocytic leukemia 6/50 5/50 1/50 Circulatory System: Hemangiosarcoma 0/50 0/50 5/50 Circulatory System: Hemangiosarcoma or Hemangiosarcoma, invasive 0/50 0/50 5/50 Liver: Neoplastic nodule 2/50 0/49 3/48 Liver: Hepatocellular carcinoma 0/50 1/49 3/48 Liver: Neoplastic nodule or Hepatocellular carcinoma 2/50 1/49 5/48 Pituitary: Adenoma, NOS 1/50 18/49 4/45 Pituitary: Chromophobe adenoma 20/50 0/49 0/45 Adrenal: Pheochromocytoma 3/50 1/49 0/47 Thyroid: C-cell Carcinoma 1/49 3/48 1/45 Mammary Gland: Adenocarcinoma,NOS 1/50 0/50 4/50 Mamnla.ry Gland: Fibroadenoma 4/50 29/50 24/50 Notice especially the various groupings which are employed - this is a matter of judgement. It is clear that the major effect is in the nasal cavity, but observe also the effect on fibroadenomas in the mammary gland., and the negative trend seen in the pituitary. Such negative trends are ignored. Using the combined'results in the nasal cavity, we fit the E.P.A. multistage model and find best estimates of: q0 = 2.699 x 10-2; ql = 6.876 x 10-2; q2 = 0; and obtain an upper confidence limit for ql of ql* = 8.6 x 10-2 in all cases using as doses the values 0, 10 and 40 from the experimental design. In fact, the earlier figure of a distribution of values for ql is taken from this example - you can read the probability of ql being less than any given value from that figure. -17-

44 Principles of Health and Safety in Agriculture Dis.. 128. 144, 1983. 22. Beck, B.D., Gerson, B., Feldman, H.A., and Brain, J.D., Lactic dehydrogenase isoenzymes in hamster lung lavage fluid after lung injury, To.ricol. Appl. Pharnwcol., 71, 59, 1983. 23. Beck, B.D., Brain, J.D., and Wolfthal, S.F., Assessment of lung injury produced by particulate emissions of space heaters burning automobile waste oil, in Inhaled Particles VI, Dodgson, J., Ed., British Occupational Hygiene Society, Edinburgh, Scotland, in press. 24. Hall, R.E., Crooke, M.W., and Barbour, R.L., Comparison of air pollutant emissions from vaporizing and air atomizing waste oil heaters, J. Air Pollur. Control Assoc., 33, 683, 1983. 25. Beck, B.D„ Feldman, H.A., Brain, J.D., Smith, TJ., Hallock, M., and Gerson, B., The pulmonary toxicity of talc and granite dust as estimated from an in vivo hamster bioassay, Toxicol. AppL Pharma• col., 87, 222, 1987.

9. Interspecies comparisons - practical and theoretical The measure of carcinogenic potency introduced above was roughly def.ined as the ratio of (excess tumor probability)/(dose), at low enough dose. For the E.P.A. model usually used in risk assessments: p = 1 - exp( - ( q0 + ql.d + q2.d2 + ... + qk-1•dk-1 ) } the corresponding measure is ql. When this dose-response relationship is used with real data, it is usual to use an "upper 95% confidence limit" estimate ql* of ql as the measure of potency, since such an estimate is always non-zero (while, for example, the maximum likelihood estimate is oft:e.n zero). The "upper 95% confidence limit" is with respect to the numerical uncertainties of the experiment only, and so this estimate of potency is in no sense an upper limit with respect to all the other uncertainties involved. To compare humans with animals, the approach taken is to use a similar dose-response relationship in both cases: Animal Human p = 1 - exp( - ( qO + ql.d +...)} p = 1 - exp( - (QO + Q1.D + ..)} and then the constant relative potency hypothesis suggests that Ql is proportional to q1, or to our estimate ql* of it: Ql = const. q1* where the constant depends only on which animals species is used. We expect the constant to be different for different animal species - it will presumably depend on how we measure dose, on the relative lifespans of animal and human, on relative metabolic rates, and a whole host of other factors. With enough experiments, we could measure the constant in this relationship - at least in comparing animal with animal, rather than human with animal - and (in theory) empirically determine how it vari'_ea with these factors. The graphs above suggest that the constant is not completely constant, but that there is some sort of random uncertainty built in (or at least, an uncertainty that we can treat as random), amounting to an average factor of about 5. If we are very lucky, it may be possible to find some way of measuring dose so that the constant in the above relationship is numerically equal to 1, so that the potency is equal in different species (up to the uncertainties) - just as it was possible to find such a measure in the case of the LD50• In practice, the E.P.A. assumes that the constant is exactly unity if the dose is measured as a (daily average amount)/(surface area of animal), by analogy with the LD50 case. (The'graphs shown above actually suggest that it would be better to assume an average factor of unity, with an uncertainty factor of about 5, when the dose is measured as a (daily average amout)/(bodyweight of animal)). -15-

..:::..: .. ... ..... -- . : .. . .. ... . .. ._. - _ --- - _ -.-- - - : . ..-...... . : : r. - - - - -- Rko -_ - _ - -- - _ -- - -___. -__- _- -_r~~. - - - - _~--- _~_- - - - -- -- - --_-- . , .. ...__: _. ._._..... .... -- -.. , -- - - ... .. ...~. ....-- ------ -- .. ---~- -- - --- __.... .. _ . =- ~ - - .- . - ----,... . ~~-_- .- ~--- -- -_ _.--- ..~-. - - - ~- =r:: - - - -- ---- =}=- --- --- _ -- -- -__ - - ~ -- :-{: - -~-- - --- - -- -- -- -- --- -+- - _ . . r - -- ,,.... . _. _ .. .._. _.~ ~ -- 1- - -- -- =- ~ = ~ - _ ~~ ~ _- -- - - - - - - _ - - - -- - _ - - ~-,~-- - - -_ - --~- --__~_=---~--~ - - - _ -~-- _ In particular, we can iind that value ql* such that there is 95% probability that ql < ql*. However, it is important to note that the uncertainty distribution so plotted contains only the uncertainty due to the numerical size of the experiment -- the uncertainty that arises because we used a small number of animals, instead of an infinite number. It does not include the uncertainties which must be present because of the shakiness of all our assumptions. 7. The Two Major Extrapolations The assumptions made so far have allowed us to parametrize an animal dose-response relationship, obtaining values for the parameters which are •,presumably reasonably appropriate for high doses. Strictly speaking, this parametrization of the dose-response curve only enables us to estimate the results we would expect to see at high doses in animals - the dose-resposne relationship can only be relied on to interpolate between high doses and perhaps to extrapolate a short distance outside the experimental range of doses. The problem now is to perform two extrapolations - from animals to humans, and from high dose to low dose: ~ Animal ~ Human ( ------------- ~ ~ ~ High Dose ~ Observed ....~.......:. ~ I I I V I ---------------------- ~------------------ ~------------ I I I I I Low Dose ~ ---------- ~----- >Required ( I I I ------------------------------------------------------- WGICALLY there are two distinct routes to follow in this extrapolation, since there are logically two distinct dose-response -11-

10. An example - 1,2 Dibromoethane As an example of the procedures usually adopted, let us look at the case of 1,2-Dibromoethane. What follows is by now means complete, but it indi'_cates the sort of analysis which has to be performed. This example is confined to analysing just one result out of many, in a single bioassay (of about 5). In practice, it is essential to look at all the results. The bioassay I have chosen was an inhalation bioassay in the National Toxicology Program series. A summary of the study design is: Initial number of Concentration Time on study animals ppm exposed observed (6 hrs/d, 5 d/wk) (weeks) ------------------------------------------------------------------------ Male rats control 50 0 0 104-106 :Low-dose 50 10 103 1 high-dose 50 40 88 0-1 Fema l.e rats control 50 0 0 104-106 low-dose 50 10 103 1 high-dose 50 40 91 0-1 And similarly for mice We wi11 look only at the results in female rats. First, their survival was not as good as might be desired in such an experiment, but the early mortality was probably due to the cancers appearing in the study, so it is acceptable - we can use (at least initially) the simplest analysis based on "end-of-life" data, without having to worry too much about the age dependence (this should always be backed up by further analysis, of course). TIME ON STUDY IWEEKSI a -'- ~. _ tt FEMALE RATS 'p UqTN[AT[OCOHT/Wl - O tOw00ti HWOON ~ N w ao n TIME ON STUDY IWEEKS) Firyro 2. SurviYaf Curva for Rstt Exposad to Air Cont.ining 1, 2-Dibramoath.m -16-

ror those reported in humans, with exposure at I ppm benzene and above inducing measurable cy[ogenetic dam- age(5')Women inhaling 1-9 ppm exhibited increased lym- phocyte chronnosome aberrations,CG3> and significant ele- varions in chrornasomal aberrations have been corroborated among workers inhaling benzene at mean concentrations less than 10 ppm.t64-66) Several quantitative human health risk assessments have been carried out in an attempt to define the concentrations of benzene in air that are associated with lifetime excess cancer risk,(z) but these methods are problematic, partic- ularly when attempting to extrapolate quantitative animal data to the hurnan. Notable has been their failure to in- corporate the differential metabolic disposition and known pharmacokinetic parameters for rodents(76) compared to human beings. The rodent carcinogenicity data support the designation of benzene as a known human carcinogen. Theoretical estimates of excess cancer risk can be cal- culated using any of a variety of statistical models, including the linearized "multistage" (which does not describe bio- logic initiatioNpromotion phenomena), the one-hit, Wei- bull, logit, or probit models; however, there is no current understanding of the biochemical mechanisms involved in benzene-induced leukemia and other cancers to show that any one of these methods is any more accurate than an- other. Because of the different assumptions that must be made for use of 1Ihe different models, the theoretical es- timates of excess cancer risk that result can differ by orders of magnitude. White et al.(118) used a linear, nonthreshold model to describe the benzene dose-response human carcinogenicity data and calculated that at 10 ppm benzene, 44-152 excess cases of leukemia per 1000 exposed work- ers would occur, and that at 1 ppm benzene, 5-16 such excess case would occur. The International Agency for Research on Cancer (IARC)(115) used a similar approach and published t~eoretical excess cancer risk estimates of 14-140 excess cases per 1000 individuals exposed at 10 ppm, and 1.4-14 excess cases among 1000 individuals ex- posed at 1 ppm. Crump and Allen(2,119) carried out quan- titative analyses of the epidemiologic data gathered by Rin- sky et al.,(96,97) On, el al.; 10°) and Wong et a~(1°4•I°5) After 45 years (working lifetime) exposure at 10 ppm benzene, Crump and Allen(I19) calculated 95 theoretical excess leu- kemia deaths per 1000 workers. Exposure at 1 ppm was calculated as associated with 10 theoretical excess leuke- mia deaths per 1CO) workers. Although such estimates have been preferred in the legal arena,(z) these methods remain the subjects of severe criticism.(z,12°,1-'1) Because of the acknowledged high quality of the epi- demiologic data,(') direct inspection of these data can pro- vide the basis for the benzene TLV. The Dow Chemical Company study(1')0> "demonstrates a significant fourfold increase in myelogenous leukemia for workers who had been exposed to av°rage benzene concentrations of about 5 ppm for an average of about nine years" and "two out of the four individuals in the studv who died from leukemia were characterized as having been exposed to average benzene levels below 2 ppm:'(-') The risk a.ssessment for henzene and leukemia is based on the human data. Rinsky et al.~9_) provided the most authoritative examination of the known odds of death from benzene-induced leukemia. For a worker exposed at av- erage daily benzene concentrations of 10 ppm for 45 years, the odds of death from leukemia were 290 times that of an unexposed worker. For an individual inhaling 1 ppm for 45 years, the odds of benzene-induced leukemic death were 1.7 times that of an unexposed worker. For an in- dividual inhaling 0.5 ppm for 45 years, the odds of ben- zene-induced leukemic death were 1.3 times that of an unexposed worker. Using these data, the odds of benzene- induced leukemic death at 0.1 ppm approach very nearly the odds of leukemic death for a worker who i, not ex- posed to benzene. Accordingly, a TLV-TWA of 0.1 ppm benzene is recommended. A STEL is not recommended. The reader is encouraged to review the section on Er- cursion Limits in the "Introduction to the Chemical Sub- stances" of the current TLV/BEI Booklet for guidance and control of excursions above the TLV-TWA even when the 8-hour TWA is within recommended limits. The recom- mended TLV of 0.1 ppm is less than the concentration associated with genetic damage in animals,(57) and it is less than the concentrations associated with genetic damage in human beings.(63) As calculations show that benzene der- mal absorption can contribute substantially to the total absorbed benzene dose,t'I> the skin designation is appropriate. BEt Indication Biological monitoring for human benzene exposure at ambient concentrations less than 1 ppm can be most read- ily documented by determination of urinary S-phenylmer- capturic acid (Figure 1).(122) The mercapturic acid conju- gate is formed and excreted together with phenol, catechol, hydroquinone, and hydroxy hydroquinone. It is a urinary metabolite of high specificity for occupational benzene exposure giving reliable indication of exposures at the 0.1-0.15 ppm range, whereas urinary phenol is not reliable unless gross benzene exposure has occurred.(1z-') The lowest practical detection limit, in the absence of interfering substances, has been reported at concentra- tions at least as low as 0.1 ppm. In the presence of inter- fering vapors, the accuracy and reliability of workplace air monitoring at ambient benzene concentrations even above 1.0 ppm can be questioned. References I. Svnder, R.: The Benzene Problem in Historical Perspective. Fundam. Appl. Toxicol. 4:692-699 (1984). 2. Occupational Safety and Health Administration: 29 CFR Pan 1910, Occupational Exposure to Benzene; Final Rule. Part Il, Department of Labc)r. Fed. Reg. 52(176):34460-34578 (September 11, 1987). 3. Ward, C.O.; Kuna, RA; Snyder, N.K.; et al.: Subchronic Inhalation Toxicity of Benzene in RaLti and Mice. Am. J. Ind. Med. 7:457-473 (1985). 4. l;}'eki. E.M.; Ashkar, A.E.; Shoeman, D.W.; Bi.Sel, T.V.: Acute Toxicirv of Benzene Inhalation in Hemopoietic Precursor Cells. Toxicol. Appl. Pharmacol. 40:49-57 (19?', ). 5. Gill, D.D.; Jenkins. V_L: Kempen. RE.; Ellis, S.: The Importance of 460 APPL OCCUP. ENVIRON. HYG. 50 • JULY 1990

Risk Anaf ysis in Environmental and Occupational Health September 4, 1991 Comment Mushrooms are known to contain various compounds, including hydrazine analogs, that are mutagenico in vitro and/or carcinogenic in laboratory animals under certain conditions. An extract of mushrcoms of the type tested has also been shown to be mutagenic. However, the spectrum of tumors found in this experiment on raw mushrooms was not what might be expected from the known carcinogenic compounds present in the mushrooms. Presumably there are different carcinogenic compounds are also present, or there was an interaction with other chemicals present. Referemu's Toth, B. and J. Erickson. 1986. Cancer Induction in mice by feeding of the uncooked cultivated mushroom of commerce Agaricus Bisporus. Cancer Research 46 (1986) 4007-4011.

6 TABLE E3. ANALYSIS OF PRIMARY TUMORS IN MALE MICE (Continued) Vehicle Control 500 mg/kg 1,000 mg/kg C.irculatory System: Hemangiosarcoma Overall Rates (a) 4;'50 (8%) 3/49 (6%) 1/50 (2%) Adjusted Rates (b) 10.1% 8.8% 2.6%, Terminal Rates (c) 3/38 (8c/'D 2/33 (K) 1/39 (3%) Life Table Tests (d) P=0.130\ P=0.559\ P=0.169N Incidental Tumor Tests (d) P=0.097\ P=0.408\ P=0.176N Cochran-Armitage Trend Test (d) P=0.134\ Fisher Exact Tests P=0.512\ P=0.181 \ Circulatory System: Hemangioma or Hemangiosarcoma Overall Rates (a) 4,50 (8%) 4,'49 (8%) 1j50 (2Si) Adjusted Rates (b) 10.1~'i 11.8~~ 2.6(/i Tesminal Rates (c) 3:38 (8~~) 3;33 (9ci:) I; 39 (3Si) Life Table Tests (d) - P=0.142N P=0.579 P=0.169N Ifncidental Tumor Tests (d) P=O.I ION P=0.573\ P=0.176\ Cochran-Armitage Trend Test (d) P=0.147ti 1i>her Exact Tests P=0.631 P=0.181\ if.iver: Adenoma Overall Rates (a) 0),50 (0%) 5 49 (10(/i) 13,50 (26~(-) Adjusted Rates (b) 0.0~i 13.0-&Ii 33.3~i Terminal Rates (c) 0: 38 (0-,i) 3; 33 (9cii) 13; 39 (339i) LFe Table Tests (d) P<0.001 P=0.030 P<0.001 lnr.idental Tumor Tests (d) P<0.001 P=0.023 P<0.00I Ccehran-Armitage Trend Test (d) P<0.001 Fis,her Exact Tests P=0.027 P<0.001 Liver: Carcinoma Overall Rates (a) 10; 50 (20c~) 14, 49 (29ii) 12 50 (24(/i) Adjusted Rates (h) 24.3r/i 35.9(/'i 25.8 r Terminal Rates (c) 7, 38 (18Si) 9 33 (27/j) 5, 39 (13S'c) L.ife Table Tests (c!) P=0.427 P=0.183 P=0.463 Incidental Tumor Tests (d) P=0.536 P=0.379 P=0.548K Cochran-Armitage Trend Test (d) P=0.363 Fisher Exact Tests P=0.224 P=0.405 Li.er: Adenoma or Carcinoma Overall Rates (a) 10 50 (20 (,,j) 18 49 (37ii) 23 50 (Wj) Adjusted Rates (h) 24.3ri 45.1~i 49.85"r Terminal Rates (c) 7 38 (18(/i) 12 33 (36Si) 16 39 (41(/i) Lifi: Table Tests (d) P=0.013 P=0.042 P=0.014 Incidental Tumor Tests (cO P=0.009 P=0.098 P=0.019 Cochran-Armitage Trend Test (d) P=0.004 Fisher Exact Tests P=0.052 P=0.005 Forestomach: Squamous Cell Papilloma Ovcrall Rates (a) 3 49 (6~i) 3 48 (6c;i) 9 49 (18~i) Adjusted Rates (h) 7.9~(' 9.1c:i . 23.1~i Terminal Rates (c) 3 38 (8r,(') 3 33 (9rD .9 39 (23c;(') Life Table Tests (d) P=0.038 P=0.597 P=0.065 Incidental Tumor Tests (d) P=0.038 P=0.597 P=0.065 Cochran-Armitage Trend Test (d) P=0.034 Fasher Exact Tests P=0.651 P=0.060 20 Benzyl Acetate

Endpoints Other than Cancer Brain

R6k. Assessments of EDB and Epidemiology lating from animal bioassay data to predict human cancer risks is inappropriate.(3-6) Because data from such assays are often the strongest evidence of carcinogenicity and usually the only basis available for quantification of human risks, alternative ex- planations should be investigated before drawing such a co aclusion. These should include the possibilities that the assumed exposures in the occupational study were too high, that some aspect of the model or its application to the workers was inappropriate, or that the method for scaling doses between species was incorrect. lit seems unlikely that the workers' exposures were greatly overestimated. Most of the exposures to EDB occurred decades before the measurements were taken, suggesting underestimation of exposure. While indi.vi'idual work histories or job activities were not taken into account, and while it is possible that heavily exposed short-term workers were excluded and that long-term employees experienced lower ex- posures than the TWAs, biases due to these factors are likely to be less significant than the changes in exposure over time. With risk predictions over seven- fold too high (even using the lower of two estimated TWAs), inaccuracies in exposure assessment are un- likely to explain the inconsistency between observed and predicted cancer deaths. We have therefore ex- plored other potential explanations: deficiencies of the model, problems in its application to the workers including the different route of exposure, and the interspecies conversion factor. 3. MI:THODS AND RESULTS A crude extrapolation using direct proportional- ity from the lowest dosed animals in the gavage experiment indicated compatibility between the ani- mal aiad occupational data. The low-dose gavaged rats )received 5.37 mg/kg/day (in human equivalent) and developed 60% more tumors than the control rats. The average worktime dose in the occupational study was 4.6 mg/kg/day, which amounts to only 0.35 :mg/kg/day when averaged over their lives. This represe.nts about 0.065 of the rats' dose, implying an excess risk of 0.04 for each worker (0.065 X 0.6), or about six extra cancer deaths in the cohort of 161, where three excess cancer deaths were seen. 'We further compared the observed cancer deaths among workers in the study by Ott et al.(lZ) with predictions from several linear nonthreshold models, 207 using data from both the NCI inhalation bioassay(t`) and the NCI gavage bioassay.(9) The following mod- els were used: the one-hit model and the multistage` model were fitted to the inhalation bioassay data; the multistage model incorporating time-to-tumor data, the multistage model with variable dosing, and the proportional hazards model were fitted to the gavage data. In each case the fitted extrapolation model was applied to the workers' exposure to obtain predicted cancer risks, which were then compared with ob- served cancer deaths in the EDB-exposed cohort. For the inhalation bioassay, nasal cavity malignancies in male rats represented the most sensitive site, sex, and species. To simplify comparison, we made the same exposure assumptions as Ramsey et al.,(2) with the exception that the model incorporating variable dos- ing and the Cox model do not assume that average lifetime dose is the determinant of risk. Also, while CAG used data from only the low-dose animals in the gavage study, these analyses used data from all dosed animals. 3.1. Inhalation Data: Two Models The one-hit, nonthreshold model takes the form P(d)-P(0) =1-exp(-P•d) where P(d) represents average lifetime cancer risk for an individual exposed to dose d, P(0) is the background lifetime cancer risk, and P is the un- known parameter for carcinogenic potency (i.e., mortality per unit dose) of the substance. When fitted to the inhalation bioassay data using Global 82 software,(t7) this model predicted upper limits of 1.2 and 0.7 excess cancer deaths among . the exposed workers at the Texas and Michigan plants, respec- tively, assuming EDB concentrations averaged 3.0 ppm for all workers at both plants. Since the inhala- tion experiment ran for the full two years, and since most of the animals were sacrificed at term, the calculation of partial lifetime risks for the workers was based on the exponent for time (or age) depen- dence of cancer risk obtained from the gavage data. Other researchers have estimated similar values for the age dependence of human cancer(l8•19) ; using smaller values as reported for lung cancer by Doll and Peto(1O) did not substantially alter the results. As shown in Table I and Fig. lb, when the small excess risks predicted by this model were added to the expected deaths, the resulting total predicted cancer

a reduction in mean fetal hc>dy weight at 100 ppm was ohsen,ed. No teratogenic effects were t~>und.'"' When pregnant Swi,s-Web.titer mice were exposed to 5, 10, or 20 pprn henzene in air on days 6-15 of gestation for 6 hours per day, alterations in the nunthers of hematopoietic colony-forming cells in the progeny were recorded.''j' Markeci reductions in erythroid colony-forming cells ,~vere e>bsen~ed at all benzene concentrations studied, and in- halation of 10 or 20 ppm also decreased the numbers of granuloctilic colony-forming cells. When mice, previously exposed in utero to 10 ppm benzene, were re-exposed to 10 ppm lbr 6 hours per day for 2 week.s, a marked reduc- tion in the numbers of bone marrow differentiated ervth- roid colony-forming cells occurred('i> Keller and Sny- der'-'31 interpreted these data as an indication that alterations of the murine hematopoietic system induced by neonatal benzene exposure could persist into adulthood. Ungvary and Tatrait2{) exposed CFLP mice and NZ rab- bits to benzene at 154 or 308 ppm, 24 hours per day, throughout days 6-15 of gestation. Benzene was detected in fetal blood and in amniotic fluid. At 308 ppm, retarded skeletal development and reduced fetal body weight were observed in mouse fetuses, and spontaneous abortions were reported in rabbits.t'-}> Genotox:icitw Studies Benz,ene exposure can cause chromosomal aberrations in animals and in humans. Benzene exposure induces clas- togenes:is, sister chromatid exchange, and micronuclei both in vivo and in ritro.t-'S> Benzene exposure has been shown to induce aneuploidy in dividing cells, presumably through inhibition of tubulin assembly during mitosis. However, benzene exposure ha5 failed consistently to induce point mutations, in genotoxicity test systems. Point M utation In the Salmonella ryphimurium gene mutation assay, benzene proved consistently negative for mutagenesis in plate-incorporation assays with or without microsomal en- zyme activation.(26-'9) McCarroll et a1.t30> published the only positive result using a microsuspension assay with hepatic microsomal activation such that an increase in the numbers of revertants in Salmonella strain TA100 was observed. Benzene exposure inhibited the growth of DNA repair deficient Ischerichia coli strain WP100 (uvra-, recA-, but no such effect was observed in repair proficient strains.t31> Growth inhibition was also observed in DNA repair defi- cient Bacillus subtilis strain M45 (rec-)t3'-> but benzene was considered without mutagenic activity in the E. coli PoIA asav, an indication that the DNA polvmerase activity was not critical for repair of benzene-induced damage to nucleic acidti3> Benzene was reported negative in Sac- charomyces ceretJrsiae gene conversion and mitotic cross- ing-over as:savs,t3"> however, it was considered mutagenic for S cereidsiae strains D61-M and D6.(i5) When benzene was fed to Drosophila melanokcuttv,~ at up to 2.5 percent in the diet, no evidence for a mutagenic response using the eye pigmentation as a genetic market wa.ti found.'j1'' When Drosophila were placed in air con- taining 27,000 ppm for 60 minutes (20% survival), a sig- nificant increase in spermatogonial crossing-over %.as oh- sen,ed and nmrttiom frequenc}• and trantilcxatic>n frequenc)• were increased. These data were considered indicative of the stage-specific nature of benzene-induced spermato- gonial mutagenesis in Drosophila.' i'> Benzene exposure altered gene expression as measured in the Drosophila wing morphology a5sav,(j") but results using the Droso- phila eve spot assay were judged negative,39> or at most equivocaL(+0) In grasshopper embryos, benzene exposure was associated with mitotic arrest, multipolar division, and chromosome lags.t-'1 > Benzene was tested in a colloborative study of 12 lab- oratories using a variery of cell lines and genetic mark- ersP'-> Benzene was mutagenic without hepatic enzyme (S9) activation in the mouse lymphoma L5178Y (TK+/+) assay in one laboratory, it was mutagenic in the Chinese hamster V79 cell assay at the oubain-resistant locus (NaK- ATPase defective) in one laboratory, and it was mutagenic at the 6-thioguanine resistance locus (HGPRT-) in one laboratory. Mutagenic activity was observed with S9 acti- vation in the mouse lymphoma L5178Y (TK+/+ ) assay for trifluorothymine resistance (TK-) in two of the laborato- ries, and mutagenicity was observed in the mouse lym- phoma L5178Y (TK+/+) assay for oubain-resistance in one of the laboratories. Benzene was considered muta- genic without exogenous activation for 6-thioguanine re- sistance in human AHH-1 lymphoblasts. Except for the human lytnphoblast and Chinese hamster V79 studies (which were not repeated in other laboratories), the findings for benzene point mutation could not be confirmed by other laboratories involved in the collaborative stud,v.t4Z> There- fore, potential point mutation as,sociated with benzene ex- posure in cultured mammalian cells is considered incon- clusive based on the studies published to date. Chromosomal Aberration Benzene treatment induces chromosomal structural changes and aneuploidy in cultured mammalian cells. In cultured human lymphocytes, chromosomal aberrations were observed after three hours of incubation with 9-88 µg benzene/ml with or without S9 activation.(43) Aberrations were also observed in Chinese hamster lung fibroblasts after treatment with 1100 µg benzene/ml and in Chinese hamster ovary (CHO) cells at 100 gg benzene/mt with S9 activation. Aneupoloidy was reported in Chinese hamster primary hepatocytes treated with benzene at 62.5 Wg/ml.t44i Benzene itself failed to induce sister chromatid ex- change (SCE) in cultured human lymphocytes without ex- ogenous metabolic activation (S9), but benzene metabo- lites increased SCE in a dose-dependent fashion.t45> The primary benzene metabolites (phenol, catechol, hydro- quinone) are transformed to benzo(semi)quinones, which presumably act as the ultimate genotoxic agents.45) Ca- APPL OCCUP. ENVIRON. HYG. 5/7) • JULY 1990 455

Risk Analysis in Environmental and Occupational Health September 4, 1991 ARE YOUR MUSHROOMS SAFE TO EAT? Raw commercial mushrooms, obtained from the supplier of local food stores, have been tested in a bioassay (Toth and Erickson, 1986) similar to those used for synthetic organic chemicals. We can therei`ore perform a risk assessment on raw mushrooms similar in all respects to the risk assessments performed on synthetic organic chemicals. In the mushroom experiment, there was one control group of 50 mice for each sex and one experimental group of 50 mice for each sex, the former kept on a normal diet and the latter fed the material under test at an average rate of about 1.57,000 mg/kg-day for their lifetime (assuming mice weigh 30 g). Feeding of the dosed group was ad Ilb mushrooms (without other feed) 3 days/week, normal diet 4 days/week; while the control group received the normal diet. Average mushroom consumption was 11 g/day/mouse during days on which mushrooms were the only food available (mushrooms are about 90°/) water). The experiment was continued for the natural lifetime of the animals, and no differences were seen in the lifefime of the dosed animals versus the control groups. However, the average weight of the dosed animals was substantially lower than the average weight of the control groups. There were increased incidences of tumors in several organs: TurrnDr site:type Sex Control Group Dosed Group }_Significance Bone :various F 0/50 8/50 0.003 Bone:various M 0/50 8/50 0.003 Forestomach:various F 0/50 19/50 2.3 x 10-' Forestomach:various M 2/50 14/50 0.00094 Liver:hepatoma F 0/50 4/50 0.059 Liver:hepatoma M 1/50 6/50 0.055 Lunca:All tumors F 13/50 20/50 0.1 Lung:Adenoma F 6/50 12/50 0.096 Lunci:,Adenocarcinoma F 7/50 11/50 0.22 Lung:All tumors M 17/50 31/50 0.0045 Lunci:Adenoma M 12/50 24/50 0.006 Lunc~:Adenocarcinoma M 9/50 13/50 0.23 1

TABLE B3. INDIVIDUAL ANIMAL TUMOR PATHOLOGY OF MALE MICE IN THE 2-YEAR STUDY OF BENZYL ACETATE HIGH DOSE RA.,,,.t A--- R ANIMALS NECROPSIED N o. TISSUE EXAMINED MICROSCOPICALLY + NO TISSUE INFORMATION SULMITTED Q t XI REOUIRED TISSUE NOT EXAMINED MICROSCOPICALLY TUMOR INCIOENCE C. A: NECROPSY. AUTOLYSIS NO NISTOLOGY OUE TO PROTOCOL N: 5• NECROPSY. NO AUTOLYSIS. NO MICROSCOPIC EXAMINATION ANIMAL MIS-SEXED M+ /: ANIMAL MISSING NO NECROPSY PERFORMED N MAL 0 0 0 0 0 0 0 0 0 0 0 0 NUMlER 2/ 21 I 21 21 31 SI 31 31 31 31 31 31 31 31 41 41 41 41 41 41 41 41 41 41 5 1 TOTAL M K 0 1 .f f f 1 I 0 f 0 1 1 1 1 f t 0 1 0 1 1 1 t t f ITISS~ES STUDY 1 91 01 01 01 01 01 01 21 01 91 01 01 01 01 01 01 71 01 EI 01 01 01 01 01 0 1 TLPC1R5 LUNGS AMD 1RONCNI I • • + a + • • + • • . • . . • . • • • • • • + • . 1 I Sl MEPATOCELLULAR CdRCIMONA. METAS ALVEOLARI/RONCNIOIAR AOEHOMA I 1 ' % X % X I 1 I c i ALVEOlARi1ROHCXIOIAR CARCINOM4 I TRACHEA 1 • • • . • + . + . • • • • . a • • . - ff i MEHAIUYOIEIIG SYSTEM DONE MARRON I ~ + . . I I SPLEEN 1 • • 4 • • . + . + • . • . . . + 41 I HENANOIOSARCOMA I t 1 LYMPH MODES + • . • . • • • . . • . + . . • . • • . . • . • • i SI I MALIGNANT LYMPHOMA. MIXED TYPE t I THYMUS _ - -- ~ + • • • a a • - - - . . . _ • a • + + r • • - . <1 I CIRCULATORY SYS ~ /IEART ; . • . . . • . 1 I I SI / / DTF,ESTIVE SYSTEM 1 I SALIVARY OLAND I • + + . . . . • • . • • < I I LIVER 1 ' + + + . • • . . . . . . . . . . . • . • • • • . • . • .1 Sl 1 NEOPLASM. N0S 1 I 1 I HEPATOCELLULAR AOENOMA I Y X X X X X % X X XI 13 I HEPATOCELLULAR CARCINOMA 1 I IILE DUCT • ( + • + • • • + r + a + • .I 5/ I GALLILADDER 3 COWIOM DIIE DUCT.. ' 1 + N _+ • M +• +_ ::1 Sf• I ` I PANCREAS lll •- +• a •+1 c• ~ ESOPHACUS ! + . . • • + . . + . . + • . . . . - . . . . - .I <. j SYOMACH . • ~ • • • • + • • . • + • • • + • . - • • • • . ai 1 N i SGUAMOUS CELL PAPIIIOMA Y X % T x 1 1 I SQUAMOUS CELL CARCINOMA I SlWLL INTESTINE I . - . . - . . . <7 I I IARGE INTESTINE • I • + + + . • • + . + • - • • + + • + cc I N I RIDNEY • • . . • . . . . . • • + • . • • . • •' 51 ~ TUlULAR-CELL ADEMOlU I X 1 t f TUIULAR-CELL ADENOCARCINOMA Y 111 ./ URINARY ILADDER ~ ' . . • • - • . + . • .1 1 c9 I ENDbCRINE -ST3 I PITUITARY I . + + + o - • • + • • ADRENAL 4 • + + • a • • • • a 4 + 4 + + + • . • r • . . I N 1 OANGLIONEUROMA X _ I + I THYROID FOLLICULAR-CELL AOENOMA r + • • • • + • • + . . . • . . . • - • • . • - -/ i 47 PARATHYROID i • + • + • + • + + + - • + - - • - . cc I PANCREATIC IsLETS 1 + . . . + . + . • + a . . . . . . . - . • . • . 41 H 1 ISLET-CELL ADENOMA X 2 1 / MAMMARY OLAND 1 -• I TESTIS ' INTERSTITIAL-CELL TUMOR • • . • . . . . • • • . . . . . • • . . . • . . .I S. 1 I I PROSTAT 1 a - + .1 <9 / NERV0U3 SYSTEM 1 I I IRAIH / ' + • i • . • • + . . • . . . . . . . . • •i 51 i SPECIAL 3FFSE b kGAi3 1 HARDERIAN OLAND - i N N N N N M N N N N M N N N N N N M N N N N N N NI SD I ADENOMA. N0S I X . % 3 I MESENTERY I N N N M N N M N M N N N N N N N N N N N N N N N Nj I SO• ( MEPATOCELLULAR CARCINOMA. METAS I I 1 1 N SYSTEMS 1 MULTIPLE OROANS NOS I N N N N N N N N N N N N N N N N N N N N N N M N NI SO• 1 MEPATOCELLULAR CARCINOMA. METAS / / 1 I MALIGNANT LYMPMOMA. M0S / I 1 I MALIG.LYMPNOMA. LYMPNOCYTIC TYP I Y / 1 1 ; 5~Y ,2S e f-klz 6"0 n, ~ c~4), ~u. 1A34 c~es e-6 rm.+.P of wI Q Ics ,x

tub and compare the predicted risks to the observed cancer mortality. 2. THE EIDB DATA AND THE REPORTED DISCREPANCY Cancer deaths among workers exposed to EDB were reported as not significantly elevated, unless a small group with additional exposure to arsenic was included.tlz> However, in a long-term gavage bioas- say,(4) and in two inhalation studies(1a.t6) published subsequent to the risk assessment of CAG, EDB proved highly carcinogenic. In the two assays (one gavage and one inhalation) conducted by the Na- tional Cancer Institute/National Toxicology Pro- gram (NCI/NTP), more than 50% of the high-dose animals exhibited contact-site tumors (squamous cell carcinomas of the stomach from gavage administra- tion, nasal cavity malignancies of several types from inhalation). Both low- and high-dose animals had s,atistically significant excesses of contact-site tumors aad a variety of tumors remote from the site of administration (i.e., systemic tumors). A one-hit, nonthreshold model was fitted by the CAG00> to the rat data from the NCI gavage bioas- say to assess risk from ingestion of EDB-contami- nated food. The CAG used squamous cell carcinomas of the stomach in male rats and an interspecies conversion based on surface area equivalence. In developing its risk assessment for public exposure via ingestion, the CAG specified that the parameter estimates were applicable only for intubation ex- posure. Different parameter estimates were recom- m+.r(ded for dietary exposure and for inhalation ex- pasuret10j (in a later risk assessment of EDB, the CAG scientists developed a more sophisticated model to deal with the irregularities in the gavage bio- assay(lt)). Etamsey et al.tz> applied the one-hit model fitted by the CAG to a cohort of 161 employees involved in the manufacture of EDB J12> Exposure was estimated by (i) assuming all workers were exposed to time- weighted average (TWA) concentrations based on measurements made during the 1970s at one of the two plants,0Z> and (ii) converting to a continuous lifetime equivalent dose using an average weight of 70 kg. Additionally, it was assumed that both potency and brologically effective dose were the same for inhalation as for intubation, i.e., no adjustments were made for route of exposure. I Hertz-Picciotto, Gravitz, 9nd [Yeft The risk of an EDB-induced cancer death wa calculated for each worker in the study by Ott et atO4 These were then summed to obtain the number d excess cancer deaths predicted by the model. In tht cohort of 161 workers, this model predicted over 80 excess cases of cancer from an exposure of 3.0 ppm, or about 50 cases from 0.9 ppm exposure.(Z> Thex predictions are for the partial lifetimes of the workett. Given that only eight cancer deaths were observed, with a 95% upper bound of 16, these predictions are clearly inconsistent with the observed mortality. Figure la displays a comparison of (a) observed cancer deaths, (b) expected cancer deaths based on U.S. white male age-specific rates, and (c) cancer deaths predicted by this model. Predictions are shown for each of the two assumed exposure levels. Since measurements were taken at the Michigan plant only, results for the two plants are presented separately. In light of this discrepancy between predicted and observed cancer mortality in EDB-exposed workers, some authors have suggested that extrapo- s0 55 0 30 to 0 a 1 KEY: } OBSERVED I 95'b CONFIDENCE INTERVAL ~ EXPECTED (AGE ADJUSTED U.S. WHITE MALE RATE) • PREDICTED AT 3.Oppm o PREDICTED AT .9ppm 0 0 + gl A TEXAS MICHIGAN B TEXAS MICHIGAN GAVAGE BIOASSAY INHALATION BIOASSAY Fig. 1. Observed, expected, and predicted cancer deaths using ~ one-hit model and two animal bioassays of ethylene dibromide. 0

Thursday, September 5 Cancer 8:30 - 9:30 What is 9:30 - 9:45 9:45 - 1]l:.15 11:15 - ].].:30 & Cancer Model ine Cancer? Refreshment Break Cancer Modeling Break Applications of Expert Judament 11:30 - 12:15 Applications of Expert Judgment in Risk Analysis 12:15- 1:15 Lunch Exposure Assessment 1:15 - 2:45 The Respiratory System as an Entry for Exposure 2:45 - 3:13 Refreshment Break Upton Cohen Moeller Valberg 3:15 - 4:4:i Assessment of Exposures Ryan

Risk Assessments of EDB and Epidemiology predictions of this model are fully compatible with the observed mortality in the EDB-exposed cohort. The proportional hazards model, also sensitive to both the survival times and the variable dosing, gav: rather high risks, particularly if one assumes its effect to be on lung cancer, the site of contact for occupational exposure. This was due to the depen- dence of the model predictions on background rates in humans, when the model parameters were esti- mated from an animal experiment in which the back- ground rate was zero. It is also rare for an agent that inducs lung cancer to affect the background rates, which are primarily due to smoking, in a multiplica- tive way. Asbestos is a notable exception.('9) On the other hand, at the low exposure estimate (0.9 ppm), assuming EDB's effect is on stomach cancer, the predictions are compatible with the observed cancer deat#ls, at the one plant where measurements were taken (Michigan). Without knowledge of the site of EDE's carcinogenic activity in humans, it is difficult to say whether this model is compatible with the epidemiologic data. Tlae multistage model with time-to-tumor yields very similar risk estimates as the model that also incorporates the variable dosing pattern, suggesting that ~the variable dosing was not an important factor in th : potency of EDB in the gavage bioassay. With the 3.0 exposure estimate, overall predictions were aboun ithree times the observed mortality among the work .rs. Thus, because the workers' exposure begins comparatively late in life, the additive models which incorhorate information on time since exposure be- gins provide a far better fit to the worker data than those models that do not. NVlule numerous other models (probit, Weibull, logit, etc.) have been advocated for extrapolation, we have limited our analyses to those with the property of being linear at low doses. Such curves will yield an upper bound for the risk at low doses(3°-32) and thus provide a health-protective basis for regulatory deci- sion making. Itt should be emphasized that risk assessment makes no claim to providing precise predictions, but rather seeks to generate ball-park estimates. These estimates are intended as plausible upper bounds of risk. Restricting discussion to models used with the gavage data, the one-hit model predicts implausible risks, as does the proportional hazards model using lung cancer deaths, while both the variable dosing and the time-to-tumor forms of the multistage model are compatible with the epidemiologic data. We conclude '211 that the gavage data are not inherently incompatible with the workers' cancer mortality experience. The inhalation bioassay was not fraught with the complications -of high early mortality and a re- duced latency period. This may partially explain why the one-hit model (without an adjustment for latency) applied to the inhalation data predicted cancer risks that were fully compatible with the epidemiologic study. The importance of the latency period is under- scored by the fact that most of the tumors observed in the inhalation bioassay were discovered at termi- nal sacrifice. If we were to exclude such tumors, the predicted excess risks from this bioassay would be halved. 4.2. Route of Exposure When the doses in the inhalation study were converted to units of mg/kg/day, they were in fact larger than the doses of the gavage study (see Table VI). The time on the study was about double the duration of the gavage bioassay, while the probabil- ity of developing tumors was comparable to the probability in the gavage study if one includes the tumors found at terminal sacrifice (104 weeks in the inhalation bioassay). Thus, the gavage study is dis- tinguished from the inhalation study by lower doses, 'rable VI. A Comparison of Two Carcinogenicity Bioassays: Gavage and Inhalation Controls Low dose High dose Gavage Dosc° 0 5.37 6.66 Response (crude) 0/20 30/50 25/49 P (response)b 0.0 0.61 0.75 Weeks on study Mean 53 45 31 Median 49 47 36 Inhalation Dose° 0 9.56 38.24 Respotsse` (crude) 0/20 2/48 24/48 39/49 P(response)b 0.0 0.05 0.65 0.91 Weeks on study Mean 103 98 76 Median 104 104 80 °Doses are in mg/kg/day averaged over the lifetime, convened to human equivalent using surface area as the basis for interspecies scaling. "Kaplan-Meier probabilities. `Response rates for low-dose rats in the inhalation study were derived excluding (2/48) and including (24/48) tumors found at terminal sacrifice.

What is Cancer? Upton

212 a higher carcinogenic potency manifested as much shorter latencies, and greater subchronic toxicity. The high carcinogenic potency in the gavage bioassay was driven largely by the shortened laten- cies. Adjusting for latency was especially crucial for accurately estimating risks to those whose exposure began late in life. Since even those models incorpo- ra'.ing latency period produced larger gavage-based risk estimates than models fitted to the inhalation data, it is possible that extrapolating from the inhala- tion bioassay underestimates the effects of EDB in humans. That is, humans are potentially as sensitive as the most sensitive site, sex, and species observed in a bioassay using any route. When data from more than one route are available, a risk assessment for human exposure can be based on data from the same route of exposure as that of the humans, if this same route leads to a positive dose-response and other factors are equal (e.g., statistical power of the study, species or strain sensitivity). If, however, bioassay da.ta using a route different from that of the human exposures show a higher potency, these data should not be rejected outright. In the interest of protecting the public health, the bioassay data showing the higher potency should be considered carefully, in conj unction with pharmacokinetic data that may shed light on species and route differences. 4.3. ]lnterspecies Conversion As demonstrated in Table VI, the mg/unit surface area basis for scaling doses between animals and man yields higher risks for humans than the use of mg/kg body weight. Thus, even the one-hit model would have predicted risks compatible with the workers' mortality, had body weight been used as the scaling factor. In a comprehensive review of the interspecies scaling issue, Davidson et a1.(33) conclude that surface area scaling is most likely to provide the correct scaling for carcinogenicity because toxico- logic,, metabolic, and pharmacokinetic data correlate best when body weight is raised to the power 2/3 or 3/4. With respect to EDB, two lines of argument lead to the conclusion that surface area is likely to be the appropriate basis. (i) EDB acts as a carcinogen at sites af art from point of contact; based on experi- mental data, at least one pathway involves activation by the cytochrome P-450 mixed function oxidase system.j34) (ii) Contact-site tumors were the most sensitive site for EDB, and the surface area of this target site is proportional to the body surface area. Hertz-Picciotto, Gravitz, and Neutra 5. GENERAL IMPLICATIONS While the relationship between the quantitative aspects of laboratory animal carcinogenesis and hu- man carcinogenesis remains to be delineated, there is evidence that the two may not be far apart for at least some agents. Rowe and Springer(3i) showed animal-based estimates of asbestos-induced carcino- genic potency to be within the range of human-based estimates from several studies. Similarly, animal- and human-based estimates derived for the carcinogenic potency of benzene(3b) and gasoline(37) were re- markably close. An analysis similar to the one pre- sented here indicated compatibility between a risk assessment for ethylene oxide based on rat mono- nuclear cell leukemias and leukemias observed in two cohorts of workers involved in the manufacture of ethylene oxide.08) Crouch and Wilson compare potency estimates based on human data to estimates based on rat and mouse data, and found that in about 2/3 of the comparisons, the estimates differed by less than one order of magnitude.(39) These find- ings are in direct contradiction to the claims of some scientists<3•4•',IIj that animal-to-human extrapolations have no scientific basis. In response to reports of a high correlatioL, between rat and mouse carcinogenic potencies,(39-41) Bernstein et a1.(42) have shown that rat-mice potency correlations are an artifact of the way doses are determined for the bioassays. Since human doses are not established by experimental protocol, similar potencies for animals and humans are unlikely to be an artifact. The present paper adds further empirical evi- dence that quantitative data from animal carcinogen- icity studies are a reasonable basis for estimating human cancer risks, and that linear nonthreshold additive models provide a practical means for such risk estimation. This should not be construed to mean that human and animal data will necessarily be consistent. Species differences for some compounds are supported on both theoretical and empirical grounds. When comparing bioassays of the same chemical performed in different species, potencies may differ by more than an order of magni- tude.(39-a1,a3•`w) On the other hand, even if the true carcinogenic potencies for two species are close, the estimated potencies may not be. This is because assumptions are required wherever the data are lack- ing. Well-conducted epidemiologic studies of those occupationally exposed to compounds present at X

Risk Assessments of EDB and Epidemiology 209 T'slae Il. Numbers of Cancer Deaths Predicted by Multistage race-, sex-, and calendar-year-specific rates Model Adapted for Variable Dosing Fitted to Gavage Data for U.S. males. Number of Cancer Deaths c. The predicted excess probability of a cancer Observed Excess predicted by model" death for worker j at dose = d was derived (95% CI) Expected° 0.9 ppm 3.0 ppm using Hj(0), the estimate for fl, and the pro- T:xas 3` (0.6-8.8) 3.6 2.84d (3.49)° 8.65 (10.42) portional hazards assumption: MSchigan 5 (1.6-11.7) 2.2 1.95 (2.33) 4.84 (5.53) "From U.S. white male age- and calendar-year-specific mortality rates. hMttltistage model adapted for variable dosing (Ref. 11). `Includes one arteriosclerotic heart disease death with lymph node malignancy (see Ref. 2). dMeximum likelihood estimates. 'Numbers in parentheses are upper 95% confidence fimits. wfaere P(d, [) represents the risk by time = t, for a dosing pattern d, and X(d, u) represents the in- st,antaneous hazard at time u due to dose = d„ (_= dose between 0 and u). Furthermore, this model assumes that the hazard for an exposed individual is proportional to the base-line hazard at all times: X(d, t) =f(d )'a(0,r) (Dose can be a function of time or not.) This model has been used recently as a basis for developing and' comparing potencies from animal carcinogenicity bioassays. t16-11' The function f(d) is taken to be linear: (1 +R• d). The following steps implemented this model: a. The parameter ~B was estimated from the animal gavage data. b. For each worker j, the integrated base-line hazard, Hj(0), was determined using age-, P(dj) + P(0) =1-exp{ -I,8dj] - [H,(0)I} d. The predicted number of excess cancer deaths for the cohort was obtained by summing the risks over all workers in the cohort. Unlike the previously discussed models, the Cox proportional hazards model assumes that the excess risk is a function of the background rates. Since EDB is not expected to affect all cancer sites uniformly, two potential sites were selected for extrapolation: lung and stomach. The excess lung cancer deaths predicted by the Cox model assuming that EDB concentrations aver- aged 3.0 ppm were 46.8 and 22.1 for the Texas and Michigan plants, respectively; predicted excess stom- ach cancer deaths were 17.8 and 9.5. These predict- ions are maximum likelihood estimates; because of the strong monotonicity of the gavage data, the vari- ance was too unstable to derive a reliable confidence interval. The high lung cancer predictions were simi- lar to those of the one-hit model fitted to the gavage data. Tables IIY and IV summarize the risk predic- tions from all of the models discussed above. For simplicity of presentation, Table III assumes 0.9 ppm exposure, and Table IV assumes 3.0 ppm exposure. Table IA. Total' Cancer Deaths Predictedh by Several Models for EDB-Exposed Workers'` Models fitted to gavage data Texas Michigan Overall Observed cancer deaths 3 5 Proportional hazards Multistage with Multistage with One-hit Stomach Lung time-to-tumor variable dosing 38.6 10.4 31.4 7.7 7.1 21.2 5.9 13.0 3.5 4.5 59.8 16.3 44.4 11.2 11.6 Models fitted to inhalation data One-hit: terminal sacrifice tumor: Omitted Included 3.8 2.3 6.1 4.0 2.4 6.4 'Total cancer deaths s Iexpected+predicted excessl. °Otlter assumptions are described in the text. Values in the table represent upper 95% confidence limits, except for the proportional hazards modcl. : whidh the variance estimates were too unstable to derive an upper confidence limit. `tissurnes 0.9 ppm exposure during time employed. 2025545876

42 Princihltas of Health and Safety in Agriculture LDt LD 2 LD 3 0 40001 S LL w Q 3000 .'}~ a J T ~ J 2000 ~.. E Y I 0001 z ~ ~ a LD 4 LD 5 015 ff'IGURE S. Comparison of LD isoenzyme pattems from hamster perito- neal P:vL'~s and from lung lavage fluid of hamsters exposed to 3.75 mg iron oxide per 100 g body v:eight. (Adapted from Beck. B. D., Gerson, B., Feldman. H. A.. and 13r3in, J. D., Toxicot. AppF_ Piwrmacol., 71, 59, 1983. With permission.) LD could be comin€, from PMNs. Macrophages have a similar LD composition, sc, they also may be a source. AUTOMOBILE WASTE OIL COMBUSTION PRODUCTS This assay is particularly suited for analyzing new complex agents which are just being introduced into the environment. We have recently investigated the pulmonary toxicity of respirable particulatcs from an air-atomizing oil space heater using automobile waste crankcase oil (AWO).23 A combus- tion sample was prepared from AWO from a service station by Dr. R. E. Hall of the 1U. S. Environmental Protection Agency, using an air-atomi2:ing oil bumerrated at 250,000 BTU/h heat input. Respirable patticulates were collected from a dilution tunnel by electrostatic precipitation using a massive air vol- ume sampler.24 Analysis of the particles showed certain met- als were present at relatively high levels, for example: Pb, 75.6 mg/g; Zn, 23.0 mg/g and Fe, 5.3 mg/g. At I d postexposure, there was extensive pulmonary injury as demonstrated by cellular and biochemical indicators in BAL: (1) elevated levels of albumin, (2) increased extracellu- lar glucosaminidase, and (3) impaired pulmonary macroph- age phagocytosis. The, injury was often greater than that seen in response to toxic a-quartz. Some of the data obtained are shown in Figures 6 xo 8. However, assays of BAL up to 14 d post-AWO exposure demonstrated that most indicators rapidly approached control values. This is in c:ontrast to the persistent inflammation caused by a-quartz. As shown in Figure 9, LDH values approached control values at 2 weeks after intratracheal instil- lation of AWO. Following quartz exposure, the LDH level remains elevated. This suggests that the toxic effects of AWO stem from soluble components which are rapidly cleared. AWO may be less likely to cause chronic pulmonary disease than a-quartz unless exposure persists. Acute injury as mani- fested by bronchitis or increased susceptibility to infection may be a more likely outcome than fibrosis. 0 75 mg DUST INSTILLED/100g BODY WEIGHT AWO T a-QUARTZ e 3 75 FIGURE 6. Concentration of albumin in the cell-free supematant of BAL fluid.The effects of iron oxide, a-quartz, and combustion products of A WO are shown I d after intratracheal instillation. Values are mean ± standard errors. (Adapted from Beck, B. D., Brain, J. D., and Wolfthal, S. F., lnhaled Particles IV. Dodgson, J., Ed., British Occupational Hygiene Socieiy, Edinburgh, Scotland.) FIGURE 7. Concentration of (3-N-glucosaminidase in the cell-free su- pematant of lavage fluid after exposure to iron oxide, a-quartz, and AWO. Values are mean ± standard errors. (Adapted from Beck, B. D., Brain, J. D., and Wolfthal, S. F., Inhaled Particles 1V, Dodgson, J., Ed., British Occupa- tional Hygiene Society, Edinburgh, Scotland.) By comparing the response to AWO with the response to the same doses of toxic a-quartz and nontoxic iron oxide, we conclude that the AWO combustion products have a high potential to cause acute lung injury. Both soluble and insolu- able components of AWO can produce lung injury. Some, but not all, of these effects are due to acidity and divalent cations, such as lead, which are present at high levels, CONCLUSION Experimental pathology has frequently advanced because of the addition of new diagnostic tools. During the last decade, BAL has emerged as a very useful tool in the assessment of lung injury. It is applicable to both animal models exposed to inhaled particles and gases in a laboratory and to humans

Toxikok>gL,che Knsicht. Arch. Exp. Pathc l. Pharmakc l. 138:65 (1928). ti'. Erdem• S.; DinCul, G.: Ixukemia in Shoe-workers Exposed C:h,-oni alh• to Benzene. Blood -i:83'-8-f 1( I9'-t ). riH. A,k,uv. M.: Erdem. S.: DinCul. G.: T}pes of Leukemia in Chronic Benzene Poisoning. A Study in Thirty-four Patients. Acta Ilaematctl. 55 a:5-"3 ( 19'6 ) 89. :1ksov, M.: Different Types of Malignancies due to Oc•cupati< nal Ex- posure to Benrene. A Review of Recent Observations in Turkey. Emirc n. Res. 23:181-190 ( 1980). 90. A.ksov. M.: Malignancies due to Occupational Ex}x sure to Benzene. Am. J. Ind. Med. 7:395--I02 (1985). 91. Alc,tn•, M.: Benzene as a Leukemogenic and Carcinogenic Agent. Am. J. Ind. Med. 8:9-20 (1985). 92. Vigliani, E.C.: Leukemia Associated with Benzene Exposure. Ann. N.Y. Acad. Sci. 271:143-151 (1976). 93. Infante, P.F.; Rinskv, R.A; Wagoner, J.K; Young, RJ.: Leukemia in B(:n?ene Workers. Lancet 2:76-78 (1977). 94. I nfante, P.F.; Wh ite, M.C.: Chu, K.C.: Assessment of Leukemia Mortality Associated with Occupational Exposure to Benzene. Risk Anal. 4:9-13 (1984). 95. Van Raalte, H.G.S.; Grasso, P.; Irvine, D.: Tackling a Very Difficult ProE lem. Risk Anal. 4:1-2 (1984). 96. Rinsln•, RA; Young, RJ.; Smith, AB.: Leukemia in Benzene Workers. Am. J. Ind. Med. 2(3):217-245 (1981). 97. Rinslry, RA; Smith, A.B.; Hornung, R; et al.: Benzene and Leukemia. An Epidemiologic Risk Assessment. N. Eng1. J. Med 316(17):1044-1050 (1987). 98. Yin, S.N.; Li, G.L; Tain, F.D.; et al.: Leukemia in Benzene Workers: A Re:rospective Cohort Study. Br. J. Ind. Med. 44:124-128 (1987). 99. Yin, S.N.; Li, Q.; Tian, F.; et al.: Occupational Exposure to Benzene in China, Br. J. Ind. Med. 44:192-195 (1987). 100. On:, ~G.M.: Townsend, J.C.; Fishbeck, WA; Langner, RA: Mortality among Individuals Occupationally Exposed to Benzene. Arch. En- viron. Health 33(1):3-10 (1978). 101. Occupational Safetv and Health Administration: Occupational Ex- posure to Benzene: Proposed Rule and Notice of Hearing. Fed. Reg. 50:50512-50586 (December 10, 1985). 102. Bond, G.G.; McLaren, EA; Baldwin, C.L; Cook, RR.: An Update of Mot•cdity, among Workers Exposed to Benzene. Br. J. Ind. Med. 430O):685-691 (1986). 103. Decbufle, P.; Blattner, WA; Blair, A: Mortaitvamong Chemical Workers Expv,ed to Benzene and Other Agents. Environ. Res. 30:16-25 (1983). 104. Wong, 0.: An Industry Wide Mortality Study of Chemical Workers Occupationally Exposed to Benzene. i. General Results. Br. J. Ind. Med. 44(6):365-381 (1987). 105. Wong, 0.: An Industry Wide Mortality Study of Chemical Workers Occupationally Exposed to Benzene. 11. Dose-Response Analysis. Br. J. 1nd. Med. 44(6)382-395 (1987). 106. Thomas, T.L; Waxweiler, RJ.; Moure-Eraso, R; et al.: Mortality Pat- terns ,unong Workers in Three Texas Oil Reftneries. J. Occup. Med. 24:135-141 (1982). 10". Wen. C.P.: Tsai, S.P.: McClellan. ~X'A: Gibson, RL.: Long-term Mor- talit, • Study of Oil Refinen• Workers. l. Mortalin e f Hc urlv and Sal- aried Workers. Am. I. Epiderniol. 112i:526-5-+2 (1983). 108. Theriault, G.: Guulct• L: A 11ortality Studr of Oil Relinet,• Workers. 1. Occup. Med. 2 t:36,-3?0 (19'9) 109, Hanis. N.M.: Stavraky. K.M.: Fowler. i.L.: Cancer Mortalin in Oil Re hnen• Wt>rkers. 1. Occup. Med. 21:16'-17-i ( 19'9). 110. Hanis. N.M.; Holmes. T.M.: Shallenherger, L.,;.: Iemes. KE.: Epide- miology Study of Refinery and Chemical Plant W'orkers.l. Occup. Med. 24:203-212 (1982). 111. Schottenfeld, D.; Warshauer, M.E.; Zauber, AG.: A Prospective of Morbiditv and Mortality in Petroleum Industry Employees in the United States. A Preliminary Report. In: Quantification of Occupa- tional Cancer, pp. 247-265. MA Schneiderman, Ed. Banbury Report No. 9. Cold Spring Harbor, New York (1981). 112. Rushton, L; Alderson, M.R.: A Case-Control Study to Investigate the As.Sociation Between Exposure to Benzene and Deaths from Leu- kemia in Oil Refinery Workers. Br. J. Cancer 43:77-84 (1981). 113. Joyner• R.E.: Leukemia at Shell Wood River and Deer Park Manufac- turing Complexes. Letter and Report of Studies to J.D. Millar, Di- rector, NIOSH, dated July 28, 1983. OSHA Docket H-059B, Ex. No. 142-13. Occupational Safety and Health Administration, Washington, DC. 114. Tsai, S.P.; Wen, C.P.; Weiss, N.S.; et al.: Retrospective Mortality and Medical Surveillance Studies of Workers in Benzene Are-as of Refin- eries. J. Occup. Med. 25:685-692 (1983). 115. International Age,xv for Research on Cancer: Benzene. IARC Mono- graphs on the Evaluation of Carcinogenic Risk of Chemicals to Hu- mans, Vol. 29, Some Industrial Chemicals and Dyestuff.s, pp. 93-148. IARC, Lyon, France (1982). 116. Thorpe, Jj.: Epidemiological Survey of Leukemia in Persons Ptxen- tially Exposed to Benzene. J. Occup. Med. 16:375-382 (1974). 117, Arp, E.W.; Wolf, P.H.; Checkoway, H.: Lymphocytic Leukemia and Exposures to Benzene and Other Solvents in the Rubbet Industry. J. Occup. Med. 25:598-602 (1983). 118. White, M.C.; Infante, P.F.; Chu, KC.: A Quantitative Estimate of Leu- kemia MortalityAssociated with Occupational Exposure to Benzene. Risk Anal. 2:195-204 (1982). 119. Crump, KS.; Allen, B.C.; Howe, RB.; Crocket, P.W.: Time-related Factors in Quantitative Risk Assessment. J. Chronic Dis. 40 (Suppl.2):1015-1115 (1987;. 120. Chandler, J.LR: Benzene and the One-hit Model. Risk Anal. 4:7-8 (1984). 121. Hoel, D.G.; Kaplan, N.L; Andersen, M.W.: Implication of Nonlinear Kinetics in Risk Fstimation in Carcinogertesis. Science 219:1032-1037 (1983). 122. Stommel, P.; Mueller, G.; Stucker, W.; et al.: Determination of S-Phenylmercapturic Acid in the Urine-An Improvement in the Biological Monitoring of Benzene Exposure. Carcinogenesis 10:279-282 (1989). 463 APPL OCCUP. ENVIRON. HYG. 5(71 • JULY 1990

Risk Anal}-sis, Vol. 8, No. 2, 1988 How Do Cancer Risks Predicted From Animal Bioassays Compare with the Epidemiologic Evidence? The Case of Ethylene Dibromide Irva Hertz-I'icciotto,l Norman Gravitz,2 and Raymond Neutra2 Received June 20, 1985 Cancer risks for ethylene dibromide (EDB) were estimated by fitting several linear non- threshold additive models to data from a gavage bioassay. Risks predicted by these models were compared to the observed cancer mortality among a cohort of workers occupationally exposed to the same chemical. Models that accounted for the shortened latency period in the gavaged rats predicted upper bound risks that were within a factor of 3 of the observed cancer deaths. Data from an animal inhalation study of EDB also were compatible with the epidemiologic data. These findings contradict those of Ramsey et al. (1978), who reported that extrapolation from animal data produced highly exaggerated risk estimates for EDB- exposed workers. This paper explores the reasons for these discrepant findings. KEY WORDS: Ethylene dibromide; risk assessment; cancer; occupational exposure. 1. INTRODUCTION In the absence of adequate human data, quanti- tative cancer risk assessments have relied heavily on extrapolations from animal bioassays conducted at comparatively high doses.0) The validity of such extrapolations has, however, been a source of con- troversy.(2-s) A case in point is that of ethylene dibromid: (EDB), a fumigant that, until recently, was widely used on grain and citrus products. Re- sults of an animal bioassay(9) showed EDB to be an extremely potent carcinogen when administered by gavage. For regulatory purposes, the Carcinogen As- sessment Group (CAG) of the U.S. Environmental Protection Agency used these bioassay data to esti- mate human risks from consumption of EDB re- sidues in I'ood."°•it> I California Public Health Foundation, 2151 Berkeley Way, Room 515, Berkeley, California 94704. 2California 1Department of Health Services, Berkeley, California 94704. Ramsey and associates(2) applied the risk ex- trapolation model used in an early report of the regulatory agency(10) to a cohort of workers at two chemical manufacturing plants who were exposed to EDB by inhalation, and whose mortality was under study.(lZ) The results of Ramsey et al. suggested a wide discrepancy between the observed mortality and the risks predicted from the animal gavage data by a low-dose-linear extrapolation model. These results have been cited as evidence that extrapolations from animal bioassays to human real-world exposures are implausible and hence contraindicateVz-6•13) In this paper we investigate the reascns for the apparent discrepancy. Other nonthreshold models which are linear at low doses are fitted to the gavage data, including the one used in the final risk assess- ment of the CAWiI) Models are also fitted to data from a more recent inhalation bioassay.(14) The use of safety factors is not considered because of the limitations of this approach.(15) We then apply each fitted model to the cohort of EDB-exposed workers 205 0272-4332/88/0604o205SO6.0o/I CJ 1988 Society for Risk Analysis

Lifestyle Stress cO besity T- Hypertension f r w ~ -- let rireased ncPlate Aggregability Cigarette Smoking Myocardial Electrical Instability I-OGSVS'SZOz

0 Principles of Health and Safety in Agriculture. Edl by J.A. Dosman and D.W. Cockcroft. Boca Raton, CRC Press. 1989. pp. 39-44. Use of Biological Assays in Short- -rrerm Assessment of Inhaled Substances Joseph.D. Brain INTRODLIC°TLQ'd Workers in the agricultural industry are exposed to an exceptionally wide varaety of inhaled particles. These include fertilizers, pesticides, and herbicides as well as resuspended soil. Moreover, the composition of the soil (for example, the fraction which is free silica) varies from place to place. Other workers are exposed to complex grain dusts, such as that coming frotn various cereal grains (wheat, barley, rye, oats, corn), as well as various contaminants such as insects, mites, rodent debris, and fungi. This wide array of complex dusts presents problems in assessing the potential risk of various occupational exposures in agriculture. In order to understand such exposures, it is possible to measure responses at the molecular, cell, organ, ororganismic level. All approaches reflect the need to evaluate the toxicol- ogy of mate:rials to which agricultural workers are exposed so that we can take appropriate preventive action. Government, unions, and industry now face the difficult task of assessing the toxicology of a wide variety of new chemicals and espe- cially complex mixtures. The creativity of chemists who synthesize new compounds, the availability of new technolo- gies, and final',~y the competitiveness of agriculture ensure that there will bi: a continuing stream of new aerosol exposures whose potential for damage must be assessed. Since they are new, epidemiology fails to provide information about health effects. New;rtheless, a guide to potential toxicity is needed to help design both appropriate control strategies and medical surveillance studies for humans employed in agriculture. How can the risk of human pulmonary disease caused by exposure to complex and often poorly characterized dusts in the agricultural industry be predicted? Risk assessment may include: (1) air monitoring and physical and chemical charac- terization of collected dusts; (2) epidemiologic studies of humans; (3) controlled experimental exposures of humans in the laboratory; (4) chronic lifetime animal studies; (5) short- term animal b:ioassays; and (6) in vitro tests of mammalian cells. This paper emphasizes the fifth method of analysis and discusses the use of short-term animal bioassay systems to determine the health effects of inhaled particulates. Animal studies have numerous advantages since ethical problems are minimized. The possibility of more serious disease can b.- assessed, and there are few limits to the invasiveness o1'the diagnostic procedures used. For example, long-term inhalation exposures of animals, followed by func- tional or histopathological studies of their lungs, have been used to study asbestos,' crystalline silica,2 and coal dust.3 A problem is that such studies are costly and time consuming. A typical lifetime study in rodents costs between 0.5 and 3 million dollars and may take 3 to 5 years to plan and complete. It is also difficult to obtain quantitative estimates of toxicity using standard pathological analyses. Morphometric meas- ures based on extensive sampling of lung tissue as well as physiological or biochemical assessment may be required. Clearly, there is a need for short-term tests. If large num- bers of materials are to be analyzed, it is essential to have assays that are relatively inexpensive and that yield results in weeks or months, not years. Many investigators have pro- moted the use of in vitro assays to assess the potential toxicity of inhaled aerosols ¢' In vitro systems have advantages of reproducibility, cost, and specificity. Several tissue culture systems have been developed.8•9 However, because the human pulmonary response to inhaled particles is the result of com- plex interactions involving many different cell types within the lung, the results obtained may be spurious. For example, inflammation involves recruitment of neutrophils, platelets, and serum proteins to the injured lung. Fibrogenesis involves the action of fibrogenesis-stimulating factors secreted by one cell (e.g., a macrophage) on another cell (a fibroblast). These essential interactions are rarely reproduced in any in vitro system. Short-term in vivo assays can be considered as an alterna- tive to short-term in vitro tests, because the short-term re- sponse of small animals to dusts is sufficiently similar to the human response to have predictive values when properly calibrated and interpreted. The major mechanisms of lung injury10 are common to most mammals. THE HAMSTER BIOASSAY The hamster bioassay features the use of bronchoalveolar lavage (BAL). During the last decade, BAL has been used increasingly to assess lung injury in animals and man. BAL has been employed to discriminate among toxic agents such as metal salts or mineral dusts.""2 Key issues in the application of BAL to inhalation toxicology are the specificity and sensi- tivity of the procedure. What is the smallest amount of dust which causes a measurable response? More important, what is the ability of BAL to discriminate among dusts of varying toxicities and those producing different resulting lesions? To what extent does BAL have predictive value? Can one exam- ine acute events and describe long-term irreversible chronic changes? We have developed a short-term (1 to 30 d postexposure) animal bioassay system in which the toxicity of a particular dust may be estimated by comparison to known dusts with a 39 N ~ N N, ® - .~

Risk An~dysis in Environmental and Occupational Health September 4, 1991 From these results we can construct the following estimates for potency (q, and q,*) in mice. ~ Tumor site:type Sex q, (kg-d/mg) q,* (kg-d/mg) Elone:Various ; F 1.1 x 1e 1.9 x 10, Elone:Various M 1.1 x 10, 1.9 x 1e F'orestomach:Various F 3.0 x 1e 4.4 x 1e Forestomach:Various M 1.8 x 10-1 2.9 x 10' Lung:Total F 1.3 x 10-' 2.9 x 10' Lung:Total M 3.5 x 1e 5.8 x 10' Using the EPA methodology, the value chosen from these would be the highest value of q,* that corresponds to a statistically significant result - 5.8 x 1V kg-d/mg - and this value would then have to be extrapolated to humans using a surface area factor of (70 kgI30 g)"' = 13.26. Such an approach leads to an upper bound estimate of carcinogenic potency in humans of 7.7 x 10"6 kg-dlmg. What does this imply for eating; raw mushrooms in your salad? (1) A upper bound estimate of potency of 7.7 x 10' kg-dlmg implies that the dose rate required to give an upper bound estimate of risk of 10' is 0.013 mg/kg-d, or about 23 g (0.82 oz) per lifetime. (2) A consumption of 1 ozlmonth (13.3 mgikg-d) of raw mushrooms corresponds to an upper bound estimate of lifetime risk of 1 x 1V. (3) According to Toth and Erickson (1986), estimated annual US consumption of these mushrooms was 340 x 106 kg in 1984-1985. This was an annual average per capita consumption of about 55 mg/kg-d, corresponding to an upper bound estimate of lifetime risk of 4.3 x 1V. Presumably not all the mushrooms would be eaten raw, but we have no idea. what would be the effect on the carcinogenicity of the mushrooms of cooking them. (4) With the figures given in (3), the upper bound estimate of the annual number of cancers expected in the US to be due to mushrooms is about 8500! 2

TABLE B3. INDIVIDUAL ANIMAL TUMOR PATHOLOGY OF MALE MICE IN THE 2-YEAR STUDY OF BENZYL ACETATE HIGH DOSE 1 f 1 ! i I Benzyl Acetate N MAL D 0 G D G G I MW1)ER I 21 21 21 21 SI SI SI 31 SI SI 5 I 31 S1 L S{ 41 L/ LI t 61 LI 4 I 4 I L 4 I LI 41 S I I I 1OTAL I 0 t f t t t t 0 t 0 t t t t t t 0 t 0 t t t t f t (TIS5UE51 STUDY ( 91 01 !I 11 $1 11 01 21 01 9/ 11 01 01 01 01 01 71 01 EI 01 01 01 01 01 0 1 7UMOR51 LUNGS AMD 6RONC1II ~ * • • • . • + • • + . o i ~ • • • • • • • • * + • • ~ 56 I NE?ATOCEILULAR CARCIMGMA. METAS I X I t I ALVEDLAbEROMC1tI0LAR AOEMORI ALVEOLAR/1RONCHIOLAR CARCINOMA I I X X X 1 • I TRACNEA I ( a • + + + a a • • a i + + i + • + • _ • + • a + a l f9 I HuuroPOtEilc SYSTEM /ON£ 14lRRON tI / • . • . • / I SPtEE11 Ii + i • • • • • • • • + + + • • • • • - • * + + + + { 49 I MEIWMGIOSARCDMA f ! t i LYMPN 1IODES + . + * • . + • • • + 1 I 5! I 14LIDMANT LYIPMOMA. MI%ID TYPE I t I I TNY?RI$ { • + • • + + + i • • • • • • + * + 4 - • + • • • • 1 I 49 ,I i ULA T FSTEM - [- I HEART { • • • • • • • • + * • • • + • • • • • • • • • • • 1 SI I I V ' I SALIVARY OLAMD / + + 4 ~ LIVER MEOPLASN. NOS { NElATDCELLULAR AD£MOMA I HElATOCELLULAR CARCINOtI X f X X X X X X X X XI tS 1 t I I 11ILE DUCT * * • • + e • • • • . j OALL)LADDER i CD1lION IILE DUC'T ` + a ~ . • • • + . . j PAMCREAS ± • - • •1 49 % ( ESOPM.IGUS • a • • • • • * • • • + • • • • • - • • • + - + 5T0MACM • • • ~ • • • • • • + • • • + • • • • - • • • • + •1 49 { SQUAMOtIS CELL PAPILLDl4 X X X X X I ! I . SOWMOUS CELL CJRCINDMA { SMALL LITESTIME I * * • * • + . - • a + . . • - a • . + ( 47 I LAROE IMTESTIME 1 * • + • • • • * • - • • • • • • • + - • • • • • 4{ 1 RIDMEY ' • • • + e • • • • • • * • • • • • . . • • • • • •1 ~ 51 (' 7UIULAR-C£LL ADEMOMA ' TU3ULUL-CELL ADEMOCJtCIMCMA 1 z X 1 t 1 t I OtSTXART lLADDER I • • • • • + + • • • • • • • + • • • - • • • • • • 49 ( { M Y I PITUITAR't I • • • • - + • • • 1 44 1 ADREMAI ' • • • • L! W N NGLIONEURpU ' GLIONEU RpU X ' t THYROID I'{ FOLLICULAR-CE1L ADEMOMA + . • • . • • * • + • • e + + • • + - • • ~ ' PARATMYROID I . * + • • + i • • • • + + - • • • • - . _ + + - • ' PANCREaTIC ISLETS 1 • • • • • • + • • e • • • • • • • • - • • • • • •{ 49 ISLET-CEIL ADEMCItA f X 2{ MAMMARY OLAND 1 M j TESTIS • • + + • • • + • * a • + r a + • • + a I S! IMTERS7ITIAL-CELL TUNOR X ' Y 2 1 g t15TATE 1 + • • L9 j 0U5 STSTEN ; I 1 /RAIM i • • • • • • + • * • • • • • • + • + • + • • • • • S~ 1) M MARDERIAM OLAMD ' ~ M M M M M M N M M N N M M M M M M N N N M M N M Nr SB ADEN014. M0S X X 1i 3 /ODI CAVITIES MESEMERY 1 N M M M N N M M M M M N M M M N N N It M N M M N N ~ SON ( NEPATOCELLULAR CARCINGMA. METAS ; ~ t { ER S75TE11S MULTIPLE OROAMS NOS i M M N M M M M N N M N M M M M M M M p N M M M N Nj ~ S0u I HEPATOCELLULAR CARCIMD74t. MEfAS MALIGNANT LTMPHOMA. N05 ' t / 1 MALIG.LYKPHOMA. LYMPHOCTTIC T1T X 7 f M AMIMALS MECROPSIED •. TISSUE EXAMINED MICROSCDPICALLY NO TISSUE INFORMATIOM SUDMITTED ~ REOUIRED TISSUE N0T EXAMINED MICROSCOPICALLY Ct NECROPSY. N0 MIST0L04Y DUE T0 PROTOCOL X TUifOR INCIDENCE A: AUTOLYSIS Ms S+ NECR0PSY. k0 AUTOLYSIS. NO MICRGSCDPIC EXAMINATIOM AkIMAL MIS-SEXFD M: D= ANIMAL MISSING MO NECRDPSY PERFORMED 1F ;z ~ .~ $ ~ ! !

214 9. National Cancer Institute (NCI), "Bioassay of 1,2- Dibromoethane for Possible Carcinogenicity," NCI Technical Report Series No. 86, Bethesda, Maryland (1978). 10. Environmental Protection Agency, "The Carcinogen Assess- ment Group's Risk Assessment for Ethylene Dibromide (EDB)" (1978). 11. Environmental Protection Agency, Office of Pesticide Pro- grams "Ethylene Dibromide" Position Document 4, Washing- ton, D.C. (1983). 12. M. G. Ott, H. C. Scharnweber, and R. R. Langner, "Mortality ,.xperience of 161 employees exposed to ethylene dibromide in ;wo production units." British Journal of Industrial Medicine :37, 163-168 (1980). 13. G. W. Gribble, "Ethylene Dibromide Uproar" (Letter), Chem- ical and Engineering News 62, 63 (1984). 14. National Cancer Institute (NCI), "Carcinogenicity Bioassay of 1,2-Dibromoethane in F344 Rats and B6C3F1 Mice (Inha- lation Study)," NCI Technical Report Series No. 210, Re- search Triangle Park, North Carolina and Bethesda, Maryland (1982). 15. I). Krewski, C. Brown, and D. Murdoch, "Determining 'Safe' Levels of Exposure: Safety Factors or Mathematical Models?" 1'undamentals of Applied Toxicology 4, S383-S394 (1984). 16. C. K. Wong, J. M. Winston, C. B. Hong, and H. Plotnick, "Carcinogenicity and Toxicity of 1,2-Dibromoethane in the P'.at;" Toxicology and Applied Pharmacology 63, 155-162 (1982). 17. K. S. Crump, "An Improved Procedure for Low-Dose Carcinogenic Risk Assessment from Animal Data," Journal of Environmental Pathology, Toxicology, and Oncology 5, 339-345 (1984). 18. P. Armitage and R. Doll, "The Age Distribution of Cancer and a Multistage Theory of Carcinogenesis," British Journal of Cancer 8, 1-12 (1954). 19. R. Doll, "The Age Distribution of Cancer: Implications for Models of Carcinogenesis," Journal of the Royal Statistical S xiety 134, 133-166 (1971). 20. R. Doll and R. Peto, "Cigarette Smoking and Bronchial Carcinoma: Dose and Time Relationships Among Regular Smokers and Lifelong Nonsmokers," Journal of Epidemiology and Community Health 32, 303-313 (1978). 21. IC, S. Crump, "Dose Response Problems in Carcinogenesis," Biometrics 35, 157-167 (1979). 22. L. Zeise and E. A. C. Crouch, "Reconciling the Results of Carcinogenesis Bioassays with Benzo(a)pyrene for the Purpose of Risk Assessment," Society for Environmental Toxicology and Chemistry Annual Meeting, Arlington, Virginia (1983). 23. T. W. Thorslund, "Estirnation of the Effects of Exposure to a Carcinogen that Fluctuates over Time on the Lifetime Risk of Cancer Death," AAAS Pacific Division 63rd Annual Meeting (1982). 24. K. S. Crump and R. B. Howe, "The Multistage Model with a 'lirne-Dependent Dose Pattern: Applications to Carcinogenic Risk Assessment," Risk Analysis 4, 163-176 (1984). 25. 1. I). Kalbfleisch and R. L. Prentice, The Statistical Analysis of Failure Time Data (Wiley, New York, 1984). 26. C. Sawyer, R. Peto, L. Bernstein, and M. C. Pike, "Calcula- tion of Carcinogenic Potency from Long-Term Animal Carcinogensis Experiments," Biometrics 40, 27-40 (1984). 27. R. Peto, M. C. Pike, L. Bernstein, L S. Gold, and B. N. Ames, "'i'he TD,,: A Proposed General Convention for the Numeri- Hertz-Picciotto, Gravitz, and Neutra cal Description of the Carcinogenic Potency of Chemicals in Chronic-Exposure Animal Experiments;' Environmental Health Perspectives 58, 1-12 (1984). 28. L. S. Gold, L. Bernstein, J. Kaldor, G. Backman, and D. Hoel, "An Empirical Comparison of Methods Used to Estimate Carcinogenic Potency in Long-Term Animal Bioassays: Life- table vs. Summary Incidence Data," Fundamentals of Applied Toxicology 6, 263-269 (1986). 29. E. C. Hammond, I. J. Selikoff, and H. Seidman, "Asbestos Exposure, Cigarette Smoking and Death Rates," Annals of the New York Academy of Science 330, 473-490 (1979). 30. R. Peto, "Carcinogenic Effects of Chronic Exposure to Very Low Levels of Toxic Substances," Environmental Health Per- spectives 22, 155-159 (1978). -31. K. S. Crump, D. G. Hoel, C. H. Langley, and R. Peto, "Fundamental Carcinogenic Processes and their Implications for Low Dose Risk Assessment," Cancer Research 36, 2973-2979 (1976). 32. P. Armitage, "The Assessment of Low-Dose Carcinogenicity:" Biometrics Suppl. 38, 119 -129 (1982). 33. I. W. F. Davidson, J. C. Parker, and R. P. Beliles, "Biological Basis for Extrapolation Across Mammalian Species;" Regul Toxicol Pharmacology 6, 211-237 (1986). 34. P. J. van Bladeren, J. J. Hoogeterp, D. D. Breimer, and A. van der Glen, "The Influence of Disulfiram and Other Inhibi- tors of Oxidative Metabolism on the Formation of 2-Hydroxy- ethyl-mercapturic Acid from 1,2-Dibromoethane by the Rat " Biochemical Pharmacology 30, 2983-2987 (1981). 35. J. N. Rowe and J. A. Springer, "Asbestos Lung Cancer Risks: Comparison of Animal and Human Extrapolations," Risk Analysis 6, 171-180 (1988). 36. California Department of Health Services, "Health Effects of Benzene, Report to the Scientific Review Panel, Part ' (1984). 37. P. E. Enterline, "A Method for Estimating Lifetime Cancar Risks from Limited Epidemiologic Data," Risk Analvsis 7, 91-96 (1987). 38. 1. Hertz-Picciotto, R. R. Neutra, and J. F. Collins, "Carcino- genicity of Ethylene Oxide: A Comparison of Dose-Response Data in Animals and Humans" (letter), Journal of the Ameri- can Medical Association 257, 2290 (1987). 39. E. Crouch and R. Wilson, "Interspecies Comparison of Carcinogenic Potency; " Journal of Toxicology and Environmen- tal Health 5, 1095-1118 (1979). 40. E. A. C. Crouch, "Uncertainties in Interspecies Extrapolations of Carcinogenicity," Environmental Health Perspecives '50, 321-327 (1983). 41. D. W. Gaylor and J. J. Chen, "Relative Potency of Chemical Carcinogens in Rodents," Risk Analysis 6, 283-290 (1986). 42. L Bernstein, L. S. Gold, B. N. Ames, M. C. Pike, and D. G. Hoel, "Some Tautologous Aspects of the Comparison of Carcinogenic Potency in Rats and Mice," Fundamentals of Applied Toxicology 5, 79-86 (1985). 43. D. M. Siegel, I. Hertz-Picciotto, M. Lipsett, and R. Neutra, "Health Effects of Cadmium. Part B. Report to the Air Resources Board." California Department of Health Services (1986). 44. I. Hertz-Picciotto, "The Role of Assumptions in Quantitative Cancer Risk Estimation," American Public Health Association 114th Annual Meeting (1986).

3 5, a-A concentration gradients 6. Diffusing capacity (carbon monoxide uptake) C. Measurement of pathology by radiologic techniques 1. Atelectasis 2. lFibrosis, emphysema, etc. 3. l3ronchography (Tantalum) 4. Focal lesions D. M:ucociliary transport (in vitro and in vivo) 1. Nasal 2. Asirways 3. Mucus studies 4. Cilia studies E. Luni; lavage parameters 1. Surfactant: quantity, composition 2. Cell numbers, appearance, and viability 3. Cell differential counts: RBC's, PMN's, monocytes, macrophages, lymphocytes 4. Proliferation: production of colony-forming units (CFU's) by lavaged cells, uptake of tri- t.iated thymidine 5. Mucus constituents 6. Biochemistry: albumin, hemoglobin, hydroxyproline, elastase, collagenase, LDH, myelo- peroxidase, antiproteases, lysosomal enzymes, active oxygen species, chemotaxins, proli- ferative factors, and inflammatory mediators (histamine, prostaglandins, leukotrienes) 7. In vitro functional assays of macrophage activity: trypan blue dye exclusion, oxygen con- sumption, ATP levels, lactate production, migration, chemotactic responsiveness, phago- cytosis, killing of microorganisms, release of mediators F. Morphology 1. Gough sections 2. Reid index 3. Mo:rphometric approaches: airway and alveolar dimensions 4. Cell types: connective tissue, inflammatory, neoplastic 5. Proliferation and cell turnover measures 6. Vascular changes G. Renewal of lung constituents observed in tissue sections 1. Metaphase counts - colchicine 2. Uptake of tritiated thymidine 3. Collagen and elastin breakdown and synthesis H. Lung clearance

t NON-CANCER ENDPOINTS There are perhaps three considerations that distinguish the health risk evaluation process for cancer endpoints from that for non-cancer endpoints. I. While carcinogenic effects are thought to be linear with dose all the way to zero dose, for non- cancer endpoints there exists a threshold dose level below which no adverse health effects occur. This level is typically called the reference dose (RfD), allowable intake chronic (AIC), or no observed adverse effect level (NOAEL). II. The contrast between target tissues and the rest of the body is generally more sharply drawn than in carcinogenesis. That is, with non-cancer endpoints the target tissue f organ is often exquisitely susceptible to harm in comparison to other body tissues. Calculation of health effects Often calls for the use of physiologically-based pharmacokinetics (PB-PK), so that dose to target tissues can be more closely estimated. III. Non-c-mcer endpoints of injury are much more widely varied and toxin-specific than in cancer, where we believe there is primarily one endpoint, genetic damage, and one outcome, death, that we seek to avoid. Because of the diversity in non-cancer endpoints, it would be impossible to present an overall survey, and one example will be discussed in some depth. Many of the principles can be extrapolated to other organ systems. Assessment of Risk for Inhaled Airborne Material There are many methods available to assess the toxicity of inhaled agents. As summarized below, these tests ra.nge from studies in human populations, to measures of lung function in whole animals and histopathological studies of lungs from exposed animals, to in vitro measures of pulmonary macrophage fimction (phagocytosis, viability), etc. The following outline describes various categories of lung injury and types of assays for indicating onset of tissue- damage. I. Inhalation toxicology data development A. Air fronitoring and characterization of collected dusts. B. Epidemiologic studies of previously-exposed populations. C. Ctin.cal trials using controlled exposures of humans. D. Animals, chronic lifetime studies. E. Short term animal bioassays. F. In vit)-o tests on mammalian or non-mammalian cells. G. In vitro examination of molecular interactions with phospholipids, enzymes, nucleic acids, etc.

Methodologies in Respiratory Occupational Surveillance 43 sensitive, The use of BAL in short-term animal assays can be an important source of information regarding the toxicity of new and poorly characterized inhaled particles. mg DUST INSTILLED/100q BODY WEIGHT REFERENCES e2 3 1. Brody, A,R. and DeNee, P.B., Biological activity of inorganic CONTROL F 0 a-OUARTZ particles in the lung, CRC Crir. Rer. Toxicol., 7, 277, 1981. 2. Gross, P., DeVilliers, A.J., and deTrevelle, R.T.P., Experimental silicosis, Arch. Parhol., 84, 87, 1967. 3. Busch, R.H„ Filipy, R.E., Karagianes, M.T., and Palmer, R.F., Awo Pathologic changes associated with experimenal exposure of rats to 0 75 3 75 FIGURE 8. The fraction of gold particles, lambda ingested by macrophages in situ. is shown. Measurements were made I d after exposure to iron oxide. a-quartz, and AWO. Values are mean ± standard errors. (Adapted from 13e:k, B. D.. Brain, J. D., and Wolfthal. S. F., Inhaled Particles N, Dcdlson, J., Ed., British Occupational Hygiene Society, Edinburgh, Scotland.) 3001 E 250~ 2001 a-QUARTZ Fe203 AWO l l~- --~ CONTROL 0 5 10 15 DAYS AFTER INSTILLATION FIGURE 9. Time course for LDH in the extracellular supematant fraction of lung lavage fluid after exposure to 3.75 mg iron oxide, a-quartz, or AWO per 100 g body weil;ht. Values are mean ± standard errors. (Adapted from Beck, B. D., Brain, J, D., and Wolfthal, S. F., Inhaled Particles IV. Dodgson, J., Ed., British Occepational Hygiene Society, Edinburgh, Scotland.) encountering exposures to the same agents in occupational and urban environments. Information can be gathered from BAL relating to the extent and type of lung injury and the mechanisms involved. Needed are more extensive compari- sons of injury as judged by other approaches with the results of BAL. For example, short-term bioassay results can be integrated withL industrial hygiene and epidemiology results as was done in a recent study of talc and granite dusts.25 It is also likely that other constituents of BAL can be quantified which will help makl: bioassays utilizing BAL more specific and coal dust, Environ. Res., 24, 53, 1981. 4. Dean, J.H., Boorman, G.A., Luster, M.I., Adkins, B., Jr., Lauer, L.D., and Adams, D.O., Effect of agents of environmental concern on macrophage functions, in Mononuclear Phagocyre Biology. Volkman, A., Ed., Marcel Dekker, New York, 1984, 473. 5. Liu, W.K., Tsao, S.W., and Wong, J.W.C., In vitro effects of fly ash on alveolar macrophages, Consen-ation RecYcling, 7,361, 1984. 6. Snella, M.-C., Manganese dioxide induces alveolar macrophage chemotaxis for neturophils in vitro, To.ricology, 34, 153, 1985. 7. Hatch, G.E., Boykin, E., Graham, J.A., Kewtas, J,, Pott, F., Loud, K., and Mumford, J.L., Inhalable particles and pulmonary host defense: in vivo and in vitro effects of ambient air and combus- tion particles, Environ. Res., 36, 67, 1985. 8. Kaw, J.L., Tissue culture in pneumoconiosis, CRC Crit. Re •. Toxi- col., 5, 103, 1977. 9. Miller, K., The effects of asbestos on macrophages, CRC Crir. Rer. Toxicol., 5, 319, 1978. 10. Fantone, J.C. and Ward, P.A., Mechanisms of lung parenchymal injury, Am. Rev. Respir. Dis., 130, 484. 1984. 11. Henderson, R.F., Rebar, A.H., Pickrell, J.A., and Neulton, G.J., Early damage indicators in the Iung.llI. Biochemical and cytological response of the lung to inhaled metal salts. Toxicol. Appl. Pharma- col., 51, 123, 1979. 12. Beck, B.D., Brain, J.D., and Bohannon, D.E., An in vivo hamster bioassay to assess the toxicity of particulates for the lungs. Toxicol. Appl. Pharmacol., 66. 9, 1982. 13. Brain, J.D., Knudson, D.E., Sorokin, S.P., and Davis, M.A., Pulmonary distribution ofpanicles given by intratracheal instillation or by aerosol inhalation. Environ. Res., 11. 13, 1976. 14. Weissman, G., Smolin, J.E., and Korchak, H.M., Release of inflammatory mediators from stimulated neutrophils, t\'. Engl. J. Med., 303, 27, 1980. 15. Hook, G.E,R„ Extracellular hydrolases of the lung. Biochemistn, 17, 520, 1978. 16. Bell, D.Y., Haseman, J.A., Spock, A., McLennan, G., and Hook, G.E.R., Plasma proteins of the bronchoalveolar surface of the lungs of smokers and nonsmokers, Am. Rev. Respir. Dis., 124, 72, 1981. 17. Merrill, W., O'Hearn, E., Rankin, J., Naegel, G., Matthay, R.A., and Reynolds, H.Y., Kinetic analysis of respiratory tract proteins recovered during a sequential lavage protocol, Am. Rev. Respir. Dis.. 126, 617, 1982. 18. Chichester, C.O., Palmer, K.C., Haves, J.A., and Kagen, H.M., Lung lysyl oxidase and prolyl hydroxylase: increases induced by cadmium chloride inhalation and the effect of beta-aminopropioni- trile in rats, Am. Rev. Respir. Dis., 124, 709, 1981. 19. Brain, J.D. and Corkery, G.C„ The effect of increased particles on the endocytosis of radiocolloids by pulmonary macrophages in t•irro: competitive and cytotoxic effects, in Inhaled Particles Il', Walton. W.H., Ed., Perfamon, New York, 1977, 551. 20. Beck, B.D., Brain, J.D., and Bohannon, D.E., The pulmonary toxicity of an ash sample from Mt. St. Helens volcano, Exp. Lung Res., 2, 289, 1981. 21. Martin, T.R., Chi, E.Y., Covert, D.S., Hodson, W.A., Kessler, D.E., Moore, W.E., Altman, L.C., and Butler, J., Comparative effects of inhaled volcanic ash and quartz in rats, Am. Re v. Respir.

From: Science, Vol. 245, No. 4915, pages 269 - 276 (21 July 1989). The Rat as an Experimental Animal THOMAS ,(. GILL III,~ GARRY J. SMITH, ROBERT W. WISSLER, HEINZ W. KUNZ The development and characterization of many inbred, congcnic, and, recombinant strains of rats in recent years has led to the cietailed genetic description of this species, especially in re3;ard to its major histocompatibility com- plex. This information has contributed substantially to the study of camparative genetics and has greatly en- hanced the utility of the rat in a variety of areas of biomedical res4arch. This article focuses on the use of the rat in immun,ogenetics, transplantation, cancer-risk as- sessment, cardli6vascular diseases, and behavior. HE RAT IS A, MAJOR EXPERIMENTAL ANIMAL IN TRANSPLAN- tation, imm.unology, genetics, cancer research, pharmacolo- gy, physiology, neurosciences, and aging. The strains and randomly bred stocks that have been used almost exclusively arc derived from the Norway rat (Rattus norvegicus), which is thought to have originated in the area between the Caspian Sea and Tobolsk, extending as far c:azt as Lake Baikal in Siberia. It spread to Europe and the United States with the development of commerce in the 18th century, and by the middle of the 19th century it was being century by H. H. Donaldson, W. E. Castle, and their colleagues for studies in basic genetics and in cancer research (1). Further develop- ment and genetic characterization of inbred, congenic, and recombi- nant strains occurred in the United States, Japan, and Czechoslova- kia (2), and several reviews have documented these developments in detail (3-5). In addition to its experimental uses, the rat has a worldwide economic and medical impact, since it destroys one-fifth of the world's crops each year, carries many diseascs that are pathogenic for humans, and kills many children by direct attack (6). This review will focus on current work utilizing the rat in immunogenctics, transplantation, cancer-risk assessment, cardiovas- cular diseases, and behavior. In these areas of research, the rat has the advantage of being a well-characterizcd, intermediate-sized rodent without the disadvantages, both scicntific and cconomic, of larger animals and without many of the technical disadvantages of smaller rodents. T. J. Gill III is the Maud L. Mcntcn Professor of Experimenal Pathology and professor of human genetics and H. W. Kunz is associateprofessor of pathology at the Univcr- siry of Pittsbur~h Schoo! of Medicine Pittsburgh, PA 15261. G. J. Smith is associate professor and director of the Cardnogenesis Research Unit at the School of Pathology of the Universiry of New South Walcs, Keuington, New South Wales, 2033 AustzaLa. R. W. Wissler is the Donald N. PritzScer Distin guWxd Service Professor of Pathol g . g q , s at t e n vers ty used extensively for studies in anatomy, physiology, and nutrition. The first inbredlinis were developed at the begirlning of the 20th Chica o IL 60637 active emcritu of Chin o Sc1 of Medicin i h i U 'To whom correspondence should be addressed. 21 JULY 1989 ARTICLES 269

210 Hertz-Picciotto, Gravitz, and Neutra Table IV. For Total° Cancer Deaths Predicted° by Several Models for EDB-Exposed Workers` Texes MicEug,an Overal l Observed cancer deaths 3 5 8 Models fitted to gavage data Proportional hazards Multistage with Multistage with One-hit Stomach Lung time-to-tumor variable dosing 56.6 21.4 50.4 15.5 14.0 34.2 11.7 24.3 6.1 7.7 90.8 34.5 74.7 21.6 21.7 Models fitted to inhalation diu One-hit: terminal sacrifice tumor Omitted Included 4.2 2.5 6.7 4.9 2.9 7.8 °Tot,al cancer deaths=(expected+predicted excessl. ^ Other assumptions are described in the text. Values in the table represent upper 95% confidence limits, except for the proportional hazards model, which the variance estimates were too unstable to derive an upper confidence limit. `Assumes 3.0 ppm exposure during time employed. Table V. Effect of Interspecies Dose Conversion Factor on Cancer Risk Predictions° a Dose equivalence by Predicted excess cancer deaths mg/kg/day mg/kg2/3/day Texas 0.9 pm 0.8 4.1 3.0 ppm 2.6 11.9 Michil att 0.9 ppim 0.3 1.3 3.0ppm 0.8 3.9 °Time-tatumor model fitted to gavage data. Values in the table represent upper 95% confidence limits. 3.3. Int:rspecie§ Scaling Factor As noted, the gavage analyses used surface area to scale the doses from animals to man. To investi- gate the role of the interspecies scaling factor we repeated the analysis that fitted the multistage with time-to-tumor model to the NCI gavage bioassay data, utsiing mg/(kg body weight)/day equivalence rather than surface area equivalence. A compari- son of these two analyses is shown in Table V. Sur- face area equivalence yielded risk estimates that were about five times larger than those based on mg/kg/day equivalence. 4. DISCUSSION The large uncertainty in risk assessment due to extrapolati,ng between high and low doses is of con- siderable concern. Unfortunately, the only feasible way to conduct a sensitive animal bioassay is to use high doses, since the risks at low doses generally cannot be detected unless many thousands of animals are treated; nevertheless, such risks may be of con- siderable public health concern if exposures are widespread. Thus, the uncertainty of high-to-low dose extrapolation is unavoidable. One result of this inves- tigation was to develop a means of narrowing the range of uncertainty by comparing model-based estimates derived from animal data to the observa- tions in epidemiologic studies. All the linear nonthreshold additive risk models considered here for extrapolating human cancer risks from animal bioassay data performed well when validated against the mortality of EDB-exposed workers. That is, the predictions are compatible with the reported cancer deaths in the occupational study of Ott and colleagues.t12> Even a crud`e extrapolation using direct proportionality, which is equivalent to drawing a straight line between zero excess risk at zero dose and the excess tumor rate in the low-dose rats, gave plausible risk estimates when compared to the epidemiologic data. This compatibility contrasts sharply with the findings of Ramsey and co- workers.S2> We discuss the reasons for this difference and the implications of our findings. 4.1. Choice of Model On theoretical grounds, the multistage model for variable dosing and time-dependent risk appears to be the most appropriate of the additive models for the analysis of the EDB gavage data. This is because it utilizes the full information on both survival times and dosing pattern and makes no assumptions re- garding dose rate. Considering the uncertainties in- volved in risk extrapolation it is apparent that the

20255457S9 h, ~ ~.I~~ . _j'-~~ • -; --r -rr rn * LCL ~ ~ ;_r-..-~ ~ EEi4 , (.' 1 1 ! !. • . ~ I , - ~ 1 y 1 I - . I _..J. _...,_ . -I ' _ 1 . . _, . . {-_- :~__~_i-h. ~ T"-'-•.1-' - ~ ~ T l i - •. . •--~ ----r t , ,_.._~_ - _~- ' f~~ ri _L L.: '_ .'t .. . , _... i - ~~ i -'-'- . _.~J_~... . :. - _ - - ~ ~ - - -- - - - ` - - - ._ --- - '- ;: - - . -i , - , - - " - -- -'- ~ - -- - '- - .l.l~_.__ .. ~ . _. ._- .. , . . . ~ - --- , L_._E_..__......_ . _.... .t..._•.- ~ . . ; ..•_.._~....L:..:. , . _' :. ~:'... :~.~._,....•.. ... ~ ~:... _...... -.... ~~.. , , ... _ ........... ,. .: . ..--•- . . _._._.__._,_,._... , . .,. I . „ ..~._._._I_...._ • ~ . . . ' --f- ....... .. -'• . • . ..... _. ...-._. ~ . ......-., . , . ..... ._.,._.. .. . . ~ . . . . . . . . _.....-t__..... _ . _ . . ...~1._ .~.. ...i . . .' i __ •_i..i_ . , t- . . i , . . , __..... ..~... _ _. . , . • . . .. . ...t ,. . . . • .. .- ..-.._.. _ _.._ _ . : ...._.... ... . .. .. - . . . .. .- .. . .. . _ ...... ..__: .. .. - - -;- - _ :' - - - . :... : .,._. - --- __ : ~.t....__.... .. . ._....- . .. _ . . . . . --- - ...., . ._ . - -.._..... . .- - - '~ ., _._ ..._.__ . ; ' .. _._. - . . .. _ - .._.. _... . . .... •- - - , , . , • .. ~ , ~• ~ : - i.•, t l . . ._. - _ .. • . ., ., t . . _ : , `. , .,. . .., . . . ., .. ~ t ~• .. . ~;~. _: . . i ... •_-- . ' .___..._...-- . --- - "--- . , . . ._~ .._ y_...._.- . . _ .. .. . . _ ,. .. , .. _ . .. : - • . , • " . . . .. .... •- _........._. . ..... ... ~ ... .. -.. . _ . _ _ . _ _.. ..._ - •_... . . .. ... _ - _ _. .. . ....._ _ _ _ _ _ .... _ _ - . .. .. . _-- , ~ t ., . . , _.J_ . . - , . . , . . •--: .., ... ._... _.. „ t , .. , .t. __ _ . . . . .._. . _- , .., - -i ; .._ - - . .... .. . .. .~ r - ; .. . ... . . ._ _ . .... , t- -' - • _. . . . . ...__. .. _ _ ._. . ! __.. .. - - - ._._. _ . ,..., - -_-. - ., . .. . , __........ ~ .. . ------ .. . ~ , ; .:. ..,_ ----r- - - , . , . • .,. ~ , .._.. ' - - - - - ; _: - - . . - -- - --'-'----- - ... ----_--__.__. . . ~_,_._-__J_ ._ ...._ .._ . . . .. __....... ..._-_.._ _ _ - - --- : , ,. . _;. • _._. _ _L.' • . .... _ - --- -- - ` ~ t _ !_ _ ~ .~ / E~~ - - - --- _ - ... . .. .... .. . _ ._ ._ ..- . _ _... -... . _ , t _... .. . -..._ _ 1 ~ .... . . . .. __ .. . . t- ~ ---..__._+.-, _. t _!~.. ._.. ._1..._.__ , _ - - r'- , i _ t _ ..1- -- r- t _ -- - _ _ . - -L` - ----- -... ~I -- ' t.. - ---~- - - . •- - ---- ---- --- = L ...-- - -- -- ~..__.. . .. _. ., . -•-- ~r- -° - _ . -- t •--.. ' -- _. ~ ' ! -•~ . ' : .. ..- ~_ ~ -' _ ~.. -- --- - - --'----- -I - -' ; --r--•--r--• - • ,, ....l-= '~- - -.- . . . . .. .. ,.. -I_ -; , -- - .. _ _ . . . . .... __ - -- ' - -- - ._ .. _ .. . .. - ---LL.;._i . . .: -- -- ~ - - - ; . - -- - - _ .--_.._ ._-...,_ _>_:_ ._ . _ ._•. . ._..;.. ._-._.._ _ . .,._- __.. , . _ _ ___ ... __ ._. :.. :. .. .. ,•. ;. . ~...- - „ _. .. -. _ - _ .... . ..... .. _ _ . . __. _ . _ _ _ { .. - . . . , . - . • ~ . .•--:- - . . - , -, ' . . t ~T' _.. ~911 1AY ...,. I . .~, ~ ..~ •;.. : -' -- .,, . ,. . .. ... .. -; ' - : ' _~_ _ . _. _7/rIOY~Yy/ .., , . ... . . / •..~~~.)'r7ll~ l d/Yd' iYf1WJ;YY,Y17~3!l ..9~J? • ~ - --..._- -;' , ..,, : _ . .. . ..... . . . . . ' . I . ., - -- - - . . . - ••- -- ._~_ :~. ... -... . . ... .. ~ . .. _ _ • - ~ , ~- . . . .. - , . ~ - -i I I • .-- -:----' - ._ . . .., _ ..- - - . -_ -- - ~ ----- - -'- -~---- .__ . _:_I_ ' - . ._.. .. I/ _L - . . . . . . . . . _-. . .-.... - _ - ~ . . .... . ' . ~. _.. .. .. . .._ ._ . _ _ _. _ .1 ..- _ .-. . _ . _ . ..... - . . . _ . . Xlk ~ - ' I i ~ I_ . _ ~.; .i . -.. . . ..... . I i _. .~ i i . . . I _ . - _. . . . .:. . I . . , .... .! . . . i .... _ . . . _ . . ~ . ~...•_L~ _ ~ , . . -- . . . . ~-• .. . . _ -- .. . . _... .. . I : . ~..• - . - - • - -- . .. .. _ ._ .. . . - - .. , . _,__. ~.. ~~ . _ ._ ! . . . . ....;. -

acids.1'6" Althou,jlh bone marrow enzymes are not efficient for henzene metabolism, phenol can he meta(xolized in marrow via mveloperoxidase.(") Benzene -netabolism to phenol, formatiun of water-soluble phenyl glucuronide and sulfate conjugates, and conjugation with glutathione and urinary elimination of benzene as the phenylmercap- turic acid are considered detoxication pathways. Micro- some ring-opening reactions giving rise to the reactive mucondialdehyde yield muconic acid, a pathway consid- ered responsible for at least some aspects of benzene tox- icity. Hydroxylation of phenol generates hydroquinone; dehydrogenation of benzene dihydrodiol generates ca- techoLt78,79> Hydroquinone and catechol can accumulate in bone marrow and lymphoid tissues;t80> hydroquinone can oxidize spontaneously in vitro to para-benzoquinone under physiologic conditions.(si,sz) Catechol does not ox- idize spontaneously under these conditions; however, it can be metabolized (presumably the cytochrome P-450 system) to 1,2,4-benzenetriol.(H3> The toxicity of hydro- quinone and 1,24-benzenetriol involves free radical for- mation via superoxides; covalent binding of the semiqui- nones to DNA, RNA, and other cellular components; and direct alkylation of sulfhydryl groups by para-benzoqui- none or its der.5vatives. Hydroquinone and benzoquinone were the most toxic metabolites to cultured bone marrow stromal cells, where catechol and benzenetriol inhibited colony growth only at very high benzene doses to male B6C3F1 mice.ts'» Injury to bone marrow stromal cells has been implicated as a precursor step to benzene hemato- toxicity!1~0> A recent symposium on benzene metabolism, toxicity, and carcinogenesists`'> provides an authoritative summary on benzene biotransformation and the implica- tion for human health risk assessment. Human Studies As an acute poison, benzene produces narcotic effects comparable to those of toluene. Benzene is considered very toxic; probable human oral lethal dose would be between 50-500 mg/kg (1 tsp to 1 oz).(135) Human inha- lation of approximately 20,000 ppm (2% in air) was fatal in 5-10 minutes.t86> AkSo,v et al ts7-H9> studied 28,500 Turkish shoe and hand- bag production workers who inhaled an average of 150- 210 ppm when b°nzene-containing adhesives were used and 15-30 ppm at other times. Peak benzene exposures varied between 210 and 640 ppm, and the duration of exposure was estimated to average 9.7 years. Of the 44 cases of pancytolxnia, 23 (52%) experienced remission of the aplastic anernia, 14 (32%) died from complications of aplastic anemia or pancytopenia, and 6 (14%) later died from leukemia. Of 42 leukemia cases, 26 percent were preceded by a 6-month to 6-year period of pancytopenia prior to the onsei: of leukemia. Akso}(90,9'> reported an update to the above cohort to the year 1983, wherein a total of 73 patients chronically exposed to benzene were examined. Fiftv-one of the 73 had leukemia, 12 had ma- lignant Ivmphoma, 4 had multiple mveloma, and 6 had lung cancer. Among the 51 leukemic patients, 20 were afflicted with acute mveloblastic leukemia.' were consid- ered preleukemic, 20 were diagnosed with acute etyth- roleukemia, 5 had acute mvelomonoc-ytic leukemia, and 1 was diagnosed as an acute undifferentiated leukemia. Thir- teen of the 51 leukemic patients had suffered pancyto- penia; the average duration of benzene exposure was 9.93 years. Vigliani 02) studied groups of workers employed in ro- togravure plants, shoe factories, and other industrie,s where benzene was used as a solvent. Benzene concentrations in air near the rotogravure machines were 200-400 ppm, with peak values as high as 1500 ppm. Sixty-six cases of benzene hemopathy were observed, and of the 18 deaths in this group, 7 died of aplastic anemia and 11 died of leukemia. In a second group of workers where ambient benzene ranged from 25-600 ppm, 135 workers with ben- zene hemopathy were studied. Of the 135, 16 died (3 from aplastic anemia and 13 from leukemia). Infante et al t93> reviewed death certificates for a cohort of 748 white male workers who had been occupationally exposed to benzene from 1940-1949; exposures are not known precisely but ranged up to 100 ppmt94> Other,05> cite reports that peak exposures may have been as high as 200-350 ppm. Vital status was followed up to 1973. A fivefold excess risk of all leukemias was reported, and a tenfold excess of deaths from myelogenous and monocytic leukemias was observed. In a follow-up through June 30, 1975, Rinsky et al.f96> reported 7 deaths from leukemia versus 1.25 expected (standardized mortality ratio [SMR] = observed no. deaths/expected no. deaths = 560). When compared by length of employment, there was a significant excess of leukemia observed among workers employed five or more years, but not among those employed less than five years. Two workers died from leukemia among the group employed less than five years compared to 1.02 expected (not statistically significant). Among those em- ployed for five or more years, five died from leukemia compared to 0.23 expected (SMR = 2100). Short-term area samples measured between 1946 and 1976 indicated that most benzene levels were below 100 ppm and some were above 100 ppm.t95,56) Rinsky et al.t96> cite documents in- dicating that these workers were required to wear respi- rators (efficiency not stated) when exposed (even mo- mentarily) to concentrations greater than the TWA (ranging to a maximum allowable concentration of 100 ppm in 1941 to an 8-hour TWA of 10 ppm from 1969 on). For those individuals with more than ten years of employment, three leukemia deaths were observed as compared to 0.09 ex- pected (SMR = 3300). Cumulative benzene exposure was calculated for each member of the benzene cohort in ppm- years, and the cohort follow-up was extended to 1982.( ') A total of 1165 white males with at least one ppm-day of cumulative benzene exposure (to December 31, 1965) were included in the cohort for a total of 31,612 person-years at risk. Fifteen deaths in this cohort were observed from lymphatic and hemopoietic cancers versus 6.6 expected (SMR = 227). Nine cases of leukemia were observed com- 458 APPL OCCUP. ENVIRON. HYG. 50 - JULY 1990

2 II. Mechanisms of lung injury As a consequence of inhaling toxic gases and particles, a number of pathological processes may be set into motion. None are specific to the lung, but their expression and consequences depend on the unique architecture and physiological role of the respiratory system. Major pathological mechanisms to be discussed are: A Pulmonary edema: Transudation of fluid, altered alveolar stability, impaired gas exchange, and respiratory distress B. Inf7ammation: Irritation leading to mucosal edema, increased mucus production and bron- chitis, appearance of neutrophils and inflammatory mediators, increased cell renewal C. Immunologic reactions: Asthma, hypersensitivity lung disease, extrinsic allergic alveolitis, anaphylaxis - D. Altered susceptibility to infection: Cytotoxic and competitive effects on macrophage func- tion, altered mucociliary transport because of changes in cilia or the quantity or rheological character of mucus E. Infection: Bacterial, viral, or fungal pneumonia F. Proteolysis: Destruction of elastin and collagen leading to emphysema, obstructive lung disease G. Fibrosis: Increased connective tissue scarring, excessive collagen, restrictive lung disease H. Degenerative changes: Necrosis, calcification, and autolysis I. "Pulmonary carcinogenesis: bronchogenic carcinoma, oat cell carcinoma, adenocarcinoma, me:sothelioma" III. Measwre.ment of lung injury If the lr.ing is injured by inhaled toxic gases and particles, how can the lung injury be detected and quantified? What repertoire of approaches can be used? Approaches and Parameters or Methods: A. Mechanical properties (pulmonary function) 1. Resistance 2. Compliance: pressure-volume curves 3. Lu:ng volumes: VC (spirometry), TLC, RV, and FRC (measured by helium dilution, Boyle's law) 4. FEV1 o and Full or Partial flow-volume curves B. Gas exchange, Adequacy of ventilation, Distribution of ventilation and perfusion 1. Alveolar gas tensions (PAC42, PA®2) 2. Arterial paC42, Pa02 3. Ventilation homogeneity: N2 washout 4. Ventilation (133~ce) or Perfusion (67Ga) scans

4 ].. DTPA-measured lung epithelial permeability 2. Clearance of radioactively-labelled inhaled particles 3. Clearance of magnetic inhaled particles 4. Macrophage motile activity measured by inhaled magnetic particles I. Microbicidal activity 1. Recognizable experimental pulmonary infections (morbidity and mortality studies) 2. Bacterial aerosol models, in vivo models 3. rn vitro killing 4. Phagocytosis: in vitro and in vivo J. Identifying pulmonary carcinogens 1. E xperimental pulmonary carcinogenesis (Saffiotti model) 2. t:hromosome abnormalities 3. ~~mes mutagenesis assay IV. Bioassays for measuring toxicity of particles and components of particles A. Whole animals B. In 1,vitro cell culture systems C. Ce1I homogenates V. Questiors s to be considered in the interpretation of data A. Species extrapolation. Are human and animal toxicities equivalent ? B. Dose extrapolation. Are the doses given to animals comparable to human exposures ? C. Time extrapolation. At what stage is the injury being measured, and how does it compare to the time course of disease development in humans ? D. Correlation of disease mechanism with bioassay result E. Spe<:ificity of bioassay result: Is result unique to the agent tested ? Is the result generaliz- able to a class of agents ? If the agent is a complex mixture, what are the active com- poner,:ts ? How does the bioassay result agree with disease outcomes in cases where human data are avaiIable ? N ~ N BIBLIOGRAPHY ~ f I. Mechanisms w1d Measurement of Lung Injury ~ 1. Allison, A.C:. Mechanisms of macrophage damage in relation to the pathogenesis of some lung diseases. In: ~ Respiratory Defense Mechanisms. (Lung Biology in Health and Disease., Monograph 5). Brain, J.D., Proctor, ~ D.F., Reid, L., Eds. Marcel Dekker. New York. 1977. 1075-1102. w~ 2. Brain, J.D. :Macrophage damage in relation to the pathogenesis of lung diseases. Environ. Health Perspectives. IPA 35:21-28, 1980.

6 28. Wolf, A.F. Occupational Diseases of the Lungs. Part III. Pulmonary disease due to inhalation of noxious gases, aerosols, or fumes. Ann. Allergy. 35:165-171,1975. III. PL~lmonary Bioassays 29. Hende:rson, R.F., E.G. Damon, and T.R. Henderson. Early damage indicators in the lungs: Lactate dehydrogenase activity in the airways. Toxicol. Appt. Pharmacol. 44:291-297, 1978. 30. Kavet, RI., and Brain, J.D. Methods to quantify endocytosis: a review. J. Reticuloendothel. Soc. 27:201-221, 1980. 31. Beck, E.D., Brain, J.D., and Bohannon, D.E. An in vivo hamster bioassay to assess the toxicity of particulates for the lungs: Toxicol. Appl. Phannacol. 66:9-29, 1982. 32. Smith, 'C T., Beck, B.D., Brain, J.D., Hinds, W.C., Baron, S.G., and Weil, L. Prediction of pneumoconiosis risk by bioassays of particulates from occupational exposures. In: Inhaled Particles, V, Walton, W.H. ed., Oxford: Per- gamon ]Press, pp. 163-176, 1982. Also in Ann. Occup. Hyg. 86:163-176,1982. 33. Beck, B„D., Gerson, B., Feldman, HA., and Brain, J.D. Lactic dehydrogenase isoenzymes in hamster lung lavage fluid after lung injury. Toxicol. Appl. Pharmacol. 71:59-71, 1983. 34. Henderso:a, R.F. The use of bronchoalveolar lavage to detect lung damage. Environ. Health Perspect. 56: 115-129, 1984. 35. Brain, J.D., and Beck, B.D. Broncboalveolar lavage. In: Toxicology of Inhaled Materfais, Handbook of Experimen- tal Pharr,7acology, Vol. 75, Witchi, H. and Brain, J.D., eds. Berlin: Springer Verlag, pp. 203-226, 1985. 36. Brain, J.D, and B.D. Beck. Bioassays for mineral dusts and other particulates. In: In Vitro Effects of Mineral Dusts., Beck, E.G. and Bignon, J., eds. NATO ASI Series Vol. G3. Berlin: Springer Verlag, pp. 3?3-335,1985. 37. Henderson, R.F., J.M. Benson, F.F. Hahn, C.H. Hobbs, R.K Jones, J.L. Mauderly, R.O. McClellan, and JA.Pickrell. New approaches for the evaluation of pulmonary toxicity: Bronchoalveolar lavage fluid analysis. Fund f!p,nl. Toxicol. 5: 451-458, 1985. 38. Beck, B.D., E.J. CIabrese, and P.D. Anderson. The use of toxicology in the regulatory process. Principles and Methodc of Toxicology, 2nd Edition. (A.W. Hayes, Editor), Raven Press Ltd., New York, pp. 1-28. 1989.

Rislk Assessments of EDB and Epidemiology muc:h lower levels in the environment provide crucial information to environmental health professionals and risk assessors. Even when such studies yield negative results in a hypothesis test, they can serve as a check on the plausibility of animal-based risk estimates. Clearly inappropriate models or assump- tions can be discarded, and greater confidence can ther be placed in the final risk assessment. Towards thi"is end, occupational studies require more attention to exposure estimation than has generally been the case in the past, and continuing foflow-up of exposed cohorts. Twelve years have passed since the closing date of follow-up in the study by Ott et a02) While we urge the collection and analysis of such data, we would also emphasize that even in the absence of human data, the continued use of animal data is appropriate. The field of carcinogenic risk assessment is in its infancy. The primitiveness of methodology echoes the lack of a clear theory of carcinogenesis. However, the gaps in knowledge and the uncertainties in meth- ods do not constitute sufficient justification for abandoning efforts to provide the public with plausi- ble upper bounds for cancer risks due to environ- mental chemical exposures. For a large number of such exposures, these estimates will necessarily be based on animal data. When quantified human ex- posure data are available and are related to cancer risk, these data can be useful either as a basis for exarapolation or as a standard for assessing the plausibility of risk estimates based on animal data alone. 6. SUMMARY Critics of cancer risk projections based on animal bioassays frequently make reference to negative epi- de niologic findings, and to reports such as that by Ram sey et aL(2) The analyses presented here demon- strate that low-dose extrapolations using linear non- thre;,hold additive models are not intrinsically dis- crepant with epidemiologic observations of cancer mortality. Additive risk models fitted to data from both gavage and inhalation bioassays predicted risks that were plausible when compared to published data from. an epidemiologic study of EDB-exposed work- ers. However, in rats, EDB is a more potent carcino- gen by gavage than by inhalation, with the higher potency manifested in shortened latency periods. Be- caus: of the shortened latency period, only models incorporating age at start of exposure were ap- propriate for the purpose of applying a risk assess- ment based on the gavage data to workers whose exposure began late in life. Application of a multi- plicative model gave implausibly high risk estimates when using lung cancers, though this may have been due to the choice of the wrong target site in humans. Thus, the previously reported overestimate of risk to workers occupationally exposed to EDB was due to a failure to consider their age at start of exposure when extrapolating from an animal bioassay with an ex- ceedingly short latency period. In the absence of viable alternatives, the results of this investigation support continued use of animal extrapolations to predict human cancer risks from environmental chemicals. Epidemiologic data with quantified exposure estimates can serve as an em- pirical standard for assessing the plausibility of ex- trapolation models. Linear nonthreshold additive models have been shown to provide plausible upper bounds when applied with due consideration to the quality of the data from the animal bioassays. ACKNOWLEDGMENTS We would like to thank Dow Chemical for pro- viding the mortality data for the employees exposed to ethylene dibromide, and Dr. Todd Thorslund for reviewing the history of the CAG's development of a risk assessment for EDB. REFERENCES 213 1. E. L. Anderson and the Carcinogen Assessment Group of the U.S. Environmental Protection Agency. "Quantitative Ap- proaches in Use to Assess Cancer Risk" Risk Analvsis 3, 277-295 (1983). 2. J. C. Ramsey, C. N. Park, M. G. Ott, and P. J. Gehring. "Carcinogenic Risk Assessment: Ethylene Dibromide," Toxi- cology and Applied Pharmacology 47, 411-414 (1978). 3. B. N. Ames, "Cancer and Diet-Reply," Science 21.4, 668-670, 757-760 (1984). 4. F. J. Stare, "Controversy About the Risks of EDB (Letter)," New England Journal of Medicine 310, 1387 (1984). 5. W. R. Havendar, Editorial, "EDB and the Marigold Option," Regulation, AEI Journal on Government and Society 16, ~ 13 -17 (1984). O 6. P. J. Gehring, "The Chemical Industry's Record in Environ- .~~ mental Health," Journal of Enuironmental Health 47, 58-61 (1984). ~ 7. D. A. Freedman and H. Zeisel, "From Mouse to Man: The ~ Quantitative Assessment of Cancer Risks," Technical Report ~ No. 79, Department of Statistics, University of California. Berkeley, California (1986). ~ 8. B. N. Ames, R. Magaw, and L. S. Gold, "Ranking Possiblc ~ Carcinogenic Hazards," Science 236, 271-280 (1987). ~..~'

1~1PtER1C~1 ~ ~CANTM3 Estimated number of new cancer cases in 1989 by states, total:1,010,000' (excluding Puerto Rico). 'Excluding non-melanoma skin cancer and carcinoma in situ. BASED ON RATES FROM NCl SEER PROGRAM (1983-1985}. OREGON 11,800 CALIF. 101,000 WAS H. 17,300 NEV. ",100 IDAHO 3,500 ARIZ. 13,000 WIS. 20,200 ILL. 48,000 LA. 17,500 " MICH. Z`j 37,400 IND. 23,200 16,800 TENN. 21,000 MISS. 12,000 ALA. 18,000 N.Y. 77,500 PA. 59,000 N.C. 24,500 GA. 22,500 S.C. 13,000 FLA: 65,500 i~..i, ~~r~_~~~~-~~ r ni~[~ ~~, ~~~~~' UTAH 3,500 ALASKA 1,000 N.H. 4,000 MAINE VT• 1 5,500 2,300 1 4,900 CONN. 14,400 N.J. 36,500 DEL • MD. 2,800 19,300 D.C. 3,200 PUERTO RICO 6,000 MASS. 28,400 202554S910

5 3. Brain, J.D. Toxicological aspects of alterations of pulmonary macrophage function. Ann. Rev. Pharmacol. Tazzcol. 26:547-565, 1986. 4. Doull, J., Klaassen, C.D., and Amdur, M.O. Taricology: The Basic Science of Poisons. New York: MacMillan, 1980. See particularly Chapter 12, "Toxic Responses of the Respiratory System," by D.B. Menzel and R.O. McClellan, and Chapter 24, "Air Pollutants," by M.O. Amdur. 5. Fishm:art, A.P. Pulmonary Diseases and Disorders., volumes 1 and 2. New York: McGraw Hill, 1980. 6. Gadek, T.E., Fells, GA., Crystal, R. Cigarette smoking induces functional antiprotease deficiency in the lower respiratory tract of humans. Science. 206: 1315-1316, 1979. 7. Harington, J.S., Allison, A.C. Tissue and cellular reactions to particles, fibers, and aerosols retained after inhala- tion. I:n: Handbook of Physiology, Section 9: Reactions of Environmental Agents., Falk, H.L., Murphy, S.D., Eds. Bethesda: American Physiological Society, 1977, pp. 263-283. 8. Harris, C.C. Pathogenesis and Therapy of Lung Cancer. New York: Marcel Dekker, 1978. 9. Janoff, .A., Carp, H., Lee, D.K., Drew, R.T. Cigarette smoke inhalation decreases alpha-l-antitrypsin activity in rat lung. Science. 206: 1314-1315, 1979. 10. Kirkpatrick, C.H., Reynolds, H.Y., eds. Immunologic and Infectious Reactions in the Lung. ( LungBiology in Health and Disease., Monograph 1). New York: Marcel Dekker, 1976. 11. Kuhn, C,1II, Senior, R.M. The role of elastase in the development of emphysema. Lung. 155:185-197, 1978. 12. Litwin, S.D., Ed. Genetic Determinants of Pulmonary Disease. New York: Marcel Dekker, 1978. 13. Snider, G.L., Lucey, E.C., and Stone, P.J. Animal models of Emphysema. Am. Rev. Respir. Dis. 133:149-169,1986. 14. Turino, G.M., Rodriquez, J.R., Greenbaum, L.M., Mandl, I. Mechanisms of pulmonary injury. Am. J. Med. 57:493-.505,1974. 15. Wahl, L.P1, et al. Collagenase production by lymphokine-activated macrophages. Science. 187: 261-263, 1975. 16. White, R., P,in, H.S., Kuhn, C. III. Elastase secretion by peritoneal exudative and alveolar macrophages. J. Exp. Med. 146: 802-808, 1977. 17. West, J.B, j?ulmonary Patltophysiology: 77ie Essentials. Baltimore: Williams & Wilkins, 2nd edition, 1982. IL Occupational Lung Diseases 18. Brooks, S.M., Lockey, J.E., Harber, P., eds. Clinics in Chest Medicine: Occupational Lung Diseases I. Philadelphia: W.B. Saunders Company, 1981, 19. Brooks, S.:M., Lockey, J.E., Harber, P., eds. Clinics in Chest Medicine: Occupational Lung Diseases II. Philadel- phia: W.:B. Saunders Company, 1981. 20. Dosman, JA., and Cotton, D.J., eds. Occupational Pulmonary Disease: Focus on Grain Dust and Health. New York: Academic Press, 1977. 21. Key, M.M., et al. eds. Occupational Diseases.•A Guide to TheirRecognition. Dept. of Health, Education, and Wel- fare Publication No. 77-181, Washington, D.C., U.S. Government Printing Office, 1978. esp. Chapter V, "Diseases of the Airways," by W. Keith, C. Morgan, and N. LeRoy Lapp. 22. Kusnetz, S., Hutchinson, M.K., eds. A Guide to the Work-Relatedness of Disease. Dept. of Health, Education, and Welfare Pulblication No. 79-116, Washington, D.C., U.S. Government Printing Office, 1979. 23. Morgan, W.K.C., Seaton, A. Occupational Lung Diseases. Philadelphia: W.B. Saunders Co., 2nd edition, 1984. 24. Parks, W.R. t)ccupational Lung Disorders (2nd ed.). London: Butterworths, 1982. ' 25. Clayton, C.D.,, and Clayton, F.E. Patty's Industrial Hygiene and Toazcology. Vols. 2A, 2B, 2C.• To.zzcology., 3rd Rev. Ed. New'York: John Wiley and Sons, 1982.9. Wagner, W.L., Rom, WA., Merchant, JA., eds. Health Issues Related to Metal and Nonmetallic Mining. Boston: Butterworth Publishers, 1983. 26. Wolf, A.F. Occupational Disease of the Lungs. Part I. Ann. Allergy 35:1-6, 1975. 27. Wolf, A.F. Occupational Diseases of the Lungs. Part II. Inhalation diseases due to inorganic dust. Ann. Allergy. 35:87-92, 1975.

C A N C E R F A C T S A N D F I G U R E S 7 9 8 9 BASIC DATA HOW CANCER WORKS Normally, the cells that make up the body reproduce themselves in an orderly manner so that worn-out tissues are replaced, injuries are repaired and growth of the body proceeds. Occasionally, certain cells undergo an abnormal change and begin a process of uncontrolled growth and spread: One cell divides into two, those redivide into four, and so on. These cells may grow into masses of tissue called tumors-some benign and others malig- nant (cancerous). The danger of cancer is that it invades and destroys normal tissue. In the beginning, cancer cells usually remain at their original site, and the cancer is said to be localized. Later, some cancer cells may invade neighboring organs or tissue. This occurs either by direct extension of growth or by becoming detached and carried through the lymph or blood systems to other parts of the body. This is called metastasis of a cancer. This spread may be regional-confined to one region of the body-when cells are trapped by lymph nodes. If left untreated, however, the cancer is likely to spread throughout the body. That condition is known as advanced cancer, and usually results in death. Because a case of cancer becomes progressively more serious with each stage, it is important to detect cancer as early as possible. Aids to early detection include cancer's Seven Warning Signals and the cancer risk factors. TRENDS IN DIAGNOSIS AND TREATMENT The diagnosis and treatment of cancer has become increasingly individualized. Early detection is followed by more precise staging, and the use of more than one kind of therapy, often in combination. Some cancers, which only a few decades ago had a very poor outlook, are often being cured today; acute lymphocytic letikemia in children, Hodgkin's disease, Burkitt's lymphoma, Ewing's sarcoma (a form of bone cancer), Wilms' tumor (a kidney cancer in children), rhabdomyosarcoma (a cancer in certain muscle tissue), choriocarcinonla (placental cancer), testicular cancer, ovarian cancer and osteogenic sarcoma. Other cancers have not yet yielded to effective treatment, and are the focus of continuing research. An outstanding example of progress is the improve- ment in the management of testicular cancer in young men. More prec,ise diagnostic tools and staging allow better selection of treatment. The use of combinations of cancer drugs has resulted in remarkably improved survival. In 20 years, the 5-year survival rate of testicular cancer rose from 63% to 91%. The following developments indicate the directions of current and luture research: • New ways have been found to use an old drug, 5- fluorouracil, more effectively against metastatic colon cancer. By combining it with leukovorin it is a much more potent inhibitor of colon cancer cells. * Analysis of onc:ogene products is a promising new means of predicting which tumors are likely to recur after surgery. o Use of potent growth factors stimulates normal bone marrow cells to withstand very high doses of che- motherapeutic: drugs. • P,A genetic fusing; of cancer cells with normal cells can produce disease-fighting "monoclonal antibodies"- specific antibod:ies tailored to seek out chosen targets on cancer cells. Their potential in the diagnosis and treatment of can,cer is under study. • New understanding of the causes of pain in cancer patients has increased the options for controL Regular use of oral pain medicines, infusions or injections of analgesics, procedures to interrupt pain pathways, are among the effective approaches available. • Studies with agents like synthetic retinoids (cousins of vitamin A), and other substances are being under- taken to see if recurrences of certain cancers can be prevented. Another step is to see if these agents can reduce cancer in high risk groups. • New approaches to drug therapy use combination chemotherapy and chemotherapy with surgery or radiation. New classes of agents are being tested for their effectiveness in treating patients resistant to drug therapies now in use. • Many patients with primary bone cancer now are treated successfully by removing and replacing a section of bone rather than by amputating the leg or arm. Drugs and radiation therapy are being used effectively after bone cancer surgery, resulting in dra- matic improvement in survival. • New high technology diagnostic imaging techniques have replaced exploratory surgery for some cancer patients. Magnetic Resonance Imaging (MRI) is one example of such technology under study. It uses a huge electromagnet to detect tumors by sensing the vibrations of the different atoms in the body. Com- puterized tomography (CT scanning) uses X rays to examine the brain and other parts of the body. Cross- section pictures are constructed which show a tumor's shape and location more accurately than is possible with conventional x-ray techniques. For patients undergoing radiation therapy, CT scanning may enable the therapist to pinpoint the tumor more precisely to provide more accurate radiation dosage while sparing normal tissue. • Immunotherapy holds the hope of enhancing the body's own disease-fighting systems to control cancer. Interferon, interleukin-2 and other biologic response modifiers are under study. Recently, interferon was made available as the treatment for hairy cell leukemia, a rare blood cancer of older Amer- icans. Interleukin-2 is under very active research in the treatment of kidney cancer and melanoma. 4

C A N C E R F A C T S A N D F 1 G U R E S 1 9 8 9 CANCER: BASIC DATA BASIC DATA What is cancer? Cancer i:5 a large group of diseases characterized by uncontrolled growth and spread of abnormal cells. If the spread is not controlled or checked, it results in death. However, many cancers can be cured if detected and treated promptly. How is cancer treated? By surgery, radiation, radioactive substances, chem- icals, hormones and immunotherapy. Who get<.; cancer? Cancer strikes at any age. It kills more children 3 to 14 than a ny other disease. And cancer strikes more frequently with advancing age. In the 1980's, there were estimated over 4.5 million cancer deaths, almost 9 million new cancer cases, and some 15 million people under medical care for cancer. How many people alive today will get cancer? About 76 million Americans now living will even- tually have cancer; about 30%, according to present rates. Over the years, cancer will strike in approximately three out of four families. How many ipeople alive today have ever had canceir? There are over 5 million Americans alive today who have a history of cancer, 3 million of them with diag- nosis five or more years ago. Most of these 3 million can be considered cured, while others still have evi- dence of cancer. By "cured" is meant that a patient has no evidence of disease and has the same life expectancy as a person who never had cancer. The decision as to when a patient may be considered cured is one that must be made by the physician after examining the individual patient. For most forms of cancer, five years without symptoms following treat- ment is the accepted time. However, some patients can be considerect cured after one year, others after three years, whereas some have to be followed much longer than five years. How many new cases will there be this year? In 1989 about 1,010,000 people will be diagnosed as having cancer.* How many people are surviving cancer? In the early 1900's few cancer patients had any hope of long-term survival. In the 1930's less than one in five was alive at least five years after treatment. In the 1940's it was one in four, and in the 1960's it was one in three. Today, about 405,000 Americans, or 4 out of 10 patients who get cancer this year, will be alive 5 years after diagnosis. The gain from 1 in 3 to 4 in 10 represents about 67,000 persons this year. This 4 in 10, or about 40% is called the "observed" survival rate. When normal life expectancy is taken into consideration (factors such as dying of heart disease, accidents and diseases of old age) 49%~o will be alive 5 years after diagnosis. This is the "relative" survival rate, and is considered a more accurate yardstick of our battle against cancer. Could more people be saved? Yes. About 178,000 people with cancer will probably die in 1989 who might have been saved by earlier diagnosis and prompt treatment. How many people will die? This year about 502,000 will die of the disease-1,375 people a day, about one every 63 seconds. Of every five deaths from all causes in the U.S., one is from cancer. In 1988 an estimated 494,000 Americans died of cancer. In 1987 it was 483,000; in 1986 the figure was 469,376. What is the national death rate? There has been a steady rise in the age-adjusted** national death rate. In 1930 the number of cancer deaths per 100,000 population was 143. In 1940 it was 152. By 1950 it had risen to.158 and in 1986 the number was 171. The major cause of these increases has been cancer of the lung. Except for that form of cancer, age-adjusted cancer death rates for major sites are leveling off, and in some cases declining. Can cancer be prevented? Some cancers, not all. Most lung cancers are caused by cigarette smoking, and most skin cancers by fre- quent overexposure to direct sunlight. These cancers can be prevented by avoiding their causes. Certain can- cers caused by occupational-environmental factors can be prevented by eliminating or reducing contact with carcinogenic agents. See Prevention section, pp.18-22 'These estimates of the incidence of cancer are based upon data from the National Cancer Institute's SEER Program (1983-1985). Non- melanoma skin cancer and carcinoma in situ have not been included in the statistics. The incidence of non-melanoma skin cancer is esti- mated to be over 5f)0,000 cases annually. "Age-adjusted-a m,ethod used to make valid statistical comparisons by assuming the same age distribution among different groups being compared. 3

C A N C E R F A C T S A N D F 1 G U R E S 1 9 8 9 BASIC DATA This research area will take many years to find the proper role of these agents in cancer treatment. • Many cancers are caused by a two-stage process through exposure to substances known as initiators and promoters. Research scientists are exploring ways of interrupting these processes to prevent the development of cancer. • New technologies have made it possible to use bone marrow transplantation as an important treatment option in selected patients with aplastic anemia and leukemia. Bone marrow transplantation for other cancers is under study. The administration of larger doses of anti-cancer drugs or radiation therapy may be tolerated by some patients if their bone marrow is storecl and later transplanted to restore marrow function (autologous bone marrow transplants). • Hyperthermia is a way to increase the heat or tem- perature of the entire body or a part of the body. It is known that heat can kill cancer cells. A cell tem- perafure of'45 degrees kills cancer cells. A temperature of 42-43 degrees makes the cell more susceptible to damage by ionizing radiation (X rays). Studies are underway to learn if hyperthermia can increase the effect of radiation or chemotherapy. • With medical progress producing longer survival periods for many cancer patients, clinical concerns are expanding to include not only patients' physical well-being but also their psychosocial needs. The patient's and family's reactions to the disease, sexual concerns, employment and insurance needs, and ways to provide psychosocial support, have emerged as important areas of research and clinical care. • Improvements in cancer treatment have made possible more conservative management of some early cancers. In early cancer of the larynx, many patients have been able to retain their larynx and their voice; in colorectal cancer, fewer permanent colostomies are needed; and the surgery required in many cases of breast cancer is often more limited. • Prostatic ultrasound, a rectal probe using ultrasonic waves producing an image of the prostate, is currently being investigated as a potential means to increase the early detection of occult, or not clinically sus- pected, prostate cancer. • Neoadjuvant chemotherapy has been successful against certain types of cancers. This involves giving chemotherapy to shrink the cancer and then removing it surgically. CANCER DEATH RATES* BY SITE, UNITED STATES, 1930-85 . ~ ; LUNG .• . : : .• : .•' : .~ . BREAST ATE i COLON r `x s UTERUS E~t VE~t U V . $ C ~ ~. 10 ~~ STOMA -,_ ~y.W P NCRla- - - ° I . -- LEUKEMIA BL0 ECTUM YEAR 'Rate for the population standardized for age on the 1970 U.S. population. Sources of Data: National Center for Health Statistics and Bureau of the Census, United States. Note: Rates are for both sexes combined except breast and uterus female population only and prostate male population only. 5

C A N C E R F A C T 5 A N D F I G U R E 5 1 9 8 9 SELECTED CANCER SITES LUNG CANCER Incidence: An estimated 155,000 new cases in 1989. The incidence rate in white males rose from 82.7 per 100,000 in 1982 to 84.2 in 1984. The incidence rate in white females and in black males and females also rose. Mortality: An estimated 142,000 deaths in 1989. The age-standardized lung cancer death rate for women is higher than that of any other cancer. It has surpassed breast cancer which for over 50 years was the number one cancer Ici:ller of women. Warning Signals: A persistent cough; sputum streaked with blood; chest pain; recurring attacks of pneumonia or bronchitis. Risk Factors: Cigarette smoking; history of smoking 20 or more years; exposure to certain industrial sub- stances such as asbestos, particularly for those who smoke. Involuntary smoking increases the risk. Expo- sure to radiation may also contribute to lung cancer. Early Detecl,ion: Lung cancer is very difficult to detect early; symptoms often don't appear until the disease has advanced considerably. If a smoker quits at the time of early precancerous cellular changes, the damaged bronchial lining often returns to normal. If a smoker continues the habit, cells may form abnormal growth patterns that lead to cancer. Diagnosis may be aided by such procedures as the chest X ray, sputum cytology test and fiberoptic bronchoscopy. Treatment: Treatment depends on the type of, and stage of lung cancer. Surgery, radiation therapy and chemotherapy are all options. For many localized cancers, surgery is usually the treatment of choice. Since the majority of patients with lung cancer have tumor spread, radiation therapy and chemotherapy are often combined with surgery. In small cell cancer of the lung, chemotherapy alone or combined with radiation has largely replaced surgery as the treatment of choice, with a large percentage of patients experiencing remission- in some cases, long-lasting remission. Survival: Only 13% of lung cancer patients (all stages, whites and blacks) live five or more years after diag- nosis. The rate is 33% for cases detected in a local- ized stage; but only 24% of lung cancers are discovered that early. Rates have improved only slightly over a recent 10-year period. COLON AND RECTUM CANCER Incidence: An estimated 151,000 new cases in 1989, including 107,000 of colon cancer and 44,000 of rectum cancer. Their combined incidence is second only to that of lung cancer i;excluding common skin cancers). Mortality: Aii estimated 61,300 deaths in 1989, second only to lung cancer. This includes 53,500 for colon cancer and 7,800 for rectum cancer. Warning Signals: Bleeding from the rectum, blood in the stool, change in bowel habits. Risk Factors: Personal or family history of colon and rectum cance, personal or family history of polyps in the colon or rectum; inflammatory bowel disease. Evidence suggests that bowel cancer may be linked to the diet. A diet high in fat and/or low in fiber content may be a signii.ficant causative factor. Early Detection: The ACS recommends three tests as valuable aids in detecting colon and rectum cancer early in people without symptoms. The digital rectal examination is performed by a physician during an office visit. The ACS recommends one every year after age 40. The stool blood slide test is a simple method of testing the feces for hidden blood. The specimen is obtained by the patient at home, and returned to the physician's office, a hospit:al or clinic for examination. The ACS recommends the test every year after 50. Proctosigmoidoscopy, known as the "procto," is an examination in which a physician inspects the rectum and lower colon with a hollow lighted tube. As the site of most colorectal cancers appears to be shifting higher in the colon, longer, flexible instruments are being used as well as the rigid scope. The ACS rec- ommends a procto every 3 to 5 years after the age of 50, following two annual normal exams. If any of these tests reveals possible problems, a physician may recommend more extensive studies, such as colonoscopy and a barium enema. Colono- scopes view the entire colon. Treatment: Surgery, at times combined with radia- tion, is the most effective method of treating colorec- tal cancer. Chemotherapy is being studied to determine its possible role in treating advanced cases. In cases of colon cancer, a permanent colostomy, the creation of an abdominal opening for the elimination of body wastes, is seldom needed, and is infrequently required for patients with rectal cancer. One report found permanent colostornies necessary for only 15% of patients whose rectal cancers are detected early. For those who do have permanent colostomies, the Society has a special patient assistance program. (See p. 25) Survival: When colorectal cancer is detected and treated in an early, localized stage, the 5-year survival rate' is 87% for colon cancer and 79% for rectal cancer, compared with 40% and 31% respectively, after the cancer has spread to other parts of the body. 9

C A N C E R F A C T S A N D F 1 G U R E S 1 9 B 9 SELECTED CANCER SITES BREAST CANCER Incidence: An estimated 142,900 new cases in the United States during 1989. About one out of 10 women will develop b reast cancer at some time during her life. Mortality: An estimated 43,300 deaths (43,000 females; 300 males) in 1989, in females, second only to lung cancer, now th,e foremost site of cancer deaths in women. Warning Sij;rdals: Breast changes that persist such as a lump, thickening, swelling, dimpling, skin irrita- tion, distortion, retraction or scaliness of the nipple, nipple discharge, pain or tenderness. Risk Factors: Over age 50; personal or family history of breast cancer; never had children; first child after age 30. Early Detection: The American Cancer Society rec- ommends the monthly practice of breast self-exami- nation (BSE) by women 20 years and older as a routine good health habit. Most breast lumps are not cancer, but only a physician can make a diagnosis. The A.merican Cancer Society and the National Cancer Institute, in their joint Breast Cancer Detection Demonstration program, found that mammography- a low-dose x-ray examination-could find cancers too small to be felt by the most experienced examiner. Besides its effectiveness in screening women without symptoms, mammography is recognized as a valuable diagnostic technique for women who do have findings suggestive of breast cancer. Once a breast lump is found, mammography can help determine if there are other lesions in the same or opposite breast which are too small to be felt. All suspicious lumps should be biopsied for a definitive diagnosis-even when the mammogram is described as normal. The Society recommends a mammogram every year for asymptomatic women age 50 and over, and a baseline mammogram for those 35 to 39. Asymptom- atic women 40 to 49 should have mammography every 1-2 years. In addition, a professional physical exam- ination of the breast is recommended every three years for women 20 to 40, and every year for those over 40. Treatment: Several methods may be used, depending on the individual woman's preferences and medical situation-surgery varying from local removal of the tumor to mastectomy, radiation therapy, chemotherapy or hormone manipulation. Often two or more methods may be used in combination. Patients should discuss with their physicians possible options available con- cerning the specific management of their breast cancer. New techniques in recent years have made breast reconstruction possible after mastectomy, and the cos- metic results are good. Reconstruction now has be- come an important part of treatment and rehabilitation. (See p. 25) Survival: The 5-year survival rate for localized breast cancer has risen from 78% in the 1940's to 90% today. If the breast cancer is not invasive (in situ), the survival rate approaches 100%. If the cancer has spread, how- ever, the rate is 60%. Despite an increasing incidence of breast cancer, longer survival has helped to stabilize mortality rates over the last 50 years. UTERINE CANCER Incidence: An E=.stimated 47,000 new invasive cases in 1989, including 13,000 cases of cancer of the cervix, and 34,000 cases of cancer of the endometrium or body of the uterus. Invasive cervical cancer incidence has steadily decreased over the years, while cancer in situ has risen in all groups. Cervical cancer is most common today among low s,ocioeconomic groups but all groups are at risk. Endom etrial cancer afflicts mostly mature women, and diagnosis usually is made between the ages of 55 and 69. Mortality: An e;>timated 6,000 deaths in 1989 from cervical cancer, 4,000 from endometrial cancer. Overall, the death rate from uterine cancer has decreased more than 70% during the last 40 years, due mainly to the Pap test and regular checkups. Warning Signal's; Intermenstrual or postmenopausal bleeding or unusual discharge. RLk Factors: Fo r cervical cancer: early age at first intercourse, multiple sex partners. For• endometrial cancer: history of infertility, failure of ovulation, pro- longed estrogen therapy and obesity. Early Detection: The Pap test, an examination under a rnicroscope of cells from the cervix and body of the 10 uterus, is a simple procedure which can be performed at appropriate intervals by physicians as part of every pelvic examination. For cervical cancer, women who are or have been sexually active, or have reached age 18 years, should have an annual Pap test and pelvic examination. After a woman has had three or more consecutive satisfactory normal annual examinations, the Pap test may be performed less frequently at the discretion of her physician. The Pap test is highly effective in detecting early cancer of the uterine cervix; it is only 50% effective in detecting endometrial cancer. Women at high risk of developing endometrial cancer should have an endo- metrial tissue sample at menopause. The hormone estrogen frequently is given to women during and after menopause to make up for the decline in estrogens normally produced by the ovaries. Estrogen helps to control menopausal symptoms such as hot flashes or thinning of the vaginal lining causing painful sexual intercourse. For mature women, there are certain risks associated with such treatment, including an increased risk of endometrial cancer. Women and their physicians should carefully discuss

CONTENTS CANCER: BASIC DATA .................................... 3 Basic Data ................................................. 3 How Cancer Works ...................................... 4 Trends in DiagTosis and Treatment .................... 4 Cancer Death Rates by Site, U.S., 1930-1985" ........... 5 New Cancer Cases-1989* .............................. 6 Cancer Deaths--1989* .................................... 7 Estimated New Cancer Cases and Deaths by Sex for A11 Sites-1989* ........................... 8 SELECTED CANCER SITES .............................. 9 Lung Cancer .............................................. 9 Colon and Rectu.m Cancer .............................. 9 Breast Cancer ............................................. 10 Uterine Cancer ........................................... 10 Ovarian Cancer ........................................... 11 Oral Cancer ............................................... 11 Cancer Incidence and Deaths by Site and Sex-1989 Estimates* ............................ 12 Prostate Cancer ...............:.......................... 12 Bladder Cancer ........................................... 13 Skin Cancer ............................................... 13 Pancreatic Cancer ........................................ 14 Leukemia ................................................. 14 Five-Year Cancer Survival Rates for Selected Sites" ......................................... 15 How to Estimate Cancer Statistics Locally* ............ 15 CANCER BY AGFs AND RACE ............................ 16 Black Americans .......................................... 16 The Economically Disadvantaged ....................... 16 Hispanic-Americans ..................................... 16 Children ................................................... 16 Trends in Survival by Site of Cancer, by Race* ........ 17 PREVENTION ............................................... 18 Primary Prevention ...................................... 18 Secondary Prevention ................................... 18 Cancer-Related Checkup Guidelines ................... 19 Colorectal Cancer: Early Detection Is the Key ......... 19 Breast Cancer: A Program of Action .................... 20 Tobacco Use .............................................. 20 Nutrition and Cancer: A Common Sense Approach ........................ 21 THE AMERICAN CANCER SOCIETY .................. 22 Profile ..................................................... 22 Public Education ......................................... 22 Professional Education .................................. 23 Service and Rehabilitation ............................... 24 Costs of Cancer ........................................... 25 Allocation of ACS Funds, 1987-1988* ................... 25 RESEARCH .................................................. 26 The ACS and Research ................................... 26 Cancer and the Environment ........................... 27 Cancer's Seven Warning Signals ...................... .. 28 30-Year Trends in Age-Adjusted Cancer Death Rates* .......................................... 29 Summary of Research Grants and Fellowships* .......................................... 30 Comprehensive Cancer Centers ........................ 31 Chartered Divisions of the ACS .............. Back Cover 'Table/Chart SOURCES OF STATISTICS @ ncicCence Since there is no national office which records every new cancer case, there is no way of knowing exactly how many new cases of cancer are diagnosed each year. In the past, estimates of cancer incidence were made by extrapolating from the experience of the few population-based cancer registries. Estimates of incidence in Facts & Figures editions prior to 1974 were based on data from two state cancer registries. The issues from 1974 through 1978 used information from the National Cancer Institute's Third National Cancer Survey (1969-1971) of nine major areas of the United States. Then in 1973, NCI began a new and larger program, gathering data from 11 population-based registries. It is called SEER, standing for Surveillance, Epidemiology and End Results. Beginning with the 1979 edition of Facts & Figures, SEER incidence information has been used. Each time a new data base is introduced, there may be some sharp changes in figures, due to the more accurate data. The changes do NOT indicate either a cancer epidemic or miracle cure. For valid comparisons between years, incidence statistics from the 1974 through 1978 editions of Facts & Figures may be compared with one another, while those from the 1979 to 1984 editions may be compared. The latest available information for this 1989 edition is SEER data from the years 1983-1985. Mortality The source for mortality statistics has remained constant over the years: the National Center for Health Statistics, Department of Health and Human Services. The 1989 figures are estimates based on the latest available information, which includes mortality data through 1985. Beginning with the 1981 edition of Facts & Figures, mortality rates per 100,000 population were age-adjusted to the 1970 census population, rather than the 1940 census population. Comparing these charts and figures with those of previous years may indicate false trends. Survival Because of the 5-year waiting period, survival statistics take longer to compile. In this edition, we show the latest survival rates for cases diagnosed in the period 1979-84 in the SEER program. C1989, American Cancer Society, Inc. All rights reserved, including the right to reproduce this publication or portions thereof in any form. For written permission, address American Cancer Society, 1599 Clifton Road, N.E., Atlanta, GA 30329. 2

human, and die rat has certainly prospered in an otherwise hostile environment (6). Current thinking assigns a central role to class I antigens in the presentation of foreign antigens to the host immune system and to class II antigens in the recognition offoreign antigens. If these are, indeed, the primary functions of the MHC antigens, then either thi- specificities of their antigen-recognizing structures are much broader than those of the antibody combining sites or the extent of their antigen-recognition repertoire is not reflected in their serological polymorphism. There is also the relevant, and intriguing, observation that the MHC polymorphism in the protochordate Botryllus is the same as that in the mouse and the human (29). Why? Only more extensive structural studies of MHC antigens at both the protein and DNA levels will provide the crucial insights into the biological significance of MHC antigen polymorphism. Transplantation The rat is the animal most often used in organ transplantation studies: its size makes surgical procedures feasible, provides large amounts of ~As and serum, and allows serial biopsies of the transplanted organ to assess the rejection process. The advances in rat immunogenedcs over the past two decades have enhanced considerably its usefulness in transplantation research. The rejection times of variou:; organs in different strain combinations have been documented (5), and the roles of the different MHC and non-MHC antigens in this process (30) have been examined by the use of different combination of inbred, congenic, and recombinant strains. Such transplantation studies have been done with skin (7, 30), kidney (31), heart (32), bone marrow (33), liver (34), small bowel (35, 36), pancreas (37), and brain (38, 39). There are four major areas of current interest in experimental transplantation research, and the rat is thc: crucial animal in each of them: allotransplantation of the small bowel, heart, and liver; neural transplantation; xeno- grafting; and reproduction. Allografiing. Inn systemic allotransplantation, grafting of the small bowel is the most pressing area of study (35, 36). Loss of function in this organ occurs in a variety of situations and at all stages of life: for example, congenital abnormalities, necrotizing enterocolitis, mesen- teric artery thrombosis, and trauma. The problems encountered include the prcap+:r preservation and restoration of the physiological function of this delicate organ. The immunological problems are those of the host-versus-graft reaction by the recipient's immune system and the graft-versus-host reaction by the lymphoid tissue in the Peyer's patches of the graft. In this sense, small bowel grafting presents the sar.ne type of tissue matching problems as bone marrow grafting, but tlie offending T cells cannot be removed from the bowel graft as easily as they can from the bone marrow graft. Two other important areas of research in allografting are heart grafting and liver grafting. The most critical issue in the long-term survival of cardiac transplant patients is the development of athero- sclerosis in the coronary arteries of the transplant (40). In humans, this process can lead to the loss of the transplant in 5 to 7 years, so an understanding of its pathogenesis will provide a cogent insight into its therapy. In hr:Irnan liver transplantation, the role of histocompati- bility (HLA) matching in the survival of the transplant has not been clarified, and there is the suggestion that under certain circum- stances matching can reduce the survival of the graft (41). The liver transplantation model has been well developed in the rat (34), and it should provide th e appropriate system in which to explore these questions. Neural transplantation. The rat has been an important animal in the study of allogeneir and xenogeneic neural transplantation. Embry- onic neural tissue can be transplanted i_nto neonatal and adult brains where it can mature and integrate into the host brain. Both allografts and xenografts can survive for prolonged periods, but they are always susceptible to immune rejection either spontaneously or after challenge by related antigens or by mechanical trauma to the central nervous system (38). In the rejection process, however it is precipitated, the host astrocytes are induced to express MHC class I and class II antigens, and the control of such expression may be central to the acceptance of the neural transplant. Cyclosporine A can effectively prolong neural grafts (42). Recent studies in humans (43) suggest that grafts of neuroectodermal origin can be performed, but such grafts have not yet proven to be clinically useful for any significant period of time. The critical factors that affect the success of a neural transplant are the technique and site of the transplant, the amount of disruption of the blood-brain barrier, the size and source of the donor tissue, the vascularization of the transplant, the age of the host and of the donor at the time of transplantation, and the immunogenetic difference between host and donor. Studies in rats have shown that such transplants can reduce cognitive defects due to frontal cortex lesions (44), improve impair- ment of motor function in aged animals (45), and make functional connections in an allogeneic or xenogeneic setting (46). These studies are also providing insight into the immunological status of the brain and the immune reactivity in this organ and into the pathogenesis of focal neurodegenerative diseases (38). The potential value of neural grafts in clinical medicine lies in replacement of damaged neural circuits and in the replacement of cells making chemicals that modulate neural function. Neural circuit replacement might be used to treat trauma in adults and congenital neurological defects in children, and it is in the latter that long-term possibilities for the therapeutic use of neural grafting lie. The use of transplanted cells as a substitute for chemical replacement therapy is complicated by the fact that many of the diseases causing such deficits may have an autoimmune basis, so the transplanted cells themselves may fall victim to the underlying disease process. Much basic work must be done to clarify the immunological and neuro- physiological aspects of neural transplantation, the development of specific immunosuppressive regimens for neural transplants, and the pathogenesis of the neurodegenerative diseases for which it might be used as therapy. The effort is worthwhile, since transplantation of tissue into the brain is one of the most promising approaches to have come from experimental neurobiology as potential therapy for a variety of disorders involving damage to the central nervous system. Finally, the use of neural xenotransplants in humans is a distinct possibility (38), and the ethical dilemmas raised by this procedure must be examined. Xenografis. The use of grafts from animals of different families and genuses, xenografting, has been explored sporadically (47) and has recently had a resurgence because of the interesting'basic immuno- logical questions that it raises and because of the possibility of the use of such grafts as neural transplants (38) and as temporary expedients ("bridging grafts") in humans. Each xenograft system has its own peculiarities (47): thus, it is not possible, at the present time, to generalize about the nature of the immune response to xenografts. In order to explore systematically the immunobiology and immunogenetics of xenografting, three areas of resarch should be developed. First, xenoantigens should be identified and characterized. The relative immunogenicity of various xcnografts should be studied in one donor-recipient combination in order to develop a coherent body of knowledge that can serve as a paradigm for other systems. The rat-mouse combination will be the most useful one to study initially, because both species are immuno- logically and genetically well defined. This research should explore (i) the possible existence of unique xenoantigenic systems, (ii) the role of donor MHC antigens in eliciting an immune response to the 2I JULY 1989 ARTICLES 271

C A N C E R F A C T S A N D F 1 G U R E S SELECTED CANCER Sal°ES 1 9 8 9 CANCER INCIDENCE AND DEATH S B Y SITE AND SEX-1989 ESTIMATES CANCIER INCIDENCE BY SITE AND SEXt CANCER DEATHS BY SITE AND SEX SKIN I 3% 3% SKIN SKIN 2yo 1% SKIN ORAL ~ 4% 2% ORAL ORAL ORAL LUNG [20% 28% BREAST LUNG 357% 18% BREAST COLON &~14oJo 11% LUNG RECTUM l. COLON & 11% 21% LUNG RECTUM 15% COLON & PANCREAS [3% RECTUM 13% COLON & PANCREAS 5% RECTUM PROSTATE [t1°Jo 3% PANCREAS PROSTATE j1% rJ°Jo PANCREAS 4% OVARY URINARY C'0°Jo 5% OVARY URINARY nj /0 0 LEUKEMIA & 7 l;°JQ g°Jo UTERUS LYMPHOMAS I_ LEUKEMIA & 9pofO 4°lo UTERUS LYMPHOMAS 4% URINARY ALL OTHER 3°lo URINARY ALL OTHER 20% 7oJO LEUKEMIA & 9oJ0 LEUKEMIA & LYMPHOMAS LYMPHOMAS l4°Jo ALL OTHER 20°lo ALL OTHER fiExcluding non-melanoma skin cancer and carcinoma in situ. ® PROSTATE CANCER Incidence: An estimated 103,000 new cases in the United States durilg 1989. About one out of 11 men will develop prost2 te cancer at some time during his lifetime. The third highest incidence of cancer in men, next to skin cancer and lung cancer. Mortality: An estimated 28,500 deaths in 1989, the third leading cause of cancer deaths in men. Warning Signals: Most signs or symptoms of pros- tate cancer are nonspecific, and do not distinguish from benign conditions such as infection or prostate enlarge- ment. These include weak or interrupted flow of urine; inability to urinate or difficulty in starting urination; need to urinate frequently, especially at night; blood in the urine; urine flow that is not easily stopped; painful or burning urination; continuing pain in lower back, pelvis or upper thighs. Risk Factors: Incidence increases with age through the most advanced ages; about 80% of all prostate cancers are diagnosed in men over the age of 65. The disease is more common in northwest Europe and North America; rare in the Near East, Africa, Central and South America. Black Americans have the highest rate of incidence in the world for reasons not currently known. There is sor.ae familial association, but it is unclear whether thi:> is due to genetic or environmental association. Dietary fat may be a factor, based on studies conducted internatio,nally. Workers who work with cadmium are found ta be at slightly higher risk. Studies of migrating populations have suggested that environ- mental factors, such as diet and lifestyle, may play an important role in the risk of developing cancer of the prostate. Early Detection: Every man over 40 should have a rectal exam as part of his regular annual physical checkup. A new technique, prostate ultrasound is being investigated for the early detection of small non- palpable cancers. This new approach may be of special benefit:to high risk men.,Men over 40 should be alert to changes such as urinary difficulties, continuing pain in lower back, pelvis or upper thighs, and should see their physician immediately should any occur. The key to saving lives from prostate cancer is early detection and treatment. Treatment: Surgery, alone or in combination with radiation and/or hormones, and anticancer drugs are all options available in the treatment of prostate cancer. Surgery or radiation therapy may be the treatment chosen to cure prostate cancer if it is found in an early localized state. Hormone treatment and anticancer drugs also may control prostate cancer for long periods by shrinking the size of the tumor and greatly relieving pain. Survival: Sixty-four percent of all prostate cancers are discovered while still localized within the general region of the prostate; 84% of all patients whose tumors are diagnosed at this stage are alive 5 years after treatment. Survival rates for all stages combined have steadily improved since 1940, and in the last 20 years have increased from 48% to 71%. 12

C A N C E R F A C T 5 A N D F I G U R E S 1 9 8 9 SELECTED CANCER StT'ES PANCREATIC CANCER Incidence: An estimated 27,000 new cases in the United States in 1989.1'ancreatic cancer is the 5th leading cancer killer. The incidence rate of pancreatic cancer among U.S. blacks is about 1.5 times higher than for whites. li4ortality: An estimated 25,000 deaths in 1989 due to pancreatic cancer. From 1954 to 1984, the death rates for pancreatic caj.lcer in the United States rose 12% to 10.2 deaths per 100,000 men. During the same period, the death rates for women rose 26% to 7.2 deaths per 100,000 women. Warning SignadLs: Cancer of the pancreas is a "silent" disease, one that occurs without symptoms until it is advanced. Itisk Factors: F:isk increases with age after age 30, with the highest frequency of incidence occurring between ages 65 a nd 79. Smoking is a major risk factor, incidence is more than twice as high for smokers versus nonsmokers. The disease is 30% more common in men, and occurs about 50%o more frequently in black, versus white Americans. Some studies, as yet unconfirmed, suggest an association with chronic pancreatitis, dia- betes and cirrhosis. High-fat diets may be a risk factor, countries with higher fat consumption levels have higher pancreatic cancer rates. Coffee has been inves- tigated as a possible risk factor, but no conclusive evi- dence is currently available. Early Detection: Research has focused on ways to diagnose pancreatic cancer before it is advanced enough to cause symptoms. Ultrasound and CT scans are being tried, but to date only a biopsy yields a certain diagnosis. Prevention: Very little is known about what causes the disease, or how to prevent it. Treatment: Surgery, radiation therapy and anti-cancer drugs are used to treat pancreatic cancer, but so far have had little influence on outcome. In 59% of cases, diagnosis is so late that none of these is used. Survival: Only 4% of patients live more than 3 years after diagnosis. The 2% of patients whose cancers occur in the insulin-producing cells, and not the duct cells of the pancreas tend to live longer; about 30% of these patients live more than 3 years after diagnosis. LEUKEMIA Incidence: An estimated 27,300 new cases in 1989, about half of them acute leukemia, and half of them chronic leukemia, Although it is often thought of as primarily a childhood disease, leukemia strikes many more adults (25,000 cases per year compared with 2,300 in children). Acute lymphocytic leukemia accounts for about 1,800 of the cases of leukemia among children, whereas in adults the most common types are acute granulocytic (about 8,000 cases annually), and chronic lymphocytic (9,6(N) cases annually). 1Vlortality: An es,timated 18,100 deaths in 1989. Warning Signals: Symptoms of acute leukemia in children can appear suddenly. Early signs may include fatigue, paleness, weight loss, repeated infections, easy bruising, nose bleeds or other hemorrhages. Chronic leukemia can progress slowly and with few symptoms. Risk Factors: Leukemia, a cancer of the bloodforming tissues, strikes both sexes and all ages. Causes of most cases are unknown. Individuals with Down's syndrome (mongolism) and certain other hereditary abnormalities have higher than normal incidence of leukemia. It has also been linked to excessive exposure to radiation and certain chemicals such as benzene. Early Detection: Leukemia may be difficult to diag- nose early because symptoms often appear to be those of other less serious conditions. When a physician does suspect leukemia, a diagnosis can be made through blood tests and am examination of bone marrow. Treatment: Chemotherapy is the most effective meth- od of treating leuk _mia. Today, continuing research in leading U.S. medird centers is yielding new and better drugs for treating leukemia patients. A variety of anti- cancer drugs are used, either in combinations or as single agents. To prevent persistence of hidden cells, therapy of the central nervous system has become standard treatment, especially in acute lymphocytic leukemia. Under appropriate conditions, bone marrow transplantation may be useful in the treatment of certain leukemias. When leukemia occurs, millions of abnormal, imma- ture white blood cells are released into the circulatory systems. These abnormal cells crowd out normal white cells to fight infection, platelets to control hemorrhaging and red blood cells to prevent anemia. Transfusions of blood components and antibiotics are used as supportive treatments. Survival: The overall, average 5-year survival rate for white patients with leukemia is 33%, due partly to very poor survival of patients with some types of leukemia such as acute granulocytic. The 5-year survival rate for black patients is 28%. Over the last 30 years, however, there has been a dramatic improvement in survival of patients with acute lymphocytic leukemia: From a 5- year survival of 4% for white males diagnosed in the early 1960's to 27% in the early 1970's to 46% around 1980; for white females diagnosed in the same time periods, from 3% to 29%a to 52%. In white children, the improvement has been from 4% to 68%. Moreover, in some medical centers, optimum treatment has raised survival of children with acute lymphocytic leukemia up to 75%. 14

l. A N l. G K r A L i J A h U r i l, U K [ J j y b y CANCER BY AG E AN D RACE* BLACK AMERICANS A study of cancer rates over several decades shows that the cancer incidence rate for blacks is higher than for whites, and that the death rate is also higher. Over a 30-year period, black male cancer death rates rose by 77% compared to a 10% increase in black females. Incidence rates inn blacks also have increased in both males and females. The overall cancer incidence rate for blacks went up 27%, while for whites it increased 12%. Cancer mortality has increased in. both races, but the rate for blacks is greater than for whites. The rates were virtually the same 30 years ago. Since then, cancer death rates in whites have increased 10%, while black rates have increased almost 50%. Cancer sites where blacks had significantly higher increases in incidence and mortality rates included the lung, colon-rectum, prostate and esophagus. Esopha- geal cancer, long considered mainly a disease of males, remained about the same in whites and rose rapidly in blacks of both sexes. The incidence of invasive cancer of the uterine cervix dropped in both black and white women, although the incidence in blacks is still double that in whites. However, the rate for endometrial cancer-or cancer of the body of the uterus-for white women is almost double that of black women. Survival rates for patients diagnosed between 1974 and 1982 were compared for whites and blacks. More whites than blacks had cancer diagnosed in an early, localized stage when the chances of cure are best: 39% for whites versus 33% for blacks. In a survey done for the ACS by the Gallup Orga- nization in December 1987, the public's awareness and use of cancer tests was determined. The survey showed that 93% of white women knew of the Pap test and that 88% had had the test at some time, while 92% of black women knew of it and 79% had had it. For proctoscopic exams, 60% of the white population were aware of the procedure and 29% had had it at some time. For blacks, only 49% were aware of it and 22% had had it. THE ECONOMICALLY DISADVANTAGED A 1986 ACS Special Subcommittee report, "Cancer in the Economically Disadvantaged" found that cancer survival, and in some cases incidence, are related to socioeconomic factors such as the availability of health services. The report also found that ethnic differences in cancer are secondary to socioeconomic factors, and that there are higher rates of cancer mortality for patients of low socioeconomic status compared to those in higher brackets. Estimates suggest that at least half of the differences in survival rates are due to late diagnosis among economically disadvantaged patients, pointing up the need for more effective early detection programs and better access to treatment among this segment of the American population. HISPANIC-AMERICANS A recent ACS-sponsored report described Hispanic attitudes toward cancer, cancer risk reduction and early detection. The study, conducted for the Society by the firm of Clark, Mart;ire and Bartolomeo, Inc., underscored an urgent need for extensive cancer education and information programs directed to Hispanic-Americans. Survey findings shcwed that Hispanic-Americans are not adequately aware of most of the warning signals of cancer or of ways to reduce cancer risk, and that they tend not to seek early detection or treatment. The study identified the key psychological, cultural and economic barriers hindering the,fight against cancer in the Hispanic-American community. CHILDREN Incidence: An estimated 6,600 new cases in 1989, making it rare as a childhood disease. Common sites include the blood andd bone marrow, bone, lymph nodes, brain, nervous system, kidneys and soft tissues. Mortality: An estimated 1,800 deaths in 1989, about half of them from leukemia. Despite its rarity, cancer is the chief cause of death by disease in children between the ages of 3 and 14. Mortality has declined from 8.3 per 100,000 i~n 1950 to 3.5 in 1986. Early Detection: Cancers in children often are dif- ficult to recognize. Parents should see that their children have regular medical checkups, and be alert to any unusual symptoms that persist. They include: unusual mass or swelling; unexplained paleness and loss of energy; sudden tendency to bruise; persistent, localized pain or limping; prolonged, unezplained fever or illness; frequent headaches, often with vomiting; sudden eye or vision changes; and excessive, rapid weight loss. Some of the main childhood cancers are: Leukemia: See preceding section. Osteogenic Sarcoma and Ewing's Sarcoma are bone can- cers. There may be no pain at first, but swelling in the area of the tumor is often a first sign. 16

Table 9. Amino acid homologies between MHC class I and class II antigens of the rat and those of the mouse and the human (3, 14, 101). Percentage homologies Approximate number of Type Comparison Rat compared to Allelic and interlocus homologies* serologically defined alleles* Mouse Human Rat Mouse Human Rat Mouse Human Class I Signal p.pdde 85 50 68-73 (A) 32-69 (K) 85-95 (A) 12 (A) 92 (K) 24 (A) at domain 71-73 68 34-57 (D) 93 (B) 2 (E) 63 (D) 52 (B) a2 domain 71-78 67 97-98 (A:E) 36-69 (K:D) 79-85 (A:B) 4(C) 2 (L) 11 (C) a3 domain 87 72 Transmcmbrane- cytoplasmic domain 38-46 40 Class II • 80-91 73-81 56-59 (B:D) 52-60 (A:E) 64-66 (DR:DQ) 10 (B,D) 74 (A) 20 (DR) 72 (E) 9 (DQ) 6 (DP) ® *Lccvs or loci compared given in parentheses. ascending thoracic aorta, and the additive influence of multiple risk factors (83). In subsequent studies this model was used to define the influences of various kinds of food fats (84) and of metabolic manipulations (85) and to delineate the ultrastructural features of these lesions (86). :[n the latter studies, the lesions resemble the foam cell lesions of the ra,bbit and of other animals in which the blood cholesterol had very high values and in which there was some degree of endothelial injury (87). The availability of a wide variety of genetically defined strains of rats will now allow studies such as these to be designed to explore the genetic basis of the various risk factors involved in atheroge.nesis. Two inbred strains of rats are particularly useful for studying the pathogenesis of cardiovascular diseases: the SHR (spontaneously hypertensive) strain (88) and the BB strain, which spontaneously develops insulin-dependent diabetes mellitus (89). The SHR rats develop hypertension that increases with age; is more severe in males; leads to cerebral, myocardial, vascular, and renal lesions; and is responsive to antihypertensive agents. The hypertension is a genetically transtnitted trait that is most likely polygenic, and in well-maintained colonies all of the animals develop hypertension between 5 and 10 weeks of age. The inbred, genetically related WKY strain is often used as the normotensive control for the SHR strain. Stroke-prone (90) and obese (91) substrains of the SHR strain have been developed, but they are difficult to select and maintain because tllese phenotypic traits most likely have a polygen- ic basis. The onset of diabetes in the BB rats is rapid, occurs around 90 days of age, a9ects both males and females, and is under polygenic control, one component of which is linked to the MHC. The clinical syndrome consists of hyperglycemia, hypoinsulinemia, ketosis, polyuria, glycosuria, and weight loss. Pathologic examina- tion shows selective inflammatory destruction of the beta cells of the islets of Langerhans in the pancreas, and the inflammatory process has a substantial immunological component. various pharmacological agents, including alcohol (99) and narcotics (100), on behavior have been explored. These studies have provided insights into behavior and into its anatomic and physiologic basis and have led to the development of the field of experimental psychology. However, the lines of rats used were not developed according to the standard rules of genetic inbreeding, and they generally led, at best, to populations with a restricted genetic composition, relative to a randomly breeding population of rats, in which a certain phenotypic characteristic was prominent. This situation has complicated the more detailed genetic interpretation of much of the experimental literature on behavior, and it is particularly acute when examining the relative roles of heredity and environment in learning. One possible approach to developing appropriate strains of rats for behavioral studies may be to select partially inbred rats for their behavioral characteristics and then to breed them for these traits in the context of a mating scheme that would also continue the inbreeding. Concluding Remarks The rat is a major experimental animal in all fields of biomedical research and technology, and studies with it have provided much basic and applied knowledge. Its greatest utility has been in those fields broadly classified as experimental pathology and experimental surgery. The extensive work done on the immunology and genetics of the rat in recent decades has greatly enhanced its utility and has contributed substantially to the body of knowledge in immunoge- netics. As the constraints on the use of larger animals grow, the rat should provide an excellent alternative to their use. Such a change would also have the advantage of allowing more sophisticated studies to be designed, since so much is known about the biology of the rat, and this would greatly enhance the value of the experiments done. Behavior The rat has been used for studies in behavior since the turn of the century, and a substantial literature has emerged from these studies (92, 93). The investigation of the hereditary and environmental aspects of learning be gan with the introduction of maze experiments by Small (94) and le:d to the development of "maze-bright" and "ma2,e-du1P' lines of r.ats by selective breeding (95). Various emo- tional characteristics have been developed in rats by selective breeding (93, 96), arid the role of different areas of the brain in behavior has been investigated by stimulation and by extirpation experiments (44, 45,97). Finally, the effects of aging (93, 98) and of REFERENCES AND NOTES 1. H. H. Donaldson, J. Acad. Nat. Sci. Philadelphia 15, 365 (1912); W. E. Castle, Proc. Nail. Acad, Sci. U.S.A. 33, 109 (1947). 2. R. D. Owen, Anti. N.Y. Acad. Sci. 97, 37 (1962); J. Paltn, ibid., p. 57; O. Stark, V. Kren, B. Frenzl, Folia Biol. (Prague) 13, 85 (1967); B. Heslop, Aust. J. Exp. Biol. Med. Sci. 46,479 (1968); H. W. Kunz and T. J. Gill III, J. Immunogenet. 1, 413 (1974); T. Natori et al., Transplant. Proc. 11, 1568 (1979). 3. T. J. Gill III, H. W. Kunz, D. N. Misra, A. L. Cortese Hassett, Transplantation 43, 773 (1987). 4. T. J. Gill III, Physiologist 28, 9 (1985). 5. J. B. Calhoun, The Ecology and Sociology of the Norway Rat (Department of Health, Education and Wdfare, Bethesda, MD, 1962); R. Robinson, Genetics of the Norway Rat (Pergamon, New York, 1965); Inbred and Genetically Defined Strains of Laboratory Animals, part 1, Mouse and Rat, P. L. Altman and D. D. Katz, Fds. (Federation of American Societies for Experimental Biology, Bethesda, MD, 274 SCIENCE, VOL. 245

C A N C E R F A C T 5 A N D F I G U R E s 1 9 8 9 SELECTED CANCER SITES the use of postmenopausal estrogens in terms of the benefit an.d risk to the individual patient. Treatmiealt: Uterine cancers generally are treated by surgery or radiation, or by a combination of the two. In precancerous (in situ) stages, changes in the cervix may be treated by cryotherapy (the destruction of cells by extrem e cold), by electrocoagulation (the destruction of tissue through intense heat by electric current) or by local surgery. Precancerous endometrial changes may be treated with the hormone progesterone. Survival: The 5-year survival rate for all cervical cancer patients is 66%. For patients diagnosed early, however, the rate is 80-90%. Cancer in situ is virtually 100%. The figures for endometrial cancer are 83% all stages, 91% early and virtually 100% for endometrial precancerous lesions. During a recent 10-year period, there was moderate improvement for both uterine sites. ®VARIAN CANCER Incider ce: An estimated 20,000 new cases in the United Si:ates in 1989. It is estimated that about 1.4% or one out of every 70 newborn girls will develop ovar- ian cancer during her lifetime. It accounts for 4% of all cancers among women and 27% of the cancers of the female reproductive system. Mortality: An estimated 12,000 deaths in 1989. Although ovarian cancer ranks second in incidence among gynecological cancers, it causes more deaths than any other cancer of the female reproductive system. Warning Signals: Ovarian cancer is often "silent," showing no obvious signs or symptoms until late in its development. The most common sign is an enlarged abdomen caused by the collection of fluid. Rarely will there be abnormal vaginal bleeding. In women over 40, vague digestive disturbances (stomach discomfort, gas, distention) which persist and cannot be explained by any oth er cause may indicate the need for a thorough checkup for ovarian cancer. Risk Factors: Risk for ovarian cancer increases with age, with highest rates for women 65-84. Women who have never had children are twice as likely to develop ovarian cancer as those who have. A number of inter- related, reproductive factors,, such as age at first live birth, age at first pregnancy, and number of pregnan- cies are all involved in varying degrees. In addition, years of ovulation, the product of a number of other interrelated factors such as length of pregnancies and oral contraceptive use (which may themselves actually decrease risk), are also tied to an observed increased risk. Breast and endometrial cancer increases a.woman's chances of developing ovarian cancer twofold. Patients with colorectal cancer are at increased risk of ovarian cancer, although risk decreases over time following diagnosis of their colorectal cancer. Some rare genetic disorders are associated with increased risk. Incidence rates are higher in North America and Northern Europe, and lower in Asia and Africa. Rates are significantly higher for nuns, Jewish women, and women who have never been married. Early Detection: Periodic, thorough pelvic examina- tions are important. The Pap test, useful in detecting cervical cancer, does not reveal ovarian cancer. Women over the age of 40 should have a cancer-related checkup every year. Treatment: Surgery, radiation therapy and drug ther- apy are all options in the treatment of ovarian cancer. Surgical treatment usually includes the removal of one or both ovaries, the uterus (hysterectomy) and the fallopian tubes. In some very early tumors, only the involved ovary may be removed, especially in young women. In advanced disease, an attempt is made to remove all intra-abdominal disease to enhance the effect of chemotherapy. Survival: If ovarian cancer is diagnosed and treated early, about 85% of such patients live 5 years or longer. However, when diagnosed in an advanced stage, the survival rate drops to 23%. It has improved with mod- em chemotherapeutic agents. Overall, the survival rate for ovarian cancer is 38%. ORAL CANCER Incidence~~ An estimated 31,000 new cases in 1989. Incidence is more than twice as high in males as in females, and i.; most frequent in men over age 40. Cancer can affect any part of the oral cavity, from lip to tongue to mouth and throat. Mortality: An estimated 8,650 deaths in 1989. Warning Si:gnals: A sore that bleeds easily and doesn't heal; a lump or thickening; a reddish or whitish patch that peisists. Difficulty in chewing, swallowing or moving tongue or jaws are often late changes. Risk Factors: Cigarette, cigar and pipe smoking; use of smokeless tobacco; excess use of alcohol. Early Detection: Dentists and primary care physi- cians have the opportunity, during regular checkups, to see abnormal tissue changes and to detect cancer at an early and curable stage. Treatment: Principal methods are radiation therapy and surgery. Chemotherapy is being studied as an aid to surgery in advanced disease. Survival: Five-year survival rates vary substantially depending on the site. Rates range from 32% for cancer of the pharynx to 91% for lip cancer. Overall, 5-year survival for oral cancer patients is about 51%. 11

40 Principles of Health and.Safety in Agriculture demonstrated range of toxicities for human pulmonary dis- ease. We employ hamsters exposed to dusts by either inhala- tion or by intratracheal instillation" and quantify the response by measuring biixhentical and cellular indicators in BAL fluid. The parameters meas.ured represent a wide spectrum of possible responses to inhaled particles, including inflamrrta- tion, pulmonary edema, cellular damage, cellular secretion, and endocytic capacity of pulmonary macrophages. We have calibrated the system with dusts for which there is consider- able human experience. Cellular and biochemical changes were measured in BAL of hamsters afterexposure to a-quartz, iron oxide, and atuiminum ozide.12 a-Quartz is a highly toxic, fibrogenic mineral dust, whereas aluminum oxide and iron oxide are both of h»v toxicity. One day after exposure, the levels of ~-N-acetylglucos- aminidase were significantly elevated by exposure to the 0.75- and 3.75-mg doses of all three dusts (see Figure 1). However, the response to a-quartz was greater than the response to the otller two dusts, especially at the highest dose. J3-tV-acetylglocosarninidase is an example of a lysosomal enzyme that is released from cells during phagocytosis, cell injury, or cell death.'d Polymorphonuclear neutrophils (PMNs),j° macrophages, and type II cellst5 all contain acid hydrolases. Excessive release of lysosomal enzymes may elicit unwanted proteolysis from cathepsins or membrane destruction by phospholipases, a-Quartz also elcvated albumin levels in lavage fluid at both 0.75- and 3.75-mg doses as shown in Figure 2. The highest dose caused a more than 40-fold increase above control levels. Aluminum oxide and iron oxide were also associated with an increase at 3.75 mg, but albumin levels clearly distinguished IXtween these relatively nontoxic dusts and the highly fibrogenic a-quartz. Albumin is primarily a 0 a-twARTZ 250 ~ _..._.._ ~ r -'. .. =.. ~- ~ -~--'--•------"----•• CONTROL 0 015 S r 0.75 3.75 mq W:iT INS?iLLEO 11009 BOUY WEIGHT FIGURE i, Dose-response curve for f~-N-acetylglucosaminidase I d after instillation of particles. p<0.01 for all points except 0.75 mg iron oxide and 0.15 mg aluminum oric'e (p <0A5), and 0.15 mg a-quartz (not significant). Values are mean ± standard errors. (Adapted from Beck, B. D., Brain, J. D., and Bohannon, 1). E., Exp. Lung Res., 2, 289, 1981. With permission.) FIGURE 2. Dose-response curve for albumin in extracellularsupematant of lung lavage fluid 1 d after exposure to iron oxide, aluminum oxide, or a-t}uartz The Wilcoxon rank-sum test was used to compare experimentals and saline-only controls. p <0.01 for all points except 0.75 mg aluminum oxide and all 0.15-mg samples (not significant). Values represent mean ± standard errors. (Adapted from Beck, B. D., Brain, J. D., and Bohannon, D. E., Exp. Lung. Res., 2, 289, 198). With permission.) serum protein whose presence in BAL is due to passage across damaged endothelial and epithelial barriers. Albumin is usu- ally the most abundant protein in BAL.'b~t' Elevated albumin levels indicate pulmonary edema, a common manifestation of acute pulmonary injury.1z,'S Figure 3 illutratres that a-quartz also causes depressed macrophage function. The lambda values shown are the fraction ofrad'toactive gold colloid which was ingested 90 min after it had been instilled through the trachea. Brain and Corkeryf9 provide details of this assay which estimate the endoc}tic activity of macrophages in situ. At a dose of 3.75 mg of a-quartz, less than 30% of the gold was ingested; iron oxide and aluminum oxide have no significant effect on lambda. The full bioassay includes a number of other parameters such as peroxidase, elastase; hemoglobin, as well as the numbers of erythrocytes, neutrophils, and macrophages. An essential aspect of bioassays like this is to compare the responses of unknown dusts with other well-characterized standards. Both positive and negative controls should be used. The best calibrating materials would be those for which there is a considerable experience in humans such as the dusts shown in Figures 1 to 3. Then the type and intensity of response for a new unknown dust could be compared to these standards. A key feature of assays vtilizing lung lavage is the time course of the response. Some agents will yield similar re- sponses when examined soon after exposure. However, the more toxic material may frequently exhibit a more persistent change in the cellular and enzymatic parameters than nontoxic controls. For example, there was a prolonged elevation in the numbers of macrophages and PMNs with quartz, but not with iron oxide. PMN numbers in the lung lavage fluid were

81, 384 (1952). 81. R. W. Wissla, Fros. Inst. Med. Chicago 19,79 (1952); _., M. L. Eilcrt, M. A. Schroeder, L. Gohxt, A.M.A. (Am. Med. Assoc.) Arch. Pathol. 57,333 (1954). 82. M. R. Malinow, D. Hojtnan, A. Pellegrino, Acta CardioL 9, 480 (1954). 83. L. C. Filllos, S. B. Andrus, G. V. Mann, F. J. Starc, J. F_xp. Med. 104, 539 (1956); G. F. Ii'ill7am, ibid. 109, 293 (1959); S. Naimi, R. Goldstein, M. M. Nothman, G. F. Wilgratn, S. Prager, J. Clin. Invest. 41, 1708 (1962); W. J. S. Still and R. M. lYNeal, Am. J. Pathol. 40, 21 (1962). 84. C. R. Seskind, M. ~~.'. Schroeder, R. A. Rasmussen, R. W. Wissler, Proc. Soc. Exp. Biol. Med. 100, 631 (1959); C. R. Scskind, V. R. Whcadry, R. A. Rasmussen, R. W. Wissler, ibid.a0:2, 90 (1959). 85. M. S. Moskowitz, A. A. Moskowitz, W. L. Bradford, R. W. Wissler, Arch. Pathol. 61, 245 (1956); R. W. Priest, M. T. Schroeder, R. Rasmussen, R. W. Wisslcr, Proc. Soc. Exp. B a1. Med. 96,298 (1957). 86. I. Joris, T. Zand, J. J. Nunnari, F. J. Kralikowski, G. Majno, Am. J. Pathol. 113, 341 (1983). 87. W. J. S. Still, Anit. Pathol. 89,392 (1970). 88. K. Okamoto, Int. Rev. Fxp. Pathol. 7,227 (1969); "Spontaneously hypertensive (SHR) rats: Guidelines for breeding, care and usc," ILAR News 19, G1 (1976). 89. A. A. Like, E. Kislauskis, R. M. Williams, A. A. Rossini, Science 216,644 (1982); R. D. Guttmann, ]-1. Colic, F. Michel, T. Scemryer, J. Immunol. 130, 1732 (1983); M. Angelillo et al., ibid. 141, 4146 (1988). 90. K. Okamoto, Cin. Res. Suppl. 1(1972), p. 143. 91. S. Koletsky, Exp. Mol. Pathol. 19, 53 (1973). 92. G. M. Hatrington, i4ehav. Cnut. 11, 445 (1981). 93. R. E. Wimer and C. C. Wimcr, Annu. Rev. Psychol. 36, 171 (1985); G. E. McClearn and T. T. Foch, in Stevens Handbook oJExperimental Psychology, R. C. Atkinson, R. J. Hcrmstcin, G. Lindzcv, R. D. Lucc, Eds. (Wiley, New York, 1988), pp. 677-764. 94. W. S. Small, Am. J. Psychol. 11, 80 (1900). 95. R. C. Tyson, 39th Yemb. N-atl. Soc. Stud)' Educ. 1, 111 (1940). 96. C. S. Hall, in Handbook oJExperimental Psychology, S. S. Stevens, Ed. ( Wilev, Ncw York, 1951), pp. 304-329; C. Gucnairc, G. Fcghali, B. Senault, J. Delacour, Physiol. Behav. 37, 423 (1986); R. L. Comtnissaris, G. M. Harrington, A. M. Ortiz, H. J. Altman, ibid. 38, 291 (1986). 97. J. Olds and P. Milner, J. Comp. Physiol. Psychol. 47,419 (1954); N. E. Miller, Am. Prycho(. 13, 100 (1958). 98. M. Auroux, Teratology 27, 141 (1983). 99. F. R. George, Phannacol. Biochem. Behav. 27, 379 (1987); M. A. Linseman, Psychopharmacology 92, 254 (1987). 100. T. Suzuki, Y. Koike, S. Yanaura, F. R. George, R. A. Meisch, Jpn. J. Pham,acol. 45,479 (1987); T. Suzuki, K. Otani, Y. Koike, M. Misawa, ibid. 47,425 (1988). 101. R. Sodoyer et al., EMBOJ. 3,879 (1984); E. D. Albert, M. P. Baur, W. R. Mayr, Eds., Histocompatibility Testing 1984 (Springcr-Verlag, New York, 1984), pp. 333- 341; "Nomenclature for factors of the HLA system, 1987," Immunogenetics 28, 391 (1988); A. Radojcic et al., ibid. 29, 134 (1989). 102. The work in the autftors' laboratories was supported by grants from the National Institutes of Health [CA 18659, HD 09880, HD 08662 (T.J.G. and H.W.K.) and HL 33740, HL 07237, LM 0009 (R.W.W.)]; the Tim Caracio Memorial Cancer Fund, the Beaver County Cancer Society, and the Pathology Education and Research Foundation (T.J.G. and H.W.K.); the New South Wales State Cancer Council and a Yamigawa-Yoshida Memorial International Cancer Study Grant from the International Union Against Cancer (G.J.S.); and the Nutrition and Heart Disase Study (R.W.W.). 276 SCIENCE, VOL. 2¢5

C A N C E R F A C T S A N D F/ G U R E S 1 9 8 9 PREVENTION/THE AMERICAN CANCER SOCIETY 3. Eat more high-fiber foods such as whole grain cereals, fruilLs and vegetables. Regular consumption of cereals, fresh fruits and vegetables is recommended. Studies suggest that diets high in fiber may help to reduce the risk of colon cancer. Furthermore, foods high in fiber content are a whole- some substitute i'or foods high in fat. 4. include focKls rich in vitamins A and C in your daily diet. People should include in their diet dark green and deep yellow frecshh vegetables and fruits, such as carrots, spinach, sweet potatoes, peaches, and apricots as sources of vitamiin A; and oranges, grapefruit, straw- berries, green and red peppers for vitamin C. These foods may help 'lower risk for cancers of the larynx, esophagus and the lung. The excess use of vitamin A supplements is not recommended because of possible toxicity. S. Include cruciiferous vegetables in your diet. Certain vegetables in the cruciferous family- C A N C E R F A C T S A N D F I G U R E 5 cabbage, broccoli, brussels sprouts, kohlrabi and cau- liflower-may help prevent certain cancers from developing. Research is in progress to determine how these foods may protect against cancer. Cruciferous vegetables have flowers with four leaves in the pattern of a cross. 6. Eat moderately of salt-cured, smoked and nitrite-cured foods. In areas of the world where salt-cured and smoked foods are eaten frequently, there is more incidence of cancer of the esophagus and stomach. The American food industry has developed new processes to avoid possible cancer-causing by-products. 7. Keep alcohol consumption moderate, if you do drink. The heavy use of alcohol, especially when accom- panied by cigarette smoking or smokeless tobacco, increases risk of cancers of the mouth, larynx, throat, esophagus and liver. 1 9 8 9 THE AMERICAN CANCER SOCIETY PROFILE The ACS trace:~, its origins to 1913, when a group of ten physician s and five laymen met in New York City and founded the American Society for the Control of Cancer. Its stated purpose at the time was to "disseminate knowledge concerning the symptoms, treatment, and prevention of cancer, to investigate conditions under which cancer is found; and to com- pile statistics in regard thereto." Later renamed the American Cancer Society, it is today one of the oldest and largest voluntary health agencies in the United States, comprised of 2.5 million Americans united to conquer cancer through balanced programs of research, education, patient service and rehabilitation. Organization: The American Cancer Society, Inc. is composed of a National Society, with 57 chartered Divisions and 3,232 Units. The National Society: A 206-member House of Delegates provides a basic representation from the 57 Divisions and additional representation on the basis of population. It elects and is governed by a Board of Directors of 124 •voting members, approximately half of whom are members of the medical or scientific professions. The National Society is responsible for overall plan- ning and coordination, provides technical help and materials to Divisions and Units, administers programs of research, medical grants and clinical fellowships, and carries out public and professional education on the national level. The 57 Divisions: These are governed by members of Divisional Boards of Directors, both medical and lay, in all the states plus five metropolitan areas, the District of Columbia and Puerto Rico. The Units: These are organized to cover the counties in the United States. There are thousands of community leaders who direct the Society's programs at this level. The Programs: The Society maintains its priorities and goals through activities developed by the depart- ments of Research, Professional Education, Public Education, Public' Information, Epidemiology and Statistics, Service and Rehabilitation, Public Affairs, and Crusade. PUBLIC EDUCATION The American Cancer Society has a strong and long- standing commitment to educating the public about ways of preventing or reducing the risk of developing cancer. Because each year thousands of lives could 22 be saved through cancer prevention, risk reduction and early detection practices, the Society's Public Ed- ucation programs are designed to inform people about cancer, tell them what they can do to protect

excluded children with excess lead, or plumbism: Prior to exclusion, with N=167, the lead effect t=-1.51 (g =.133, 2 - sided); after exclusion, with N= 171, t=-2.56 (p =.011). This suggests the presence of high IQ's in the plumbism group. In the present follow-up report, the previously excluded cases who agreed to participate were incorporated in the analysis, including, in separate descriptive summaries, ten of the plumbism cases. Five of these plumbi;am cases had reading disabilities, and three out of seven failed to graduai;e high school. These high proportions of adverse outcomes seem to corroborate the hypothesized lead effect. However, in view of the apparently contradictory IQ data described above, a summary of the IQ scores of all 16 plumbism cases would be helpful in assessing the implications of the findings. ~r`..~a..t re `3 FYk 4 01~

208 Hertz-Picciotto, Gravitz, and Neutrlt Table L Numbers of Cancer Deaths Predicted by Linear Nonthreshold Models Fitted to Inhalation Data Excess predicted by one-hit model Observed Onutted With tumors found at terminal sacrifice: Included (95% Cl) Expected° 0.9 ppm 3.0 ppm 0.9 ppm 3.0 ppm Texas 3` (0.6-8.8) 3.6 0.13° (0.17)` 0.44 (0.57) 0.30 (0.38) 1.00 (1.25) Michigan 5 (1.6-11.7) 2.2 0.08 (0.10) 0.26 (0.33) 0.17 (0.22) 0.58 (0.72) °From t1.S. white male age- and calendar-year-specific mortality rates. hModels were fitted to inhalation data using Global 82 software published by Crump and Howe (1982). Proportion with tumors was based on life table adjustment. Animals who died prior to the first tumor were not included. `Included one arteriosclerotic heart disease death with lymph node malignancy (see Ref. 2). dMaximum likelihood estimates. `Numbers in parentheses are upper 95% confidence limits. mortality was close to the observed mortality among the EDB-exposed workers. 'The multistage model generalizes the one-hit model by allowing for nonlinear terms in the hazard rate of cancer death. This model is of the form P(d)-P(0)=1-exp[ -(iBt.d+RZ.d2+ .-..)] with E3; the unknown parameters. When fitted to the inhalation data, the linearized multistage model pre- dicted identical risks, that is, the strong linearity in the data dictated that the best fit was the one-hit model. 3.2. Gavage Data: Three Models The gavage bioassay was marked by severe early mortalhy from both toxic and carcinogenic effects of the high doses of EDB: one-third of the high-dose animals died by the 15th week:'The two-year bioas- SQY Wils, iht'.Bt'ft)rC tt'fn)inJlCtf hefi!(C /I):r rilj!# 11` 1})X' I/r t;, l,~ 1'1, !udlw.t joir 1Ir, 0u,rlr•ru'll (Ilhsprul~ sttr4 4Nii1 , 1111, 1 # t///i N, /t ri. 11 'l////Nt{oN 'if (hl' lf//1~~14t:)j!l tt/~n/i 1/ti~ ~~tlutttlflft~ tht,• Sltfv/vaI t/rr1cS of the animals (tlcnoted tttultistage with time-to-tumor model) was fitted to these data using Weibull 82 software. The form of this model is P(d)--P(0) =1-expl-'(Pt-d+P2•d' + ..-)(t-to)r~ where t represents time since first exposure and to represents the latency period. Thus, (3i and to are parameters to be estimated. Using this model, the upper-limit predictions were two to three times the observed cancer deaths assuming exposures of 3.0 ppm: 11.9 and 3.9 excess cancer deaths at the Texas and Michigan plants, respectively, In response to the early mortality, dosing was stopped for the high-dose animals, and subsequently a variable dosing pattern was instituted for this group. Zeise and Crouch,t1I Thorslund,t23t and Crump and Howetz4> developed a special case of the Armitage- Doll multistage model for carcinogenesis, which in- corporated such a variable dosing regimen. This model was used in the CAG's final risk assessment for EDB,tII> and took the form P(d) -P(0) =1-expt-(fl•d)[(t-s)'' -(t- f )Y11 where s is the age at start of exposure, f is the age at end of exposure, t is the age at end of observation period, and d is the daily dose from age s tn f. Y/Jtxtr t/tt~•f t/, tdo; pvage data, this model gaVe predictiorts that were similar to those of the ti.me-to- turrurr rru,del: 10.4 and 5.5 excess cancer deaths at the Texas and Michigan plants, respectively, assum- ing concentrations of EDB averaged 3.0 ppm (see Table II). Thus, a variable dosing schedule did not significantly influence the risk projections. The Cox proportional hazards model differs from the previously described models by treating the in- crease in risk as a multiplicative rather ~ than an additive effect. The model takes the form P(d, t) =1-expI- f tX(d,,. u) 3.1 0

FIVE-YEAR CANCER SURVIVAL RATES* FOR SELECTED SITES ORAL COLON- RECTUM PANCREAS LUNG MELANOMA FEMALE BREAST CERVIX UTERI IJTERI OVARY PROSTATE TESTIS BLADDER LEUKEMIA el% 0 20 60 : .. . . ..... . .. --. PL--a kl % `'M ALL STAGES ® LOCALRED REG{ONAL 100 QDtSTANT `Adjusted for normal life expectancy. Source: Surveillance and Operations Research Branch, fhis chart is based on cases diagnosed in 1979-1984. National Cancer Institute. HOW TO ESTIMATE CANCER STATISTICS LOCALLY ommunity Population Estimated No. Who Are Alive, Saved from Cancer Estimated No. Cancer Cases Under Medical Care in 1989 Estimated No. Who Will Die of Cancer in 1989 Estimated No. of New Cases in 1989 Estimated No. Who Will Be Saved from Cancer in 1989 Estimated No. Who Will Eventually Develop Cancer Estimated No. Who Will Die of Cancer if Present Rates Continue 1,000 10 5 1 3 1 280 180 2,000 20 11 4 7 3 560 360 3,000 30 16 5 10 4 840 540 4,000 40 21 7 13 5 1,120 720 5,000 50 26 9 16 6 1,400 900 10,000 100 52 18 33 12 2,800 1,800 25,000 250 131 45 79 30 7,000 4,500 50,000 500 262 90 158 59 14,000 9,000 100,000 1,000 525 180 325 122 28,000 18,000 200,000 2,000 1,050 360 650 244. 56,000 36,000 500,000 5,000 2,625 900 1,575 590 140,000 90,000 NOTE: The figures can only be the roughest approximation of actual data for your community and should be used with caution. It is suggested that every effort be made to obtain actual data from a Registry source. 15

C A N C E R F A C T S A N D F I G U R E S 1 9 8 9 PREVENTION BREAST CANCER: A PROGRAM OF ACTION About one cut of every 10 women in the United States will develop breast cancer during her lifetime. And until the disease ca)1 be prevented, the best way women can protect themselves is through early detection and prompt treatment. Today, with modern technologies, breast cancer can be detected at very early stages of development, when the chance of cure is highest. The risk of breast cancer increases as a woman grows older, and genetic and lifestyle variances-a history of breast cancer ir1 a close family relative, giving first birth after age 30, never giving birth, and obesity (body weight 40% above normal)-may increase risk further. The American Cancer Society recommends that women develop a three-part, personal plan of action, in cooperation with their doctors for early detection of breast cancer. (See page 19 for Checkup Guidelines.) A clinical breast exam should be performed by a doctor as part of a regular health checkup. This includes a visual inspection of the breasts, looking for changes in shape or size or skin dimpling, followed by a thor- ough inspection of the breast, chest and armpits. Women should ask their doctors how often they should have a clinical breast exam. A mammogram is a low-dose breast X ray that can identify cancers too small to be felt. Follow the ACS guidelines for recommended frequency, depending on age and health history. Recent improvements have reduced the amount of radiation necessary for high- quality mammograms. The Society recommends that all women over the age of 20 perform breast self-examination once a month. BSE is important because breast cancer symptoms may develop and be found between clinical breast exams or mammography. Through regular self-examination women become familiar with their breasts, making any changes more likely to be noticed. TOBACCO USE The American. Cancer Society estimates that cigarette smoking is respcnsible for 85% of lung cancer cases among men and 75% among women-about 83% overall. The cancer death rate for male cigarette smokers is more than double that of nonsmokers, and the rate for female smokers is 67% higher than for nonsmokers. The American Car cer Society estimates that 40% of male smokers and 28% of female smokers die prematurely, or about 35% overall. The higher ca nc:er rates for men reflect the fact that in the past, more men than women smoked, and smoked more heavily. In recent years, however, the gap between male and female smoking has been narrowing. Smoking also has been implicated in cancers of the mouth, pharynx, larynx, esophagus, pancreas, cervix uteri and bladder. Smoking accounts for about 30% of all cancer deaths, is a major cause of heart disease, and is linked to conditions ranging from colds and gastric ui<cers to chronic bronchitis and emphysema. Smoking is relai:ed to 390,000 deaths each year. A September 1985 study by the U.S. Congress Office of Technology Assessment estimates the cost of smoking to the economy from $38 billion to $95 billion, with a middle estimate of $65 billion. This amounts to $2.17 in lost productivity and the treatment of smoking- related diseases for each pack of cigarettes sold. A Decline in Smoking A September 198;' tobacco report of the U.S. Depart- ment of Agriculture estimates cigarette output in 1987 at 654 billion, down 1.0% from 1986, about the same decrease as the previous year. From 1976 to 198' 7, adult male smokers (20 years and older) dropped from 42% of the population to 33%, while women smokers decreased from 32% to 28%, according to the National Center for Health Statistics. Overall, the percentage of adult smokers in the population had dropped to 30%. A 1987 report from the Office of Smoking and Health says that 26.5% of Americans now smoke. Per capita cigarette consumption among adults has fallen-from 4,141 in 1974 to 3,121 in 1988-reflecting a growing number of ex-smokers. This is the lowest per capita consumption since 1944. From 1965 to 1987, the proportion of adult male ex-smokers (20 years and older) in the total U.S. population increased from 20% to 31%, while female ex-smokers rose from 8% to 19%. A survey supported by the National Institute on Drug Abuse indicated that the percentage of high school seniors (aged 17 and 18) who smoked cigarettes daily decreased from 29% in 1976 to 19% in 1987. It is now estimated-from past national surveys and data from the Cancer Prevention Study 11-that there are about 40 million ex-cigarette smokers in the U.S. today and about 50 million smokers. At the same time, however, the average smoker appears to be smoking more heavily. The U.S. Office on Smoking and Health reports that the proportion of adult male smokers (20 years and older) consuming 25 or more cigarettes per day increased from 30.7% to 32% between 1976 and 1985, and female smokers from 19.0% to 21%. Figures from the U.S. Department of Agriculture show that a total of 567 billion cigarettes were 'con- sumed in 1988, down from 575 billion in 1987. Nicotine Addiction The Surgeon General released a report on nicotine addiction in May 1988. The report points out that all tobacco products contain substantial amounts of nicotine. Nicotine is absorbed readily from tobacco smoke in the lungs and from smokeless tobacco in the mouth or nose, and is rapidly distributed throughout 20

Program on Chemical Safety Collaborative Study on In Vitro Asa)s, pp. 397-q 11. J. A;hby, F.J. de Serres, M. Draper, et al., Edti. Elsevier, Amsterdam (1985). -t5. Morimoto, K; Wolff, S.; Koizumi, A: Induction of Sister-Chromatid Exchanges in Human Lymphocyt-s by Microsomal Activation of Ben- zene Metabvlites Mutat. Res. 119:355-360 (1983). 46. Morimoto, K; Wolff, S.: Increase of Sister Chromatid Exchanges and Perturbations of Cell Division Kinetics in Human Lymphocytes by Benzene Metabolites. Cancer Res. 40:1189-1193 (1980). 47. Morimoto, K: induction of Sister Chromatid Exchanges and Cell Division Delay;> in Human Lymphocytes by Microsomai Activation of Benzene. Cancer Res. 43:1130-1334 (1983). 48. Erexson, G.L; Wilmer, J.L; Kligerman, AD.: Sister Chromatid Ex- change Induction of Human Lymphocytes Exposed to Benzene and Its Metabolites in t'itro. Cancer Res. 45:2471-2477 (1985). 49. Tice, RR; Vogt, TP.; Costa, D.L: Effect of Sex, Strain, Age and Route of Exposure on Benzene-induced Sister Chromatid Exchange (SCE) in Murine Bone Marrow. Environ. Mutagen. 3:338-339 (1981). 50. Glauert, H.P.; Kennan, W.S.; Sattler, G.S.; Pitot, H.C.: Acsays to Measure the Induction of Unscheduled DNA Synthesis in Cultured Hepato- cvtes. In: Evaluc,tion of Short Term Tests for Carcinogens: Report of the Internationai Program on Chemical Safety Collaborative Study on In Vitro Assays, pp. 371-373. J. Ashby, FJ. deSerres, M. Draper, et al., Eds. Elsevier, Amsterdam (1985). 51. Pellack-Walker, P.; Blumer, J.L.: DNA Damage in L5178YS CelLs Fol- lowing Exposure o:) Bertzene'Metabolites. Molec. PhartnacoL 30:42-47 (1983). 52. Tice, RR; Sawe,v, M].; Drew, R.T.; Cronkite, E.P.: Benzene-induced Micronuclei in the Peripheral Blood of Mice: A Retrospective Anal- ysis. Environ. Mutagen. 6:421 (1984). 53. Luke, CA; Tice, RR; Drew, RT.: Duration and Regimen Induced Micronuclei in the Peripheral Blood of Mice Exposed Chronically to Benzene. Environ. Mutagen. 7(Suppl. 3):29 (1985). 54. Choy, W.N.; MacG:.egor, J.T.; Shelby, M.D.; Maronpot, RR: Induction of Micronuclei in ihe Peripheral Bkwd of Mice Exposed Chronically to Benzene. Mutar.. Res. 143:55-59 (1985). 55. Gad-El-Karim, M.Dd.; Harper, BJ.; Legator, M.S.: Modification in the Myek>clastogenic Effect of Benzene in Mice with Toluene, Phentr barbital, 3-Methylcholanthrene, Arochlor 1254 and SKF-525A ( Proad- ifen Hydrochloride). Mutat. Res. 135:225-243 (1984). 56. Styles, JA; Richardson, C.R: Cytogenetic Effects of Benzene: Dosi- metric Studies on Rats Exposed to Benzene Vapour. Mutat. Res. 135:203-209 (1984). 57. Erexx>n, G.L; Wilmer, J.L; Steinhagen,. W.H.; Kilgerman, AD.: In- duction of Cytogenetic Damage in Rodents after Short-term Inha- lation of Benzene. Environ. Mutagen. 8:29-40 (1986). 58. McGregor, D.; Aslaby,J.: Summary Report on the Performance of the Cell Transformation Assays. In: Evaluation of Short-Term Te.scs for Carcinogen.s: Report of the International Program on Chemical Safen• Collaborative Study on In Vitro Assays, pp. 103-115. J. Ashby, FJ. de,Serres, M. Draper, et al., Eds. Elsevier, Amsterdam (1985). 59. Post, G.B.; Snyder, R; Kaif, G.F.: Inhibition of RNA Synthesis and Interleukin-2 Pr;>dluction in Lymphocytes in tftro by Benzene and Its Metabolite.s, Hydroquinone and p-Benzoquinone. Toxicol. Lett. 29:161-167 (1985). 60. Forni, A.M.; Capellini, A; Pacifico, E.; Vigliani, E.C.: Chromosome Changes and Their Evolution in Subjects with Past Exposure to Ben- zene. Arch. Environ. Health 23:385-391 (1971). 61. Tough, LM.; Cocrt•Brown, W.M.: (~hromosome Aberrations and Ex- posure to Ambieni Benzene. Lancet 1:684 (1965). 62. Tough, LM., Smith, P.G.; Court-Brown, W.M.; Hamden, D.G.: Chro- mosome Studies on Workers Exposed to Atmospheric Benzene. The Pos.tiible Influence of Age. Eur. J. Cancer 6:49-55 (1970). 63• Watanabe, T.; End(.,, A; Kato, Y.; et al.: Cytogenetics and Cy2okinetics of Cultured 1-ymph;xlrtes from Benzene-exposed Workers. Int. Arch. Occup. Ern•iron. Health 46:31-41 (1980). 64. Killian, I).J.; LY,tnier, RL.: Cytogenetic Study of Workers Exposed to Benzene in the ife.+as DivLsion of Dow Chemical L'SA Februan• 29, 1978. OS11A Lk c. No. H-059, Exhibit No. 230 X-2. Occupational Safety and Health Administration, VPatihingtan, DC ( 19'tt). 65. Picciano, D.J.: Cytogenetic Study of Workers Exposed to Benzene, Environ. Res. 19:33-39 (19'9). 66. Picciano. D.(.: Monitoring Industrial Populations by Cytogenctics Pro- cedures. In: Proceedings of a Workshop on Methodology for A- sesing Reproductive Hazards in the Workplace, pp. 293-306. P.F. Infante and M.S. Legator. Eds. Ci.S. Government Printing Office, Washington. i)C (1980). 67. Ding, Xf.: Li, Y.; Ding, Y.: Chromosome Changes in Patients with Chronic Benzene Poisoning. Chinese Med. J. 96:681-685 (1983). 68. Sasiadek, M.; Jagielski, J.: Localization of Break-points in Chromo- somes of Workers Occupationally Exposed to Benzene. Clinical Genet. 28:462 (1985). 69. Sarto, F.; Cominato. I.; Pinton, AM.; et al.: A Cytogenetic Study on Workers Exposed to Low Concentrations of Benzene. Carcinogenesis 5:827-832 (1984). 70. Franz, TJ.: Percutaneous Absorption of Benzene. In: Proceedings of the Symposium: The Toxicology of Petroleum Hydrocarbons, pp. 108-114 H. MacFarland, C. Holdsworth, J. MacGregor, et al., Eds. American Petroleum Institute, Washington, DC (1983). 71. Blank, I.H.; McAuliffe, D. J.: Penetration of Benzene Through Human Skin. J. I nvest. Dermatol. 85:522-526 (1985). 72. Susten, AS.; Dames, B.L; Niemeier, R.W.: In t*,o Percutaneous Ab- sorption Studies of Volatile Solvents in Hairless Mice. 1. Description of a Skin Depot. J. Appl. Toxicol. 6:43-46 (1986). 73. Sabt>urin, PJ.; Chen, B.T.; Lucier, G.; et al.: Effect of Dose on the Absorption and Excretion of ('•Cj-Benzene Administered Orally or by Inhalation in Rats and Mice. Toxicol. Appl. Pharmaco1.87:325-336 (1987). 74. Kalf, G.F.; Post, G.B.; Snyder, R: Solvent Toxicology: Recent Advances in the Toxicology of Benzene, the Glycol Ethers and Carbon Tet- rachloride. Ann. Rev. Pharmacol. Toxicol. 27:399-427 (1987). 75. Gonasun, LM.; Witmer, C.; KocsLs, JJ.; Snyder, R.: Benzene Metab- olism in Mouse Liver Microsomes. Toxicol. AppL Pharmacol. 26:398-406 (1973). 76. Medinsky, MA; Sabourin, PJ.; Lucier, G.; et al.: A Physiological Model for Simulation of Benzene Metabolism by Rats and Mice. Toxicol. Appl. Pharmacol. 99:193-206 (1989 )- '7. Sawahata, T.: Rickert, D.E.; Greenlee, W.F.: Metabolism of Benzene and Its Metabolites in Bone Marrow. In: Toxicology of the Blex>d and Bone Marrow, pp. 141-148. RD. Irons, Ed. Raven, New York (1985). 78. Jerina, D.; Daly, J.; Witkop, B.; et al.: Role of Arene Oxide-Oxepin S.stem in the Metabolism of Aromatic Substrates. 1. In tdrro Con- version of Benzene Oxide to a Mercapturic Acid and a Dihydrodiol. Arch. Biochem. Biophys. 128:176-183 (1968). 79. Tunek, A: P1at, KL; PrpzybbyLski, M.; Oe,sch, F.: Multistep Metabolic Activation of Benzene. Effect of Superoxide Dismutase on Covalent Binding of Microsomal Macromolecules, and Identification of Glu- tathione Conjugates Using High Pressure Liquid Chromatography and Field Desorption Mass Spectrcxnetn% Chem. BioL Interact. 33:1-17 (1980). 80. Gaido, KW.; Wierda, D.: in ritro Effects of Benzene Metabolites on Mouse Bone Marrow Stromal CelLs. Toxicol. AppL PharmacoL 76:45-55 (1984). 81. Greenlee, W.F.: Sun, J.D.; Bus, J.S.: A Proposed Mechanism of Ben- zene Toxicin: Formation of Reactive Intermediates from Polyphenol Metabolites. Toxicol. Appl. Pharmacol. 59:187-195 (1981). 82. Irons, RD.: Neptun, DA; Pfeifer, RW.: Inhibition of Lymphocyte Transformation and Microtubule As,sembly by Quinone Metabolites of Benzene: Evidence for a Common Mechanism. J. Reticuloen- d >thel. Soc. 30:359-372 (1981). 83. Irons. R.D.: Neptun. DA.: Effects of the Principal Hydroxymetabo- lites of Benzene on Microtubule Polvmerization. Arch. Toxicol. 45:297-305 (1980). 84. Goldstein, B.D.; et al.: S}miposium on Benzene Metabolism, Toxicity and Carcinogenesis. Environ. lfealth Perspect. 82:3-310 (1989). 85. Gosselin, RE.; Smith, R.P.; Hodge, H.C.: Clinical Toxicology of Com- mercial Products, 5th ed., pp. 11-151. Williams & Wilkins, Baltimore (1984). 86. Flun•, F.: Mexterne gewerbliche vergiftungen in Pharmakologische 462 APPL OCCUP. ENVIRON. HYG. 5/7) - JULY 1990

References 1. Needleman HL, Schell A, Bellinger D, Leviton A, Allred EN. The long term effects of exposure to low doses of lead in childhood. N Eng J Med 1990;322: 83-8. 2. McMichael AJ, Baghurst PA, Wigg NR, Vimpani GV, Robertson EF, Roberts RJ„ Port Pirie cohort study: Environmental exposure to lead and children's abilities at the age of four years. N Eng J Med 1988;319: 4l3EI-75. 3. Fergusson DM, Fergusson JE, Horwood LJ, Kinzett NG. A longitudinal study of dentine lead levels, intelligence, school performance and behaviour II. Dentine lead and cognitive ability. J Child Psychol Psychiatry 1988;29:793-809. 4. Ernhart CB, Morrow-Tlucak M, Wolf AW, Super D, Drotar D. Low level lead exposure in the prenatal and early preschool periods: Intelligence prior to school entry. Neurotoxicol Teratol 1.,989;11: 161-170. 5. Caldwell BM, Bradley R. Home Observation for the Measurement of the Ernriironment. Unpublished manuscript. Little Rock: Univ of Arkansas at Little Rock, 1984. 6. Po11ansky NA, Borgman RD, De Saix C. Roots of Futility. San Francisco: Jossey-Bass, 1972. 7. Dietrich KN, Krafft KM, Pearson DT, Harris LC, Bornschein RL, Hammond PB, Succop PA. Contribution of social and developmental factors to lead exposure during the first year of life. Pediatrics 1985;75:1114-9.

C A N C E R F A C T 5 A N D F t C U R E 5 1 9 8 9 THE AMERICAN CANCER SOCIETY themse;!ves, and demonstrate related health habits and lifestyle s. The Society places its major educational focus in two areas: 1) primary cancer prevention which includes smoki;ng control and the relationship between diet, nutrition and cancer, and 2) the importance and value of periodic, cancer-related checkups and specific cancer early detection tests. Prompt action in the event that one of cancer's seven warning signals occurs, is also encouraged. Six cancer sites offer the greatest opportunity for the prevention or cure of cancer: colon and rectum, lung, breast, uterus, oral cavity and skin. These sites account i'or the majority of cancer cases and about half of all ca ncer deaths. The Society's Public Education planning strategy places emphasis on these six sites where prevention, risk reduction and early detection practices realize the greatest return in terms of lives saved. Educating the Young and Old ACS Public Education programs are divided into two major audience categories: adults and youth. Adults are reached through their worksite, healthsite and community. Programs for adults are carried out in small group settings or on a one-to-one basis, involving two-way communication and interaction. Whenever possible, volunteers are selected on the basis of skills that can be readily adapted to Society work, such as ex-smokers with group experience who can help in smoking cessation programs, and nurses who can teach breast seLf-examination to groups of women. The Society reinforces its Public Education messages with a variety of audio-visuals, pamphlets and posters. Youth programs are organized according to age- level to reach children and youth on the pre-school, elementary, intermediate and secondary levels. The program fo:r youth is a scientific, comprehensive cancer education program with promise of significant impact on cancer risk. Educational strategies are designed to teach young people good health habits, help them to make health-enhancing lifestyle decisions and under- stand health behavior as it relates to cancer risk re- duction. Materials are available as coordinated com- ponents or prograr4 packages and are implemented through existing school curricula or as a basic intro- duction to health. Youth programs are usually. con- ducted in the nation's schools and often include activities to be used in the home and community. Reaching More People In 1987-88, American Cancer Society Public Education programs, carried out at local levels, reached 23 million adults and 27 million young people for a total of 50 million. In the decade of the 1980's, the Society, as its goals, hopes to encourage more Americans to have tests for colorectal cancer, reduce the number of smokers, and increase the number of women who have breast cancer detection tests and who practice monthly breast self- examination, get Pap tests and have endometrial tissue samples taken. To help achieve its education objectives and priorities, the Society has launched a number of programs including "Taking Control" and "Eating Smart" for a healthier life of reduced cancer risk; "Special Delivery, Smoke Free" for pregnant women who are smokers; "Starting Free, Good Air For Me" for preschool children; "Where There's No Smoke..." on involuntary tobacco smoke; and an educational emphasis on breast cancer detection awareness, "Special Touch." In addition to the Society's intensive, person-to- person educational outreach, broader ACS programs blanket the nation with lifesaving messages. During the Society's annual door-to-door fund-raising Crusade, volunteers make personal home visits, informing individuals on how to protect themselves against cancer. PROFESSdONAL EDUCATION ACS Professional Education programs bring the latest developments in cancer control and management to health professionals, especially primary care providers. Professional Education's National conferences, clin- ical awards, materials, professorships and scholarships provide information and training in the prevention and early detection of cancer, and in the treatment and rehabilitation of cancer patients. Breast Cancer Detec- tion Awareness, Colorectal Health Check and Tobacco- Free Young America are among the major initiatives offered by Divisions and Units as interdepartmental collaboration promoted by Professional Education. Recruitment and involvement of health professionals into Professional Education remains a major objective, particularly primary care providers. Audiovisuals, )ournals, and Other Publications Videotapes, films, slide sets, audiotapes, publications and exhibits are available for physicians and other health professionals as well as for programs in hospitals, medical, dental and nursing schools. The Society pub- lishes several texts and pamphlets dealing with various cancer issues along with proceedings of its conferences and workshops. Audiovisuals and other publications are distributed through ACS Divisions and Units. Ca A Cancer Journal for Clinicians, (470,000 circulated) is directed to update health professionals about cancer. The Society supports the publication of Cancer, directed to those specializing in cancer research and in the care of the cancer patient. 23

ESTIMATED NEW CANCER CASES AND DEATHS BY SEX FOR ALL SITES-1989• ESTIMATED NEW CASES ESTIMATED DEATHS ~ Total Male Female Total Male Female ALL SITES 1,010,000' 505,000' 505,000' 502,000 266,000 236,000 Buccal Cavity & Pharynx (ORAL) 30,600 20,600 10,000 8,650 5,775 2,875 Lip 4,200 3,700 500 100 75 25 Tongue 6,000 3,900 2,100 1,950 1,300 650 Mouth 11,700 7,000 4,700 2,600 1,600 1,000 Pharynx 8,700 6,000 2,700 4,000 2,800 1,200 Digestive Organs 227,800 115,200 112,600 123,000 64,400 58,600 Esophagus 10,100 7,200 2,900 9,400 6,900 2,500 Stomach 20,000 11,900 8,100 13,900 8,200 5,700 Small Intestine E 2,700 1,400 1,300 900 500 400 Large Intestine ~ (COLON-RECTUM) 107,000 50,000 57,000 53,500 26,000 27,500 Rectum 44,000 23,000 21,000 7,800 4,000 3,800 Liver & Biliaxy Passages 14,500 7,500 7,000 11,400 5,800 5,600 Pancreas 27,000 13,000 14,000 25,000 12,500 12,500 Other& Unspecified Digestive 2,500 1,200 1,300 1,100 500 600 Respiratory System 171,600 114,000 57,600 147,100 96,900 50,200 Larynx 12,300 10,000 2,300 3,700 3,000 700 LUNG 155,000 101,000 54,000 142,000 93,000 49,000 Other & Umapecified Respiratory 4,300 3,000 1,300 1,400 900 500 Bone 2,100 1,200 900 1,300 700 600 Connective Tissue 5,600 3,000 2,600 3,000 1,400 1,600 SKIN 27,000" 14,500" 12,500" 8,200t 3,200 3,000 _ BREAST 142,900"' 900"' 142,000'°` 43,300 300 43,000 _ GenitalOrgaws 181,800•" 109,900 71,900•" 52,200 29,100 23,100 Cervix Uteri ~ (UTERUS) 13,000"' - 13,000•°' 6,000 - 6,000 Corpus, Endometrium 34,000 - 34,000 4,000 - 4,000 Ovary 20,000 - 20,000 12,000 ~ 12,000 Other & Unspecified Genital, Female 4,900 - 4,900 1,100 - 1,100 Prostate 103,000 103,000 - 28,500 28,500 - Testis 5,700 5,700 - 350 350 - Other & Unspecified Genital, Male 1,200 1,200 - 250 250 - Urinary Organs 70,200 49,000 21,200 20,200 12,900 7,300 Bladder 47,100 34,500 12,600 10,200 6,900 3,300 Kidney & Other Urinary 23,100 14,500 8,600 10,000 6,000 4,000 Eye 1,900 1,000 900 300 150 150 Brain & Central N ervous System 15,000 8,200 6,800 11,000 6,000 5,000 Endocrine Glands 12,600 3,700 8,900 1,750 775 975 Thyroid 11,300 3,000 8,300 1,025 375 650 Other Endocrine 1,300 700 600 725 400 . 325 Leukemia 27,300 15,200 12,100 18,100 9,800 8,300 Lymphocytic Leukemia s13, 00 7,500 5,500 7,000 3,900 3,100 Granulocytic Leukemia V,300 7,200 6,100 10,600 5,600 5,000 Monocytic Leukemia 1,000 500 500 500 300 200 Other Blood & Lymph Tissues 51,800 27,000 24,800 27,400 14,100 13,300 Hodgkin's Disease 7,400 4,200 3,200 1,500 900 600 Non-Hodgkin's Lymphomas 32,800 16,800 16,000 17,300 8,900 8,400 Multiple Myeloma 11,600 6,000 5,600 8,600 4,300 4,300 All Other & Unspiecified Sites 41,800 21,600 20,200 36,500 18,500 18,000 NOTE: The estimates of new cancer cases are offered as a rough guide and should not be regarded as definitive. Especially note that year-to-year changes may only represent improvements in the basic data. ACS six major sites appear in boldface caps. 'Carcinoma in situ and non-melanoma skin cancers are not included in totals. Carcinoma in situ of the uterine cervix accounts for more than 50,000 new casEs annually, and carcinoma in situ of the female breast accounts for about 10,000 new cases annually. Non-melanoma skin cancer accounts for more than 500,000 new cases annually. °•tvlelanoma only. '••Invasive cancer only. iNCIDENCE ESTIMATES ARE BASED ON RATES FROM NCI SEER PROGRAM 1983-85. tMelanoma 6,000; other skin 2_)00 8

8. Hunt TJ, Hepner R, Seaton KW. Childhood lead poisoning and inadequate child care. Am J Dis Child 1982;136:538-542. 9. Bradley RH, Caldwell BM, Rock SL, Ramey CT, Barnard KE, Gray C, Hammond MA, Mitchell S, Gottfried AW, Siegel L, Johnson DL. Home environment and cognitive development in the first 3 years of life: A collaborative study involving six sites and three ethnic groups in North America. Dev Psychol 1989;25:217-35. 10. He,as RD, Holloway SD. Family and school as educational institutions. In„ Parke RD, ed. The Family. Chicago: Univ. Chicago Press, 1984. 11. Schroeder SR, Hawk B. Psycho-social factors, lead exposure and IQ. In: SR Schroeder (Ed.) Toxic Substances and Mental Retardation: Neurobehavioral Toxicology and Teratology. Washington, D.C.: AAMD Monograph Series, 1987 12. Needleman HL, Gunnoe C, Leviton A, Reed R, Peresie HD Maher C, Barrett P. (1979). Deficits in psychological and classroom performance in children with elevated dentine lead levels. N Eng J Med 1979;300: 689-95. 13. US I:nvironmental Protection Agency. Independent peer review of sele:cted studies concerning neurobehavioral effect of lead exposures in nominally asymptomatic children: Official report of findings and rec;camendations of an interdisciplinary expert review cannittee. (EPA-600/8-83-028A). 14. Needleman HL. Appendix to the ECAO critique. Unpublished manuscript, on file with the Environmental Protection Agency, 1984.

NEW CANCER CASES-1989 Estimated New Cancer Cases for All Sites Plus Major Sites, by State-1989 _ ALL SITES* MAJOR SITES Number of Female Colon & Skin STATE Cases Breast Rectum Lung Oral Uterus Prostate Melanoma Pancreas Leukemia Alabama 18,000 2,400 2,300 2,800 450 950 2,100 400 500 450 Alaska 1,000 150 125 150 40 20 100 50 20 10 Arizona 13,000 1,800 1,700 1,900 350 550 1,500 300 350 375 Arkansas 11,300 1,100 1,500 1,900 200 400 1,000 350 350 300 California 101,000 14,200 13,500 15,400 3,500 5,000 10,000 3,200 2,600 2,800 Colorado 9,400 1,500 1,5()O 1,200 225 450 1,100 400 250 250 Connecticut 14,400 2,200 2,300 2,000 450 600 1,400 40a 375 375 Delaware 2,800 400 450 500 40 125 275 70 70 80 Dist. of Columbia 3,200 450 400 450 250 150 450 60 90 60 Florida 65,500 8,300 10,200 10,600 2,200 2,800 7,600 1,800 1,700 1,600 Georgia 22,500 2,800 2,900 3,700 850 1,100 2,400 600 550 600 Hawaii 3,300 350 450 450 150 150 275 80 90 80 Idaho 3,500 500 475 475 60 125 425 150 80 125 Illinois 48,000 •7,000 7,800 7,400 1,400 2,400 4,700 950 1,300 1,300 Indiana 23,200 3,200 3,700 3,800 650 1,200 2,200 600 550 550 Iowa 12,700 1,800 2,100 1,800 400 550 1,500 300 375 375 Kansas 9,900 1,400 1,600 1,600 300 450 1,200 200 300 325 Kentucky 16,800 2,100 2,500 3,000 450 850 1,600 350 375 425 Louisiana 17,500 2,200 2,200 3,000 550 750 1,700 300 500 400 Maine 5,500 750 950 850 150 250 600 100 150 150 Maryland 19,300 2,700 2,900 3,000 650 800 1,900 550 400 450 Massachusetts 28,400 4,800 4,500 3,800 800 1,000 2,600 800 700 700 Michigan 37,400 5,500 5,300 5,800 1,000 1,700 3,600 900 950 1,000 Minnesota 16,400 2,300 2,600 2,100 400 650 2,00D 400 450 450 Mississippi 12,000 1,100 1,400 1,800 300 600 1,200 250 325 300 Missouri 23,500 3,200 3,900 3,700 700 1,200 2,000 600 650 700 Montana 3,200 500 425 425 60 150 400 90 100 100 Nebraska 6,400 900 1,100 900 150 350 750 175 225 175 Nevada 4,100 500 500 750 150 175 375 150 90 70 New Hampshire 4,000 650 650 600 80 200 400 125 125 90 New Jersey ^ 36,500 5,500 6,200 5,300 1,200 1,800 3,500 950 1,000 850 New Mexico 4,500 600 600 550 125 175 550 80 150 150 New York 77,500 12,100 13,200 10,900 2,500 4,000 7,900 2,100 2,300 2,000 North Carolina 24,500 3,400 3,200 4,000 900 1,300 2,700 750 700 700 North Dakota 2,700 400 450 325 70 125 450 40 90 90 Ohio 49,000 6,800 7,700 7,900 1,400 2,200 4,700 1,200 1,300 1,300 Ok9ahoma 14,000 1,800 1,900 2,500 400 550 1,400 500 425 450 Oregon 11,800 1,700 1,600 2,000 300 425 1,200 350 325 350 Pennsyfvania 59,000 8,800 10,000 8,600 1,700 2,500 5,300 1,600 1,500 1,500 Rhode Island 4,900 700 900 700 200 200 500 150 150 100 South Carolina 13,000 1,900 1,700 2,000 500 750 1,500 400 350 250 South Dakota 2,900 425 500 375 40 125 350 80 90 100 Tennessee 21,000 2,600 2,800 3,500 700 950 2,200 500 550 600 Texas 54,500 7,300 7,200 8,800 1,800 2,600 5,000 1,600 1,400 1,700 Utah 3,500 550 450 350 100 200 600 100 100 125 Vermont 2,300 350 375 350 80 150 250 50 50 90 Virginia 23,500 3,300 3,400 3,800 800 1,100 2,500 700 600 600 Washington 17,300 2,500 2,300 2,800 550 830 1,800 500 500 450 West Vitrginia 8,900 1,200 1,200 1,500 200 375 800 200 250 250 Wisconsin 20,200 3,100 3,200 2,700 450 900 2,300 400 550 650 Wyoming 1,300 225 200 204 30 30 150 50 30 30 United States 1,010,000 142,000 151,000 155,000 31,000 47,000 103,000 27,000 27,000 27,000 Puerto Rico 6,000 450 450 350 425 750 400 500 100 175 oDoes not include carcinoma in situ or non-melanoma skin cancer. iheae estimates are offered as a rough guide and shou{d not be regarded as definitive. They are calculated according to the distribution of estimated 1989 canc cv deaths by state. Especially note that year-to-year changes may only represent improvements in the basic da:ta. 6

CANCER DEATHS-1989 Estimated Cancer Deaths for Al! Sites Plus Major Sites, by State-1989 ALL SITES MA JOR SITES _ Number Death Rate Skin of per 100,000 Female Colon & Mela- STATE Deaths Population` Breast Rectum Lung Oral Uterus Prostate noma Pancreas Leukemia Alabama 8,900 214 700 950 2,600 125 200 550 100 425 300 Alaska, 500 221 30 50 150 10 10 20 10 25 10 Arizxta 6,500 180 550 700 1,800 100 50 400 70 300 225 Arkansas 5,600 190 350 600 1,800 60 100 300 80 300 225 Calitornia 50,000 181 4,400 5,500 14,100 950 900 2,600 700 2,500 1,800 Colorado 4,700 141 450 600 1,200 60 70 275 80 250 200 Connecticut 7,200 217 650 950 1,800 125 125 375 90 400 275 Delaware 1,400 250 125 175 450 10 20 60 10 60 50 Dist. of Columbia 1,700 264 175 175 400 70 60 125 10 100 50 Florida 32,500 182 2,500 4,100 9,800 600 490 2,100 400 1,600 1,000 Ge)r;;ia 11,200 202 850 1,200 3,400 250 250 700 150 500 400 Hawaii 1,700 191 100 175 375 50 20 80 20 80 50 tdafio 1,800 158 150 175 400 20 25 125 30 100 80 Illinois 24,000 201 2,100 3,100 6,600 450 600 1,300 200 1,300 900 Indiana 11,500 217 950 1,500 3,500 175 300 600 125 550 425 Iowa 6,400 190 550 850 1,600 125 125 400 70 350 300 iCamas 4,900 171 425 650 1,300 90 100 350 50 275 225 Kenv.ky 8,400 207 650 1,000 2,800 125 175 425 80 375 300 Louisiana 8,800 212 650 900 2,800 150 175 475 80 450 300' Maine 2,800 199 225 400 800 40 60 175 20 150 90 Maryland 9,600 244 800 1,200 2,700 175 175 500 125 425 300 Massachusetts 14,100 220 1,500 1,900 3,500 250 275 750 175 650 475 Michigan 18,600 226 1,600 2,100 5,300 275 400 1,000 200 900 650 Minnesota 8,100 181 700 1,100 2,000 125 125 550 90 450 350 Misshsippi 5,100 186 325 .500 1,700 80 100 350 60 300 225 Misscuri 11,800 196 950 1,500 3,400 175 250 550 125 550 450 Montana 1,600 186 150 175 375 20 30 100 20 100 70 Nebraska 3,300 173 300 450 800 40 70 200 40 200 175 Nevada 2,100 216 150 200 600 40 20 100 30 80 40 New IHampshire 2,100 197 200 250 550 30 40 90 30 100 70 New JetSey 18,100 230 1,600 2,500 4,900 325 375 950 225 900 550 New wlexico 2,300 168 200 250 500 30 40 150 20 125 70 New 1'ork 38,500 200 3,800 5,400 9,800 750 950 2,200 475 2,100 1,400 North Carolina 12,200 203 1,000 1,300 3,700 225 275 750 175 550 425 North Dakota 1,300 171 125 175 300 20 20 125 10 90 60 Ohio 24,000 227 2,100 3,100 7,300 400 600 1,300 250 1,200 850 Oklahoma 7,000 163 550 800 2,300 100 100 375 100 325 275 Oregon 5,900 198 500 650 1,800 100 75 350 70 300 225 Pennsylvania 29,500 221 2,600 4,000 7,800 475 700 1,500 350 1,400 1,000 Rhode Island 2,500 227 250 350 650 60 40 125 30 125 70 South Carolina 6,500 209 550 650 1,900 125 125 425 90 325 175 South Dakota 1,500 180 125 200 325 10 30 125 20 100 80 Tennessee, 10,400 202 800 1,100 3,300 200 200 600 125 500 375 Texas 27,000 155 2,200 2,900 8,100 475 500 1,400 350 1,300 1,000 Utah 1,800 118 175 175 275 20 30 175 30 90 80 Vermont 1,200 196 100 150 275 20 30 70 10 60 50 Virginia 11,700 219 950 1,400 3,500 225 225 650 150 500 375 Washington 8,600 181 750 900 2,600 150 150 500 100 425 300 West Virginia 4,400 202 350 500 1,400 60 100 225 50 200 175 Wisconsin 10,000 197 950 1,300 2,500 125 175 650 90 500 425 Wyomin;; 700 128 70 75 175 10 10 30 10 40 30 United States 502 000 204 43,000 61,000 142,000 8,700 10,000 28,500 6,000 25,000 18,000 Puerto Rico 3,500 150 200 250 400 175 150 300 400 80 150 `Adjusted to the age distribution of the 1970 US. Census Population. 7

C A N C E R F A C T S A N D F i G U R E S 1 9 8 9 PREVENTION PRIMARY PRE VENTION REFERS TO STEPS THAT MIGHT BE TAKEN TO AVOID THOSE FACTORS THAT MIGHT LEAD TO THE DEVELOPMENT OF CANCER. SMOKING Cigarette smoking is responsible for 85% of lung cancer cases among men and 75% among women-about 83% overall. Smoking accounts for about 30% of all cancer deaths. Those who smoke two or more packs of cigarettes a day have lung cancer mortality rates 15 to 25 times greater than nonsmokers. SUNLIGHT Almost all of the more than 500,000 cases of non-melanoma skin cancer developed each year in the U.S. are considered to be sun-related. Recent epidemiological evidence shows that sun exposure is a major factor in the development of melanoma and that the incidence increases for those living near the equator. (See Selected Cancer Sites: Skin Cancer) ALCOHOL Oral cancer and cancers of the larynx, throat, esophagus, and• liver occur more frequently among heavy drinkers of alcohol. (See Selected Cancer Sites: Oral Cancer) SMOKELESS Increased risk factor for cancers of the mouth, larynx, throat, and esophagus. Highly habit TOBACCO forming. (See Selected Cancer Sites: Lung Cancer and Oral Cancer) ESTROGEN For mature women, certain risks associated with estrogen treatment to control menopausal symptoms, including an increased risk of endometrial cancer. Use of estrogen by menopausal women needs careful discussion by the woman and her physician. (See Selected Cancer Sites: Uterine Cancer) RADIATION Excessive exposure to radiation can increase cancer risk. Most medical X rays are adjusted to deliver the lowest dose possible without sacrificing image quality. The ACS believes there is a potential problem of radon in the home. If levels are found to be too high, remedial actions should be taken. OCCUPATIONAL Exposure to a number of industrial agents (nickel, chromate, asbestos, vinyl chloride, etc.) I-iAZARDS increases risk. Risk factor greatly increased when combined with smoking. NUTRITION Risk for colon, breast and uterine cancers increases for obese people. High-fat diet may be a factor in the development of certain cancers such as breast, colon and prostate. High- fiber foods may help reduce risk of colon cancer, and can be a wholesome substitute for high-fat diets. Foods rich in vitamins A and C may help lower risk for cancers of larynx, esophagus, stomach and lung. Eating cruciferous vegetables may help protect against certain cancers. Salt-cured, smoked and nitrite-cured foods have been linked to esophageal and stomach cancer. The heavy use of alcohol, especially when accompanied by cigarette smoking or chewing tobacco, increases risk of cancers of the mouth, larynx, throat, esophagus, and liver. (See above) SECONDARY PREVENTION REFERS TO STEPS TO BE TAKEN TO DIAGNOSE A CANCER OR PRECURSOR AS EARLY AS POSSIBLE AFTER IT HAS DEVELOPED. COLORECTAL The ACS recommends three tests for the early detection of colon and rectum cancer in people TESTS without symptoms. The digital rectal examination, performed by a physician during an office visit, should be performed every year after the age of 40; the stool blood test is recommended every year after 50; and the proctosigmoidoscopy examination should be carried out every 3 to 5 years after the age of 50 following two annual exams with negative results. (See Selected Cancer Sites: Colon and Rectum Cancer) PAP TEST For cervical cancer, women who are or have been sexually active, or have reached age 18 years, should have an annual Pap test and pelvic examination. After a woman has had three or more consecutive satisfactory normal annual examinations, the Pap test may be performed less frequently at the discretion of her physician. BREAST CANCER DETECTION The ACS recommends the monthly practice of breast self-examination (BSE) by women 20 years and older as a routine good health habit. Physical examination of the breast should be done every three years from ages 20-40 and then every year. The ACS recommends a mammogram every year for asymptomatic women age 50 and over, and a baseline mammo- gram between ages 35 and 39. Women 40 to 49 should have mammography every 1-2 years, depending on physical and mammographic findings. 28

C A N C E R F A C T S A N D F I G U R E S 1 98 9 SELECTED CANCER SITES BLADDER CANCER Incidence: An estimated 47,000 new cases of bladder cancer in 1989; 34,500 in males, 12,500 in females. Blad- der c:ancers account for 7% of the new cancer cases diagnosed each year in men and 3% in women. Bladder cancer is the 5th most common form of cancer in males and 10th most common form of cancer in females in this country. McM:ality: An estimated"10,200 deaths in 1989 from bladder cancers, the 8th leading cause of cancer deaths in males and 14th in females. Warning Signals: Blood in the urine. Usually asso- ciated with increased frequency of urination. Risk Factors: Smoking is the greatest risk factor in bladder cancer, with smokers experiencing twice the risk of nonsmokers. Smoking is estimated to be respon- sible ;Eor about 49% of the bladder cancers among men and 10% among women. Overall, the incidence rate of bladder cancer is 4 times as great among men as women, and higher in whites than in blacks. People living in urban areas, and dye, rubber and leather workers also are at higher risk. Coffee and artificial sweeteners have been found to increase cancer risk in a few studies but most studies have not found an increased risk. Diagnosis: Diagnosis of bladder cancer is achieved by examination of the bladder wall with a cystoscope, a slender tube fitted with a lens and light that can be inserted into the tract through the urethra. Treatment: Surgery, alone or in combination with other treatments, is used in 92% of cases. Survival: The 5-year survival rate for bladder cancer is 88% when detected in an early stage. For those can- cers more advanced, the survival rate drops to 41%. SKIN CANCER Incidence: Over 500,000 cases a year, the vast major- ity of which are highly curable basal or squamous cell cance!rs. They are more common among individuals with lightly pigmented skin, living at latitudes near the equator. The most serious skin cancer is malignant melanoma, which strikes about 27,000 persons each year. The incidence of melanoma is increasing at the rate of ,3.4% per year. Mortality: An estimated 8,200 deaths this year, 6,000 from malignant melanoma, and 2,200 due to other skin cancers. Warning Signals: Any unusual skin condition, especially a change in the size or color of a mole or other darkly pigmented growth or spot. Scaliness, oozing, bleeding or the appearance of a bump or nodule, the spread of pigment beyond the border, a change in sen sation, itchiness, tenderness -or pain are all warning signs of melanoma. Risk Factors: Excessive exposure to the sun; fair compaexion; occupational exposure to coal tar, pitch, creosote, arsenic compounds or radium. Among blacks, because of heavy skin pigmentation, skin cancer is negligible. One study has found that severe sunburn in childhood carries with it an excessive risk of mel- anoma in later life. Prevention: Avoid the sun between 10 a.m. and 3 p.m. when ultraviolet rays are strongest, and use protective clothing. Use one of a number of sunscreen preparations, especially those containing such ingre- dients as PABA (para-aminobenzoic acid). They come in varying strengths, ranging from those that permit graduzil tanning to those allowing practically no tanning at all. Children, in particular, should be protected from traumatiic sunburns. Early Detection: Early detection is critical. Recog- nition of changes in or the appearance of new skin growths is the best way to find early skin cancer. Basal and squamous cell skin cancers often take the form of a pale, waxlike, pearly nodule, or a red scaly, sharply outlined patch. A sudden or continuous change in a mole's appearance should be checked by a physician. Melanomas often start as small, mole-like growths that increase in size, change color, become ulcerated and bleed easily from a slight injury. There is a simple ABCD rule that will help individuals remember the warning signs of melanoma: A is for asymmetry. One half of the mole does not match the other half. B is for border irregularity. The edges are ragged, notched or blurred. C is for color. The pigmentation is not uniform. D is for diameter greater than 6 millimeters. Any sudden or continuing increase in size should be of special concern. Adults should practice skin self-examination once a month. Treatment: There are four methods of treatment: surgery (used in 90% of cases), radiation therapy, electrodesiccation (tissue destruction by heat), or cryosurgery (tissue destruction by freezing) for early skin cancer. For malignant melanoma, adequate surgical excision of the primary growth is indicated. Nearby lymph nodes may be removed. The microscopic examination of all suspicious moles is essential. Advanced cases of melanoma are treated on an individual basis. Survival: For basal cell and squamous cell cancers, cure is highly likely with early detection and treatment. Malignant melanoma can spread to other parts of the body quickly. However, when detected in its earliest stages, with proper treatment, it is highly curable. The overall 5-year survival rate for white patients with malignant melanoma is 80% compared with 95% for patients with other kinds of skin cancer. The 5-year survival rate for localized malignant melanoma is 89%; however, the survival rate, once melanoma has spread, is 39%. 13 202554591202

C A N C E R F A C T 5 A N D F 1 G U R E 5 1 9 8 9 CANCER BY AGE AND RACE Neurobtastoma can appear anywhere but usually in the abdomen, where a swell.irtg occurs. Rhabdomiiosarcoma, the most common soft tissue sarcoma, can occur in the head and neck area, genito- urinary area, trunk and extremities. Brain Czncers in early stages may cause headaches, blurred or double vision, dizziness, difficulty in walking or handling objects, and nausea. Lymphomcrs, and Hodgkin's Disease are cancers that involve the lymph nodes, and also may invade bone marrow and other organs. They may cause swelling of lymph nodes in the neck, armpit or groin. Other symptoms nnay include general weakness and possibly fever. RetinoblasEoma, or an eye cancer, usually occurs in' children under the age of four. When detected early, cure is possible with appropriate treatment. Wilms' Tumor, a kidney cancer, may be recognized by a swelling or lump in the abdomen. Treatment: Childhood cancers can be treated by a combination of therapies, coordinated by a team of experts. They include oncologic physicians, pediatric nurses, social workers, psychologists and others who assist children and their families. Survival: Five-year survival rates vary considerably, depending on the site. Among those for white children: bone cancer, 48%; neuroblastoma, 56%; brain and cen- tral nervous system, 56%; Wilms' tumor (kidney), 82%; and Hodgkin's disease, 91%. (Data for black children is insufficient.) 'Figures for carcer incidence are from the National Cancer Institute National Surveys, 1947, and the NCI SEER Program, 1973-1985; those for cancer mortality are from the National Center for Health Statistics, 1953-55 to 1983-85. TRENDS IN SURVIVAL BY SITE OF CANCER, BY RACE Cases Diagnosed in 1960-63, 1970-73, 1974-76, 1977-78, 1979-84 11 _ WHITE BLACK RELATIVE 5-YEAR SURVIVAL RELATIVE 5-YEAR SURVIVAL SITE 1960-631 1970-731 1974-762 1977-782 1979-84= 1960-63' 1970-731 1974-76z 1977-782 1979-842 All Sites 39% 43% 50% 50% 50% 27% 31% 38% 38% 37% Oral Cavity & Pharynx 45 43 54 53 54 - - 35 35 31 Esophagus 4 4 5 6 7 1 4 4 2 5 Stomach 11 13 14 15 16' 8 13 15 16 17 Colon 43 49 50 52 54' 34 37 45 44 49 Rectum 38 45 48 50 52' 27 30 40 40 34 Liver 2 3 4 3 3 - - 1 1 5 Pancreas 1 2 3 2 3 1 2 2 3 5 Larynx 53 62 66 69 66 - - 58 59 55 Lung & Bronchus 8 10 12 13 13' 5 7 11 10 11 Melanoma of Skin 60 68 78 81 80' - - 62## - 61iF Breast (females) 63 68 74 75 75' 46 51 62 62 62 Cervix Uteri 58 64 69 69 67 47 61 61 63 59 Corpus Uteri 73 81 89 87 83• 31 44 61 58 52* Ovary 32 36 36 37 37' 32 32 41 40 36 Prostate Gland 50 63 67 70 73' 35 55 56 64 60' Testis 63 72 78 86 91' - - 77ii - 82a Urinary Bladder 53 61 73 75 77' 24 36 47 53 57' Kidney & Renal Pelvis 37 46 51 50 51 38 44 49 54 53 Brain & Nervous System 18 20 22 23 23 19 19 27 24 31 Thyroid Gland 83 86 92 92 93 - - 88 92 95 Hodgkin's Disease 40 67 71 73 74' - - 67M 79# 69 Non-Hodgkin's iLymphoma 31 41 47 48 49• - - 47 46 49 Multiple Myeloma 12 19 24 24 24 - - 28 30 29 Leukemia 14 22 34 37 32 - - 30 31 27 Source: Surveillance snd Operations Research Branch, National Cancer Institute. ' Rates are based on IEnd Results Group data from a series of hospital registries and one population-based registry. I Rates are from the SEER Program. They are based on data from population-based registries in Connecticut, New Mexico, Utah, Iowa, Hawaii, Atlanta, Detroit, Seattle-Puget Souna' and San Francisco-Oakland. Rates are based on follow-up of patients through 1985. ° The difference in rates between 1974-76 and 1979-84 is statistically significant (p <.05). , fl The standard error of the survival rate is between 5 and 10 percentage points. i» The standard error of the survival rate is greater than 10 percentage points. -Valid survival rate could not be calculated. 17

Letter to the New England Journal of Medicine [September, 1990] To the Editors: In their January 11 article Needleman et al. t report strikingly large effects of low lead levels on several late adolescence outcomes. For example, an estimated 7.4-fold increased odds of school failure was attributed to childhood lead dentin levels above 20 ppm. Such massive effects sizes contrast sharply with results of other studies relating low lead level to earlier developmental outcomes 2-4. The authors argue that the estimated effects represent causal relationships because their analysis controlled for ten socio- demographic covariates. This conclusion of causality may be premature, however, because the covariate set did not include measures of the quality of child care (i.e., parental responsitivity, involvement with the child, provision of books, suitable playthings, etc.), a primary confounder in previous studies of develop.,nental lead effects. Thus the reported lead effects may be partly due to spuriou,s association induced by variations in the caretaking environment. Indices of child care quality such as the HOME S and the CLL 6 have repeatedly been found to be strongly related to lead-level in poor and working class children2~}~~e Quality of child care is also strongly associated with developmental outcome 9, including school performance through adolescence 10 . These confounding effects are conceptually distinct from and only partly accounted for empirically by socio-demographic variables such as maternal IQ and parental education'1 , which were included as covariates by Needleman et al. The fact that none of the reported lead effects were attenuated by inclusion of their covariates, as is usually the case in observational studies of low lead levels, indicates that confounders such as child care may not have been fully controlled. On another matter, the present report is a follow-up of a 1979 report 12 which troubled reviewers 13 , in part, because many cases were excluded after testing. In a written response to the review 14, Needleman reported data indicating that a key IQ analysis was substantially affected by 16 of the

C A N C E R F A C T S A N D F 1 G EJ R E S 1 9 8 9 RESEARCH THE ACS AND RESEARCH The American Cancer Society is the largest source of private cancer research funds in the United States, second only to the National Cancer Institute, an agency of the Federal government. The Society's overall investment in research each year has grown steadily from $1 million in 1946 to over $86 million* today. This sum represents nearly a third of the total ACS budget. To date, the Society has invested close to $1 billion in cancer research. The research program focuses primarily on inves- tigator-initiated projects, rather than directed research undertaken on a contract basis. With the exception of staff and facilities to carry out its epidemiological studies, the AC S neither hires staff researchers nor operates its own laboratories. This gives the Society the freedom to place its grants where the most innovative and promising ideas are being explored. A key factor in the role of the Society in cancer research is providing qualified scientists with altema- tive funding sources to carry out their work. The Society believes it can make the most effective use of its research funds by supporting investigators working in established medica,l and other scientific institutions across the country. In this way there is a minimum of overhead and a nnaximum of flexibility to make sure that research money has the highest probability of yielding results tha- will benefit people. Applications for ACS grants are put through a rigorous process of evaluation, beginning with careful study by the appropriate one of 12 scientific review committees and then by two additional groups of experts. They must be given final approval by the National Board of Directors. Kinds of Grants The Society's research program is diverse in concept and recipients. It provides support both for established scientists and those starting out on their own inde- pendent research. It funds postdoctoral training for promising young investigators and stimulates new ideas in cancer research among those working in uni- versities, institutes and teaching hospitals. Overall, the program offers five types of grants: (1) Research and Cl.'mical Investigation Grants to finance investigator-initiated!d research; (2) Institutional Re- search Grants to uni!.versities, institutes and hospitals to support pilot studies and the work of young investigators in cancer, (3) Research Personnel Grants to outstanding scientists and fellows specializing or planning to specialize in cancer research; (4) Research Development Program Grants to provide rapid funding for priority projects; and (5) Special Institutional Grants for Cancer Cause and Prevention Research to provide longer term fundin g for interdisciplinary projects for which support is not readily available through the Society's other programs. Research Professcrships. The Research Professorship program, unique inn the field, has been in existence since 1957. The Society supports 25 of the nation's most gifted scientists for long periods of time, until their retirement. These are people devoting their lives' work to cancer research. Freed of major administrative responsibilities and other restrictions, they can con- centrate on their fields of scientific investigation. Clinical Research Professorships. This novel and unique program is a new initiative of the Society to provide support for clinicians and scientists who are able to facilitate advances in clinical cancer research by bridging the gap between basic science and clinical medicine. Three awards have been made since the inception of the program in 1987. Physicians' Research Training Fellowships. Unique in the research world, this type of Research Personnel Grant was inaugurated in 1981 because of a dearth of MD's in the research field. It provides an opportunity for physicians to take three years from their medical careers to train as researchers. Research Development Program. Established to identify and provide rapid funding for high priority projects, approved applications can be funded in less than three months. This compares with the 10 to 18 months required by the Federal government before a new application can be funded. The kinds of research projects eligible under the Research Development Program include: (1) unique research opportunities which cannot wait for the normal lengthy funding procedures; (2) unanticipated needs relating to research already under way; (3) program coordination, especially that involving clinical trials and the dissemination of research results to community hospitals; and (4) program integration between the American Cancer Society and other health organizations. All applications are evaluated for merit, qualifications and productivity of the investigator, relevance, need for rapid funding, and probability of the project's eventual contribution to cancer control. More than $13 million has been appropriated so far to the Research Development Program, over half of which has been for interferon research. Interferon Research. Interferons, a group of natural body proteins, were discovered as antiviral agents, and later found to have some anticancer activity. In 1978, the Society invested an unprecedented $2 million to purchase interferon for clinical trials. At the time, interferons were extremely scarce and expensive, since they were obtained from human blood cells. Interferons work dramatically to improve certain diseases such as hairy cell leukemia and some lym- phomas and papillomas. In these the frequency of improvement approaches 90%. In other diseases, such as kidney cancer and Kaposi's sarcoma, there . are dramatic responses, but they are far less frequent - on the order of 10-30%; in lung and colon cancer, interferon rarely causes improvement. The thrust of current research with interferons is to attempt to improve their effectiveness by combining them with *Subject to audit 26

Cancer Investigation, 7(3), 267-276 (1989) CONTROVERSIES IN BASIC SCIENCE The Question of Thresholds for Radiation and Chemical Carcinogenesis Arthur C. Upton Institute of Environmental Medicine New York University Medical Center New York, New York INTRODUCTION Selection of the dose-incidence model that is appro- priate for predicting the risks of cancer from low-level exposure to a given carcinogen is among the most con- tentious issues in public health. Although the existence of a threshold in the dose-effect relationship is well documented for many, if not most, types of toxicological effects, the exi stence of a threshold for the mutagenic effects of ionizing radiation (1-3) and of certain chemicals (4,5) has been questioned since the middle of the century. More recently, the existence of a threshold for car- cinogenic effecm also has been seriously questioned, since carcinogenesis may, likewise, be envisionei to result from effects on indivichlat cells rather than groups of cells (6-8). Because in principle it is not possible to prove or disprove the existence of a threshold for carcinogenesis, the argument for or against the threshold hypothesis must be based on theoretical as well as empirical evidence (7,8). Some of the cogent data and concepts are surveyed in the following. Copyright © 1989 by Marcel Dekker, Inc. 267 BIOLOGY OF CARCINOGENESIS Monoclonal, Multicausal, Multistage Nature of Cancer The evidence that cancer usually originates from a single transformed cell (9-11) implies that appropriate damage to one cell alone may suffice to increase the prob- ability of neoplasia in a suitably susceptible individual. A single alteration, however, apparently does not suffice to convert a normal cell into a cancer cell. On the con- trary, cancer typically appears to evolve through a suc- cession of stages; for example, initiation, promotion, and progression (12,13). The mechanism of initiation remains to be established, but some type of mutational change is implicated by evidence that: (i) the initiating event is relatively prompt and irreversible (14,15); (ii) most ultimate carcinogens are mutagens (16); (iii) the frequency of cell transforma- tion that is induced by a given carcinogen is usually max- imal if exposure to the agent occurs just before or during 2025545949

Methodologies in Respiratory Occupational Surveillance 41 0 ~ O TS 5 75 me DuST IIISTILLED/ t00p DODt WEIGNT FIGURE 3. DDse-response curve for lambda assay I d after exposure to iron oxide, aluminum oxide, or a-quanz. The Witcoxon rank-sum test was used to compare c xperimenWs and saline oniy controls. p<0.01 for 0.75 and 3.75 mg a-quartr.,0.75mg afuminumoxide:p c0.05 for0.15 mg'tron oxide. Values represent mean ± standard errors. Udrxed from Be.c@. B. D., Brain, J. D., and Bohannon, D. E.. Exp. L1+nR. Rrs., 2, 289, 198 1. With permission.) highest 4 d afterexposure to a-quartz, although after2 weeks they still had not approached control levels.12 A somewhat different pattern was observed for lactate dehydrogenase (LDH) in lavage fluid. This is a cytoplasmic enzyme involved in energy metabolism; its extracellular re- lease is assoc:i,aed with cell injury ordeath. LDH levels in lung lavage fluid were highest I d after exposure to both iron oxide and a-quartz.:[n time, LDH levels declined signficantly in the quartz-exposed animals and only slightly in the iron oxide- exposed animals. Nevertheless, the levels in the quartz ani- mals remained higher than those in the iron oxide-exposed animals at al'li times. These effects were observed at relatively low levels of quartz compared to levels used in animal models of chronic silicosis. Application of this system to dusts produced by the erup- tion of Mt. St. Helens volcanic ash suggested that volcanic ash has low to moderate toxicity20 We concluded that adverse health effects in human populations are unlikely except with high or prolonged exposure. Surfactant levels in BAIL in rats after quartz and Mt. St. Helens volcanic ash exposure have been studied by Martin and co-workers.21 Quartz causes a prolonged elevation in PMN numbers and surfactant levels. The effects were much less marked with volcanic ash than with quartz. These observations are consistent with histopa- thological studi es of lungs of exposed animals which demon- strated much greater fibrogenicity of a-quartz than of vol- canic ash. These studies show the usefulness of BAL in providing a rapid evaluation of the toxicity of poorly charac- terized samples. Useful results can be obtained even when chemical analyses of epidemiological studies are not avail- able for toxicity estimates. IDENTIFYING SOURCES OF DAMAGE INDICATORS: LDH ISOENZYMES We are searching for other ways of making the assay more interpretable. As discussed earlier, LDH is released from cells in response to toxic particles. However, if LDH is recovered in the cell-free supernatant of lung lavage fluid, where does it come from? Is the source inflammatory cells (macrophages or PMNs), serum, epithelial cells, or endothelial cells? Beck et al,u have used isoenzyme analysis to infer the sources of LDH. To differentiate among types of injury, we monitored changes in LDH isoenzyme patterns in BAL after a range of injuries: a-quartz, hyperoxia, the detergent Triton X- 100, and SO2. The LDH isoenzyme patterns in BAL were evaluated and compared with patterns from hamster lung homogenates, red blood cells, macrophages,PMNs, type II cells, and serum. The isoenzyme pattern in BAL from quartz-exposed animals resembled that of the PMNs and macrophages, suggesting phagocytic cell death. In contrast, BAL from Triton X- 104-treated animals had an isoenzyme pattern similar to that of the lung homogenate and red blood cells. Exposure to 100% 0 2 for 4 d produced an isoenzyme pattern similar to serum, an observation consistent with the demonstrated effects of Oz on the capillary endothelium. Figure 4 presents graphically the percentage of each LD isoenzyme from serum or from lung lavage fluid of Syrian golden hamsters exposed to 100% 02 for 96 h. The distribu- tion of the five LD isoenzymes is similar and consistent with the hypothesis that oxygen toxicity caused damage to the air- blood barrier. Serum LD and other serum proteins leaked into alveolar spaces and were subsequently recovered by lavage. In Figure 5 the LD pattern is shown for: (1) supernatant from BAL recovered from hamsters exposed to iron oxide aerosol and (2) hamster peritoneal PMNs. The LD patterns shown in Figure 5 are markedly different from those seen in Figure 4. For example, there is little LDI (<3%), but a great deal ofLDS (-60%). The similarity in pattern suggests that the 20 wi 0 ~ J ~ I- LDI LD2 LO 3 LD4 LD5 • FIGURE 4. Comparison of LD isoenzyme patterns from hamster serum and from lung lavage fluid of hamsters exposed to 100% 07 for 96 h. (Adapted from Beck, B. D.. Gerson, B., Feldman. H. A.. and Brain. J. D.. Toxicol. Appl. Phornwco(..71, 59, 1983. With permission.) 2@Htr,-M~,~:~,~.~®~® ®~ ~,

C A N C E R F A C T S A N D F I G U R E S 1 9 8 9 PREVENTION CANCER-RELATED CHECKUP GUIDELINES Guidelines for the early detection of cancer in people without symptoms are recommended by the American Cancer So- iety as follows: A cancer-related checkup: • every 3 years for those 20-40 years of age. • every year for those 40 and over. The Society advises that you talk with your doctor. Ask how rhe guidelines apply to you. The checkup should always include health counseling (such as tips on quitting smoking) and examinations for cancer of the thyroid., testes, prostate, mouth, ovaries, skin and lymph nodes. In particular. • Ages 20-40-For breast cancer, an examination by a physician every three years, a self-exam every month, and one baseline breast X ray between the ages of 35 and 39. For i:ervical cancer, women who are or have been sexually active, or have reached age 18, should have an annual Pap test and pelvic examination. After a woman has had three or more consecutive satisfactory normal annual examinations, the Pap test may be per- formed less frequently at the discretion of her physician. • Ages 40 and over-For breast cancer, a professional exam every year, a self-exam every month and a breast X ray every 1-2 years for those 40-49; every year for those 50 and over. For cervical cancer, women who are or have been sexually active, or have reached age 18 years, should have an annual Pap test and pelvic examination. After a women has had three or more consecutive satisfactory normal annual examinations, the Pap test may be performed less frequently at the discretion of her physician. For women at risk, an endometrial tissue sample at menopause should be taken. For colon and rectum cancer, a digital rectal exam every year after 40, and a stool blood test every year after 50 as well as a procto exam every 3-5 years after two initial negative tests one year apart. Some people are at higher risk for certain cancers and may need tests more frequently. (See pp. 9-14 for high risk factors.) COLORECTAL CANCER: EARLY DETECTION IS THE KEY When cancer of the colon and rectum is found and treated in an early, localized stage, the 5-year survival rate is 90% for colon cancer and 80% for rectal cancer. However, survival figures drop to 40% and 31%, re- spectively, after the cancer has started to spread to other parts of the body. Because colorectal cancer develops over a period of time, detection of the disease is possible long before symptoms appear. Early detection of small cancers and polyps reduces the likelihood of major surgery and the need for a coi~.ostomy-an abdominal opening created for the elimination of wastes. In fact, permanent colos- tomies are rare in cases of colon cancer, and are neces- sary in only l:i% of rectal cancer cases. Colorectal cancer is second only to lung cancer in terms of incidence. Currently, about 151,000 new cases develop each year; about 61,000 people die from the disease annually. The incidence of colorectal cancer tends to increase with age, starting at 40 years. More than 94% of all cases occur after the age of 50. Colorectal cancer occurs about equally in both sexes. Anyone with a personal or family history of colorectal cancer, polyps in the colon. or inflammatory bowel disease, is at particularly high risk for the disease and should be examined care ~ully. Evidence suggests that bowel cancer may be linked to a diet high in fat and/or low in fiber content. Projected `.i-year survival rates for colorectal cancer show that earl'.y detection saves lives. Currently, the 5-year survival'. rate is estimated at 55%. With the use of early detection techniques, such as the digital rectal exam, the stool blood test and sigmoidoscopy, and with appropriate ma.nagement, the survival rate for patients with colorectal cancer could be increased from 55% to 85%. This means that, over a period of time, 125,000 lives, versus the current 80,000, could be saved each year. It is recommended that the following procedures, all part of a cancer-related checkup, be performed at designated intervals: • A digital rectal examination every year after age 40. • A stool blood test every year after age 50. • A procto every three to five years after the age of 50, following two annual negative examinations. These guidelines apply only to people without symp- toms. Persons with rectal bleeding, cramping abdom- inal pain, or a change in bowel habits should see their physicians immediately. A 1987 study of men and women age 40 and over, conducted for the Society by the Gallup Organization, revealed a number of important findings concerning Americans' attitudes toward detection measures for colorectal cancer. There has been some increase in public awareness of the 3 tests recommended to detect the disease, but there is much room for improvement. The study found, for instance, that the percentage of Americans who ever had a digital rectal examination increased slightly since 1983, from 51% to 56%. Likewise, the percentage of Americans who ever had a stool blood test rose, from 28% in 1983 to 40% in 1987. And while the percentage of men and. women 50 and over .who ever had a proctoscopic examination rose from 31% in 1983 to 42% in 1987, 60% of Americans who should have the examination (according to the ACS guidelines) have not had it. The survey also showed that 24% of those individuals in the 40-plus age group have ever asked their doctor to examine their colon or rectum. And of this group, more than half did so only because something was bothering them. On the promising side, the survey showed that almost 50% of all Americans would be interested in learning more about this form of cancer. 19

C A N C E R F A C T S A N D F 1 G U R f S 1 9 8 9 RESEARCH a two-part program of environmental cancer research. This involves (1) Cancer Prevention Study II, an epi- demiologic study to examine the relationship of en- vironment and lUestyle to cancer development; and (2) support of extrarnural cancer cause and prevention research projects. The American Cancer Society's Cancer Prevention Study 66 One of the largest research studies ever carried out in the United States was launched in 1982. Cancer Pre- vention Study II,,a long-term prospective study, is ex- amining the habi:s and exposures of more than one million Americans to learn how lifestyles and environ- mental factors affeci; the development of cancer. Modeled after the first ACS Cancer Prevention Study (1959-72), CPS-II is s~imilar in method but wider in scope and involves more participants. Over 77,000 volunteers enrolled 1.2 million men and women in the study. These volunteer researchers distributed a four-page confidential questionnaire to participants who were asked about their exposures to certain environmental conditions, their history of disease and their liiestyles. The questionnaires were designed to elicit more than 500 pieces of information each, which were computerized for statistical analysis. Many of the questions focus on health issues of current concern. These include risks of certain .drugs, foods and various occupational exposures; low-tar and nicotine cigarettes; consumer products; long-term exposure to low-level radiation; and the health effects associated with air and water pollution. For a period of six years, the volunteers will keep track of the status aild whereabouts of study partic- ipants. Various suspected relationships will be tested by comparing mortality rates of differently exposed groups. The goal of the study is to identify those factors that increase a person's chances of developing cancer, those that carry little or no risk, and those that actually may help prevent cancer. So far, five papers have been published from the analyses of data on the original questionnaires. One showed massive changes in American smoking habits compared to 23 years earlier in CPS I. Among men, 24% smoked, half as many as in the earlier study. More than twice as many had quit cigarette smoking. Among women, the percent who had ever smoked rose 10%, but the percent of ex-smokers quadrupled. More than one-third of male smokers and one-half of female smokers smoked brands with less than 12 mg. of tar. Another paper from CPS II showed that smoking in physicians is now down to 16%, about 14% in dentists and 23% in nurses. A third paper showed that a greater percentage of women who used artificial sweeteners gained weight over a one-year period than nonusers. An additional five papers have been completed and submitted for publication. Another paper shows that death rates from all causes were 81% higher in obese and underweight people than those of average weight and that degree of exercise was negatively correlated with cancer death rates. Since the first study, new factors in our environment have been identified that may be related to cancer. The Society decided to initiate a second study to respond to the concerns of the public and scientific community -about suspected carcinogens. Without the use of ACS volunteers, the cost of carrying out CPS II would total more than $100 million. With volunteers to collect the data, the study is esti- mated to cost only about $9 million to complete. CANCER'S SEVEN WARNING SIGNALS 1. Change in bowel or bladder habits 2. A sore that does not heal 3. Unusual bleeding or discharge 4. Thickening or lump in breast or elsewhere 5. Indigestion or difficulty in swallowing 6. Obvious change in wart or mole 7. Nagging cough or hoarseness If you have a warning signal, see your doctor. 1 28

C A N C E R F A C T S A N D F{ G l1 R E S 1 9 8 9 RESEARCH more conventional chemotherapeutic drugs and by using mixtures of different interferons. Lzrl;e quantities of certain types of interferon can be produced now using the techniques of recombinant DNA. They are far cheaper and purer than the original human blood substances, and have recently been approved for marketing. The new technology is also extrem,ely valuable for producing complex drugs and chem.icals to benefit mankind. Furthermore, other substances, called biologic response modifiers, which use immune means to combat cancer are being developed at a rapid pace, including interleukin-21 lymphokine-activated killer cell (IL-2/LAK) reagents which appear to shrink such cancers as kidney and melarioma. Some of these reagents are very potent and quite toxic, and the search is on to find effective and safer ways to use them in patients. Research in the 80's In acid ition to ongoing interferon studies, ACS-funded researc:h ers continue to investigate broad areas of cancer research in this decade. For example, they are exploring: Genetic engineering. One method in this new technology, recombinant DNA, is already being used to produce interferon. It has among its potential uses the manufacture of powerful new drugs, correcting impaired immune systems, even modifying heredity by transplanting foreign genes. It is hoped that the process will yield other anticancer activities. Some that appear quite promising at the moment are tumor necrosis Eactor (TNF), interleukin-2, and certain bone marrow growth regulators. Monoclonal antibodies. Tailor-made, highly specific monoclonal antibodies can be produced that will preferentially recognize cancer cells, and thus be able to detect cancer early, when the disease is most curable, before clinical signs appear. Monoclonal antibodies already have been used to deliver drugs directly to tumors, 'killing them but sparing healthy tissue. Mecha nisms of carcinogenesis. Investigators are approaching these key questions from many angles. One model, as found in animals, shows that cancer in humans develops in a two-step process - initiation and promotion. Other questions include: Are there proto-oncogenes, normal genes serving as master switches for early tissue development, which induce normal cell',s to become cancerous later in life? If so, what turns them on? Can they be programmed to stay off? Do viruses, already known to cause cancer in animals, also cause cancer in humans, perhaps by activation of these proto-oncogenes? Conversely, a normal gene that appears to suppress cancer devei- opment has been isolated recently. Does this gene produce a substance that stops normal cells from dividing before they become cancerous? Many of these questions are now being answered. Chemoprevention. There is strong evidence that perhaps people can be protected from cancer by what they eat or drink, or by other substances or lifestyles that serve as defense mechanisms. Clues are being pursued by ACS researchers studying such agents as vitamin A; retinoids (synthetic forms of vitamin A); vitamin C; vitamin E; the chemical element selenium, found in the soil; and other naturally occurring sub- stances in brussels sprouts, cabbage, and certain other foodstuffs. This is a new and important area which needs further research so that recommendations can be developed on how people should change their life- styles to reduce their chances of getting cancer. Still other ACS investigators are looking for ways to detect cancer earlier by tracing a cell's biochemical markers. They are exploring evidence that the outbreak of the rare cancer, Kaposi's sarcoma, frequently found in patients with AIDS, is linked to a breakdown in the individual's immune system. And they are testing the hypothesis that certain chemicals enhance a tumor's responsiveness to radiation therapy. The -Financial Research Picture In fiscal 1988, the ACS made 818 grants to major institutions in this country and to scientists working here and abroad. The total amount, subject to audit, was over $83 million. This does not include some $3 million granted directly by ACS Divisions. The following table-covering the years 1985-1988 inclu- sive-lists the number of applications received, the total number of dollars required, and those actually funded by the ACS National Office.* Requested Funded Year Number Amount Number Amount 1985 2,096 $273,968,261 712 $63,703,751 1986 2,438 364,065,882 775 73,896,704 1987 2,385 368,645,879 810 77,516,363 1988 2,281 357,408,459 818 83,936,347 CANCER AND THE ENVIRONMENT Most cancer cases in the United States are believed to be environmentally related, that is, associated in some way with our physical surroundings, personal habits or liJ~estyles. Occupational hazards, although associated with only a small percentage of cancers, are under close surveil- lance. Virtually every suspected major chemical and other substance in the workplace presumed to be a health risk is under investigation. Each study can require years and hundreds of thousands of dollars to complete. Some environmental causes of cancer are well known. About 30% of all cancers are directly related to the use of tobacco, either alone or in conjunction with excessive consumption of alcohol. Other causes are harder to determine. Diet is sus- pected as an important element in cancer risk, some say as much as 35% of all cancer deaths. There is much research underway on the role diet and nutrition play in the development of cancer. To help identify environmental factors in human cancer, the American Cancer Society has undertaken 27 *Subject to audit

Thresholds for Radiation and Chemical Carcinogenesis P(d) 30-5 '+ 1Figure 8. Estimated risk of liver cancer, p(d), in relation to the dose of aflatoxin, d, as de:ermined with different dose-incidence models; i.e., OH, one-hit model; MS, multistage model; W, Weibull modelt MH multihit model; and MB, Mantel-Bryan (log-probit model). (From Ref. 56.) Table 3 Estimated Risk cf Cancer of the Human Urinary Bladder from Dail,v Ingestion of 0.12 g of Saccharin Method of transspecies scaling and of high- to low-dose extrapolation Lifetime cases per million exposed Rat dose adjusted to human dose by surface area rule Single-hit model 1,200 Multistage model (with quadratic term) 5 Multihit model 0.001 Mantel-Bryan probit model 450 Rat dose adjusted to human dose by mg/kg/day equivalence Single-hit modet 210 Multihit model 0.001 Mantel-Bryan prot» t model 21 Rat dose adjusted to human dose by mg/kg/lifetime equivalence Single-hit model 5,200 Multihit model 0.001 Mantel-Bryan probit model . 4,200 Source: From Ref. 60. 273 g:ner.)forrn all killing atterwrttes F (D) F(D) -(ap + at D+aZD2IeKP(-dt D-QZD2) Dose, D F(D)=aa+nZD2 WMrauc Dou,D F(D)-aa+etD tineu Oose, 0 F{D)=ca+crD.c~D~ hner auadytic Figure 9. Dose-response curves for four different mathematical models relating cancer incidence to radiation dose which were evaluated by the National Academy of Sciences Advisory Committee on the Biological Effects of Ionizing Radiation. (From Ref. 33.) - high dose rate - - - iow dose rate DOSE -~- Figure )[0. Diagrammatic representation of char.tcterisdc dose-response curves, relating the incidence of tumors in laboratory animals to the dose and dose rate of high-LET (-) radiation and low-LET (---) radiation. (Reproduced from Ref. 69.) at which a significantly increased incidence has been observed and the baseline (zero dose) incidence is general- 1y thought to overestimate the risk at low doses (8,56), and thus to provide an "upper limit" estimate of risk, with the lower limit of the range extending to zero.

C A N C E R F A C T 5 A N D F t G U R E 5 1 9 S 9 THE AMERICAN CANCER SOCIETY Reach to Recovery. This program; the largest of the Society :~s patient visitor programs, addresses the many needs oi;~ women who have or have had breast cancer. Carefully selected and trained volunteer visitors provide support and information, with the approval of the ar:tending physician. The program is designed to help women meet the physical, emotional, and cosmetic needs related to their disease and/or its treatment. In addition, literature and services to help husban.ds, children and friends of breast cancer patients are avaihtble. Laryngectomy Rehabilitation. Spearheaded by the International Association of Laryngectomees (IAL), this program brings the message that a laryngectomee can return to a normal life. Coordinated through more than 325 clubs, laryngectomee visitors provide pre- and/or postoperitive support to patients who have recently undergone removal of the larynx. Ostomy Rehabilitation. Some patients with intestinal or urinary cancers must have abdominal ostomies (surgically constructed openings for elimination of body wastes). Trained volunteers who have experienced this same type of surgery offer help on a one-to-one basis. Cooperating with the United Ostomy Association and enterostomal therapists, patients are assisted in their physical and psychological adjustment. Patient and Family Education Programs The Society sponsors group and individual education programs, distributes pamphlets and booklets and provides audiovisual presentations for patients of all ages and their families to help them understand and deal with the complexities of the disease. I Can Cope. Information is provided on cancer therapy, treatment, side effects, nutrition, resource availability and other topics of interest to cancer patients and their families. ALLOCATION OF ACS FUNDS BASED ON TOTAL 1987-1988 BUDGET-$331,365 MANAGEMENT AND GENERAL FUND RAISING COMMUNITY SERVICES $49,205 or 15% $21,772 or ' 7% _ I $28,535 $42,424 or 13% PATIENT SERVICES $33,208 or 10% ® ~® PROFESSIONAL EDUCATION COSTS OF CANCER A study by the National Center for Health Statistics (NCHS) puts overall medical costs for cancer at $71.5 billion for i'.9,'35; $21.8 billion for direct costs; $8.6 billion for so-called morbidity costs (cost of lost productivity), and $41.2 billion for mortality costs. The figures show that cancer accounts for 10% of the total cost of disease in the U.S. and that its share of the total cost of pre- mature death is about 18% of all causes of death. Individuals have several sources of help in paying for cancer costs: third-party payers such as Blue Cross 25 RESEARCH PUBLIC EDUCATION Figures taken from 1987 Annual Report (000's omitted) and private insurance companies, public agencies and private health organizations. Cancer is covered by personal insurance plans either under narrowly defined cancer policies or through catastrophic illness provi- sions in comprehensive insurance programs. The Third National Cancer Survey showed that for patients under 65 years, Blue Cross and private insurers were the source of payment in over 77% of the cases. For patients over 65, Medicare paid expenses in nearly 88% of the cases. $94,078 or 28% $62,143 or 19% I

C A N C E R F A C T 5 A N D F 1 G U R E 5 1 4 8 4 PREVENTION the body. The conclusions were: 1) Cigarettes and other forms of tobacco are addicting; 2) Nicotine is the drug in tobacco that causes addiction; and ~) The pharmacologic and behavioral processes that determine tobacco addiction are similar to those that determine addiction to drugs such as heroin and cocaine. Lowelr Tar & Nicotine Resea:rch has shown that there is no such thing as a "safe" cigarette, but that those who are not yet able to quit would be well advised to switch to brands with the lowest possible tar and nicotine (T/N) content. Moreover, low T/N smokers find it easier to quit altogether than high T/N smokers. In an ACS study conducted from 1960 to 1972, the average mortality of low T/N smokers was 16% lower than that: of high T/N smokers, and the comparable figure for lung cancer mortality was 26%. It is important to remember that besides tar and nicotine, cigarette smoke contains a host of other poisonous gases such as hydrogen cyanide, volatile aromatic hydrocarbons, and especially carbon monox- ide-possibly a critical factor in coronary heart disease and fetal growth retardation. While some hazards are reduced slightly by cigarette filters, certain filtered brands have been found to actually deliver more carbon monoxide than those without filters. Involuntary Smoking Hazards There axe hazards for nonsmokers who breathe the smoke of others' cigarettes. Several scientific studies, including a recent study by the 'American Cancer Society, have found an increased risk of lung cancer among nonsmoking wives of cigarette smokers. Although some studies have not shown an effect, evidence continues to grow indicating that involuntary smoking is a hazard. Two major reviews in 1986 by the Surgeon General and the National Academy of Sciences state that involuntary smoking is a health hazard. Another NAS report, also in 1986, states that the amount of smoke inhaled on airplane trips constitutes a hazard, partic- ularly to airline personnel, and recommended that cigarette smoking on airlines be banned. The Society's Cancer Prevention Study II, involving more than one million Americans, will include a careful assessment of cancer risk and other diseases among smokers and involuntary smokers. Smokeless Tobacco There has been a recent resurgence in the use of all forms of smokeless tobacco-plug, leaf and snuff- but the greatest cause for concern centers on the increased use of "dipping snuff." In this practice, tobacco that has been processed into a coarse, moist powder is placed between the cheek and gum, and nicotine, along with a number of other carcinogens, are absorbed through the oral tissue. "Dipping snuff" is a highly addictive habit, one that exposes the body to levels of nicotine similar *to those of cigarettes. A 1986 report of the Advisory Committee to the Surgeon General, outlining the health consequences of smoke- less tobacco use, concluded that there is strong sci- entific evidence that the use of snuff causes cancer in humans, particularly cancer of the oral cavity. Oral cancer occurs several times more frequently among snuff dippers compared to non-tobacco users, and the excess risk of cancer of the cheek and gum may reach nearly 50-fold among long-term snuff users. Smokeless tobacco is becoming a problem large in scope; the report found that in 1985 smokeless tobacco was used by at least 12 million people in the United States, and half of these were regular users. The use of smokeless tobaccos is increasing among male adolescents and young male adults. Industrial Hazards Industrial workers are especially susceptible to lung diseases due to the combined • effects of cigarette smoking and exposure to toxic industrial substances such as fumes from rubber, chlorine and dust from cotton and coal. Exposure to asbestos in combination with cigarette smoking increases an individual's lung cancer risk nearly 60 times. NUTRITION AND CANCER: A COMMON SENSE APPROACH Extensive research is under way to evaluate and clarify the role diet and nutrition play in the devel- opment of cancer. At this point, no direct cause-and- effect relationship has been proved, though statistics show that some foods may increase or decrease the risks for certain types of cancer. Evidence indicates that people might reduce their cancer risk by observing the following recommendations: 1. Avoid obesity. Individuails 40% or more overweight increase their risk of colon, breast, prostate, gallbladder, ovary, and uterine cancers. People with weight problems should consult their physicians to determine their best body weight, since their medical condition and body build must be taken into account. Physicians can recommend a suitable diet and exercise regimen to help maintain an appropriate weight. 2. Cut down on total fat intake. A diet high in fat may be a factor in the development of certain cancers, particularly breast, colon and pros- tate. In addition, by avoiding fatty foods, people are better able to control body weight. 21

Immunogenetics Considerable eflot has been expended in recent years to develop and characterize inbred, congenic, and recombinant strains of rats, and a wide variety of these genetic resources is now available (3, 4, 7-9). Several compiCations of basic data have been assembled (5), and current developments are regularly updated in the Workshops on Alloantigenic Systems e f the Rat (10) and in the Rat Newsletter (11). This work has also provided insight into the comparative genetics of the major histocompatibility complex (MHC) and of MHC-linked genes affecting grovnh and development. The level of polyrnor- phism of MHC antigens in the fat is very low compared to that of other species; the caass I antigens have been most extensively studied. Nonetheless, the resistance to disease, reproductive capaci- ty, and ecological stability of the rat do not differ from those of other species. Hence, the biological significance of MHC polymorphism remains a mystery. The structure of die MHC in the rat (RT1) based on data from serological, molecular, and functional studies is shown in Fig. 1 (3, 12, 13). The general organization of the class I and class II loci is the same as in the mou:e but different from that in all other species studied: the class I7: loci are interspersed between class I loci rather than following them sequentially (14). This observation indicates that (i) the rat and the mouse formed separate genuses after the divergence of the prototypic Muridae, (ii) the evolutionary conser- vation of the MHC persists despite internal rearrangements, and (iii) the function of rhese loci does not depend, at least to a first approximation, on their specific order or on their polymorphism. The RT1.A and RTI.E loci encode classical class I transplantation antigens and appear w be the homologs of the mouse H-2K and H- 2D loci. There are several other class I loci in the vicinity of RT1.A, and the best definc:d are the diallelic RT1.F and Pa (pregnancy- associated) loci (3, 13, 16). The antigen encoded by the Pa locus was first identified on the surface of the basal trophoblast in the allogeneic WF(u) x DA(a) mating by alloantisera and by monoclo- nal antibodies made by the WF mother (17). This antigen carries an epitope that is broadly shared among other class I antigens, but does not have the allele-sp:cific epitope of a classical class I transplanta- tion antigen. Immunohistochemical and electron microscopic stud- ies (18) showed that both the Pa and Aa antigens are also on most somatic tissues and that they are carried by separate molecules. The mapping of the A, F, and Pa loci is based on the use of various combinations of inbred, congenic, and recombinant strains; a number of monoclonal antibodies; and specifically designed al- loantisera. No recombinants among these loci have yet been found, but inununoprecipitat on and peptide mapping studies have demon- strated that they are: separate molecules: hence, the order of these loci in Fig. 1 must be considered tentative. The RT1.G and RT1.C loci encode class I avigens that appear to be homologous to the mouse QaIIZ antigens, but these loci have not yet been well characterized (19). The class II loci RTI.B and RT1.D were detected serologically and by molecular analysis (3), whereas RT1.H has been detected only by molecular anadysis (12). The B and D loci appear to be homologous to the mause A and E loci, and the H locus appears to be homologous, in part, to the mouse qfA(33 pseudogene and the human HLA-DP locus. The growth and reproduction complex (grc) is closely linked to the MHC (20). In t1le homozygous state, it is semilethal in males and females, causes s•mall body weight in both males and females (dw-3), and causes ma!e sterility and reduced female fertility ( ft). These defects are similar to some of those associated with the t haplotypes in the mouse, but the grc is not homologous to the t genes since it does not cause segregation distortion or suppression of recombination (3, 20). The fertility defect occurs at the same stage of gametogenesis in both males and females: there is complete arrest of spermatogenesis at the primary spermatocyte stage, and a partial defect in the maturation of the primary ovarian follicle. The grc acts at an early stage of meiotic prophase I; it is not associated with any known chromosomal or hormonal abnormality; and it increases susceptibility to chemical carcinogens in both males and females (21). Its effects are probably due to the deletion of a segment of the chromosome close to the MHC (22). If so, then the increased susceptibility to cancer may be due to the loss of cancer suppressor genes, or anti-oncogenes, as in retinoblastoma and Wilms' tumor in humans (23). Hence, these animals may provide a unique system in which to study the genetics of susceptibility to cancer. The homozygousgrc genotype (20 to 25% in utero mortality) can interact with the heterozygous Tal/+ gene, which is a recessive lethal gene on a different chromosome. The Tal gene is not lethal in the heterozygous state but, when homozygous, causes the death of all embryos at 10 to 14 days of gestational age (24). This demonstra- tion in mammals of a lethal epistatic interaction, which is the interaction between genes on different chromosomes, provides a useful system in which to study gene interaction during develop- ment. Molecular analysis has delineated the major regions of the rat MHC on the basis of restriction fragment length polymorphisms (RFLPs) (13, 22, 25). There are approximately the same number of class I-hybridizing fragments of DNA as in the mouse (26), despite the much lower level of serological polymorphism in the rat (3). The class II loci have not been examined in any detail yet, but there is a "hotspot" of recombination in the RT1.H region. The biochemical comparisons among the rat, mouse, and human MHC class I and class II antigens are summarized in Table 1. The amino acid sequences of the rat class I and class II antigens are more homologous to those of the mouse than to those of the human, although both levels of homology are fairly high. The homology among antigens encoded by the same class I locus is the same in the rat and the mouse, and both are lower than in the human. The homology between antigens encoded by different class I loci of the same haplotype is much higher in the rat than in the mouse or the human, whereas the interlocus homology for the class II antigens is approximately the same for all three genuses. When one compares the rat with the mouse and the human the most striking difference is in the number of serologically defined class I and class II antigens. This difference has been documented most extensively for the class I antigens in both inbred (3) and wild (27) populations; it has been less extensively studied for the class II antigens. The class I and class II antigens present in both the inbred and wild populations are scrologically and functionally indistinguishable, and there is a high degree of linkage disequilibrium among the loci in the MHC of the rat (27). The difference between the rat and the mouse and human in the serological polymorphism of their class I antigens stands in contrast to the similarity of their RFLP patterns (20 to 36 class I- hybridizing fragments) (3, 22, 25). This observation might reflect a similarity in the total number of class I genes in all three genuses but a difference in the number of functional genes. The situation with the class II loci in the rat appears to be the same: their serological polymorphism is very low but their RFLP is high (3, 12). Thus, the rat is an extremely useful animal in which to study the control of the functional activity of MHC loci and the biological consequences thereof. The limited MHC antigen polymorphism in the rat raises the question of what the biological significance of MHC polymorphism is (28). Neither the host defense mechanisms nor the reproductive capacity of the rat appear to differ from those of the mouse and the 270 SCIENCE, VOL. 245

270 Upton Table 2 Age-Standardized Lung Cancer Death Rates as Affected by Cigarette Smok-ing, Occupational Erposure to Asbestos Dust, or Botha Exposure to asbestos History of cigarette smoking Death rate Mortality difference Mortality ratio No No 11.3 0.0 1.00 Yes No 58.4 +47.1 5.17 No Yes 122.6 + 111.3 10.85 Yes Yes 601.6 +590.3 53.24 aAge-standardizz.d lung cancer death rates are rates per 100.000 man-years standardized for age on the distribution of the man-years of all the asbestos workers. Number of lung cancer deaths based on death certificate information. Source: From Ref. 59. Because of the multicausal, multistage nature of carcin- ogenesis and the fact that the mechanism of carcinogenesis is not the same for all cancers and all agents, some diver- sity of dose-incidence relationships is to be expected. The neoplasms that a;re induced by a given chemical in dif- ferent tissues or in animals of different species also may vary in dose-incidence relationships because of pharmaco- genetic and pharmacokinetic differences affecting the dosage of carcinogen to different target cells (47). The observed age- md tissue-dependent variations in dose- incidence relationhips among radiation-induced neoplasms are largely unexplained as yet (41), but differences in cell proliferation kinetics and homeostatic ability (including capacity to repair DNA damage) may constitute poten- tial sources of such variation (20). To explore the dose-incidence curve for carcinogenesis at low doses, a number of large-scale experiments have been carried out vtith laboratory animals. In the largest of these to date, the incidence of hepatomas in mice was observed to increase with the concentration of 2-AAF in the diet even at the lowest dose level tested (Fig. 3), whereas the dose•-incidence curve for tumors of the urinary bladder was quasithresholded (Fig. 3). This con- trast in dose-incidence curves may have resulted from dif- ferences between thie liver and the bladder in the metabol- ism of 2-AAF among other explanations. Because a given carcinogen may influence the probabil- ity of neoplasia through more than one type of effect, at least at high dose levels, its dose-incidence curve can reflect differing combinations of initiating effects, pro- moting effects, and anticarcinogenic effects, depending on the dose and other circumstances. The combined effects of multiple agents may, likewise, be additive, synergistic, or antagonistic, depending on the agents in question and the conditions of exposure. At low to moderate dose levels, the effects of a complete carcinogen can general- ly be accentuated by appropriate tumor-promoting stimuli, which unmask initiating effects that would otherwise re- main unexpressed (Fig. 4). It is noteworthy, moreover, that under conditions in which initiating effects are pro- moted to full expression they often increase as a linear nonthreshold function of the dose of the initiating agent (Fig. 4). Furthermore, whereas the carcinogenic effec- tiveness per unit dose of x-rays and gamma rays tends to 100r- 75 ~ Liver I / 1 ~ / °.~ ~50 .E Bladder 25 - / / / 0 i ®0 50 100 150 Concentration of 2-AAF in the Diet (ppm) , Figure 3. Cumulative incidence of tumors of the Gver and of the urinary bladder in female BALB/c mice exposed to 2-acetylaminofluorene (2-AAF) at various concentrations in the diet for up to 33 months. (From Ref. 63.). ,

Thresholds for Radiation and Chemical Carcinogenesis nonthrehsold relationship are: (i) a 25-50% excess of leukemia in c;hildren exposed to diagnostic x-rays in utero, in whom the radiation dose is estimated to have averaged less than 50 mGy (35,36); (ii) an excess of thyroid tumors in persons who received therapeutic irradiation of the scalp in childhood for tinea capitis, in whom the dose to the thyroid gland is estimated to have averaged no more than 60-80 mGy ;37,38); (iii) a dose-dependent excess of breast cancer, of essentially the same magnitude for a given dose, in: (a) women exposed to A bomb radiation, (b) women given therapeutic irradiation of the breast for postpartum mastitis, (c) women who received multiple fluoroscopic examinations of the chest during the treat- ment of pulmonary tuberculosis with artificial pneumo- thorax, and (d) women exposed to external gamma radia- tion in the painting of luminous clock and instrument dials (33,39); and (iv) a dose-dependent excess of leukemia in A bomb survivors, which is evident at doses below 300 mGy (33,34). [n each of the above populations, the dose- incidence data in low-to-intermediate dose range are com- patible with a linear nonthreshold relationship for the neoplasms in question. Comparable data, moreover, are available for certain radiation-induced neoplasms in laboratory animals (8,32,40,41). As concerns the car- cinogenic effects of chemicals, quantitative dose-inci- dence data for humans are extremely limited, with few exceptions. A iloteworthy exception is cigarette smoke, the major cause of lung cancer. In cigarette smokers, the incidence of lung cancer increases as a function of the number of cigarettes smoked per day raised approximately to a power of 1.8 (42). Furthermore, the absence of any clear indication of a threshold in the dose-incidence curve soo ANNUAL 400 ENCIDENCE sta,auai:.d fer ,se t'ER 100000 300 MEN 200 too t9 20 30 40 DoSE RArE tcila,atta Wnok.dp.reayl Figure 1. Annual in:idence of lung cancer in regular cigarette smokers, in relation to the number of cigarettes smoked per day. (From Ref. 61.) 269 100f- ffW C tii = z W > ~ p40 ~ 20 T T 1 T 1 2 3 4 5 LENGTH OF EXPOSURE (yrs) Figure 2. Cumulative incidence of cancer of the urinary bladder in 78 distillers of /3-naphthylamine and benzidine. (From ReF. 8, based on data from Ref. 62.) (Fig. 1) is consistent with epidemiological data implying that the risk of lung cancer can be increased even in nonsmokers by passive exposure to cigarette smoke over prolonged periods (43). Other populations for which the dose-incidence data are compatible with a nonthreshold type of response in- clude groups of chemists who were employed as distillers of 2-naphthylamine. In one such group, the cumulative incidence of cancer of the urinary bladder was observed to increase with the duration of occupational exposure, approaching 100% in workers who were exposed for five years or longer (Fig. 2). In asbestos workers, likewise, the rates of lung cancer and mesothelioma appear to increase linearly with the in- tensity and duration of exposure (44). Furthermore, in asbestos workers who smoke cigarettes, the combined car- cinogenic effects of asbestos and cigarette smoke appear to be multiplicative rather than merely additive (Table 2), implying that the two agents exert their effects through complementary rather than similar mechanisms. With respect to the mechanism of cigarette smoke- induced carcinogenesis, it is noteworthy that the excess of lung cancer in ex-smokers stops rising relatively promptly after cessation of smoking (45), suggesting that cigarette smoke affects primarily late stages of car- cinogenesis. The carcinogenic effects of cigarettes thus stand in contrast to those of radiation (33) and asbestos (46), which continue to become manifest for decades after exposure.

UP'I'ON: THRESHOLDS FOR CARCINOGENESIS? 865 differentiation; and 9) modified responses to other growth-controlling factors.16 Such effects have also been demonstrated, in vivo and in vitro, in cells of various other species, including man. As yet, however, it is not known whether any of these responses is critical for tumor promotion. Furthermore, although all tumor- p:"omoting agents induce cellular proliferation in their respective target tissues, each appears to be relatively tissue-specific. The capacity of the promoters to induce pleiotropic effects at nanomolar concentrations and their discrete struc- ture-activity relationships implicate a hormone-like mode of action.1-19 Agents that possess initiating activity as well as promoting activity can cause neoplasia by themselves if given in sufficient doses. The effects of such "com- plete" carcinogens can be enhanced, however, by various other agents that are not active by themselves but can potentiate the effects of carcinogens if given simultaneously with them.20 Such "co-carcinogens," which include certain phe- ncls, aliphatic hydrocarbons, and aromatic hydrocarbons, are prevalent in the environment and appear to act by altering the uptake, distribution, andlor metabo- lism of carcinogens, or by enhancing the susceptibility of the target cells or host.20 Tumor promoters have thus traditionally been considered to act predominantly through epigenetic mechanisms.'$ Recently, however, the production of indirect da~,nage to DNA, resulting in mutations and chromosome aberrations, has been implicated in a growing number of instances21-23 ; for example, target organ-spe- cific DNA adducts have been identified in association with the carcinogenic ef- fects of diethylstilbesterol on the hamster kidney.24 Tumor Progression Tumor progression is a process through which successive alterations in neo- plastic cells give rise to increasingly autonomous clonal derivatives.u The precise nature of such alterations remains to be determined, but mutations and chromo- some aberrations have been tentatively implicated.7 Tumor progression may be acc,elerated by repeated exposure of neoplastic cells to carcinogenic stimuli or by sele ction pressures that favor the outgrowth of increasingly autonomous subpopu- lations of cells. EPIDEMIOLOGIC DATA ON DOSE-INCIDENCE RELATIONSHIPS IN HUMANS In contrast to the hundreds of chemicals that have been observed to possess oncogenic activity in laboratory animals, less than three dozen are known to induce cancer in man.26 In few cases, moreover, are the relevant epidemiological data adequate to characterize the relationship between cancer incidence and the dose of a given carcinogen, except in a semiquantitative way. Analysis of the dose-incidence relationship is less difficult with ionizing radia- tion than with carcinogenic chemicals because dosimetry with radiation is not complicated in the same way by pharmacokinetic variables. Furthermore, inci- dence data for irradiated populations are available over a wide range of radiation doses,z'.zII whereas comparable dose-incidence data for chemicals are generally lacking. In no case, however, do the data suffice to define the dose-incidence relationship in the low dose domain or to exclude the possibility of a threshold. Hence assessment of the carcinogenic risk associated with low-level exposure to

272 15 38cGy/mm 0O86cOf/m,n JANUS NEUTRONS SLOPE (:Gy-'1 59fix10'1 53.0x10-1 IOTV2 CELLS COR COEF 0.990 0 998 SLOPE RATp(-/el 1 8.89 0 086cGy/min ' 38CGy /min LL 0 20 40 60 DOSE,cGy 100 Figure 6. Frequency of neoplastic transformation in C3H IOTI/2 cells exposed to fission-spectrum neutrons. Dashed lines indicate linear regres- sions fitted to the initial portions of the dose-effect curves. (Reproduced from Ref. 67.) Interpolation and Extrapolation Models Although the relation between the incidence of neoplasia and the dose of carcinogen is known to vary with the type of neoplasm, the carcinogen, and other variables, the dose-incidence relationships at low doses is not known precisely for any neoplasm or carcinogen. The risks of low-level exposure to a cancer-causing agent can thus be assessed only through interpolation or extrapolation from effects observed at higher levels of exposure. For many of the neoplasrns induced by ionizing radiation, the dose- incidence relaticn generally conforms to the patterns il- lustrated in Figure 8, which are consistent with those to be expected if the probability of carcinogenesis could be increased in a suitably susceptible individual by an appro- priate mutation or chromosomal aberration in a single somatic cell. Under this assumption, the dose-incidence curve for high-L',ET radiation would be expected to con- form, in general, to the expression: I = C + aD (1) where I is the incidence at dose D, C is the incidence in nonirradiated can:rols, and the coefficient a is a constant; similarly, for low-LET radiation, the dose-incidence curve would conform, in general, to the expression: I = (C + aD + bD2)e-(p°+q°2) (2) where the symobls are comparable to those above, ex- cept for a different value of the coefficient a and the ad- dition of the coefficients b, p, and q (55). Upton 70 t1 G 30 C a) ~ ti 20 C O ~ a E 10 ~ ~ ~ H 0 •- Dose (Rods) Figure 7. Dose-response relationship for the induction of neoplastic transformation in mouse 10T1/2 cells by x-rays alone (o), or by x-rays followed by phorbol ester, starting 48 h after irradiation and continued for the full 6-week expression period (•). No increase in transforma- tion frequency was detected following exposure to phorbol ester alone. (Reproduced from Ref. 68.) While many of the observed dose-incidence curves con- form to the latter pattern, the curve for radiation-induced breast cancer appears more nearly linear, as noted above. To allow for uncertainty about the shape of the dose- incidence curve at low doses and thus to obtain a range of reasonable risk estimates, alternative models (Figs. 9 and 10) have been used in assessing the risks of low-level exposure to carcinogens. Most such models treat carcino- genesis as a multicausal, multistage process. Depending on the particular model that is used for interpolation or extrapolation, however, the estimated risk at low doses can vary by order of magnitude (e.g., Table 3). The linear (one-hit) model for interpolating between the lowest dose

L 268 DNA synthesis (17); (iv) the carcinogenic potency of an initiating chemical is generally correlated with the extent to which it binds covalently to DNA and with the nature of the resulting DNA adducts; and (v) DNA to which a chemical carcincgen is bound can serve as a template for DNA replication (18) which, along with subsequent cell division, is necessary to "fix" the potential for neoplastic change (19); (vi) susceptibility to cancer is increased in persons who are deficient in their capacity to repair DNA damage (20). Whatever the nature of the mutational change may be, it results in a frequency of initiation that is orders of magnitude higher than the rate of mutations at any given genf: locus (21,22), implying that multiple oncogenic sites, damage to the genome at sites unlikely to be repaired (e.;;., tandem repeats), or genetic damage other than point mutations are likely to be involved (14). The specific genes that are affected may be presumed to include antioncogenes as well as oncogenes (Table 1). Initiation can thus be envisioned to result either from the homozygous inactivation or deletion of an antioncogene, or from the aberrant activation of an otherwise normal proto-oncogene, through aneuploidy, chromosomal re- arrangement, or point mutation. For neoplastic transfor- mation, as opposed to initiation, the activation of a single oncogene alone appears to be insufficient (13). Although initiation can result from only one exposure to an appropriate: initiating agent, tumor promotion typically requires repeated and sustained exposures to an appropriate promoting agent, although low doses of the agent may suffice. In two-stage mouse skin carcinogene- sis, for example, nautomolar concentrations of 12-O-tetra- decanoyl phorbol-13-acetate (TPA) are sufficient to Table 1 Comparative Properties of Oncogenes and Antioncogenes Oncogenes Gene active Specific translocations Translocations not hereditary Dominant Tissue specificity may be broad Antioncogenes Gene inactive Deletions or invisible mutations Mutations hereditary and nonhereditary Recessive Considerable tissue specificity Especially leukemias ,and lymphomas Source: From Ref. 20. Solid tumors (e.g., Wilm's, retinoblastoma) Upton promote the effects of radiation or chemical initiators, causing concomitant stimulation of: (i) macromolecular synthesis; (ii) hyperplasia; (iii) polyamine synthesis; (iv) prostaglandin synthesis; (v) protease production; (vi) alterations of certain cell membrane enzymes and glyco- proteins; (vii) induction of sister-chromatid exchanges; (viii) altered differentiation; and (ix) modified responses to various growth-controlling factors (23). Whether any one of these changes is critical for tumor promotion, however, is not clear. Traditionally, TPA and other tumor-promoting agents have been considered to act predominantly through epigenetic mechanisms (24,25), but recent observations indicate that some of these agents can damage DNA indirectly (26-29) implying that such genotoxic effects also may be involved in promotion. Tumor progression, the process through which suc- cessive generations of neoplastic cells give rise to increas- ingly autonomous clonal derivatives (30), has been at- tributed at least in part to mutations and chromosome aber- rations (15). The process can be accelerated, however, by selection pressures that favor the outgrowth of pro- liferative subpopulations, including repeated exposure to growth-stimulating agents and carcinogens (15,30). I EMPIRICAL DOSE-INCIDENCE RELATIONSHIPS FOR CARCINOGENSIS Although hundreds of chemicals have been found to be oncogenic in laboratory animals, less than three dozen have been observed to be capable of inducing cancer in humans (31). With few exceptions, moreover, the relevant data are not sufficient to characterize the dose-incidence relationship except in a semiquantitative way (8). With ionizing radiation, for which the dosimetry is less complicated by pharmacokinetic variables than is the dosimetry for most chemicals, dose-incidence data are available over a relatively wide range of radiation doses (32,33). At best, however, the data do not suffice to define the dose-incidence relationship in the low-dose domain. Assessment of the carcinogenic risks associated with low- level irradiation must thus depend on extrapolation from observations at higher levels of exposure, based on assumptions about the relevant dose-incidence relation- ships and mechanisms of carcinogenesis. The extrapolation models that are used for estimating the carcinogenic risks of low-level irradiation generally assume a linear nonthreshold relationship between risk and dose in the low-dose domain, although the data do not exclude a threshold (8,33,34). Among the lines of epidemiological evidence that are consistent with a

864 ANNALS NEW YORK ACADEMY OF SCIENCES suitably susceptible individual. The data also indicate, however, that more than one alteration is necessary to convert a normal cell into a cancer cell, as discussed below. Multicausal, Multistage Nature of Carcinogenesis Clinical, pathological, and experimental data imply that cancer evolves through at least three successive stages: initiation, promotion, and progression.7 Initiation Initiation, which starts the process, does not itself suffice to cause neoplasia but predisposes the affected cell and its progeny to subsequent steps in carcino- genesis. The initiated cell may not be recognizable as such, however, or may never form a tumor, unless it is further altered by subsequent tumor-promoting stimulation. The mechanism of tumor initiation remains to be established; but sonne type of mutational process is suggested by the evidence that 1) initiation is relatively prompt and irreversible; 2) most ultimate carcinogens are mutagens; 3) the frequency of cell transformation induced by a given carcinogen usually is highest if exposure to the agent occurs just before or during the DNA synthetic phase of the cell cycle8; 4) DNA to which a chemical carcinogen is bound can serve as a template for DNA replication9; 5) after exposure to a carcinogen, DNA synthesis and subsequent cell division "fix" the potential for neoplastic change'o; 6) in a given biological system, the carcinogenic potency of an initiating chemical is generally correlated with the extent to which it binds covalently to DNA, and witlr, the nature of the resulting reaction products; and 7) susceptibility to cancer is increased in persons who are deficient in DNA repair.l I The frequency of neoplas- tic transformation is far higher, however, than that of single gene mutation in cells exposed to genotoxic carcinogens12•13; hence, the data implicate multiple onco- genic sites, damage of the genome at sites unlikely to be repaired (for example, tandem repeats), or genetic damage other than point mutations. i4 The specific genes that may be involved are only beginning to be defined but appear to include antioncogenes as well as oncogenes.'i It is noteworthy, furthermore, that activa- tion and expression of more than one oncogene appears to be necessary for cell transformation in vitro.1s Tumor Promotion Tunlor promotion is the process that results in the additional change, or changes, necessary to cause the neoplastic transformation of an initiated cell. In contrast to initiation, which can result from a single exposure to an appropriate tumor-initiating agent, tumor promotion requires repeated and sustained stimuli. Although tumor promotion has been demonstrated in a number of tissues, its mechanisms have thus far been studied systematically only in a few model sys- tems. In one of these, the mouse skin model, nanomolar concentrations of the tumor•promoting phorbol ester 12-0-tetradecanoylphorbol- I 3-acetate induce stim- ulation of 1) macromolecular synthesis; 2) hyperplasia; 3) polyamine synthesis; 4) prosta;eandin synthesis; 5) protease production; 6) alterations of cell membrane enzymes and glycoproteins; 7) induction of sister-chromatid exchanges; 8) altered

UlP7CONc THRESHOLDS FOR CARCINOGENESIS? 867 soo ANFCUAL 400 INCIDENCE stuod7tdirsd f9r at. PER Io00o0 00 MEN 200 100 0 10 lo ]0 40 S® DOSE RATE (citarsceas o+akad p.r dt.y) FIGURE 2. Incidence of lung cancer in regular smokers of cigarettes in relation to the number of cigarettes smoked per day. (Reproduced from Doll35 with permission.) luminous clock and instrument dials, which is similar in all four groups, irrespec- tive of the marked differences among the groups in the duration of exposurezs,34; and (4) the excess of leukemia in A-bomb survivors, which is evident at doses below 0.25 Gy.28,29 The data from each of these studies, although not adequate to precisely define the shape of the dose-incidence curve in the low dose domain, are compatible with linear nonthreshold functions for each of the neoplasms in question. SV/ VVith respect to the carcinogenic effects of chemicals, as opposed to ionizing J~- radiation, quantitative dose-incidence information for human populations is far more, limited. Nevertheless, considerable information is available for a few chemi- cals, one of them being cigarette smoke, which contains thousands of compounds, including initiating agents as well as promoting agents. In cigarette smokers, the incidence of lung cancer (FtG. 2) increases as a function of the average number of cigarettes smoked per day raised to a power of 1.8.36 Similarly, in chemists who were employed as distillers of 2-napthylamine, the cumulative incidence of cancer of the urinary bladder increases steeply with the duration of occupational exposure, approaching 100% in those who were exposed for 5 ydars or longer (FIG. 3). 100 0 FIGURE 3. Cumulative incidence of tumors of the uri- nary bladder, at 30 years after start of exposure in 78 distillers of 2-naphthylamine and benzidine, in relation to duration of occupational exposure. (Reproduced from Saffiotti3' [based on data from Williams38] with permission from the International Agency for Research on Cancer.) 1\ 00 `I 2 3 4 5~ Durotion of Ezposure (yrs)

UP;CON: THRESIiOLDS FOR CARCINOGENESIS? 869 1000 2000 3000 4000 5000 DOSE (rods) FIGURE 4. Dose-incidence curves for different neoplasms in animals exposed to external radiation: (A) myeloid leukemia in X-irradiated mice (Upton et al.43); (B) mammary gland tumors at 12 months in gamma-irradiated rats (Shellabarger et al.44); (C) thymic lymphoma in X-irradiated mice (Kaplan and Brown45); (D) kidney tumors in X-irradiated rats (Malda- gue46); i(E) skin tumors in alpha-irradiated rats (percentage incidence x 10) (Burns et al.°7); (F) sk.in tumors in electron-irradiated rats (percentage incidence x 10) (Burns et al.'~; and (G) reticulum cell sarcoma in X-irradiated mice (Metalli et al.48). (Modified from reference 49. Reproduced from Upton'-' with permission from the Elsevier/North-Holland Publishing Company.) the daily dose required to double the risk of neoplasia varies among different chemic;~ls by more than six orders of magnitude (FIG. 5); (3) with ionizing radia- tion, the dose-incidence curve for high linear energy transfer radiation generally rises more steeply with dose and is less dependent on the dose rate than is the curve for low linear energy transfer radiation54; (4) for many types of neoplasms, the incidence passes through a maximum at some intermediate dose and de- creases with further increase in the dose (FIG. 4); (5) the median time of tumor ~ ~ a TCDD FIGURE 5. Range of carcinogenic potency in male rats. (Reproduced from Gold et al.50 with permission from the National Institute of Environmental Health Sciences. ) ~AttinomycinD ~ 4llotosin Bt - Bis-(chloromethyl) ether l0onq r -- I Stenqmotocystin ~0BCP ~-- Diethyistitbestrol Procorbozine. HCJ EDB 2-AAF Auromine-0 Anil ine. HCI -rDDT .2 ,4,6-Trlchl orophenol , Metronidozole C-~FDdC RedNo.t ~ o c 109 FD 8 C Green No.I

3b$ ANNALS NEW YORK ACADEMY OF SCIENCES In asbestos workers, likewise, the incidence of lung cancer and of inesothe- lioma appears to increase linearly with the intensity and duration of exposure.39 It is noteworthy, furthermore, that in cigarette smokers who have also been exposed occupationally to asbestos, the carcinogenic effects of cigarette smoke and asbes- tos appear to interact multiplicatively rather than additively (TABLE 1). Also noteworthy is the fact that the excess of lung cancer in ex-smokers decreases rapidly after cessation of smoking,41 suggesting that cigarette smoke affects primarily on late stages of carcinogenesis, acting as a promoting agent. This situation contrasts sharply with that in irradiated28 or asbestos-exposed42 populations, in whom the risk of lung cancer persists long after exposure. EXPERIMENTAL DOSE-EFFECT DATA Carcinogenesis in LaBoratory Animals The neoplasms induced experimentally in animals of different species vary widely in dose-incidence relationships. Although neoplasms of virtually every TABLE 1. Age-Standardized Lung Cancer Death Rates for Cigarette Smoking, Occupational Exposure to Asbestos Dust, or Both Group Exposure to Asbestos? History of Cigarette Smoking? Death Rate Mortality Difference Mortality Ratio Control No No 11.3 0.0 1.00 Asbestos workers Yes No 58.4 +47.1 5.17 Control No Yes 122.6 +111.3 10.85 Asbestos workers Yes Yes 601.6 +590.3 53.24 NOTE: Age-standardized lung cancer death rates are rates per 100,000 man-years stan- d«rdized for age on the distribution of the man-years of all the asbestos workers. Number of lung cancer deaths based on death certificate information. (Adapted from Selikoff./0) type have been induced in one experiment or another, all types of neoplasms are not elicited in animals of any one species or strain. In fact, certain types of neoplasms actually decrease in frequency with increasing dose of whole-body in'adiation (FiG. 4). Among chemically induced neoplasms, the observed variations are attribut- able in part to pharmacokinetic differences affecting the dosage of carcinogen to diff-.rent cells and subcellular targets. Such an explanation cannot account, how- ever, for the observed variations in dose-incidence relations among radiation- induced neoplasms, which remain largely unexplained. Because of the multi- causal, multistage nature of carcinogenesis, and the fact that the mechanism of carcinogenic effects is not the same in all instances, some diversity of dose- incidence relationships is to be expected. Obviously, the observed dose-incidence curves cannot all be represented by the same mathematical function. Nevertheless, the following generalizations emerge from the data: (1) a carcinogenic-induced elevation in the age-specific incidence of a particular neoplasm may or may not result in an increase in the final cumulative incidence of tumors, depending on the survival of the population at risk; (2) chemicals differ greatly in carcinogenic effectiveness, with the result that

C A N C E R F A C T s A N D F! G U R E 5 1 9 8 9 THE AMERICAN CANCER SOCIETY Nursing Programs Cancer Nursing News is sent to about 90,000 nurses. It keeps nurses tip-to-date on cancer, oncology nursing, the American Cancer Society, and opportunities in continuing education. The newsletter is sent free to any nurse; requests for subscriptions should be sent to the Executive Editor, Cancer Nursing News, c/o American Cancer Society, 1599 Clifton Road, N.E., Atlanta, GA 30329. Twenty one-year nursing scholarships are awarded each year to qualified graduate students studying for a master's degme with a specialty in cancer nursing. The recipients may apply for a second year's funding. A training program to prepare nurses for Ph.D.'s in related fields was initiated with the funding of the first three candidates in 1986. Professorships in Clinical Oncology Leading experts in oncology are supported to pro- mote cancer education in medical and health profes- sional schools. Since the award's inception in 1970, the Society has funded 53 professors. Recently the program has expanded to fund its first Professor involved in Dental Oncology. Clinical Oncology Awards The ACS National Clinical Awards Program was established in 1948 to provide broad support for oncol- ogy training at qualified hospitals and institutions. Over the past 40 years, Regular Clinical Fellowships and Junior Faculty Clinical Fellowships have had consid- erable impact on the training of physicians and dentists in oncology specialties, training over 8,500 individuals to provide care to cancer patients nationwide. The program has changed somewhat over time; the original awards have been modified based on changes in oncology over time. Currently, monies are provided via the Clinicall Oncology Fellowships (COF) and Clinical Oncology Career Development Award (CDA). The former program replaces the regular Clinical Fellowship and intends to provide unique training opportunities for fellows to expand their expertise in oncology. The CDA is awarded to outstanding indi- viduals who have demonstrated a commitment to pur- sue an academic career in oncology. For the first time, a traineeship is being offered for Oncology Social Workers committed to clinical practice and research to benefit cancer patients and their families. The first awards will be made in 1989 to 24 master's and post-master's candidates. To meet the needs in cancer prevention and detection, the concept of a new career development award for primary care physicians is under consideration. When accepted, these awards will help develop academic leaders in primary care to promote lifesaving tech- niques to the critical specialties. The implementation of training program support for allied health professionals is also being studied. By broadening and expanding our efforts in oncology training, the Society's long-term goal of promoting cancer education, cancer control and cancer manage- ment among all health care providers will be advanced. Unproven Methods of Cancer Management The American Cancer Society maintains information on unproven methods of cancer management. This information is reviewed in-depth and is issued in position statements. These statements are available on request to physicians, science writers, editors and the general public, to assist in evaluating claims made for unproven methods of diagnosis and treatment. The Committee on Unproven Methods of Cancer Management has commissioned a survey to determine the prevalence of, reasons for use, and patterns of use of unproven methods by the cancer patient. The findings from this study will provide guidelines for future programs in unproven methods of cancer management. SERVICE AND REHABILITATION In 1988, over one-half million cancer patients were reached through the innovative service and rehabil- itation programs of the American Cancer Society. Because of the many volunteers at the Division and Unit levels, the Society is able to offer a wide range of services. Service Programs Resources Information and Guidance Services. Specific information is provided about cancer, as well as referral to Society services and other resources in the com- munity to meet the social, psychological and home care needs of cancer patients and their families. Home Care Items. This program provides necessary useful home care supplies, equipment, dressings and gifts for the comfort and recreation of the patient. Transportation. Through the efforts of volunteer drivers in programs such as Road to Recovery, transportation is provided to patients, enabling them to maintain their medical and continuing care programs. Rehabilitation Programs CanSurmount. This is a short-term visitor program for patients, and the families of patients, with many types of cancer. Hospital and home visits are made with the approval of the physician. The one-to-one visit by a person who has experienced the same type of cancer offers functional, emotional and social support. 24

876 ANNALS NEW YORK ACADEMY OF SCIENCES I goneral form a all killing ,~, anenutte: F(D) ~ c F(D) - (ap +c~ D+a2D2l&xp(-Qr D -AZD2) - Doa.. D Do+., D Dou, D a FIGURE 10. Dose-response curves for four different mathematical models relating cancer incidence to radiation dose. (Reproduced from Reference 28 with permission from the National Academy Press.) Criteria to aid in the evaluation of epidemiological and experimental data on the carcinogenicity of chemicals have been formulated by the International Agency for Research on Cancer,26 the Interagency Regulatory Liaison Group,102 and the Office of Science and Technology Policy.103 These criteria include defini- tions for weighing the adequacy of the data (for example, definitions of "sufficient evidence" and. "limited evidence" 26). In situations where there is sufficient evi- dence for the carcinogenicity of a chemical in'laboratory animals but not in humans, the compound is assumed to present a carcinogenic risk to humans, although the magnitude of the risk cannot be estimated with precision.26 Although Ibioassay and short-term "screening" tests may give information on the mode of action of a chemical, such tests are considered to provide no more than support- i.ng evidence of carcinogenicity and not to provide sufficient evidence by them- selves. Estimation of carcinogenic risks on the basis of animal data, however good the an imal data may be, is fraught with uncertainty. Although a chemical with carcin- ogenic potency in one species (such as aflatoxin Bi) is likely to be carcinogenic in another, the procedure for extrapolating across species involves assumptions about species differences in metabolism and appropriate scaling factors for dose and time. Various attempts have,been made to determine correct scaling factors based on pharmacokinetic data,95 but the question remains unresolved.

} panel reached the same conclusion about two of Ernhart's papers, which they also criticized for methodological flaws. While Errthart took the panel's rebuke quietly, IvTeedleman fought back. He in- sisted that the panel's conclusions were flawed, and he wrote a spirited, point-by- point refutation of the criticisms levied at his work. He blasted Grant for printing the report before sending it to him for review, accusing him of violating an agreement he said he and Grant had made. Needleman "It was-,really the data a,n alysis strategy that I t'ceoked at, and that to'rte [was] outrezgle,aus. -Sandra W. Scarr also performed some new analyses of his original data, and by the time the panel's report was presented to the EPA advisory panel that would decide on the new lead standards, both Grant and the advisory panel had made a IEIO-degree turn. Now they were convinced that Needleman's original conclusions were accurate. Indeed, those conclusions subsequently became part of the scientific basis for the revised air lead standards EPA, promulgated in 1986. But Ernhart was not daunted by this set- back; she continued to criticize Needleman's work. And her willingness to argue that the link between l,mw-level lead exposure and behavioral problems was being overstated won favor with the lead industry. As early as 1982, she had agreed to testify in favor of the industry's positio n before an EPA panel con- templating phasing out all leaded gasoline- Just last year she wrote to Senator Harry Reid (D-NV) telling him that basing legislative action on Needleman's findings would be an "egregious error.... Serious problems in the Needleman work have long been noted by scientists working in this field." And she appeared from time to time as an expert witness in cases involving lead contamination and cleanup, w.hich brought her feud with Needleman into a new arena: the courtroom. Their latest fieeoff-which has escalated beyond the hazards of lead to the high- stakes "game" of scientific fraud and mis- conduct charges--began in 1990 with a Superfund case brought by the government against Sharon Si:eel, UV Industries, and Atlantic Richfield Company. Over a period of several decades, each company had had a financial interest irt a defunct lead smelter in 23 AUGUST 1991 Midvale, Utah. The government case sought money for the cleanup of some 250 acres of tailings from a milling facility that prepared the lead ore for the smelter. The govern- ment intended to show that the tailings posed a health risk to children living in the area and hired none other than Herbert Needleman as an expert witness to testify to the dangers the tailings posed. For their part, the corporations' lawyers turned to Ernhart as an expert witness. In addition, the defense team brought in University of Virginia psy- chologist San- dra Wood Scarr, whose work fo- cuses on factors affecting child- ren's educational development. She had also served as a mem- ber of the EPA panel that had examined Needleman and Ernhart's re- search back in 1983. Although Scarr had been among the most critical of Needle- man's work then, she says she paid no fur- ther attention to it after the panel had wrapped up its business. Now, she and Ernhart felt that they could dam- age the govern- ment's case by demonstrating what they had long believed: that Needleman's 1979 paper-which they say has been "highly influential in the establishment of regulatory policies"-was seriously flawed. They asked to see Needleman's raw data for the 1979 study. He agreed to release some of the unpublished material, but not the tapes containing his raw data. Needl- eman argued, in an affidavit dated 27 July 1990 that, in part because he was in the throes of moving his lab, "it would be a substantial hardship for me to find the proper data tape for this 11-year-old study." He added that since the study had been peer reviewed and the data examined by the EPA, there had already been adequate op- portunity to establish'the legitimacy of his results. Needleman did say in his affidavit, how- ever, that he would be willing to let "any scientist who wishes to examine the complete printouts of the raw data from the study come to my laboratory in Pittsburgh for as long as he or she wants." So on 20 September last year, Scarr and Ernhart, along with de- fense lawyers in the lead smelter case, traveled to Pittsburgh to take Needleman up on his offer. When they arrived, they were directed by Justice Department attorney W. Be jamin Fisherow, who was acting for th government, to a bare room where they were given six volumes of computer print- outs containing Needleman's initial analyses of his data. Scarr and Emhart began plow- ing through the analyses, although they were hampered by the fact that the data were coded, and they were given an incom- plete key. Needleman himself would not talk to them. For his part, Needleman steadfastly insists that he will happily share his data with anyone who has a legitimate interest and will answer any questions he is asked. But, he says, "I'm just not going to make it easy for people who are going to harass me," a category to which he assigns Scarr and Ernhart. Since Scarr and Ernhart weren't able to get through all the computer printouts in one day, they returned to the lab the next morning. But this time, Fisherow asked them to sign a document saying they would 6`Serious problems in the 1Veedleman work have long been noted by scientists working in this f aeld." -Claire B. Ernh;u-t treat all the data they were being shown in absolute confidence and would discuss it only in oral testimony before the court. While such agreements are not uncommon for litigation involving private corporations, Scarr and Ernhart were appalled at what they saw as an attempt to gag them, and they refused to sign. After a few hours, lawyers for both sides decided that the visit would have to end, so Scarr and Ernhart gathered their notes and left. Scarr says that even with only one day to study the analyses, she felt she had a clear idea ofwhat had happened back in 1979. "It was really the data analysis strategy that I looked at, and that to me [was] outrageous." Ac- cording to Scarr, the printouts show that Needleman's first set of analyses failed to show a relationship between lead level anc' subsequent intelligence tests. "Not one sin€ variable came out as statistically different be- tween the top 10% [oflead-exposed children] NEWS & COMVAENT 843

30-YEAR TRENDS IN AGE-ADJUSTED CANCER DEATH RATES PER 100,000 POPULATION 1953-55 to 1983-85 SITES SEX 1953-55 1983-85 PERCENT CHANGES COMMENTS Male 175.7 203.1 + 16 Steady increase mainly due to lung cancer. ALL SITES _ Female 145.1 138.2 - 5 Slight decrease. Male 7.2 6.1 - 15 Slight decrease in recent years. BLADDER _ Female 3.1 1.8 - 42 Some fluctuations; noticeable decrease. Male 3.9 4.7 + 21 Early increase in both sexes; BRAIN ~ Female 2.6 3.2 + 23 later leveling off, reasons unknown. Male 0.3 0.2 • Constant rate. BREAST Female 26.2 27.1 + 3 Slight fluctuations; overall no change. COLON & Male 25.8 24.7 • Slight fluctuations; overall no change. RECTUM Female 24.4 17.5 - 28 Slow, steady decrease. Male 16.9 20.7 + 22 Slow steady increase, leveling in recent years. COLON Female 18.3 15.0 - 18 Slow, steady decrease. Male 8.9 4.0 - 55 Slow steady decrease. RECTUM Female 6.1 2.4 - 61 Slow steady decrease. Male 4.7 5.6 + 19 Some flucutations; small increase. ESOPHAGUS Female 1.2 1.5 ' Slight fluctuations; overall no change. Male 3.6 4.9 + 46 Steady slight increase. KIDNEY Female 2.2 2.3 ' - Slight fluctuations; overall no change. _ Male 2.6 2.7 ' Slight fluctuations; overall no change in LARYNX Female 0.2 0.5 both males and females. Male 8.2 8.4 + 2 Early increase, later leveling off and decrease. LEUKEMIA Female 5.5 5.0 - 9 Early slight increase; later leveling off and decrease. Male 6.2 4.9 - 21 Decreasing rapidly early; later leveling off. LIVER•• Female . 7.1 3.3 - 54 Some fluctuations; steady decrease. r Male 28.0 73.1 +161 Steady increase in both sexes due to LUNG Female 5.1 25.3 +3% cigarette smoking. Male 8.0 11.1 +39 Slow steady increase in CYhSPHOMAS Female 5.1 7.5 +47 both males and females. Male 6.0 5.2 ' Slight fluctuations; overall no change ORAL Female 1.5 1.8 in both males and females. OVARY Female 8.6 7.8 - 9 Steady increase; later leveling off and decrease. ! Male 9.1 10.2 * 12 Steady increase in both sexes, then leveling off, PANCREAS Female 5.7 7.2 + 26 reasons unknown. PROSTATE Male 21.3 23.2 + 9 Fluctuations throughout; overall slight increase. r Male 3.1 4.0 + 29 Slight fluctuations; slight increase. S;KIN Female 1.9 1.8 • Slight fluctuations; overall no change. Male 21.3 10.2 - 52 Steady decrease in both sexes; reasons S1'OMACFI Female 11.2 3.5 - 69 unknown. iJTERUS Female 19.0 7.1 - 63 Steady decrease. 'Percent changes not listed because they are not meaningful. •'Primziry and non-specified. 29

Thresholds for Radiation and Chemical Carcinogenesis %'isx(1iie1) o 7Y.(1/91) " g63 Y.(1t/167) (0$4) . 50 100 150 200 250 300 3S0 400 450 Radiation dose (r) Figure 4a. Cumulative incidence (in percent) of leukemia in C57BL mice in relation to the dose of whole-body x-radiation administered in a single exposure (--o-o-), with or without subsequent injections of urethane (-x-x-). (Reproduced from Ref. 64.) Figure 4b. Cumultui,ve incidence of carcinomas of the skin in mice exposed once weekly to benzo(a)pyrene (BaP), with or without subse- quent exposure to 12-C'-tetradecanoyl phorbot-13-acetate (TPA) twice weekly. Doses refer to the amount of B(a)P applied to the skin each week. (Reproduced from Ref. 65.) decrease with decreasing dose and dose rate, that of high- LET radiation tends to remain constant or even increase (Fig. 5) (32,40,4I)~. 271 400 A/n, 24 FR. (A) / 60 FR. (a) Figure 5. Life shortening (all causes) in male B6CF, mice in relation to the total dose of single, fractionated (FR), or continuous whole-body neutron- or gamma-irradiation. (Reproduced from Ref. 66.) Cell Transformation In Vitro The neoplastic transformation of cells in vitro, although not strictly analogous to carcinogenesis in vivo, provides a model system that can be helpful in identifying carcino- genic agents and exploring their mechanisms of action. Few detailed dose-response curves for cell transforma- tion have been published as yet, but the morphological transformaton of Syrian hamster embryo cells by ben- zo(a)pyrene (BAP) (48,49) is consistent with one-hit kinetics except at cytotoxic dose levels (50). A one-hit model also holds for the transformation of such cells by the combined effects of x-rays and BAP (50). With x-rays alone, the frequency of transformation per surviving cell is increased by a dose as low as 10 mGy, above which it appears to increase curvilinearly with the dose up to 1.5 Gy; however, a linear increase over the same dose range cannot be excluded (51). Although the rate of transformation per unit dose typically decreases on pro- traction or fractionation of exposure to gamma rays, it may increase on protraction or fractionation of exposure to fast neutrons (Fig. 6). In C3H101/2 cells irradiated in vitro-as well as in thyroid and mammary "clonogens" irradiated in vivo (52)-"initiation" appears to occur with a frequency as high as 0.01-0.1 per cell per Gy (53) and to increase as a linear nonthreshold function of the dose (Fig. 7). The subsequent, final transforming event in such cells is far rarer, however, occurring at a rate of only 10-6 to 10-7 per cell generation (53,54).

872 ANNALS NEW YORK ACADEMY OF SCIENCES been observed to diminish with protraction whereas that of high linear energy transfer, radiation has been enhanced.69 Few comparable split-dose experiments have been performed with chemicals. Two doses of N-acetoxy-2-fluorenylacetamine administered 2-24 hr apart, how- ever, were observed to yield a higher frequency of transformation with Syrian hamster embryo cells than the same total dose administered at once. In contrast, methyl-N'-nitro-N-nitrosoquanidine, mitomycin C, and ultraviolet light were less effective if delivered in split doses than in a single dose. The effectiveness of methyl methanesulfonate was unaffected by dose fractionation.70 The morphological transformation of C3H 10T1/2 cells in vitro, like that of cells in vivo, is not a one-step process. The first step appears to be rapid event," occurring with one-hit kinetics in a high percentage of carcinogen-exposed cells.''--74 The second step appears to be either a further qualitative change, occur- ring at a low frequency during the growth or confluence of the cells,'}-76 or an amplification of the transformed phenotype, possibly by release of the cells from inhibitory effects of neighboring nontransformed cells.72,n,78 Clearly, further data will be needed to elucidate the mechanism of in vitro transformation and its relevance to carcinogenesis in vivo. Mutations and Chromosome Aberrations In view of the putative roles of mutations and chromosome aberrations as mechanisms of carcinogenesis, the dose-response relationships for these changes must be considered in assessing the risks associated with low-level exposure to carcinogens. The changes in DNA that are induced by ionizing radiation and genotoxic chemicals, which include single-strand and double-strand breaks, base altera- tions, cross-linkage, and other modifications, can result from traversal of the cell nucleus by a single ionizing particle79 or from interaction of the DNA with a single electrophilic molecule.80,81 Although a dose of low linear energy transfer radiation that is lethal to 50% of dividing cells (that is, 2.5 Sv) causes hundreds of DNA strands breaks per cell, much of the damage is reparable, depending on the effec- tiven.ess of the cell's repair processes.79 Such homeostatic repair processes are thought to enable the average cell to repair thousands of lesions in its DNA that occur "spontaneously" each day through the effects of natural background radia- tion, free radicals, and other degradative processes.82,83 In spite of repair, how- ever, the persistence of residual damage or the occurrence of lesions resulting from misrepair can give rise to mutations or chromosome aberrations or both, the frequency of which will depend on the amount and severity of DNA damage. The frequency of mutations at the guanine (hypoxanthine) phosphoribosyl transferase locus in human lymphocytes increases as a linear, nonthreshold func- tion of the X-ray dose over the range from 50 to 220 mSv, amounting to about six mutations per 106 cells per Gy, whether the dose is delivered in several fraction- ated exposures or in a single brief exposure.84 In X-irradiated mouse spermatogonia in vivo, the frequency of specific locus mutations increases as a linear-quadratic function of the dose, amounting to approximately six mutations per 106 cells per locus per Sv at low-to-intermediate doses and dose rates; with fast neutrons, the frequency of mutations increases more steeply, as a linear nonthreshold function of the dose, and independent of the dose rate.56

Plt.rip )tent Stem Cell.s in Renzene Toxicit~'. Toxiculucn l(1:163-1-i (19N0). t). (nsn. f.l)_: Srnder, C-A.: LoBue. 1.: et aL: ~cute and Chruni-.' I))se- He,lx)nse Effect.s of Inhaled Benzene on \tultilk)tentiul Ir,marc; I'ruietic Stem t CF('-S ) and Granuloc}-te/Macn)phage Progenitor l(;Sf CFf'-C1 Cells in CD-1 Mice. Toxicol. Appl. Pharmacol. St3:-02-50i (1~)N1). Ru,c'h. G. ,v1.: Leong. 13.i:.: ts<kin. S.: Benzene ~tetah >lism. J. Toxicol. Emirun. Health 2:23-36 (19'-). 8. Cr<mlcite, E.P.: Dren, R.T.c Irn>ve. T.; Bulli;. J.E.: 13enzene Heman>- tuxicin• and Leukemugenesis. Am. J. Ind. Med. 7:-r-t7--t56 (1985). 9. Cn>nkite. E.P.: Chemical Leukemogenesis: Benzene as a M<xiel Sem- inar Hematol. 2-t:2-11 (1987). 10. Re>'oen• M.G.; Snyder, CA: Protracted Exposure o.`C57BU61 Mice to 300 ppm Benzene Depresses B- and T-Lymphoc}•te Numbers and Mitogen Responses. Evidence for Thymic and Bone Marrow Prolif- erat:on in Response to the Exposures. Toxicology 37:13-26 (1985). 11. Aovama, K.: Effects of Benzene inhalation on Lymphocyte Subpop- ulatiuns and Immune Response in Mice. Toxicol. AppL Pharmacol. 85:92-101 (1986). 12. Ruzen, M.G.; Snyder, CA; Albert, RE.: Depression in B- and T-Lym- ph<xiae Mitogen-induced Blastogenesis in Mice Exposed to Low Cuncentration5 of Benzene. Toxicol. Lett. 20:343-349 (1984). 13. Snyc'er, CA: Goldstein, B.D.: Sellakumar, A.; et al.: Hematotoxicin• of Inhaled Benzene to Sprague-Dawlev Rat,s and AKR Mice at 300 ppm. J. Toxicol. Environ. Health 4:605-618 (1978). 14. Snyder, CA; Goldstein, B.D.; Sellakumar, AR; et al.: The Inhalation Toxicology of Benzene: Incidence of Hematopoietic Neoplasms and Hematotoxicity in AKR/J and C57BU6J Mice. Toxicol. Appl. Phar- macoL 54:323-331 (1980). 15. Maltoni, C.; Scamato, C.: First Experimental Demonstration of the Carcinogenic Effects of Benzene: Long-term Bioassays on Sprague- Dawley Rats bv Oral Administration. Med. Lav. 70:352-357 (1979). - 16. Maltoni, C.; Conti, B.; Cotti, G.; Belpoggi, F.: Experimental Studies on Benzene Carcinogeniciny of The Bologna Institute of Oncology: Current Results and Ongoing Research. Am. J. Ind Med. 7:415-446 (1985) 17. National Toxicology Program: NTP Technical Report on the Toxi- cology and Carcinogenesis Studies of Benzene (CAS No. 71-43-2) in F3Y~4/N Rats and B6C3F1 Mice (Gavage Studies). NTP TR 289. DHHS (NIH ) Pub. No. 86-2545. Research Triangle Park, NC (1986). 18. Cronkite, E.P.: Benzene Hematotoxicity and Leukemogenesi•a. Blood Cell. 12:129-131 (1986). 19. Schwetz, BA: A Review of the Developmental Toxicity of Benzene. In: Advances in Modem Environmental Toxicology, Vol. IV, Carcin- ogenicity and Toxicity of Benzene, pp. 17-21. MA Mehiman, Ed. Princcaon Scientific, Princeton, NJ (1983). 20. Litton Bionetics, Inc.: Unpublished data, November 1977 and De- centb :r 1978, Kensington, MD; cited in BA Schwetz, 21. 21. Kuna, RA; Kapp, RW.: The Embryotoxic/I'eratogenic Potential of Benzene Vapor in Rats. Toxicol. Appl. Pharmacol. 57:1-7 (1981). 22. Coate, W.B.; Hoberman, A.M.; Durloo, RS.: Inhalation Teratology Studv of Benzene in Rats. Adv. Modem Environ. Toxicol. 6:187-198 (19fk4). 23. Keller, KA; Snyder, CA: Mice Exposed in utero to Low Concentra- tions of Benzene Exhibit Enduring Changes in Their Colony-forming Hematopoietic Cells. Toxicology 42:171-181 (1986). 24. Ungvary, G.; Tatrai, E.: On the Embryotoxic Effects of Benzene and Its Alk}d Derivatives in Mice, Rats and Rabbits. Arch. Toxicol. 8:425-430 (1985). 25. Dean, Bj.: Recent Findings on the Genetic Toxicology of Benzene, Toluene, Xylene and Phenols. Mutat. Res. 154:153-181 (1985). 26. Lebowitz, H.; Brusick, D.; Matheson, D.; et al.: Commonly Used Fuels and ScdvenLS Evaluated in a Battery of Short-term Bioassays. Environ. Mutagen. 1:172-173 (1979). 27. Bartsch, H.; Malaveille, C.; Camus, AM.; et al.: Validation and Com- parative Studies on 180 Chemicals with S. ttphimurium Strains and V79 Chinese Hamster Cells in the Presence of Various Metabolizing Svste.ms. Mutat. Res. 76:1-50 (1980). 2£js Nest:rnann, E.R; Lee, E.G.H.; Matula, T.I.; et al.: Mutagenicity of Con- stituents identified in Pttlp and Paper Milk Effluents I'.,ing thc• ,ur6 nroru•/kallammul`cm-Micrc)u>me Atis;t<•. Mutat. Re,. -9.2f)3-2l2 ( t9i0). 29. Shimizu, M.: Yatiui, Ya Matnumcxtl, N.: Structural Sp<cificit} c)f Aru- matiC (.Ui111k)utldS <vith Special Reference to Mutagenic Activin, in krlnrorrelkr t q~ irnttrirun: A series of Chioro- or Fluuro-Nitn tnn'r.ene Derirttives. Stutat. Res. 116:21'-238 (1983). 30. ,%1cCarrc)ll. N.E.: Piper. C.E.: Keech. B.H.: Bacterial Microsu,pe•n.ion 4s~,a}s with Benzenemxl OtherOrganicSt)Iventa. Ernirun. Mutagcn. 2:281-282 (1980). 31. McCarr<)Il, N.E.: Piper• C.C.: Keech, B.H.: An F.. cnli Micrususpen,iun A,sav for the Detection of DNA Damage Induced by Direct;tcting Agents and Promutagens. Environ. Mutagen. 3:429--t=i4 (1981). 32. McCarroll. N.E.; Keech, B.H.; Piper, C.E.: A Microsuspension Adap- tation of the Baci!ltasuhtfTis'rec As.say. Environ. Mutagen. 3:607-616 (1981). 33. Rozenkranz, H.S.; Leifer, Z: Determining the DNA Modifying Activity of Chemicals Using the DNA Polymerase-Deficient Fscilericbia coli. In: Chemical Mutagens: Principles and Methods for their Detection, Vol. 6, pp. 109-147. F.J. deSerres and A Hollaender, Eds. Plenum, New York (1980). 34. Parry, J.M.: Summary Report on the Performance of the Yeast and Aspergillus Assay. In: Evaluation ofShort-Term Tests for Carcinogens: Report of the International Program on Chemical Safety Collaber rative Study on In Vitro As,says, pp. _5-46. J. AShby, Fj. deSerres, M. Draper, et al., Eds. Elsevier, Amsterdam (1985). 35. Parry, J.M.; Eckardt, Fj.: The Induction of Mitotic Aneuploidy, Point Mutation and Mitotic Crossing-over in the Yeast, Saccharoml 'ces cere- risiae Strain D61-M and D6. In: Evaluation of Short-Term Tests for Carcinogens: Report of the International Program on Chemical tiafety Collaborative Study on In Vitro Assays, pp.261-269. J. Ashby, Fj. deSerres, M. Draper, et al.: Eds. Elsevier, Amsterdam (1985). 36. Nylander, P.O.; Olofsson, H.; Rasmuson, B.; Savahlin, H.: Mutagenic Effects of Petrol in Drosophila melanogaster. I. Effects of Benzene and 1,2-Dichloroethane. Mutat. Res• pp. 163-167 (1978). 37. Kale, P.G.; Baum, J.W.: Genetic Effects of Benzene in !h•awphila melanogaster Males. Environ. Mutagen. 5:223-226 (1983)- 38. Vogel, E.W.: Summary Report on the Performance of the Drosophila Assays. in: Evaluation of Short-Term Test,s+for Carcinogens: Report of the International Program on Chemical Safety Collaboration Study on In Vitro Assays, pp. 47-57. J. Ashby, F j. deSerres, M. Draper, et al., Eds. ELsevier, Amsterdam (1985). 39. Fujikawa, K; Ryo, H.; Kondo, S.: The Drosophrta Gene Mutation and Small Deletion As.say Using the Zeste-White Somatic Eye Colour System. In: Evaluation of Short-Term Tests for Carcinogens: Report of the International Program on Chemical Safety Collaborative Study on In Vitro Assays, pp. 319-324. J. Ashby, Fj. deSerres, M. Draper, et al., Eds. Elsevier, Amsterdam'(1985). 40. Vogel, E.W.: The Drasopbila Somatic Recombination and Mutation Assay Using the White-Coreal Somatic Eye Colour System. In: Eval- uation of Short-Term Tests for Carcinogens: Report of the Intema- tional Program on Chemical Safety Collaborative Study on In Vitro Assays, pp. 313-317, J. Ashby, Fj. deSerres, M. Draper, et al., Eds. Elsevier, Amsterdam (1985). 41. Lyang, J.C.; Hsu, T.C.; Henry, J.E.: Cytogenetic Assays for Mitotic Poi- sons: The Grasshopper Embryo System for Volatile Liquids. Murat. Res. 113:467-479 (1983). 42. Garner, RC.: Summary Report on the Performance of Gene Mutation Assays in Mammalian Cells in Culture. In: Evaluation of Short-Term Tests for Carcinogens: Report of the Intemational Program on Chem- ical Safety Collaborative Study onln Vitro Assays, pp. 85-94. J. Ashbv, Fj. deSerres, M. Draper, et al., Eds. Elsevier, Amsterdam (1985). 43. Howard, CA; Sheldon, T.; Richardson, C.R: Chromo.somal Analysis of Human Lymphocytes Exposed in t4tro to Five ChemicaLs. In: Evaluation of Short-Term Tests for Carcinogens: Report of the In- temational Program on Chemical Safety Collaborative Study on In Vitro Assays, pp. 457-467. J. Ashby, Fj. deSerres, M. Draper, et al., Eds. Elsevier, Amsterdam (1985). 44. Danford, N.D.: Tests for Chromosome Aberrations and Aneuploidy in the Chinese Hamster Fibroblast Cell Line CHI-L In: Evaluation of Short-Ternt Tests for Carcinogens: Report of the international r APPL OCCU'P. fNVlRON. HYG. 50 • JULY 1990 461

878 ANNALS NEW YORK ACADEMY OF SCIENCES TABLE 2. Estimated Human Risks from Ingestion of 0.12 G/Day of Saccharin Method of High- to Low-Dose Extrapolation Lifetime Cases per Million Exposed Cases per 50 Million per Year Rat dose adjusted to human dose by surface area rule Single-hit model 1,200 840 Multistage model (with quadratic term) 5 3.5 Multihit model 0.001 0.0007 Mantel-Bryan probit model 450 315 Rat dose adjusted to human dose by mg/kg/day equivalence Single-hit mod'eI 210 147 Multihit model 0.001 0.0007 Mantel-Bryan probit model 21 14.7 Rat dose adjusted to human dose by mg/kgl lifetime equivalence Single-hit model 5,200 3,640 Multihit model 0.001 0.0007 Mantel-Bryan probit model 4,200 2,940 NOTE: Adapted from Reference 105. occurring" cancer, then exposure to only a small dose of the chemical can be expected to increase the incidence by some finite amount.z•107-109 For this reason, the use of a nonthreshold model is generally recommended in risk assessment when the mode of action of the carcinogen in question is not known. CONCLUSIONS The possibility that there may be no threshold for the induction of some forms of cancer by ionizing radiation or certain chemicals, at least in appropriately susceptible individuals, is suggested by (1) evidence that most cancers arise from a single transformed cell; (2) the heritable nature of the transformed phenotype; (3) the association between neoplastic transformation and specific mutations or chromosomal aberrations; (4) the correlation between carcinogenicity and geno- toxicity; (5) the nature of the observed dose-response relationships for mutations, chromosomal aberrations, and cell transformation in vitro; and (6) the nature of the dose-incidence relationships for certain neoplastic lesions in vivo. At the same time, however, carcinogenesis appears to be a multistage process involving the stepwise evolution of increasingly autonomous cells in which the outcome is influenced by such variables as age, genetic constitution, physiological state, metabolism, and homeostatic interactions within and among tumor-forming cells and normal cells. Other variables that complicate analysis of dose-incidence relationships are (1) poorly defined interactions among cancer-causing agents, which may be additive, multiplicative, or antagonistic in their combined effects; (2) the fact that the human environment contains myriads of agents, many of which are known to modulate the effects of others; (3) the existence of nonlinear kinetics in the metabolism of certain chemical carcinogens; and (4) evidence that some agents act primarily through mechanisms that presumably operate only at high dose levels.

274 Although the mechanisms of action of carcinogens of different types are still to be defined precisely, the ex- isting data suggest that a linear nonthreshold interpola- tion model may be appropriate only for an initiating agent or a complete carcinogen, and that a model yielding a smaller estimate of the risk at low doses is more likely to be appropriate for a promoting agent. Similarly, for a chemical that is activated through nonlinear metabolic processes (57) or that acts through toxic effects elicited only at relatively high doses (e.g., immunosuppression) (58), a thresholad or quasithreshold dose-incidence model is likely to be more appropriate. In view, howt:ver, of the existence within the human population of individuals who vary widely in their suscep- tibility to cancer, as well as those who are at different stages of carcingenesis as a result of the action of other cancer-causing agents or risk factors, it is assumed that a carcinogen may pose some degree of risk to the popula- tion at any dose, by exerting carcinogenic effects that are additive with those which account for the "spontaneous" baseline incidence of cancer (Fig. 11). Hence, unless an agent can be shown to act through effects that are not addi- tive with those which account for the "spontaneous" baseline incidence of cancer, a nonthreshold model is generally recommended for assessing the carcinogenic risks of the agenr for public health purposes. ProbabNRy of Cancer Figtnre 11. Diagram illustrating the expected increment in risk of cancer resulting from a lowdose of a hypothetical carcinogen. Because cellular effects similar to those of the carcinogen may be produced in its absence by "background" mechanisms, the effects resulting from low doses of that carcinogen may be additive with those resulting from other "back- ground" risk factors, thu;; causing an increase in the risk that is propor- tional to the dose. (From Ref. 70.) Upton ACKNOWLEDGMENT Preparation of this report was supported in part by Grants ES 00260 and CA 13343 from the U.S. Public Health Service and Grant S[G-9 from the American Cancer Society. The author is grateful to Mrs. Lynda Witte for assistance in the prepar- tion of this report. Address reprint requests to Arthur C. Upton, Institute of Environmen- tal Medicine, New York University Medical Center. 550 First Avenue. New York, New York 10016. REFERENCES 1. Muller HJ: The manner of production of mutations by radiation. In Radiation Biology Vol. 1. High Energy Radiation. Edited by A Hollaender. McGraw-Hill, New York, 1954, pp 475-626. 2. United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR). Report of the Scientiftc Committee on the Effects of Atomic Radiatiott. United Nations, New York, 1958. 3. United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR). Genetic and Somatic Effect of Ionizing Radiation. Report to the General Assembly, with Annexes. United Nations, New York, 1986. 4. Auerbach C: Chemical mutagenesis. Biol Rev 24:355-391, 1949. 5. Ehling UH, Averbeck D, Cerutti PA, et al: Review of the evidence for the presence or absence of thresholds in the induction of genetic effects of genotoxic chemicals. International Commission for Pro- tection Against Environmental Mutagens and Carcinogens. ICPEMC Publication No. 10. Mutat Res 123:281-341, 1983. 6. Lewis EB: Leukemia and ionizing radiation. Science 125:965-975. 1957. 7. Scherer E, Emmelot P: Multihit kinetics of tumor cell formation and risk assessment of low doses of carcinogen. In Carcinogens: Identification and Mechanisms ofActron. Edited by AC Griffin, CR Shaw. Raven Press, New York, 1979, pp 337-364. 8. Zeise L. Wilson R, Crouch EAC: The dose response relationships for carcinogens: a review. Env Health Perspect 1988 73:259-306, 1987. 9. Fialkow PJ: Clonal origin of human tumors. Biochim Biophys Acta 458:283-321, 1979. 10. Ponder BAJ: Genetics and cancer. Biochim Biophys Acta 605: 368-410, 1980. 11. Sandberg AA: A chromosomal hypothesis ofoncogenesis. Cancer Genet Cytogenet 8:277-285, 1983. 12. Farber E, Sarma DSR: Biology of disease. Hepatocarcinogenesis: A dynamic cellular perspective. Lab Invest 56:4-22, 1987. 13. Nicholson GL: Tumor cell instability, diversification, and progres- sion to the metastatic phenotype: from oncogene to oncofetal ex- pression. Cancer Res 47:1473-1487, 1987. 14. Barrett JC, Crawford BD, Ts'o POP: The role of somatic muta- tion in a multistage model of carcinogenesis. In Mammalian Cell Transformation by Chemical Carcinogens. Edited by N Mishra, VC Dunkel, M Mehlman, Senate Press, Princeton Junction, NJ. 1980, p 467. 15. Farber E: Cellular biochemistry of the stepwise development of cancer with chemicals: G.H.A. Clowes Memorial Lecture. Cancer Res 44:5463-5474, 1984. 16. Barrett JC, Elmore E: Comparison of carcinogenesis and muta- genesis of mammalian cells in culture. In Handbook ofErperimental

COMPREHENSIVE CANCER CENTERS The institutions Iisted have been recognized as Comprehensive Cancer Centers by the National Cancer Institute. These centers have met rigorous criteria imposed by the National Cancer Advisory Board. They receive financial support from the National Cancer In- stitute, the American Cancer Society and many other sources. ALABAMA ILLINOIS Roswell Park Memorial Institute University of Alabama Illinois Cancer Council 666 Elm Street Comprehensive Cancer Center 36 South Wabash Avenue Buffalo, New York 14263 1918 University Boulevard, Room 108 Chicago, Illinois 60603 Phone: (716) 845-4400 Birmingham, Alabama 35294 Phone: (312) 226-2371 Phone:(2051934-6612 NORTH CAROLINA University of Chicago Cancer Duke University Comprehensive CALIFORRI IA Research Center Cancer Center University of Southern California 5841 South Maryland Avenue P.O. Box 3814 Comprehensive Cancer Center Chicago, Illinois 60637 Durham, North Carolina 27710 Kenneth Norris, Jr. Hospital & Research Phone:(312)702-6180 Phone:(919)684-6342 or286-5515 Institute 1441 Eastlake Avenue MARYLAND OHIO Los Ange{es, California 90033-0804 The Johns Hopkins Oncology Center Ohio State University Comprehensive Phone: (213) ;!26-2370 600 North Wolfe Street Cancer Center Baltimore, Maryland 21205 410 West 12th Avenue, Suite 302 )onsson CcmDrehensive Cancer Center Phone: (301) 955-8638 Columbus, Ohio 43210 10-247 Factor Building Phone: (614) 293-8619 10833 Le Conte Avenue MASSACHUSETTS Los Angeles, Ca(ifornia 90033 Dana-Farber Cancer Institute PENNSYLVANIA Phone: (213) 825-8727 44 BinneyStreet Fox Chase Cancer Center Boston, Massachusetts 02115 7701 Burholme Avenue CONNECTICUT Phone:(617) 732-3214 Philadelphia, Pennsylvania 19111 Yale Unaversity Phone: (215) 728-2570 Comprehensive Cancer Center MICHIGAN 333 Cedar Street Meyer L Prentis Comprehensive Cancer University of Pennsylvania Cancer Center New Haven, Connecticut 06510 Center of Metropolitan Detroit 3400 Spruce Street Phone:(203)785-6338 110 East Warren Avenue 7th Fioor, Silverstein Pavilion Detroit, Michigan 48201 Philadelphia, Pennsylvania 19104 DISTRICT OU C:OLUMBIA Phone: (313) 833-0710, ext. 429 Phone: (215) 662-6364 Howard University Comprehens ive Cancer Center MINNESOTA TEXAS 2041 Georgia Avenue, N.W. Mayo Comprehensive Cancer Center The University of Texas Washington, D.C. 20060 200 First Street, S.W. M.D. Anderson Cancer Center Phone:(202) 636-7610 or 636-5665 Rochester, Minnesota 55905 1515 Holcombe Boulevard Phone:(507)284-3413 Houston, Texas 77030 Vincent T. Lombardi Cancer Phone: (713) 792-6161 Research C:enter NEW YORK Georgetown University Medical Center Columbia University Cancer Center WASHINGTON 3800 Reservoir Road, N.W. College of Physicians and Surgeons Fred Hutchinson Cancer Research Center Washington, 0.,Z. 20007 630 West 168th Street 1124 Columbia Street Phone: (202) 687-2110 New York, New York 10032 Seattle, Washington 98104 Phone:(212)305-6730 Phone: (206) 467-4675 FLORIDA Papanicolaou Comprehensive Memorial Sloan-Kettering Cancer Center WISCONSIN Cancer Centei~ 1275 York Avenue Wisconsin Clinical Cancer Center University of Miami Medical School New York, New York 10021 University of Wisconsin 1475 N.W. 12th Avenue Phone: (212) 525-2225 600 Highland Avenue Miami, Florida 33136 Madison, Wisconiin 53792 Phone: (305) rrt8•4850 Phone:(608)263-6872 31

UPTON: THRESHOLDS FOR CARCINOGENESIS? 871 00 r 75 FIGURE 7. Cumulative incidence of neoplasms of the liver and urinary bladder in female BALB/c mice exposed to 2- acexylaminofluorine at various concentrations in the diet for up to 33 months. (Reproduced from Littlefield et al.s2) Cell Transformation in Vitro i ~ 8iobee P 25 i P7 0 0 50 100 150 Dose (ppm) The neoplastic transformation of cells in vitro is not a perfect model of carci- nogenesis in vivo, but the two systems have enough features in common at the ceiflular level so that cell transformation can be exploited to identify carcinogenic agents and explore their mechanisms of action. Although few detailed dose- response curves for cell transformation have been published, the effects of ben- zo[a]pyrene and ionizing radiation have been studied systematically. With ben- zo[a]pyrene, the logarithm of the frequency of morphological transformation in Syrian hamster embryo cells increases linearly with the logarithm of the dose59,60; the slope of the dose-effect curve suggests a one-hit model for this response except at the highest doses, where the deviation from linearity is attributable to cytot:oxicity.61 A one-hit model also holds for transformation by the combined effects of X-rays and benzo[ajpyrene.61 For Syrian hamster embryo cells transformed by X-rays, the logarithm of the transformation frequency per surviving Syrian hamster embryo cell appears to increase curvilinearly with the logarithm of the dose from I rad to 150 rad, but a linear response with a slope of one cannot be excluded. It is noteworthy, further- more, that an increase in the frequency of cell transformation is detectable at a dose ~of only I rad.62 Dose-response curves for X-ray-induced transformation of C3F[ 10T1/2 cells show an exponential increase in transformation frequency (foci per su)rviving cell),63,64 with a doubling dose that is higher (about 100 rad) than the doubling dose in hamster cells (about 10 rad). The effects of fractionating or protracting the dose of radiation vary with the experimental conditions in question. With a total dose of less than 100 rad, frac- tionation has been observed to enhance its transforming effectiveness,65--67 whereas the opposite effect has been observed with higher doses (300-800 rad) 63,sa.s6.s1 The transforming effectiveness of gamma radiation has generally FIGURE 8. Dose-response curves depicting the incidence of tumors in laboratory animals in relation to the dose and dose rate of high and low linear energy transfer radiation. (Reproduced from Thomson et al.s' [also in Updon et al.54}.) DosE-->

Reprinted from Living in a Chemical World Volume 534 of the Annals of the New York Academy of Sciences June 30, 1988 Are There Thresholds for Carcinogenesis?a The Thorny Problem of Low-Level Exposure ARTHUR C. UPTON Institute of Environmental Medicine New York University Medical Center New York, New York 10016 INTRODUCTIOht Few issues in health policy are more contentious than the choice of the appro- priate dose-incidence model for use in estimating the risks of cancer associated with low-level exposure to a carcinogen. The notion that there may be no thresh- old fDr carcinogenic effects-namely, that some degree of risk may be associated with the lowest dose of carcinogen-seems to contradict everyday experience that teaches us that essentially no other type of insult produces a lasting injury unless it exceeds some threshold of severity. In the past, toxicological risk assessments have traditionally been based on the concept of a no-effect level. The applicability of this concept to mutagenic effects, however, came to be questioned by the middle of the century.' Since then, the applicability of the concept to carcinogenic effects-which likewise may conceiv- ably be mediated through effects on individual cells, rather than groups of cells- also has been challenged. i[n principle, of course, it is not possible to prove or disprove the existence of an absolute threshold for carcinogenesis. Hence the argument for or against the threshold hypothesis must be based on theoretical as well as empirical evidence.2 Some of the salient lines of evidence are summarized briefly in the following. BIOLOGY OF CARCINOGENESIS Unicellular, Monoclonal Origin of Cancer The monoclonal origin of cancer is suggested by enzymological studies of human tumor cells, in which X-linked glucose-6-phosphate dehydrogenase has been used as a marker.3 Similar evidence has come from studies of chemically induced tumors of chimeric mice, in which glucose phosphate isomerase has been used as a marker.4,5 Cytogenetic analysis of tumor cells has also suggested their monoclonal nature.6 The evidence that cancer usually originates from a single precursor cell im- plies, as does the heritable nature of the malignant phenotype, that appropriate damage to one cell alone may suffice to increase the probability of the disease in a a F'reparation of this report was supported in part by Grants ES 00260 and CA 13343 from the U.SI. Public Health Service and Grant SIG-9 from the American Cancer Society. 863

species (59). Certain chemicals, notably epigenetic (non-DNA altering) ones, may affect a particular stage of the multistep carcino- genic process initiated by another chemical without being them- selves active in a long-term bioassay when tested alone. These facts, together with the increasing costs and slowness of long-term testing, have forced consideration of assays that require less time. Weisburger and Williams (59) outlined a decision-point approach to testing whereby chemicals might be analyzed in four increasingly complex classes of carcinogerucity assessment. These classes are as follows: (i) 2ulalysis of the structure of the chemical. This analysis considers the reactivity of the chemical and its metabolites based on structure (66). (ii) Short-term tests in vitro. A battery of tests is used including prokaryotic and mammalian mutagenesis systems and studies of direct effects on DNA and chromosomes. (iii) Limited bioassays in vivo. The formation of preneoplastic lesions or rapid tumor induction is assessed in selected species. (iv) Long-term bioassays in vavo. A positive result in these studies is increased overt tumor formation or tumor-induced death of the animal. For limited b:ioassay procedures, the induction of breast cancer in female Spragu•e-Dawley rats and the induction of altered foci in the rat liver may be sseful. Cellular and subcellular preparations from rat livers are also commonly used for metabolic activation of chemicals in short-term carcinogenesis and mutagenesis tests (67, 68). Cocul- ture of rat hepatocytes with liver epithelial-type cells has been reported to sustain high levels of hepatocyte, carcinogen-metaboliz- ing cytochrorne P-450 enzymes (69). Such procedures may extend the utility of r"n vitro hepatocyte cell lines in toxicity testing. The comprehensive assessment proposal of Weisburger and Williams (59) is not an established procedure (58), but rather illustrates potential future directions for carcinogenic risk assessment. The rat plays an important role in short-term in vitro tests and in limited in vivo bioassays. The rat has been the most frequently studied species in the in vivo bioassay syster,a of altered liver-focus induction. Research into the cellular events in ttie course of chemically induced tumor formation has characteriza:d many of the changes that precede malignancy (70, 71). Cell populations affected by the carcinogen generally appear as characteristically altered foci detectable by sensitive immunohis- tochemical reactions, and they appear much earlier than tumor formation. Induction of such foci is not an unequivocal indicator of ultimate malignancy, and their significance in the development of malignancy is debated (70). Nevertheless, this assay has been proposed as a l'v tuted in vivo bioassay system in carcinogenicity assessment (59, 70, 72). Pereira and Stoner (73) have reported that the rat liver focu: assay exhibited greater sensitivity and fewer false negatives that the strain A mouse lung adenoma assay [some limitations of which are discussed in (53)] in detecting genotoxic carcinogens. Parcdi et al. (74) concluded that, at least for a small group of chemica1s active predominantly in the liver, assays for liver focus and nodule formation were as accurate, and possibly more accurate, in detecting carcinogenicity than was the Ames test. Preneoplastic lesions have been studied in tissues other than the liver, but a systematic evaluation of their use in bioassays has not been reported (7S). In view of the large amount of knowledge concerning liver focus formation in the rat (72), it is clear that this species will featr.Ire prominently in potential bioassay applications. Strains of rats cartying the growth and reproduction complex (grc), which is linked to the MHC, exhibit enhanced focus formation compared to wiP•d•type rats when exposed to chemical carcinogens (21, 76), and they ,are candidates for development of highly sensitive liver-focus bioassays. In the future of carcinogenicity assessment, there is increasing interest in subdividing the carcinogenic process and studying indi- viduai stages. As more is learned about the multistep mechanisms, it may be possible to develop assays for the identification of agents that predispose cells to malignancy at specific steps in the process; one such system has already been described for the rat (61). With the increasing emphasis on genetic mechanisms in carcinogenesis, the availability of randomly bred, outbred, inbred, and congenic strains of rats (3-5) will make this species even more usefiil in risk assessment as well as in studies on the basic mechanisms of carcinogenesis. Cardiovascular DiSCmeS The extensive body of knowledge regarding nutrition, endocri- nology, metabolism, and physiology; the detailed studies on anato- my and histology; and the convenient size of the rat make it a particularly useful experimental animal for cardiovascular research. Reproducible, genetically determined abnormalities have been dis- covered in rat populations that have proven useful in examining the cardiovascular effects of hypertension, obesity, diabetes, and other metabolic diseases (4, 77) and a variety of congenital abnormalities of the cardiovascular system (78). Early studies indicated that this species was quite different from humans in its serum lipid and lipoprotein constitution and that it was very difficult to produce sustained hyperlipidemia in the rat (79). Until approximately 1950, many attempts to produce athero- matous lesions in the rat had failed in spite of the extensive knowledge about the effects of nutritional manipulation in this species. Then, in the early 1950s simultaneous reports from three laboratories indicated that this resistance could be overcome under the proper experimental conditions (80-82). Each study was de- signed to capitalize on the newly emerging concepts of risk factors for atherosclerosis, and each utilized rats whose resistance to atherogenesis was diminished by unique ways of producing hyper- cholesterolemia. Hartroft and his colleagues (80) and Wissler and his group (81) fed rats special diets designed to raise their blood cholesterol levels and then induced hypertension or renal disease or fed the rats chemicals such as propyl thiouracil and sodium cholate. Malinow and his associates (82) utilized particularly potent dietary imbalances plus thyroid-depressing agents to induce atherosclerotic lesions. Some of the major findings emerging from these studies were the greater involvement of the coronary arteries than of the aorta, the location of the aortic lesions in the proximal part of the MHC grc 3-4 cM ~ 0.4 cM O.0r cm 1-~ Neu-1 f ~....-.......~~ C2,04 ,------~ Glo-1 Acry-1 A Pa F H B D E ft dw-3 G C I (Ha,(S) (BP,, RZ) Ba (Da,s) Fig. 1. The major histocompatibility complex of the rat. 0, Class I major (classical) transplantation antigens; the dashed squares, the class I medial transplantation antigens; 0, class II antigens; !, loci controlling polymor- phic proteins (Glo-l, glyoxylase I; Acry-1, a-crystallin-1); and +, the loci of thegrc ( ft, fertility; dw-3, dwarf-3). The loci indicated by brackets have been mapped to the regions indicated (Neu-1, neuramuzidase-i; C, complement components). The evidence for this mapping is presented in (3, 12, 13). A cytogenetic study (15) has placed the MHC on chromosome 14 of the rat. 21 JULY 1989 ARTICLES 273

UPTON: THRESHOLDS FOR CARCINOGENESIS? 875 of exposure; 5) the significance of benign, as opposed to malignant, tumors; 6) the precise nature of the dose-incidence relationship; and 7) the significance of nega- tive results. On r.he basis of present knowledge, the carcinogenicity of an agent for human tissue cannot be predicted accurately by extrapolation from animal data. A chemi- cal that causes tumors in a particular organ of one species may cause tumors in another organ, or no tumors at all, in other species; for example, bioassay results in the mouse have been predictive for the rat in only about 80% of cases, and vice versa.18-99 The problem is complicated further by the fact that the human popula- tion is exposed to myriads of agents interacting in various ways, whereas animals in the standard bioassay are ordinarily exposed to only one agent at a time. ' The dose-incidence models used by national and international experts for estimating the carcinogenic risks of low-level ionizing radiation are generally of the nomthreshold type.27-29•10° The models also allow, however, for the fact that the magnitude of risk per unit dose appears to vary with the form of cancer, sex, age at irradiation, type of radiation (linear energy transfer), dose, and dose rate. In view ~of these differences, each type of neoplasm is generally considered indi- vidually, with efforts to integrate insofar as possible all relevant epidemiological and experimental data. Although the relation between incidence and radiation dose is known to vary from one type of neoplasm to another, the observed effects of dose rate and linear energy transfer on the dose-incidence relation generally conform to the patterns illustrated in FIGURE 8, which are consistent with those expected if one were to assume that carcinogenesis could be initiated in a suitably susceptible individual by a mutation or chromosomal aberration in a single somatic cell. According to this interpretation, the dose-incidence curve for high linear energy transfer radia- tion woul.d be expected to conform, in general, to the expression I = C + aDE-pD (2) where I is the incidence at dose D, C is the incidence in nonirradiated controls, and a and p are constant coefficients; for low linear energy transfer radiation, the dose-incidence curve would conform, in general, to the expression , I=(C+aD+bD2)e-(pD+qD2) (3) where the symbols are comparable to those above, except for different values of the coefficients a and p and an additional coefficient q.tol Although many of the observed dose-incidence curves conform to the pat- terns described above, the curve for breast cancer appears more nearly linear, and the curve for osteosarcomas induced by radium-226 appears more nearly qua- dratic.28 'Because of the complex, multicausal, multistage nature of carcinogene- sis, no one simple model is likely to characterize the dose-incidence relation over a wide range of doses and exposure conditions. At intermediate-to-high doses, a complete carcinogen can be expected to exert promoting effects as well as initiat- ing effects on tumor formation through alterations in cell population kinetics and other changes. At still higher doses, the response can be expected to saturate because of cytotoxicity. In vie w of uncertainty about the shape of the dose-incidence curve at low doses and low dose rates, various hypothetical models have been used in an effort to arrive at a range of estimates for assessing the risks of low-level radiation (FIG. 10) and chemicals (FIG. 11).

87® ANNALS NEW YORK ACADEMY OF SCIENCES appearance t under conditions of daily exposure tends to vary inversely with the daily dose d according to the function dt" = constant (1) where n is greater than one55; (6) because radiation or a given chemical can often influence carcinogenesis through more than one mode of action, at least at high dose levels, the dose-incidence curve may reflect a combination of initiating effects, promoting effects, and anticarcinogenic effects, depending on the particu- lar agent, dose, and exposure conditions; (7) the combined effects of different agents may be additive, synergistic, or antagonistic, depending on the agents in question and the conditions of exposure56; (8) at low-to-moderate dose levels, the effects of a complete carcinogen can generally be accentuated by appropriate tumor-promoting stimuli, which unmask initiating effects of the carcinogen that would otherwise remain unexpressed; and (9) under conditions in which initiating Dose-Response for Initioting Acirvity of Benzo(a)pyrene 1.0 C ff' 0 0001{ , d .. i -i 0.01 0.1 10 t0 t00 BaP (µ9) FIGURE 6. The yield of skin papillomas per mouse versus dose per application, after single and multiple doses of benzo- [alpyrene. After treatment, 5 µg of 12-O-tetradecanoylphorbol- 13-acetate was typically applied three times per week. (Repro- duced from Bums and Alberts' with permission from Mary Ann Liebert, Inc.) effects are promoted to full expression, they generally increase as a linear non- threshold function of the dose of the initiating agent (FIG. 6). A number of experiments have been carried out with laboratory animals to characterize the dose-incidence curve in the low dose domain. In the largest of the experiments to date, performed with BALB/c female mice exposed to 2- acetylaminofluorine in the diet, the incidence of hepatomas increased as a linear nonthreshold function of the daily dose, whereas the dose-incidence curve for t.umors of the urinary bladder approached a quasithreshold and resembled a hockey stick in shape (FIG. 7). Comparably large experiments have not been carried out with ionizing radiation, but the combined results of a number of s4zable studies in mice, rats, and dogs imply that for most types of tumors (malig- nant as well as benign) the carcinogenic effectiveness per unit dose of X-rays and gamma rays is generally reduced at low doses and low dose rates, whereas that of high linear energy transfer radiations remains constant or may even be enhanced at low doses and low dose rates (FIG. 8), arguing against the likelihood of a threshold in such instances.z'.57,5s

xenograft, (iii) the cumulative effect that weak antigenic systems have in xenograft rejection, and (iv) the genesis and nature of "naturaP' or "preforrn<:d" antibodies. As an extension of this line of work, the role that the evolutionary distance between donor and recipient plays in the magnitude of the immune response to the xenograft should be examined. Second, the immune response to the xenograft should be analyzed systematically and in detail, including an investigation of the origin and specificities of preformed antibod- ies. The latter study may provide some insight into methods for controlling their formation. Third, the mechanism of xenograft rejection should be compared to that of allograft rejection to determine whether the major differences between them are qualita- tive or quantitative. Reproductive immunology and genetics. This area has as its central theme the mechanism by which the fetal allograft survives (48). The rat is an important exp:rimental animal for examining the nature of the trophoblast antig(:ns and the genetic control of their expression. The allele-specific, class I transplantation antigens are not expressed on the trophoblast surf ice in allogeneic pregnancies, but they are on the surface in syngeneic pregnancies; in both types of pregnancies, they are present in the cytoplasm (18). The Pa antigen is expressed on the trophoblast surf tce and in the trophoblast cytoplasm in both allogeneic and syngeneic placentas; class II antigens are not ex- pressed in either type of placenta (18). This differential antigen expression may be an important factor in the maternal acceptance of the allogeneic placenita, Recent work shows that all of the class I antigens expressed in the placenta are of paternal origin, and this is the first example at the antigen level of genomic imprinting, which is a critical process in reproductive genetics (49). The very low level of MHC antigen polymorphism in the rat is crucial to the discrimina- tion needed for these types of studies. Recessive lethal ge:n:s are important causes of fetal death in experimental animals, and they may play an important role in recurrent spontaneous bortion in humans (48, 50). The grc in the rat, as discussed above, provides a unique model system in which to study these effects. This area of research is an important bridge between the aspects of reproduction of primary interest in the field of transplantation and. the broader field of developmental genetics. Risk Assessment for Potential Carcinogens The rat has been used frequently for prediction of the effects of chemicals on humans (51). For studies of teratogenesis, the advan- tages of the rat include the ease of counting corpora lutea when assessing the effects of chemicals on ovulation and implantation (52), a large litter size„ a short gestation period, and a well-studied embryology. However, the susceptibility and sensitivity of rats to particular teratogenic agents may be low when compared with the mouse and the rabbit (52), and there are significant differences from man in the effects of chemicals on the fetus (53). In mutagenesis studies, the rat appears to offer little inherent advantage over several other species (54). It is in the field of carcinogenic risk assessment that the rat has played a prominent role and will continue to do so. Prediction of carcinogenicity for a given chemical is a major concern for government, the chemical industry, and the public. The development of cancer usually involves, at some stage, an agent or agents foreign to the cell-including xenobiorics, radiation, and oncogenic viruses. Carcinogenesis is a multistep process frequently involving a genotoxic (DNA-altering) step resulting in the alteration of cell division, growth, and differentiation (55). Di$erent chemi- cals, including some with similar structures, may work by different mechanisms, and the ce:llular differences among tissues further complicate the process. Often one, or sometimes more, specific activated metabolite of a chemical may be the ultimate carcinogen (56); hence, different tissues and species of animals may respond differently to any given chemical based on their inherent metabolic patterns. The many unknown aspects of the induction of cancer, the long latency period between exposure and overt disease, and the potential for carcinogenesis at low doses of chemicals have made risk assessment an extremely difficult exercise. Ultimately, it is epidemiologic studies of humans that will confirm the ability of an agent to cause human cancer (57), but such studies are usually performed only after exposure of large popula- tions. This situation has led to the development of carcinogenic risk assessment methodologies that utilize nonhuman test systems (53). Assessment of carcinogenicity involves long-term dietary, parenter- al, or topical application of the chemical to various mammalian species (58). The rat features prominently in such studies because of a favorable combination of small body size, ease of breeding, and relatively low spontaneous tumor rates. The choice of the strain of rats that is used is important in view of the variation in spontaneous tumor rates and different responses to chemicals among inbred strains (58). More recently, it has become apparent that such long- term bioassays may occasionally produce conflicting results, as occurred initially with vinylidene chloride (59, 60), or may be used with agents such as arsenic that exhibit sufficient evidence of carcinogenicity in humans but limited evidence in animal tests (60). Furthermore, because the mechanisms of chemical carcinogenesis have become better understood and the potential for simultaneous exposure to several chemicals has become apparent, chemicals may in the future be assessed for their activity at different stages of the multistep carcinogenic process (61). The long-term application of a test chemical to animals will continue to be the fundamental method of carcinogenic risk assess- ment because short-term, and particularly in vitro, tests cannot mimic all of the aspects of animal metabolism and physiology (62). The long-term bioassays should be done over a large part of the life span of the species, starting in utero, in order to eliminate false negative results due to the prolonged latency of carcinogenic effects. In this respect, the rat is a suitable experimental animal because of its relatively short life span. In view of the important role played by metabolic enzymes in activating chemicals to reactive carcinogens, the question arises as to whether the rat is metabolically an appropriate substitute for humans. Crouch and Wilson (63), using the National Cancer Institute long-term bioassay data and a mathematical formula for carcinogenic potency, demonstrated that the ratio of potency be- tween humans and rats was, on average, within a fivefold range; however, for a given chemical it varied from 1500:1 to 0.02:I. The range of potencies was less divergent between mice and rats, although Bernstein et al. (64) have argued that this lack of divergence may be a statistical artifact inherent in the long-term bioassays. Purchase (65) analyzed 250 chemicals for carcinogenicity in rats and mice based on data from the National Cancer Institute, Internation- al Agency for Research on Cancer, and U.S. Public Health Service, and his analysis indicated that a chemical carcinogenic in one species had a 15% chance of not being carcinogenic in the other. These data emphasized the importance of testing chemicals in more than one species in long-term bioassays (58). The rat is dearly an appropriate choice for one of these species because so much is known about its metabolic and physiological patterns and because various classes of chemicals are carcinogenic for rats (53, 59). Recent studies on mechanisms of chemical carcinogenesis have demonstrated deficiencies in long-term animal carcinogenesis test- ing when it is used as the sole assessment criterion, because problems may occur with chemicals that are carcinogenic but that cause only moderate tumor incidence in a given tissue in different 272 SCIENCE, VOL. 24-5

884 ANNALS NEW YORK ACADEMY OF SCIENCES 99. HASEMAN, J. K. 1984. Statistical issues in the design, analysis, and interpretation of animal carcinogenicity studies. Environ. Health Perspect. 58: 385-392. 100. International Commission on Radiological Protection. 1977. Recommendations of the International Commission on Radiological Protection. ICRP Publication 26, Annals of the ICRP, Vol. 1, No. 3. Pergamon Press. Oxford. 101. UPrON, A. C. 1977. Radiobiological effects of low doses: Implications for radiological protection. Radiat. Res. 71: 51-74. 102. Interagency Regulatory Liaison Group. 1979. Work group on risk assessment: Scien- tific bases for identification of potential carcinogens and estimation. J. Natl. Can- cer. Inst. 63: 241-268. 103. OFFICE OF SCIENCE AND TECHNOLOGY POLICY. 1985. Chemical Carcinogens; A Re- view of the Science and Its Associated Principles. February 1985, Federal Register Vol. 50, No. 50, Thursday, March 14,: 10372-10442. 104. KREWSKI, B. & J. VAN RYZIN. 1981. Dose-response models for quantal response toxicity data. In Statistics and Related Topics. J. Sxorgo, D. Dawson, J. N. K. Rao & E. Saleh, Eds.: 201-231. Elsevier/North-Holland. New York, NY. 105. NATIONAL ACADEMY OF SCIENCES. 1978. Saccharin: Technical Assessment of Risks and Benefits. Part 1 of 1 2-Part Study of the Committee for a Study on Saccharin and Food Safety Policy. Panel I: Saccharin and Its Impurities. Assembly of Life Sciences/National Research Council and the Institute of Medicine. National Acad- emy of Sciences. Washington, DC. 106. CLAYSON, D. B. 1979. Bladder cancer in rats and mice: Possibility of artifacts. In Aspects of Cancer Research, 1971-1978, Editorials from the Journal of the National Cancer Institute. Natl. Cancer Inst. Monogr. 52: 519-525. 107. PETO, R. 1977. Epidemiology, multistage models, and short-term mutagenicity tests. In Origins of Human Cancer. H. H. Hiatt, J. D. Watson & J. A. Winsten, Eds.: 1403-1428. Cold Spring Harbor Laboratory. Cold Spring Harbor, NY. 108. GAYLOR, D. W. & R. L. KODELL. 1980. Linear interpolation algorithm for low-dose risk assessment of toxic substances. J. Environ. Pathol. Toxi~ol. 4: 305-312. 109. CRUMP, K. S. 1981. An improved procedure for low-dose carcinogenic risk assess- ment from animal data. J. Environ. Pathol. Toxicol. 5: 675-684.

UP'iCON: THRESHOLDS FOR CARCINOGENESIS? 879 Because of the complexity of carcinogenesis and the variability of dose-inci- dence relationships, it is not possible on the basis of present knowledge to extrap- olate confidently across different species, population groups, doses, and condi- tions of exposure in estimating the carcinogenic risks of a particular carcinogen for human populations exposed at low dose levels. Agents differ widely in metab- olism, potency, and mode of action, with the result that their hazards can be expeci.ed to vary greatly at low doses, whether estimated with the use of a thresh- old dose-incidence model or a nonthreshold dose-incidence model. In selecting the appropriate dose-incidence model for risk assessment, one must consider each agent individually, taking all relevant epidemiological, clinical, and experi- mental data into account. The existing evidence does not rigorously exclude a threshold for any carcino- gen, but the use of a nonthreshold model for ionizing radiation and most chemi- cals, especially those with genotoxic activity, is generally recommended on the basis oE present knowledge. The choice of a threshold model cannot be justified in the absence of evidence that the metabolism or mode of action or both of the agent varies appropriately in relation to the dose. REFERENCES 1. MULLER, H. J. 1954. The manner of production of mutations by radiation. In Radia- tion Biology. Vol. 1: High-Energy Radiation. A. Hollaender, Ed.: 475-626. McGraw-Hill. New York, NY. 2. SC:HERER, E. & P. EMMELOT. 1979. Multihit kinetics of tumor cell formation and risk assessment of low doses of carcinogen. In Carcinogens: Identification and Mecha- nisms of Action. A. C. Griffin & C. R. Shaw, Eds.: 337-364. Raven Press. New York, NY. 3. FIALKOW, P. J. 1977. Clonal origin of human tumors. Biochim. Biophys. Acta 458: >83-321. 4. IAIJNACCONE, P. M., R. L. GARDNER & H. HARRIS. 1978. The cellular origin of chemically induced tumors. J. Cell Sci. 29: 249-269. 5. PONDER, B. A. J. 1980. Genetics and cancer. Biochim. Biophys. Acta 605: 368-410. 6. SANDBERG, A. A. 1983. A chromosomal hypothesis of oncogenesis. Cancer Genet. Cytogenet. 8: 277-285. 7. FA taER, E. 1984. Cellular biochemistry of the stepwise development of cancer with chemicals. G. H. A. Clowes Memorial Lecture. Cancer Res. 44: 5463-5474. 8. BERTRAM, J. S. & C. HEIDELBERGER. 1974. Cell cyclic dependency of oncogenic transformation induced by N-methyl-N'-nitro-N-nitrosoguanidine in culture. Can- cer Res. 34: 526-537. 9. BATES, R. R., S. A. EATON, D. L. MORGAN & S. YUSPA. 1970. Replication of DNA after binding of the carcinogen 7-dimethylbenz[a]anthracene. J. Natl. Cancer Inst. 45: 1223-1228. 10. KAf:UNAGA, T. 1974. Requirement for cell replication in the fixation and expression of the transformed state in mouse cells treated with 4-nitroquinoline-l-oxide. Int. J. Cancer 14: 736-742. 11. KNUDSON, A. G. 1985. Hereditary cancer, oncogenes, and antioncogenes. Cancer Res. 45: 1437. 12. HtJBERMAN, E., R. MAGER & L. SACHS. 1976. Mutagenesis and transformation of normal cells by chemical carcinogenesis. Nature 204: 360-361. 13. PARODI, S. & G. BRAMBILLA. 1977. Relationship between mutation and transforma- tion frequencies in mammalian cells treated in vitro with chemical carcinogens. Mutat. Res. 47: 53-74. 14. BARRETT, J. C., B. D. CRAWFORD & P. O. P. Ts'o. 1980. The role of somatic mutation in a multistage model of carcinogenesis. In Mammalian Cell Transforma-

UPTON: THRESHOLDS FOR CARCINOGENESIS? In addition, a dose-response model must be used for interpolating between the lowest dose at which a significantly increased incidence has been observed and the base.line (zero dose) incidence. For this purpose, a linear, nonthreshold (one- hit) dose-incidence model is generally recommended, although such a model cannot be verified experimentally.104 This type of model gives higher estimates, however, than other models (FIG. 10 and TABLE 2). Hence it is usually thought likely to overestimate the risk at low doses and is thus often considered to esti- mate the "upper limit" of risk. Evidence concerning the modes of action of different classes of carcinogens (initiatoi,s, promoters, co-carcinogens, and complete carcinogens) suggests that a linear nonthreshold model may be appropriate only for initiating agents and com- plete carcinogens, whereas models yielding smaller estimates of risks at low doses might represent more accurately the dose-incidence relationships for other classes of carcinogens. For some types of carcinogens, thresholds might even be envisioned to exist because of relevant pharmacokinetic factors. For example, some chemicals that must be activated metabolically to become carcinogenic may be handled through nonlinear metabolic processes, with the result that thresholds for their carcinogenic effects may exist.87 In addition, some agents may act through i.oxic or systemic effects that are produced only at high doses (for exam- ple, those causing carcinogenic effects on the mucosa of the urinary bladder in association with cystitis and urinary tract calculi,106 or those acting through im- munosuppressive effects.91 If it can be shown, however, that a chemical acts through mechanisms that are shared by agents that contribute to the baseline incidence of "spontaneously P(d) g0'5 -4 10-8 -~ ( -F - 1 I 10~2 102 d FIGURE iCl. Estimated risk of liver cancer, P(d), in relation to dose of aflatoxin, d, as determined with different dose-incidence models. The models for the different curves are as follows: OH, one-hit model; MS, multistage model; W, Weibull model; MH, multihit model; MB, Mantel-Bryan (log-probit model). (Reproduced from Krewski and Van Ryzin10' with permission from the Elsevier/North-Holland Publishing Company.)

Cancer Modeling Cohen

874 ANNALS NEW YORK ACADEMY OF SCIENCES tive effects of these agents may be additive, synergistic, or inhibitory, depending on the agents in question and the conditions of exposure.56,116 In a number of instances, appropriate stimulation by a tumor-promoting agent has been observed to convert a curvilinear dose-incidence response involving a threshold into a linear response not involving a threshold (FIG. 9). HEALTH POLICY IMPLICATIONS The problem of risk assessment for purposes of public health policy is compli- cated by the fact that cancer arises through successive stages, each of which may be affected differently and in as yet unpredictable ways by a given agent. No single process is known to be applicable to all carcinogens, all types of cancer, and all persons at risk. Most multistage models assume, however, that 1) a normal cell must undergo two or more stochastic and essentially irreversible changes to become transformed into a cancer cell; 2) one or more of the changes may be inherited via the fertilized egg (zygote); and 3) it is the clonal proliferation of a 60 0 50 0 5µq TPA 2x/week ~ . ~ , ~ ~~ . . 0 10 20 30 40 50 60 70 BoP (µg/week) FIGURE 9. The incidence of carcinomas of the skin after 350 days of treatment in mice exposed to a weekly dose of benzo[alpyrene on Mon- day, with or without 5.0 Ecg of 12-O-tetradeca- noylphorbol-13-acetate on Wednesday and Friday. Doses refer to the amount of ben- zo[alpyrene given per week. The treatments were started at 56 days of age. (Reproduced from Burns and AlbertS' with permission from Mary Ann Liebert, Inc.) single cell in which all the necessary changes have occurred that ultimately gives rise to a cancer.97 According to such a model, any agent that directly or indirectly increases the probability of any one of the changes may be a carcinogen because such an agent would increase the likelihood that a cell will ultimately acquire all of the changes necessary for transformation. The model also implies that the changes necessary for malignant transformation must occur in the proper se- quence, because some carcinogenic stimuli act only on early stages while others act on later stages, and that carcinogens that affect different stages in the process can be multiplicative rather than merely additive in their combined effects. In the absence of definitive human data, risk assessment must depend on other types of evidence (for example, on the results of bioassays in laboratory animals or on short-term tests for carcinogenicity). Under such circumstances, risk as- sessment is complicated by questions about 1) the reliability of the test system for predicting risks to humans (quantitatively as well as qualitatively); 2) the repro- ducibility of the test results; 3) the influence of species differences in pharmaco- kinetics, metabolism, hoemostasis, repair rates, life span, organ sensitivity, and baseline cancer rates; 4) the influence of differences in dose, dose rate, and routes

SUMMARY OF RESEARCH GRANTS & FELLOWSHIPS AWARDED BY ACS (National Soeiety & Divisions) DURING THE FISCAL YEAR ENDED AUGUST 31, 1938 (Subject to Audit) American Health Foundation, New York, NY ( 1) $1,000,000 Medical Research Council, Cambridge, England ( I) $ 70,000 Univ. of Connecticut, Storrs ( 4) $ 679 000 Arizona State Univ., Tempe, AZ ( 1) 82,000 Michigan Cancer Fdn., Detroit 1 4) 680,500 Univ. of Delaware, Wilmington ( 1) , 63 600 Baylor College of Medicine, Houston, TX ( 5) 462,000 Michigan State Univ., East Lansing ( 3) 614,000 Univ. of Florida, Gainesville ( 3) , 311,000 Beth Israel Hosp., Boston, MA 1 3) 208,500 Miller's Children's Hospital, long Beach, CA ( 1) 250,000 Univ. of Georgia, Athens ( 7; 205,000 Boston Univ., Boston, MA ( 4) 358,500 Montefiore Hospital, Bronx, NY /1 OI,VVV Univ. of Hawaii, Honolulu ;1; 10,000 Brandeis Univ., Waltham, MA 246,UUt) Mount Sinai Sch. of Med., N- vcrk, yY 205,SO0 Univ. of Illinois, Urbana 1 5) 256,575 Brigham & Women's Hosn_ nnse...,, MA 167,000 Nat'1 Cancer Inst., Bethesda, MI) ( 1) 69,600 Univ. of Indiana, Bloomington ( 4) 353,500 Brown Univ., Providence, RI ( 5) 562,500 Nat'I Inst. of Allergy & Infectious Disease, Univ. of Kansas, Lawrence ( 3) 189,000 California Inst. of Tech., Pasadena H01 707,850 Bethesda, MD ( 1) 63,300 Univ. of Kentucky, Lexington ( 1) 30,000 California State Coll., Fullerton ( 1) 10,000 Nat'l Insts, of Health, Bethesda, MD ( 1) 69,000 Univ. of Louisville, Louisville, KY ( 1) 85,000 Carnegie Inst. of Washington, Baltimore, MD ( 3) 119,500 Nat'I Jewish Hosp. & Res. Ctr., Denver, CO ( 5) 705,318 Univ. of Maryland, Baltimore ( 5) 790,000 Carnegie-Mellon Univ., Pittsburgh, PA ( 1) 160,000 New England Med. Ctr. Hosp., Boston, MA ( 1) 80,000 Univ. of Massachusetts, Amherst ( 1) 110,000 Catholic Med. Clr. of Brooklyn & Queens, NY f 1) 96,000 New York Acad. of Sciences, New York, NY ( 1) 10,000 Univ. of Med. & Dentistry of NI, Newark, NJ ( 5) 572,000 Case Western Reserve Univ., Cleveland, OH ( 3) 343,812 New York Medical Center, Valhalla ( 1) 175,000 Univ. o(Miami, Coral Gables, FL / 3) 447,000 Children's Hospital of San Francisco, CA ( 1) 200,000 New York Univ., New York, NY 1 9) 1,710,694 Univ. of Michigan, Ann Arbor (10) 1,167,720 City Coll. of City Univ. of New York ( 1) 79,000 North Carolina State Univ., Raleigh ( 1) 68,000 Univ. of Minnesota, Minneapolis ( 9) 1,023,000 City of Hope Nat'l Med. Ctr., Duarte, CA ( 1) 131,000 Northwestern Univ., Chicago, ll ( 7) 603,920 Univ. of Nebraska, Omaha ( 5) 1,264,826 Cold Spring Harbor Lab., Cold Spring Hbr, NY ( 6) 306,500 Northern California Ca. Program, Oakland ( 1) 193,000 Univ. of New Hampshire, ( 1) 160,000 Columbia Univ., New York, NY (15) 1,398,400 Oak Ridge Nat'I Lab., Oak Ridge, TN ( 1) 103,000 Univ. of New Mexico, Albuquerque ( 31 440,000 Cornell Univ., Ithaca, NY ( 3) 354,000 Ohio Stale Univ., Columbus ( 4) 277,000 Univ. of North Carolina, Chapel Hill (10) 1,137,925 Cornell Univ., New York, NY ( 4) 357,600 Oregon Health Sciences Lab., Portland ( 2) 210,000 Univ. of North Dakota, Grand Forks ( 11 102,000 Creighlon Univ., Omaha, NE ( 1) 94,000 Oregon State Coll., Sci, Res. Inst., Corvallis ( 3) 134,987 Univ. of Oregon, Eugene ( 4) 305,000 Dana-Farber Cancer Ctr., Boston, MA (14) 1,097,500 Oregon State Univ., Corvallis ( 1) 32,000 Univ. of Pennsylvania, Philadelphia / 0) 928,643 Dartmouth Coll., Hanover, NH ( 3/ 368,875 Oxford University, England (2) 140,100 Univ. of Pittsburgh, Pittsburgh, PA 1 8) 1,290,000 Drexel Inst. of Tech., Philadelphia, PA ( 2) 320,000 Pacific Northwest Res. Fdn., Seattle, WA ( 1) 110,000 Univ. of Rochester, Rochester, NY ( 81 1,060,437 Duke Univ., Durham, NC (10) 998,855 Pennsylvania Slate Univ., Hershey ( 7) 501,000 Univ. of Rhode Island, Kingston ( 1) 43,200 Duquesne Univ., Pittsburgh, PA ( 1) 70,000 Portland State Univ., OR ( 1) 174,000 Univ. of South Carolina, Columbia ( 2) 83,000 East Carolina Univ., Greenville, NC 1 2) 241,500 Princeton Univ., Princeton, NJ (15) 1,327,513 Univ. of Southern California, Los Angeles ( 5) 578,775 Emory Univ., Atlanta, GA ( 3) 560,000 Pub, Health Res. Inst., New York, NY ( 3) 497,000 Univ. of South Florida, Tampa ( 2) 308,000 Eleanor Roosevelt Inst. for Ca. Res., Denver, CO ( 2) 70,000 Purdue Univ., Lafayette, IN ( 3) 293,000 Univ. of Tennessee, Memphis ( 4) 408,000 Foundation for Biomedical Res., Washington, DC ( 1) 10,000 Reed Coll., Portland, OR ( 1) 127,000 Univ. of Texas (Various Locations) (28) 3,044,200 Fred Hutchinson Cancer Res. Ctr., Seattle, WA ( 2) 283,000 Rockefeller Univ., New York, NY 1 7) 975,625 Univ. of Toledo, Toledo, OH ( 1) 63,000 Georgetown Univ., Washington, DC ( 1) 101,000 Roswell Park Mem. Inst., Buffalo, NY (13) 1,519,449 Univ. of Utah, Salt Lake City 1 4) 600,000 Hahnemann Med. Coll., Philadelphia, PA ( 2) 63,000 Rutgers Univ., New Brunswick, NJ ( 1) 160,000 Univ. o(Vermont, Burlington ( 2) 288,000 Harvard Medical School, Cambridge, MA (19) 1,391,703 St. Jude Children's Res. Hosp., Memphis, TN ( 5) 646,000 Univ. of Virginia, Charlottesville 1 9) 865,000 Harvard Sch. of Pub. Health, Boston, MA ( 3) 300,000 St. Louis Univ., St. Louis, MO ( 1) 40,000 Univ. of Washington, Seattle (10) 1,085,312 Henry Ford Hospital, Detroit, MI ( 1) 98,000 Salk Inst. for Biological Studies, San Diego, CA ( 2) 85,000 Univ. of Wisconsin, Madison (11) 778,726 Inst. for Cancer Res., Philadelphia, PA ( 3) 203,000 Scripps Clinic Res. Fdn., La Jolla, CA ( 3) 393,000 Univ. of Wyoming, Laramie ( 1) 20,000 Illinois Cancer Council, Chicago, IL ( 1) 100,000 Showa Univ. Res. Inst., St. Petersburg, FL ( 1) 70,000 Univ. Louis Pasteur, Strasbourg, France ( 1) 70,000 Jackson Lab., Bar Harbor, ME ( 3) 196,250 Sloan-Kettering Inst., New York, NY (30) 3,426,000 Vanderbilt Univ., Nashville, TN 1 5) 455,225 Jefferson Medical Coll., Philadelphia, PA ( 1) 18,550 Stanford Univ., Stanford, CA (21) 1,619,400 Virginia Mason Hospital, Seattle, WA ( 1) 105,500 Jewish Hospital of St. Louis, MO ( 1) 208,000 Slale Univ. of Iowa, Iowa City ( 3) 300,500 Virginia Polytechnic Inst., Blacksburg 1 1) 115,000 Johns Hopkins Univ., Baltimore, MD (19) 2,011,000 Stale Univ. of NY, Albany ( 1) 25,797 Wake forest Coll., Bowman Gray Sch. of Med., Kaiser Foundation Res. Inst., CA ( 1) 45,838 State Univ. of NY, Buffalo ( 1) 200,000 Winston-Salem, NC ( 5) 427,000 Kansas State Univ., Manhattan ( 3) 215,285 State Univ. of NY, Downstate ( 1) 151,000 Washington State Univ., Pullman ( 1) 35,000 Kirksville Coll. of Osteopathic Med., MO ( 1) 84,000 State Univ. of NY, Stony Brook ( 9) 631,695 Washington Univ., St. Louis, MO ( 7) 731,000 La Jolla Cancer Res. Clr., La Jolla, CA ( 1) 84,000 Syracuse Univ., Syracuse, NY 1 3) 223,500 Wayne State Univ., Detroit, MI 1 3) 161,000 Lehigh Univ., Bethlehem, PA 1 2) 140,000 Temple Univ., Philadelphia, PA 1 2) 218,000 Whitehead Inst., Cambridge, MA (10) 639,528 Louisiana State Univ., Baton Rouge ( 4) 503,000 Texas A&M, College Station ( 1) 90,500 Wistar Inst., Philadelphia, PA ( 9) 1,291,000 Loyola University, Chicago, IL ' ( 1) 160,000 Tufts-New England Med. Ctr., Boston, MA ( 1) 90,500 Worcester Fdn. for Exptl. Bio., Shrewsbury, MA ( 11 98,000 M.D. Anderson Cancer Clr., Houston, TX ( 1) 200,000 Tufts Univ., Medford, MA 1 3) 285,500 Woods Hole Ocean. Inst., Woods Hole, MA ( 1) 180,000 Marine Biology tab., Woods Hole, MA ( 1) 10,000 Tulane Univ., New Orleans, LA ( 2) 138,000 Wright State Univ., Dayton, OH ( 1) 149,000 Massachusetts Eye, Ear Infirmary, Boston ( 7) 87,406 Univ. of Alabama Med. Ctr., Birmingham ( 9) 953,008 Yale Univ., New Haven, CT (19) 1,716,225 Massachusetts General Hosp., Boston ( 3) 329,500 Univ. of Arizona, Tucson ( 2) 64,800 Yeshiva Univ.-Albert Einstein, The Bronx, NY (17) 1,785,000 Massachusetts Inst. of Technology, Cambridge (16) 1,015,150 Univ. of Arkansas, Fayetteville- ( 1) 40,000 Medical Biology Institute, La Jolla, CA ( 4) 536,000 Univ. of Calif. (Various Locations) (95) 9,544,063 SUBTOTAL (818) 83,936,347 Medical Coll. of Pennsylvania, Philadelphia ( 1) 108,000 Univ. of Chicago, Chicago, IL (13) 1,182,787 Division Research Grants ( 1) 3,000,000 Medical Coll. of Virginia, Richmond ( 3) 325,000 Univ. of Cincinnati, Cincinnati, OH ( 3) 300,000 Medical Coll. of Wisconsin, Milwaukee ( 4) 477,000 Univ. of Colorado, Boulder (14) 1,407,400 TOTAL (819) 86,936,347 Note: Numbers in parentheses Indicate number of grants per institution for the year ended August 31 1908; totals subject to audit. GEGS1V1_SSzoz

866 ANNALS NEW YORK ACADEMY OF SCIENCES any carcinogenic agent must depend on extrapolation from observations at higher dose levels, based on assumptions about the relevant dose-incidence relation- ships and mechanisms of carcinogenesis. The extrapolation models generally used for estimating the carcinogenic risks of low-level irradiation are of the nonthreshold type; that is, the linear non- threshold model or the "linear-quadratic" nonthreshold model is usually used.28•=9 The strongest epidemiological evidence in support of these models con- sists of (1) the large excess of acute leukemia and other juvenile malignancies that is associated with a dose as low as 1-5 rad in utero3o.31; (2) the excess of thyroid 16701 ATOMIC BOMB SURVIVORS g 200 1950•19)4 400 MASSACHUSETTS ¢ a FLUOROSCOPY W Z 150 U Q 300 e U N U f- N 6 Q ¢ 1IM1 ~ 200 m W 0 m LL W Z SA ~ v 100 Z 1 3 1 4 5 O k 5 0 1 I 1 U Y 2 - 0 Rsd U Rd1 2 100 200 300 400 500 600 100 200 300 400 5oG 600 BREAST DOSE BREAST DOSE BREAST DOSE kUMIER OF FLUOROSCOIIC EXAMINAT10H3 FIGURE 1. Incidence of cancer of the female breast as a function of dose in A-bomb survivors, in women treated with X-rays for acute postpartum mastitis, and in women subjected to multiple fluoroscopic examinations of the chest during treatment of pulmonary tubercuaosis with artificial pneumothorax. (Reproduced from Boice er a0' with permission from the Radiological Society of North America.) tumors that occurs following epilating irradiation of the scalp for tinea capitis in childhood, which is associated with an average dose to the thyroid gland of only 6-8 rad;3z•33 (3) the excess of breast cancers (FIG. 1) in (a) women exposed to A- bomb radiation, (b) women given therapeutic irradiation for postpartum mastitis, (c) women who received multiple fluoroscopic examinations of the chest during the treatment of pulmonary tuberculosis with artificial pneumothorax, and (d) women exposed occupationally to external gamma radiation in the painting of

{ UPTON: THRESHOLDS FOR CARCINOGENESIS? 873 The frequency of chromosomal aberrations in human lymphocytes irradiated iin vitro increases as a linear-quadratic nonthreshold function of the dose, approx- icnating 0.1 aberration per cell per Sv in the low-to-intermediate dose region.85 The dose required to double the frequency of aberrations in such cells can thus be calculated to approximate 0.05 Sv.86 With high linear energy transfer radiation, the frequency of aberrations increases more steeply, as a linear nonthreshold function of the dose, and irrespective of the dose rate.ss,as Dose-response relationships for chemically induced mutations and chromo- somal aberrations are less well defined than those for ionizing radiation, in part because of the greater diversity of types and mechanisms of chemically induced DNA damage. Chemical mutagenesis and clastogenesis involve complex pro- cesses, including pharmacokinetic variables (uptake, transport, distribution, and excretion), metabolic activation and detoxication, and various reactions leading to the production of DNA lesions and their subsequent repair-misrepair. Each of these steps may conceivably involve first-order kinetics at low doses and hence be linear, so that in principle the overall process may be linear and not approach a threshold. Even if mutagenesis at low dose levels involved only linear processes, the slope of the resulting dose-response relationship could be orders of magnitude shallower than the slope at high dose levels, so that the dose-response curve could appear to reach a threshold or a quasithreshold.87 In fact, nonlinear mecha- nisms are likely to operate in at least some of the transport, metabolism, elimina- tion, and repair processes that are involved in mutagenesis,8' and it is noteworthy that a single step involving a threshold in such a sequence could give the overall process a threshold. Hence, in view of the complexity of the many processes involved in chemical mutagenesis, it is not astonishing that the dose-response curves for mammalian cells exposed in vitro have been observed to include re- sponses that appear to involve linear nonthresholds as well as quasithresholds.88 Whether any of the responses truly involves a threshold, however, cannot be det:e rmined from existing data. Other factors complicating assessment of the practical implications of dose- response data for chemical mutagenesis are the fact that chemicals vary more than a millionfold in mutagenic potency and the fact that the magnitude of the variation among chemicals also differs depending on the types of cells and indices of mutagenicity in question.88,89 FACTORS MODIFYING THE DOSE RESPONSE A variety of factors are known to affect dose-incidence relationships in car- cinogenesis.4D These include, among others, variables influencing the susceptibil- ity of exposed individuals (for example, genetic background,I ' age at exposure,2S immunological reactivity,91 differences in DNA repair capacity,8S and differences in drug metabolism9z,93). The capacity to metabolize a chemical can vary among humans by more than 100-fold11 and among species by more than 1000-fold.93 In any one person, moreover, the balance between toxification and detoxification may be highly dose-dependent.95 As a result, the effective dose of a substance to its biological target may differ substantially among persons at a given ambient exposure level. Also of potential importance in modifying the dose-response relationship for a given carcinogen are the effects of other physical or chemical agents. The interac-

276 x-ray-induced oncogcnic transformation in vitro cxxcurs during cellular proliferations. Radiat Res 99:228-248. 1983. 54. Kennedy A: The condition for the modification of radiation tran.sfor- tnation in vitro by a tumor promoter and protease inhibitors. Car- cinogenesis 6:1441-1445. 1985. 55. Upton AC: Radiobiological effects of low doses: implications for radiological protection. Radiat Res 71:51-74. 1977. 56. Krewski I), Van Ryzin J: Dose response models for quantal response toxicity data. In Statistical and Related Topics. Edited by J Sxorlto, D Dawson. JNK Rao. E. Shaleh. North Holland. New York, 1981, pp 201-231. 57. Hoel DG, Kaplan NL. Anderson MW: Implication of nonlinear kinetics on risk estimation in carcinogenesis. Science 219: 1023-1031, 1983. 58. Old LJ: Cancer immunology: the search for specificity. G. H. A. Clowes Memorial Lecture. Cancer Res 41:361, 1981. 59. Selikoff JJ: Constraints in estimating occupational contributions to currenrcarcer mortality in the United States. In Quantification of Occupational Cancer. Banoury Report 9. Edited b} R Peto. M Schneid:rman, Cold Spring Harbor Laboratory. Cold Spring Harbor, 1981, pp 3-13. 60. National Academy of Sciences (NAS): Saccharin: Technical Assessmenrs c fRisks and Benefrrs. Pan I of a 2-Part Study of the Committee for a Study of Saccharin and Food Safety Policy. Panel l: Saccharini aid Its Impurities. Assembly of Life ScienceslNational Research Council and the Institute of Medicine. National Academy of Sciences,Washington. DC. 1978. 61. Doll R: An epidemiological perspective of the biology of cancer. Cancer Res 3g:3573-3583, 1978. Upton 62. Williams MHC: (kcupational tumors of the bladder. In Cancer. Edited by R\Y Raven. Butterworth. London. 1958, p 337. 63. Littlefield NA. Farmer JH. Ga)ior DW: Effects ofdou and time in a long-tcrm. low-dou carcinogenic study. J Environ Pathol Tox- icol 3: l 7-34. 1979. 64. Bcrcnblum 1. Trainin. N: New e% idence ofthe mechanism ofradia- tion leukacmogenesis. In Cellular Basis and .tetiolqei of Lrre Suniatic Effcrc~ of loni:ing Radiation. Edited by RJC Harris. Academic Pre.e, New York. 1963. pp 41-56. 65. Burns FJ, Albert RE: Mouse skin papillomas as early stage of car- cinogenesis. 1 Am Coil Toxicol 1:29-45. 1982. 66. Thomson JF, Lombard LS. Grahn D, et al: RBE of fission neutrons for life shortening and tumorigenesis. In Neutron Carcinogenesis. Edited by J Broerse, GB Gerber. Luxembourg, Commission of the European Communities. 1982. pp 75-94. 67. Hill CK, Han A. Elkind MM: Fission-spectrum neutrons at low dose rate enhance neoplastic transformation in the linear, low dose region (0-10 Gy). Int 1 Radiat Biol 46:11. 1984. 63. Little JB: Influence of noncarcinogenic secondary factors on radia- tion carcinogenesis. Radiat Res 87:240-250. 1981. 69. Upton AC: Biological aspects of radiation carcinogenesis. In Radia- tion Carcinogenesis: Epidemiologr and Biological Significance. Edited by JD Boice 1r. JF Fraumeni Jr, Raven Press, New York, 1984, pp 9-19. 70. National Council on Radiation Protection (NCRP): Contparative CarcinoXcnic•i»•ofloni;ing Radiation artd Chemicals. Washington. DC, National Council on Radiation Protection and Measurement. 1989.

Reprint Series 31 August 195'0, Volume 249, pp. 1007-1011 Cell Proliferation in Carcinogenesis SAMUEL M. COHEN AND LEON B. ELLWEIN SCIENCE Copyright 0 1990 by the American Association for the Advancement of Science

SrIENCE, Vol. 253, No. 5022, pages 842 - 844 (23 August 1991). Get-the-Lead-Out Guru Challenged A decade-old scienti f ic argument over the ef fects of low-level lead on IQ turns nasty following allegations of misconduct AS AN ENVIROhMEhTAL BOGEYhU2:, LEAD'S hard to beat. It ranks right up there with asbestos, dioxin, and nuclear waste. Vice President Dan Quayle has even suggested that lead in the drinking water at the vice presidential mans:,on might have caused the Bushes' bouts widi Graves' disease. But, irrational f:ars aside, there's no ques- tion that high lead levels can cause brain damage-it's only at low levels of exposure that there is still a debate about what amount of lead in the blood can cause detectable behavioral and medical problems. And that debate has been tainted by a festering, 10- year-old dispute over the credibility of data published by Herbert Needleman of the University of Pittsburg, a world-renowmed researcher on lead toxicity and leading ad- viser to the gove;mment on lead issues. Now, in the wake of a government lawsuit against the owners of a lead smelter in which Needleman was to have testified-but never did because the case was settled out of court-his critics h.ave filed a complaint with federal investigators alleging that Need- leman engaged irn scientific misconduct a decade ago. They accuse the government of helping cover up the flaws in his research in order to deflect criticism of its policy deci- sions. To Needleman, the charges are nothing more than old mud slung with new vigor- thoroughly debunked criticisms kept alive by a lead industry desperate to discredit his research. Regardless ofwho is right, the Needleman saga shows how hard it is to put to rest charges from persistent critics, or, con- versely, to prove misconduct against an ac- knowledged leader in a scientific field. But it also raises additional questions, widely ap- plicable to other scientific disputes, about who should have access to data collected with federal support. And, of course, it refocuses attention on a matter that is espe- cially meaningful to a lot of parents: Just how strong is the link between low-level lead exposure and :;ntelligence deficits? The story be:gins with a paper by Needleman and his colleagues in the 29 March 1979 issue ofThe New England Jour- nal ofMedicine shoAing that schoolchildren with what all would agree were "high," but not actually toxic, lead levels did significantly poorer in the classroom and had measurably lower IQs than those with "low" lead levels. In order to get a clearer picture of exposure, the researchers had looked at lead concentra- tions in the children's baby teeth, as well as the more labile measure of lead in the blood. Suzanne Binder, chief of the Lead Poisoning Prevention Branch at the Centers for Disease Control (CDC) in Atlanta, says that most people's first reaction to Needleman's study was "so what?" since the drop in IQ was only 3 or 4 points. But Binder says policy- makers came to realize that even a small drop would be important if it was affecting millions of children. Two years after the Journal article ap- peared, Claire Ernhart, a psychologist now at crude measure like IQ, except at some of the highest levels of exposure, just below what would be considered toxic. The appearance of the Pediatrics article touched off wh'at has been a decade-long personal feud between Ernhan and Needle- man. They have squared off at numerous scientific meetings with a vigor that has left observers shaking their heads. "Personal hos- tilit}• is putting it mildly," says Binder. But the Needleman/Ernhart squabble might have remained nothing more than a classic confrontation between scientists -,vith starkly opposing views had it not entered, in 1983, into a new and grander forum. The year before, the Environmental Protection Agency (EPA) had begun a major review of national air-qualitv standards for lead and r wanted to review all m a recent data on the a health effects of lead o exposures. In an ef- ~0 fort "to resolve major d E points of controversy D concerning some of the most important and controversial" studies, Lester Grant, director of the EPA's environmental criteria and assessment office, convened a special panel to look into both Needleman's and Ernhart's work (Science, 25 Novem- ber 1983, p. 906). The panel traveled Clearing the Ar. Herbert Needleman says the lead industry is behind attempts to discredit his research. Case Western Reserve University, and her colleagues fired the first shot across Needleman's bow. Writing in the journal Pediatrics, they suggested that there were serious methodological flaws in the Need- leman paper. Ernhart argued that Needleman had not done an adequate job of controlling for confounding variables-other factors such as poor schools orparental neglect that might explain the difference in IQ scores-and had performed so many comparisons that he was bound to come up with a few that were statistically significant merely by chance. Ernhart's own %.rork suggested that most lead effects were too small to be detected by a to Needleman's lab, examined some of his data, and decided there were several prob- lems with the study. Specifically, the panel members concluded that Needleman had used inappropriate measures to categorize ~~' lead exposure and had not provided suff~-s~' ~ ~ cient justification for excluding subjects, from the study. Moreover, they expressed concern about missing data, and some of the statistical analyses Needleman had em- ployed, all of which led them to conclude that the study results "neither support nor refute the hypothesis that low or moderate levels of [lead] exposure lead to cognitive or behavioral impairments in children." The 842 SCIENCE, VOL. 253

8S) ANNALS NEW YORK ACADEMY OF SCIENCES tion by Chemical Carcinogens. N. Mishra, V. C. Dunkel & M. Mehlman, Eds.: 467. Senate Press. NJ. 15. LAND, H., L. F. PARADA & R. A. WEINBERG. 1983. Cellular oncogenes and multistep carcinogenesis. Science 222: 771-778. 16. BLUMBERG, P. M. 1980, 1981. In vitro studies on the mode of action of the phorbol esters, potent tumor promoters, Parts 1 and 2. CRC Crit. Rev. Toxicol 3: 152-234. 17. BLUMBERG, P. M., S. JAKEN, B. KONIG, N. SHARKEY, K. LEACH, A. JENG & E. YEH. 1984. Mechanisms of action of the phorbol ester tumor promoters: Specific receptors for lipophilic ligands. Biochem. Pharmacol. 33: 933-940. 1K WEINSTEIN, I. B., S. GATTO-CELLI, P. KIRSCHMEIER, W. HSIAO & A. JEFFREY. 1984. Cellular targets and host genes in multistage carcinogenesis. Fed. Proc. 43: 2287-2294. lS'. WEINSTEIN, I. B. 1985. Cell culture studies on the mechanism of action of chemical carcinogens and tumor promoters. In Carcinogenesis: A Comprehensive Survey. Vol. 10: The Role of Chemicals and Radiation in the Etiology of Cancer. E. Huber- man & S. H. Barr, Eds.: 177-187. Raven Press. New York, NY. 2O. VAN DUUREN, B. L. 1976. Tumor-promoting and co-carcinogenic agent in chemical carcinogenesis. In Chemical Carcinogens. C. E. Searle, Ed.: 24-51. American Chemical Society. Washington, DC. 21. UPTON, A. C., D. G. CLAYSON, D. JANSEN, H. ROSENKRANZ & G. WILLIAMS. 1984. Report of ICPEMC task group on the differentiation between genotoxic and non- genotoxic carcinogens. Mutat. Res. 133: 1-49. 22. CERUTTI, P. A., P. AMSTAD & I. EMERIT. 1983. Tumor promoter phorbol myristate acetate-induced membrane-mediated chromosomal damage. In Radioprotectors and Anticarcinogens. O. F. Nygaard & M. G. Simic, Eds.: Academic Press. New York, NY. 23. TROLL, W. & R. WEISNER. 1985. The role of oxygen radicals as a possible mechanism of tumor promotion. Annu. Rev. Pharmacol. Toxicol. 25: 509. 24. LIEHR, J. G., K. RANDERATH & E. RANDERATH. I985. Target organ-specific covalent DNA damage preceding diethylstilbestrol-induced carcinogenesis. Carcinogenesis 6: 1067-1069. 25. FoULDS, L. 1969. Neoplastic Development. Vol. 1. Academic Press. New York, NY. 26. International Agency for Research on Cancer. (1982). IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans: Chemicals, Indus- trial Processes, and Industries Associated with Cancer in Humans. Supplement 4. International Agency for Research on Cancer. Lyons. 27. United Nations Scientific Committee on the Effects of Atomic Radiation. 1977. Sources and Effects of Ionizing Radiation. Report to the General Assembly, with Annexes. United Nations. New York, NY. 28, National Academy of Sciences, Advisory Committee on the Biological Effects of Ionizing Radiation. 1980. The Effects on Populations of Exposure to Low Levels of Ionizing Radiation. National Academy of Science. Washington, DC. 2:9,, RALL, J. E., G. W. BEEBE, D. G. HOEL, S. JABLON, C. E. LAND, 0. F. MYGAARD, A. C. UPTON, R. S. YALOW & V. H. ZEVE. 1985. Report of the National Institutes of Health Working Group to Develop Radioepidemiological Tables. U.S. National Institutes of Health Publication 85-2748. Washington, DC. 30, MONSON, R. P. & B. MACMAHON. 1984. Prenatal X-ray exposure and cancer in children. In Radiation Carcinogenesis: Epidemiology and Biological Significance. J. D. Boice, Jr. & J. F. Fraumeni, Jr., Eds.: 97-105. Raven Press. New York, NY. 31. HARVEY, E. B., J. D. BOICE, JR., M. HONEYMAN & J. T. FLANNERY. 1985. Prenatal X-ray exposure and childhood cancer in twins. N. Eng. J. Med. 312: 541-545. 32. MODAN, B., E. RON & A. WERNER. 1977. Thyroid cancer following scalp irradiation. Radiology 123: 741-744. 33. SHORE, R. E., E. D. WOODARD, L. H. HEMPLEMANN & B. S. PASTERNACK. 1980. Syngerism between radiation and other risk factors for breast cancer. Prev. Med. 9: 815-822. 34. BOICE,'J. D., JR., C. E. LAND, R. E. SHORE, J. E. NORMAN & M. TOKUNAGA. 1979. Risk of breast cancer following low-dose exposure. Radiology 131: 589-597.

UPTO:N: THRESHOLDS FOR CARCINOGENESIS? 881 35. DOLL, R. 1978. An epidemiological perspective of the biology of cancer. Cancer Res. 38: 3573-3583. 36. DOLL, R. & R. PETO. 1981. The causes of cancer: Quantitative estimates of avoidable risk of cancer in the United States today. J. Nat1. Cancer Inst. 66: 1192-1308. 37. SAFFIOTTI, U. 1973. Metabolic host factors in carcinogenesis. In Host Environment 1`.nteractions in the Etiology of Cancer in Man. R. Doll, I. Vodopija & W. Davis, E_ds.: 243-252. International Agency for Research on Cancer. Lyons. 38. WILLIAMS, M. H. C. 1958. Occupational tumors of the bladder. In Cancer. R. W. Raven, Ed.: 337-380. Butterworth. London. 39. NICHOLSON. W. J. 1985. Airborne Asbestos Health Assessment Update. U.S. Envi- ronmental Protection Agency Publication 600/8-84-003F. Washington, DC. 40. SELIKOFF, I. J. 1981. Constraints in estimating occupational contributions to current cancer mortality in the United States. In Banbury Report 9: Quantification of Occu- pational Cancer. R. Peto & M. Schneiderman, Eds.: 3-13. Cold Spring Harbor Laboratory. Cold Spring Harbor. New York, NY. 41. DO:LL, R. 1970. Cancer and aging: The epidemiologic evidence. Oncology 5: 1-28. 42. HAMMOND, E. C., I. J. SELIKOFF & H. SEIDMAN. 1979. Asbestos exposure, cigarette smoking and death rates. Ann. N.Y. Acad. Sci. 330: 473-490. 43. UPTON, A. C., F. F. WOLFF, J. FURTH & A. W. KIMBALL. 1958. A comparison of the induction of myeloid and lymphoid leukemias in X-irradiated RF mice. Cancer Res. 113: 842-848. 44. SHE.LLABARGER, C. J., V. P. BOND, E. P. CRONKITE & G. E. APONTE. 1969. Relation- ship of dose to total-body 60Co radiation to incidence of mammary neoplasia in female rats. In Radiation-Induced Cancer: 161-172. IAEA. Vienna. 45. KAPLAN, H. S. & M. B. BROWN. 1952. A quantitative dose-response study of lym- phoid tumor development in irradiated C57 black mice. J. Nati. Cancer Inst. 13: 15'5-208. 46. MALDAGUE, P. 1969. Comparative study of experimentally induced cancer of the kidney in mice and rats with X-rays. In Radiation-Induced Cancer: 439-458. Inter- national Atomic Energy Agency. Vienna. 47. BuRNs, F., R. E. ALBERT & R. D. HEIMBACH. 1968. RBE for skin tumors and hair follicle damage in the rat following irradiation with alpha particles and electrons. Ra,diat. Res. 36: 225-241. 48. METALLI, P., V. COVELLI, M. DIPAOLA & G. SILINI. 1974. Dose-incidence data for r.nouse reticulum cell sarcoma. Radiat. Res. 59: 21. 49. United Nations Scientific Committee on the Effects of Atomic Radiation. 1972. loniz- ing; Radiation: Levels and Effects. Report to the General Assembly, Official Records, 27th Session, Supplement Number 25 (A/8725). United Nations. New York, NY. 50. Gou), L. S., C. B. SAWYER, M. MAGAW, et al. 1984. A carcinogenic potency data- bas,e of the standardized results of animal bioassays. Environ. Health Perspect. 58: 9•-319. 51. Bulus!s, F. J. & R. E. ALBERT. 1982. Mouse skin papillomas as early stages of carcinogenesis. J. Am. Coll. Toxicol. 1: 29-45. 52. LIT'rI.EFIELD, N. A., J. H. FARMER & D. W. GAYLOR. 1979. Effects of dose and time in a long-term, low-dose carcinogenic study. J. Environ. Pathol. Toxicol. 3: 17-34. 53. THOMSON, J. F., L. S. LOMBARD, D. GRAHN, F. S. WILLIAMSON & T. F. FRITZ. 198'2. RBE of fission neutrons for life shortening and tumorigenesis. In Neutron Carcinogenesis. J. J. Broerse & G. B. Gerber, Eds.: 75-94. Commission of the European Communities. Luxembourg. 54. UPTON, A. C. 1984. Biological aspects of radiation carcinogenesis. In Radiation Car- cincgenesis: Epidemiology and Biological Significance. J. D. Boice, Jr. & J. F. Fraumeni, Jr., Eds.: 9-19. Raven Press. New York, NY. 55. ALBERT, R. E. & B. ALTSHULER. 1973. Consideration relating to the formulation of limil.s for unavoidable population exposures to environmental carcinogens. In Ra- dionuclide Carcinogenesis. C. L. Sanders, R. H. Bresch, J. E. Ballon & D. D. Mahlum, Eds.: 233. U.S. Atomic Energy Commission. Washington, DC. 56. United Nations Scientific Committee on the Effects of Atomic Radiation. 1982. loniz-

UPTON: THRESHOLDS FOR CARCINOGENESIS? 883 78. BERTRAM, J. S. &.I. E. MARTNER. 1985. Quantitative neoplastic transformation in C3H/IOTl/2 cells. In Assessment of Risk from Low-Level Exposure to Radiation arid Chemicals. A. D. Woodhead, C. J. Shelibarger, V. Pond & A. Hollander, Eds.: 205-222. Plenum. New York, NY. 79. CGLE, A., R. E. MEYN, R. CHEN, P. M. CORRY & W. HITTELMAN. 1980. Mechanisms of cell injury. In Radiation Biology in Cancer Research. R. E. Meyn & H. R. Withers, Eds.: 33-58. Raven Press. New York, NY. 80. SINOER, B. & J. T. KUSMIEREK. 1982. Chemical mutagenesis. Annu. Rev. Biochem. 51: 655-693. 81. SINCER, B. & D. GRUNBERGER. 1983. Molecular Biology of Mutagens and Carcino- g;eiis. Plenum. New York, NY. 82. SHAPIRO, R. 1981. Damage to DNA caused by hydrolysis. In Chromosome Damage and Repair. E. Seeberg & K. Kleppe, Eds.:` 3-18. Plenum. New York, NY. 83. SAUL, R. L. & B. N. AMES. 1985. Background levels of DNA damage in the popula- tion. In Damage and Repair: Implications for Risk Assessment. M. Simic, L. Gross- Ira.n & A. C. Upton, Eds.: 529-535. Plenum. New York, NY. 84. GROSOVSKY, A. J. & J. B. LITTLE. 1985. Evidence for linear response for the induc- tion of mutations in human cells by X-ray exposure below 10 rads. Proc. Natl. Acad. Sci. USA 82: 2092-2095. 85. Nat,ional Council on Radiation Protection and Measurements. 1980. Influence of Dose and Its Distribution in Time on Dose-Response Relationships for Low-LET Radia- tions. National Council on Radiation Protection and Measurements Report 64. Washington, DC. 86. LLOYD, D. C. & R. J. PURROTr. 1981. Chromosome aberration analysis in radiologi- cal ;protection dosimetry. Radiat. Prot. Dosimetry 1: 19-28. 87. HOEL„ D. G., N. L. KAPLAN & M. W. ANDERSON. !^'83. Implication of nonlinear kinetics on risk estimation in carcinogenesis. Science 219: 1023-1037. 88. EHL'dNG, U. H., D. AVERBECK, P. A. CERUTTI, J. FRIEDMAN, H. GRIEM, A. C. KOL.BYE, JR. & M. L. MENDELSOHN. 1983. Review of the evidence for the presence or absence of thresholds in the induction of genetic effects by genotoxic chemicals. International Commission for Protection Against Environmental Mutagens and Carcinogens Publication 10. Mutat. Res. 123: 281-341. 89. National Academy of Sciences/National Research Council. 1982. Identifying and Es- timating the Genetic Impact of Chemical Mutagens. National Academy of Sciences. Washington, DC. 90. STARR„ T. B. 1985. The role of mechanistic data in dose-response modeling. In Assessment of Risk from Low-Level Exposure to Radiation and Chemicals. A. D. Waodhead, C. J. Shellbarger, V. Pond & A. Hollaender, Eds.: 101-124. Plenum. Ne w York, NY. 91. OLD, L. J. 1981. Cancer immunology: The search for specificity. G. H. A. Clowes Memorial Lecture. Cancer Res. 41: 361. 92. MOMMSEN, S., N. M. BARFORD & J. AAGAARD. 1985. N-Acetyltransferase pheno- typ . i',n the urinary bladder carcinogenesis of a low-risk population. Carcinogenesis 6: 199-201. 93. NEBERT, D. W. & F. J. GONZALEZ. 1985. Cytochrome P-450 gene expression and regulation. Trends Pharmacol. Sci. 6: 160-164. 94. GOLDSTEIN, A., L. AoRNow & S. M. KALMAN. 1974. Principles of Drug Action: The Basiis of Pharmacology. 2nd edit. John Wiley & Sons. New York, NY. 95. DIETZ, E. K., J. C. RAMSEY & P. G. WATANABE. 1983. Experimental studies to human risk. Environ. Health Perspect. 52: 9-14. 96. UPTON, A. C. 1982. Principles of tumor biology, etiology, and prevention. In Princi- ples and Practices of Oncology. V. T. DeVita, S. Hellman & S. A. Rosenberg, Eds.: 33. J. B. Lippincott. Philadelphia, PA. 97. WHITTEMOR.E, A. S. 1978. Quantitative theories of carcinogenesis. Adv. Cancer Res. 27: 55-88. 98. PURCHASE, I. F. H. 1982. An appraisal of predictive tests for carcinogenicity. Mutat. Res. 99: 53-71.

probability that a mutation will oc- cur is sufi:iciently high that nearly all genetically susceptible individ- uals develop retinoblastoma. Inci- dences frequently are bilateral, and/or pe:rsons develop more than one tumor per eye at an early age. Retinoblasts only proliferate during development of the eye, and cell division is necessary for either of the two genetic events in the genesis of retinoblastoma to occur (unless one allele is defective be- cause of a germ line mutation). Thus, the chance of developing a retinoblastoma is eliminated once these cells stop proliferating. Sim- ilar arguments can be advanced for neuroblastoma, since neuroblasts also cease~ proliferating during childhood. Knudson's hypothesis prompted the search for other tumor suppres- sor genes (also referred to as an- tioncogenes) (13, 14). Increased susceptibility to the development of tumors in other tissues, such as osteogenic sarcomas, has been ob- served in patients with retinoblas- toma, although it remains unclear as to why tumors do not increase in all tissues. A second suppressor gene (wit:h protein product p53) might be involved with the genesis of these sarcomas. Other possible candidates for tumor suppressor genes include Wilms' tumor, renal cell carcinoma, and at least two forms of inherited colonic carci- noma. Polyposis coli (Pc) is an autoso- mal dominant, inherited suscepti- bility to adenomatous polyps and adenocarcinoma of the colon (15). Individuals with the Pc genetic de- fect (chromosome 5q) develop nu- merous co;lonic polyps that often evolve into carcinomas within a few decades. Sim.ilar genetic events oc- cur in some nonpolyposis coli pa- tients, who more commonly de- velop colon cancer at a later age. In addition, at least six other autoso- mal dominant hereditary traits predispose to colon cancer (16). Adenoma,tous polyps, which are preneoplastic lesions, exhibit in- creased proliferative capacity, pre- sumably due to enhanced prolifer- ation of the colonic crypts. Most (if not all) preneoplastic lesions in- volved in human carcinogenesis show increased proliferation com- pared with normal tissue, whether ~ 00 ~ WILD PREDISPOSED ~ TYPE PHENOTYPE HEREDITARY .. W_ TUMOR NON-HEREDITARY o.~. --40- Figure 2. Genetics of retinoblastoma. Tumors occur when defects occur in both alleles, whether caused by absence of the entire chromosome, deletion of a portion or all of the gene segment, or mutation of the gene. Individuals with hereditary retinoblastoma are born with one defective allele in all of their cells, requiring only a defect to develop in the second allele for malignancy to occur. Nonhereditary individuals must generate defects in both alleles beginning with cells having two normal alleles at conception. A - B--~ C - D-_E -= F---Po G Figure 3. Alternative explanations of multiple genetic events occurring during carcinogen- esis. lf each of the identified genetic events occurs sequentially, the upper diagram pertains. However, more likely is a mechanism similar to that presented in the lower diagram where there are multiple genetic events that will affect the proliferative rates, genetic stability, and/or the cell population sizes of A, B, or C, but are not essential for the carcinogenic process itself. However, these additional genetic effects will greatly accelerate the carcin- ogenic process overall. it comes from increased mitotic rate, blockage in differentiation, or other mechanisms. Fearon and Vogelstein (17) have recently postulated a multistep process for colonic carcinoma. However, their multiple stage model could also be consistent with only two critical events being re- quired for carcinogenesis (Fig. 3). The additional genetic alterations that they observe in other genes may enhance the proliferative ca- pacity or alter the differentiation of cells in the preneoplastic, ade- nomatous polyp. Although these CELL PROLIFERATION AND CANCER 373

Acceptable Cancer Risks: Probebilities and Beyond Paolo F. Ricci University of California, Los Angeles Los Angeles, California Louis Anthony Cox, Jr. U.S. West Advanced Technologies Englewood, Colorado John P: Dwyer University of California Berkeley, California The acceptability of cancer risk requires consideration of factors that extend beyond mere numerical repre- sentations, such as either individual lifetime risk in excess of background and excess incidence. Recently, use of these numbers has been tempered by the addi- tion of qualitative weights-of-evidence that describe the degree of support provided by animal and epide- miologic results. Nevertheless, many other factors, most of wluch are not quantitative, require incorpora- tion but remain neglected by the analyst eager to use quantitative results. In this paper we show that simple risk measures are often frau;;ht with problems. Moreover, these mea- sures do not incorporate the very essence of accept- ability, wbic:h includes notions of responsibility, ac- countability, equity, and procedural legitimacy, among others. We link the process of risk assessment to those legal and regulatory standards that shape it. These standards are among the principal means to resolve risl:-xelated disputes and to enhance the bal- ancing of competing interests when science and law meet on uncertain and often conjectural ground. We conclude the paper with a proposal for the port- folio approach to manage cancer risks and to deal with uncertain scientific information. This approach leads to the concept of "provisional acceptability," which reflects the choices available to the decisionmaker, and the trade-offs inherent to such choices. Agencies, ind.ustry, and the public demand clear standards for judging the acceptability of risks. Numerical values could reduce debate ,and ambiguity, clarify the responsibilities of businesses, and provide data for regulatory, judicial, and legislative deP,iberations.l,2 Recognizing that a single risk level is not appropriate in all contexts, it is tempting to propose specific numerical "ac- Copyright 1989-AirB: Waste Management Association ceptability" values for different classes of risks.3 For exam- ple, the average acceptable excess individual lifetime fatality probability of cancer from occupational exposures (assum- ing full disclosure and informed consent) might be set at 1 X 10-4. A level of 1. X 10-6 could be defined as acceptable for the general public experiencing involuntary exposures. A much higher risk level, such as 1 X 10-3, might be appropri- ate for sales of inherently dangerous products to fully in- formed, willing customers. Aggregate incidence could be ac- ceptable if it were less than some value, for example, unity. However, even such a range of numbers over different con- texts is neither conceptually adequate nor sufficient as a basis for responsible decisionmaking. Whether a risk is "acceptable" generally depends not only on its objective quantitative probability and the nature and severity of the consequences, but also on societal and politi- cal factors. Single numerical estimates of individual and population risks do not incorporate those qualitative aspects of risk. Protection of individual rights, the equity of risk- benefit distribution, prudence when facing uncertainty, the absence of knowledge, the legitimacy of the risk manage- ment process, and public attitudes toward and perceptions of risks do not lend themselves well to bare numerical repre- sentations.4 This paper examines these issues, assesses some current approaches to social and legal risk management, and pro- poses a risk-portfolio approach in which risk acceptability is an evolving concept. We begin with three concepts of risk. IndOvidual and Population Risks Two related concepts are useful in describing risk to an individual: the total risk to an individual of a particular adverse health consequence, such as cancer; and the concept of an attributable risk describing the incremental contribu- tion to total risk made by a particular source or cause (e.g., the contribution made by cigarette smoking to the risk of lung cancer). Finally, we discuss population risk, in which individual risk is aggregated over the population at risk. Total lndividual Risk The total risk to an individual of developing some undesir- able health response, such as death from cancer, may be defined as the probability that he will develop the response in a given year t, if he has survived until then. This probabili- ty is also called the individual's discrete time "hazard rate" 1046 JAPCA

Applications of Expert Judgment Moeller

882 ANNALS NEW YORK ACADEMY OF SCIENCES ing Radiation: Sources and Biological Effects. Report to the General Assembly, with Annexes. United Nations. New York, NY. 57. BROERSE. J. J. & G. B. GERBER, EDS. 1982. Neutron Carcinogenesis. Commission of the European Communities. Luxembourg. 58. UPTON, A. C. 1985. Biological basis for assessing carcinogenic risks of low-level radiation. In Carcinogenesis: A Comprehensive Survey. Vol. 10: The Role of Chemicals and Radiation in the Etiology of Cancer. E. Huberman & S. H. Barr, Eds.: 381-401. Raven Press. New York, NY. 59. HUBERMAN, E. & L. SACHS. 1966. Cell susceptibility to transformation and cytotox- icity by the carcinogenic hydrocarbon benzo(a]pyrene. Proc. Natl. Acad. Sci. USA 56: 1123-1129. 60. DIPAOLO. J. A., P. J. DONOVAN & R. L. NELSON. 1971. In vitro transformation of hamster cells by polycyclic hydrocarbons: Factors influencing the number of cells transformed. Nature (London), New Biol. 23®: 240-242. 61. GART, J. J.. J. A. DIPAOLO & P. J. DONOVAN. 1979. Mathematical models and the statistical analyses of cell transformation experiments. Cancer Res. 39: 6069-6075. 6 2. BoREK. C. & E. J. HALL. 1973. Transformation of mammalian cells in vitro by low doses of X-rays. Nature 243: 450-453. 63. TERZAGHI, M. & J. B. LITTLE. 1976. X-radiation-induced transformation in a C3H mouse embryo-derived cell line. Cancer Res. 36: 1367-1374. 64. HAN, A. & M. M. ELKIND. 1979. Transformation of mouse C3H/10T1/2 cells by single and fractionated doses of X-rays and fission-spectrum neutrons. Cancer Res. 39: 123-130. 65. BOREK, C. & E. J. HALL. 1974. Effect of split doses of X-rays on neoplastic transfor- mation of single cells. Nature 252: 499-501. 65. MILLER, A. & E. J. HALL. 1978. X-ray-dose fractionation and oncogenic formations in cultured mouse embryo cells. Nature 272: 58-60. 67. MILLER, R. C., E. J. HALL & H. H. Rosst. 1979. Oncogenic transformation in mammalian cells in vitro with split doses of X-rays. Proc. Natl. Acad. Sci. USA 76: 5755-5758. 68. TERZAGHI, M. & J. B. LITTLE. 1976. Oncogenic transformation in vitro after split- dose X-irradiation. Int. J. Radiat. Biol. 29: 583-587. 69. HILL, C. K., A. HAN & M. M. ELKIND. 1984. Fission-spectrum neutrons at low dose rate enhance neoplastic transformation in the linear, low dose region (0-10 Gy). Int. J. Radiat. Biol. 46: 11. 70. POPESCU, N. C., S. C. AMSBAUGH & J. A. DIPAOLA. 1984. Correlation of morphologi- cal transformation to sister chromatid exchange induced by split doses of chemical or physical carcinogens on cultured Syrian hamster cells. Cancer Res. 44: 1933- 1938. 71. BACKER, J. M., M. BOERZIG & I. B. WEINSTEIN. 1982. When do carcinogen-treated 19T1/2 cells acquire the commitment for forming transformed foci? Nature 299: 458-460. 72. HABER, D. A. & W. G. THILLY. 1978. Morphological transformation of C3H 10T1/2 cells subcultured at low densities. Life Sci. 22: 1663-1674. 73. FERNANDEZ, A., S. MONDAL & C. HEIDELBERGER. 1980. Probabilistic view of the transformation of cultured C3H 10T mouse embryo fibroblasts by 3-methylcholan- threne. Proc. Natl. Acad. Sci. USA 77: 7272-7276. 74. KENNEDY, A. R. & J. B. LITTLE. 1984. Evidence that a second event in X-ray- induced oncogenic transformation in vitro occurs during cellular proliferation. Ra- diat. Res. 99: 228-248. 75. KENNEDY, A. R., J. CAIRNS & J. B. LITTLE. 1984. Timing of the steps in transforma- tion of C3H 10T1/2 cells by X-irradiation. Nature 307: 85-87. 76. BARRETT, J. C. & E. ELMORE. 1984. Comparison of carcinogenesis and mutagenesis of mammalian cells in culture. In Handbook of Experimental Pharmacology. L. S. Andrews, R. J. Lorentzen & W. D. Flamm, Eds.: 171-206. Springer-Verlag. Berlin. 77. MORDAN, L. J., J. E. MARTNER & J. S. BERTRAM. 1983. Quantitative neoplastic transformation of C3H/10T1/2 fibroblasts: Dependence upon the size of the initiated cell colony at confluence. Cancer Res. 43: 4062-4067.

inauon of large public water systems in California. Sacramento. CA: - ~ CDHS. 1986. 133. Environmental Protection Agency. Ambient water quality criteria for benzene. Washington, DC: EP:~. 1980 [publication No. (EPA) 440/5- 80/018). 134. RcutoN HE. SCoTr LM. Benzene exposure in the United States 1978- 1983: an overvtew. Am J Ind Med 1985:7:385-393. 135. National Insritute for Occupational Safety and Health. Criteria for a recomma:nded standard. Occupational exposure to formaldehyde. Washington. DC. US Govt Print Off, 1976 (DHEW publication No. tN'IOSH) 77-1'-51. 136. tiationat Academy of Sciences. Formaldehyde: an assessment of its health effects. Prepared for the Consumer Products Safety Commis- sion. Washins:ton, DC: Natl Acad Sci Press. 1980. 137. StEGEL DM. FRANKOS VH, SCHNEIDERMAN MA. Formaldehyde risk as- sessment for occupationally exposed workers. Regul Toxicol Pharma- col 183:3:3:i5-371. 138. BERNSTEIN IZS. STAYNER LT, ELLIoY7 U, et al. Inhalation exposure to formaldehyde: an overview of its toxicology, epidemiology, monitoring and control. Am Ind Hyg Assoc J 1984:45:778-785- 139. Environmental Protection Agency. Technical document. Formalde- hyde. Washington, DC: EPA. Office of Pesticides and Toxic Sub- stances. November 16. 198 1. 140. FuNN FB. Industrial exposures to chlorinated hydrocarbons. Am J Med 1946:1:388-394. 141. KLElNFELD M. TABERSHAw IR. Trichloroethylene toxicity. AMA Arch Ind Hyg 1954;10:134-141. 142. KIMBROUGH RD, MITCHELL FL. Hotnc VN. Trichloroethylene: an up- date. J Toxicol Environ Health 1985;15:369-383. 143. StLaEROEt-D 7-- Risk assessment. Science 1987:237:1399. 144. BROwN HS. 13t:moP DR. RowAN CA- The role of skin absorption as a route of exposure for volatile organic compounds (VOCs) in drinking water. Am J Public Health I984;74:479-484. 145. New York State Department of Health. A risk assessment for ethylene dibromide. Bureau of Toxic Substance Assessment, NYSDH, Feb 21, 1984. 146. WtLLtAMs GM- WESaURGER JH. Carcinogen nsk assessment. Science 1983:221:6. 117. Internattonal Agency for Research on Cancer. Approaches to classi- fving chemical carcinogens according to mechanism of actton. Lyon. France: tARC. 1983 (technical report No. 53:001). 148. WEINS-rEtv iB. Letter to the editor. Science 1983:219:79.s-796. 149. PERERA F. The genotoxic/epigenette distinction: relevance to cancer policy. Environ Res 1984:34:175-191. 150. Office of Science and Technology Policy. Chemical carcinogens: review of the science and itf associated prnciples. Federal Register. March 14, 1985:50:t0372-I0442. 151. Californta Department of Health Services. Guidelines for chemical carcinogen risk assessments and their scientific rattonale. Sacramento. CA: CDHS, 1985. 152. KOCIBA RJ, KEYEs DG. BEYER JE. Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol Appl Pharmacol 1978:46:279-303. 153. National Toxicology Program. Carcinogenesis bioassay of 2.3.7.8- tetrachlorodibenzo-p-dioxin in Osborne Mendel rats and B6C3F, mice (gavage study). Research Triangle Park. NC: NTP. 1982 (technical report No. 209). 154. PtTOT HC, Got-DSwGRTHY T, CAMPBELL HA, et al. Quantitative evalua- tion of the promotion by 2,3,7,8-tetrachlorodibenzo-p-dioxin of hepa- tocarcinogenesis from diethylnitrosamine. Cancer Res 1980:40:3616- 3620. 155. PouND A, PALEN D. GLOVER E. Tumor production by TCDD in skin of HRS/J hairless mice. Nature 1982:300:271. 156. International Agency for Research on Cancer. Genetic and related effects: an update of selected IARC monographs, vol 1-42. IARC Monogr Eva! Carcinog Risk Chern Hum (suppl 6). In press. 157. Wti-rERs MD, STACK HF, BRADY AL et a1 Use of computerized data listings and activity profiles of genetie and related effects in the review of 195 compounds. Mutat Res 1988205Z95-312. 158. DAY NE, BROwN CC. Multistage models and primary prevention of cancer. JNCI 1980;64:977-989. _ REVIEW Vol. 80, No. 16, Ocauber 19, 1988 12S

tdeas in Pathology Pivotal Role of Increased Cell Proliferation in Human Carcinogenesis Samuel M. Cohen, David T. Purtilo, and Leon B. Ellvvein DepartmeN:s of Pathology and Microbiology and of Pediatrics, and the Eppley Institute for Cancer Research, University of Nebraska Medical Center, Omaha, Nebraska Cancer develops secondary to multiple genetic events. Each time a cell divides there is a rare chance that a genetic error related to the carc:ir ogenic process will occur. Thus, environmental agents or disease processes that produce sustained increased cell prolif- eration can enhance the likelihood of cancer development by pro- viding additional cell divisions, each with an opportunity for spontaneous genetic error. Studies of hereditary cancers and of various ]D:NA-damaging agents, such as radiation and certain vi- ruses and chemicals, have provided insight into identification of the essential genes, but many examples of carcinogenesis in hu- mans do not involve direct DNA damage. Also, most preneoplastic lesions in human carcinogenesis show increased proliferation compared with normal tissues, whether from increased mitotic rate, blocked differentiation, prolonged cell survi l A or other mechanisnis. Selected examples of proliferation-related carcino- genesis aria described, including certain infectious agents, defec- tive immune surveillance, hormonal imbalances, chronic inflam- matory-rei;enerative processes, and exposure to various chemi- cals. A common biologic mechanism for these diverse stimuli is increased cell proliferation as a prelude to cancer. This mechanism seems essential to the genesis of many cancers in humans. Key words: Cell proliferation, Carcinogenesis, Viral cancer, Chemical carcinogens, Genetics, Immune surveillance. Modern Pathology, Vol. 4, No. 3 C ancer, the second leading cause of death in the United States, may be inc,reasing, particularly in elderly individuals (1). Although progress has been made in the treatment of patients with cancer, prevention offers greater opportu- nities for reducing the death toll. Cigarette smoking, responsible for a majority of cancers of the respi- ratory tract and cancers of other organs, rem.a:ins the leading known 0893-3952/91/0403-0<71S03.00/0 MODERN PATHOLOGI' Copyright O 1991 hy',"he United States and Canadian Academy of PatholDgy, Inc. Vol. 4, No. 3, P. 371, 1991 Printed in the U.S.A. cause of cancer (2). Specific chem- icals known to be carcinogens in humans, such as 2-naphthylamine, benzidine, 4-aminobiphenyl, vinyl chloride, and diethylstilbesterol, account for only a small percentage of cancers (3). Infectious organisms have also been implicated as being etiologic agents of specific cancers (4), including enteric bacteria, par- asites, such as Schistosoma and Clonorchis, and viruses, such as Ep- stein-Barr virus (EBV), human T- lymphotropic viruses I and II, hep- atitis B virus (HBV), and human papilloma virus (HPV). Mounting evidence strongly sup- ports the contention developed in 1914 that cancer results from ge- netic alterations (5). Utilizing mo- lecular biologic techniques, numer- ous genetic alterations, including specific genes, have been identified in several cancers. However, many etiologic agents do not directly cause genetic damage. Similarly, some environmental agents asso- ciated with cancers do not directly damage DNA. Thus, although ge- netic damage is most likely an eventual common pathway to the development of cancer, other piv- otal mechanisms contribute to car- cinogenesis. That multiple events are essen- tial for the development of cancer has been demonstrated in experi- mental animal models, in in vitro systems, and in certain human can- cers. Nearly 50 years ago, Beren- blum and Shubik [6) conducted classical experiments in mouse cu- taneous carcinogenesis that re- sulted in the formulation of the two-stage carcinogenesis concept. Alfred Knudson (7) hypothesized that two genetic events occur for retinoblastomas to emerge in chil- dren. His hypothesis has been con- firmed through numerous genetic analyses and ultimately by the mo- lecular cloning of. a specific Rb ~ gene. Although cancer arises from de- ~ fective control of cell proliferation, the etiologic and pathogenetic role CA of cell proliferation has received Ll relatively little attention. Never- ~ theless, as early as 1953, Nordling (8) stated that, although genetic ~ alterations were necessary, the likelihood that certain cance*s 4~ would develop could be greatly a::g- mented by sustaining cell prolifer- CELL PROLIFERATION AND CANCER 371

tradtttonatty nas oeen rejectea as moratly ana soctaily unac- ceptabie. - A major limitation of epidemiology (and risk assess- ment) is that reliable data on human exposure to specific chemicals are frequently lacking. Therefore, by necessity, most epidemiological studies have relied on crude or in- direct measures of' exposure. A significant number of car- cinogens have been detected in drinking water, ambient air, and the food supp.y: however, reliable monitoring data ex- ist for only a small fraction of these chemicals. For ex- ample, while dozens of pesticides and industrial chemical carcinogens have been measured routinely in surface wa- ter, gi•oundwater, and drinking water, they represent only a small percentage of chemical pollutants present (10,20-23). Over 700 organic chemicals have been found to be present in the U.S. drinking water supply, including 40 known or suspected carcinogens (24). Numerous carcinogenic air pol- lutants (trace metals, polycyclic aromatic hydrocarbons, and volatile organic chemicals) have been detected in ambient air: again, there axe little or no reliable monitoring data on the majority of airborne carcinogens (25). Similarly, many carcinogenic pesticide residues have been identified in the food supply, but reliable exposure data are lacking for most (26)., Testifying to the pervasiveness of environmen- tal contamination are studies showing significant concen- trations of synthetic organic chemicals in the blood, urine, and/or adipose tissue of the U.S. population. These include 1,1,1-trichloro-2,",-bis(p-chlorophenyi)ethane (DDT), dield- rin, heptachlor epoxide, polychlorinated biphenyls (PCBs), and dioxin (27,28). Again, data are far from comprehensive; however, they do show a decline in the concentrations of DDT and PCBs as a result of regulation. Despite their l:imitations, available exposure and epi- demiologic data have served as the basis for a number of widely varying estirnates of the proportion of human cancer in the U.S. population that can be attributed to life-style, occupational exposures, or other environmental pollution. These exercises have generated significant debate, as much over the underlying assumptions as the data used to gener- ate them (17,29-3'5). Unfortunately; various such estimates (ranging, for example, from 4% to >20% for occupational exposures) have been cited as a basis for setting priorities for public health protection. This approach ignores both the underlying uncertainties, the relative preventability of var- ious risk factors (36), and the disproportionate impact on some segments of the population. For example, once rec- ognized, most chemical pollutant hazards can be reduced or eliminated by practical means. Moreover, the involuntary nature of these exlx sures necessitates control at the source, in contrast to expos ures related to life-style (e.g., diet and smoking), which can be addressed more effectively through public education regarding personal behavioral choices. An- other inherent problem with the approach of estimates is that it obscures the much higher risks to certain subpopulations. For example, if the contribution of occupational carcinogens to all cancer deaths in the United States were as low as 3%, for male industrial workers as a group, workplace carcino- gens would account for at least 25% of all identified causes of cancer (33). ?,notner toot tnat nas been usea increastngly by regula- tory agencies to set priorities and even to detennine accept- able levels of exposure to individual environmental contam- inants has been quantitative risk assessment. Here, also, the lack of good information on human exposure as well as the usual paucity of epidemiological data are compounded by uncertainties regarding the proper way to extrapolate from high to low dose and from experimental animals to hu- mans (37). To offset these uncertainties, the four major U.S. regulatory agencies, including the Environmental Protection Agency (EPA), the Occupational Safety and Health Admin- istration (OSHA), the Consumer Product Safety Commis- sion, and the Food and Drug Administration (FDA), have traditionally preferred conservative models that incorporate an assumption of low-dose linearity, regardless of the pre- sumed mechanism of action of the chemical carcinogen (19). However, in certain instances, these conservative models may underestimate cancer risk. For example, the widely ac- cepted linearized multistage model (38), considered to be one of the most conservative of the biologically plausible risk-assessment models, works on the assumption that the ex- posed population is of uniform susceptibility and that interac- tions do not occur between chemical exposures and other risk factors. Yet significant intraindividual variability has been demonstrated for human metabolism and binding of drugs and carcinogens (39-45) as well as for repair of DNA dam- age (46). Moreover, a number of epidemiological studies have demonstrated synergism between chemical exposures and host factors, such as cigarette smoking and air pollutants in the workplace and urban air (47,48). To further compli- cate the situation, although nonlinear (both superlinear and sublinear) dose-response relationships have been observed experimentally and epidemiologically, the available data do not allow low-dose linearity to be ruled out in any of these cases (49). Given these uncertainties, it is reassuring that, in a number of cases, risks observed in humans have been con- sistent with those calculated from high-dose animal experi- ments with the use of models that incorporated linearity at low dose. These include benzene, ethylene dibromide (EDB), gasoline, asbestos, and ethylene oxide (50-56). Therefore, there is general agreement that the use of quantitative risk assessment, performed with appropriate and consistent as- sumptions and models, affords the possibility of comparing risks for the purpose of setting priorities among selected can- didates for regulation. However, most scientists do not view quantitative risk assessment as capable of providing precise estimates of human risk from individual chemicals: general sources of chemical exposures are considered even less likely candidates for risk estimation by this method- Human ExposureiRodent Potency (HERP) Index Most recently, researchers at the University of Califor- nia at Berkeley and Lawrence Berkeley Laboratory have suggested still another approach to priority setting (2). They have calculated a possible hazard index for selected carcinogens by expressing the human exposure (in mil- ligramsikilogram) as a percentage of the rodent TDso dose Vol. 80, No. 16, Octob+er 19, 1988 REVIEW 1283

. r Thresholds for Radiation and Chemical Carcinogenesis 275 17. PharamcolaD. Edited by LS Andrews, RJ Lorentien, WD Flamm. Springer-Veriag, Berlin, 1984, pp 171-206. Bertram IS. Heidelberger C: Cell cyclic dependency of oncogenic 35. Tables, NIH Publication No. 85-2748. U.S. Government Print- ing Office, Washington, DC, 1985. Monson RP, MacMahon B: Pre-natal x-ray exposure and cancer 18. transfotmat:orn induced by N-methyl-N'-nitro-N-rtitrosoguanidine in culture.(Cancer Res 34:526-537, 1974. Bates RR, Eaton SA, Morgan DL, et al: Replication of DNA after in children. In Radiation Carcinogenesis: Epidemiology and Biological Signifcance. Edited by JD Boice Jr, IF Fraumeni Jr. Raven Press, New York, 1984, pp 97-105. binding of the carcinogen 7-dimethylbenz[ajanthracene. J Natl 36. Harvey EB, Boice JD Jr. Honeyman M, et al: Prenatal x-ray ex- Cancer Inst 45:1223-1228, 1970. posure and childhood cancer in twins. N Engl J Med 12:541-545, 19. Kakunaga T: 13equirement for cell replication in the fixation and 1985. expression of the transformed state in mouse cells treated with 37. Modan B, Ron E, Wemer A: Thyroid cancer following scalp ir- 20. 4-nitroquirxdiiti-l-oxide. Int I Cancer 14:736-742, 1974. Knudson A(3: Hereditary cancer, oncogenes, and antioncogenes. 38. radiation. Radiology 123:741-744, 1977. Shore RE, Woodard ED, Hemplemann LH,et al: Syngerism be- Cancer Rcs 4;5:1437, 1985. tween radiation and other risk factors for breast cancer. Prev Med 21. Hubertnan li. Mager R, Sachs L: Mutagenesis and transforma- 9:815-822, 1980. tion of norrnal cells by chemical carcinogenesis Nature 39. Boice JD Jr, Land CE, Shore RE, et al: Risk of breast cancer 204:360-361, 1976. following low-dose exposure. Radiology 131:589-597, 1979. 22. Parodi S. Brarribilla G: Relationship between mutation and transfor- 40. Broerse JJ, Gerber GB (Eds): Neutron Carrinogenesis. Luxem- mation frequencies in mammalian ceils treated in vitro with chemical carci nogens. Mutat Res 47:53-74, 1977. bourg, Commission of the European Communities, Luxembourg. 1982. 23. Blumberg P4f : In vitro studies on the mode of action of the phor-; 41. Upton AC: Biological basis for asses9ng carcinogenic risks of low- bol esters. Fotent tumor promoters, Part 1 and 2. CRC Crit Rev Toxicol 3:152-234, 1980. 1981. level radiation. In Carcinogenesis, a Comprehensive Survey. Vol. 10. 7he Role ofChenricals and Radiation in the Etiology of Cancer. 24. Weinstein Ill, ~Gatto-Cdli S, Kirschmeier P, et a]: Cellular targets Edited by E Huberman, SH Barr. Raven Press, '.ew York, 1985, and host genes in multistage carcinogenesis. Fed Proc pp 381-401. 43:2287-2294, 1984. 42. Doll R, Peto R: The cause of cancer: quantitative estimates of 25. Weinstein 113: Cell culture studies on the mechanism of action of avoidable risk of cancer in the United States today. I Natl Cancer chemical carcimgens and tumor promoters. In Carcinogenesis, Inst 66:1192-1308. 1981. a Comprehetr.rive Survey. Vol. 10. The Role of Chemicals and 43. National Research Council (NRC), Committee on Passive Smok- 26. Radiation in tlie Etiology of Cancer. Edited by E Huberman, SH Barr. Raven Press, New York, 1985, pp 177-187. Ceruni PA, )kmstad P, Emerit I: Tumor promoter phorbol- ing: Environmental Tobacco Smoke: Measuring Ezposures and Assessing Health Effecu, National Academy Press, Washington, DC, 1986. myristate-aceU:te induced membrane-mediated chrornosomu 44. Nieholson GL: Airbon, -:;bestos Health Assessment Update, damage. In Rczkoprotectors and Araicarcinogens. Edited by OF Nygaard. MG Simic. Academic Press, New York, 1983, pp 527-538. 45. publication No. EPA-60(;:8-84-003F. U.S. Environmental Pro- tection Agency, Washington, DC, 1985. Doll R: Cancer and aging: The epidemiologic evidence. Oncology 27. Upton AC, Clcryson DG, Jansen D, et al: Report of ICPEMC task 5:1-28, 1970. group on the differentiation between genotoxic and non-genotoxic 46. Hammond EC, Selikoff U, Seidman H: Asbestos exposure, 28. carcinogens. Mutat Res 133:1-49, 1984. Troll W. Wei!,ner R: The role of oxygen radicals as a possible cigarette smoking and death rates. Artn NY Acad Sci 330:473-490, 1979. mechanism of tumor promotion. Ann Rev Pharmacol Toxicol 47. Wilkinson GR: Pharmacokinetic considerations in toxicology. In 25:509-528, 1985. Drug Metabolism and Drug Toxicology. Edited by JR Mitchell. 29. Liefir JG, Randerath K, Randerath E: Target organ-specific covalent MG Horning. Raven Press, New York, 1984, pp 213-235. DNA damage preceding diethylstiibestrol-induced carcinogenesis. 48. Huberman E, Sachs L: Cell susceptibility to transformation and Carcinogenesi:~ 6:1067-1069, 1985. cytotoxiciry by the carcinogenic hydrocarbon benzo[a)pyrene- Proc. 30. Foulds L:.VeoltlasticDeveloptrtent, Vol. 1. Academic Press, New Natl Acad Sci (USA) 56:1123-1129, 1966. York. 1969. 49. DiPaolo JA, Donovan PJ, Nelson RL: In vitro transformation of 31. International Agency for Research on Cancer (IARC): On the hamster cells by polycyclic hydrocarbons: factors influencing the Evaluation of the Careinogenic Risk of Chemicals to Humans, C7temicals,14u6cstrial Processes, and Industries Associated with number of cells tratuformed. Natures New Biol 230:240-242, 1971. Cancer in Hunnns, Supplement 4. IARC Monograph. Lyon, In- 50. Gart JI, DiPaolo A, Donovan PI: Mathematical models and the ternational Agency for Research on Cancer, 1982. statistical analyses of cell transformation experiments. Cancer Res ~ 32. United Nation> Scientific Committee on the Effects of Atomic 39:6069-6075, 1979. Radiation (UNSCEAR): Sources and Effects of Ionizing Radia- 51. Borek C, Hall EJ: Transformation of mammalian cells in vitro tion. Report to the General Assembly, with Annexes. United Na- tions, New York, 1977. 52. by low doses of x-rays. Nature 243:450-453, 1973. Clifton KA, Kamiya K, Milcahy RT, et a]: Radiogenic neoplasia 33. National Academy of Sciences, Advisory Committee on the in the thyroid and mammary clonogens: progress, problems, and ~ 4. Biological EfTea.s of Ionizing Radiation. (NASBEIR): 7lu Effects on Populatr'cns of Ezposure to Low Levels of /oni:ing Radiation. National Aatd¢my of Science, Washington. DC. 1980. Ratl JE, Beebe GW, Hoel DG, et al: Report of the National Ituti- possibilities. In Assessment of Risk from Low-Level Exposure to Radiation and Chemicals, A Critical Overview. Edited by AD ~ Woodhead, CJ Shellabarger, V Bond, A Hollaender. Plenum Press, New York, 1983, pp 329-3d4. ~ tutes of Health Working Group to Develop Radioepidemiologica/ 53. Kennedy AR. Little JB: Evidence that a second event in ~

Posstble hazard (HERP %) Tab4e 2. Ranking possible carcinogenic hazards with the use of the methodo+ogy of Amcs et al. (2)' Carcinogenic exposure Average daily carcinogen dqse (70-kg adult) Potency of carcinogen TD50 (mg/kg) Comment+ Man-made chemicalr in foodr/beverages 0.02 Da,minozide in treated apples and apple juice (1987) 20 ug 1.2 (I) 0.002 Da.minozide in peanuts and peanut butter (1987) 1.9 ug 1.2 tf) 0.03 DECP in treated carrots (preregulatory, 1976) 5.1 jig 0.24 ('-) 0.003 DDT. DDD, and DDE in food (preregulatory, 29.0 pg 13 (3) 1968-1969) 0.0003 DDT. DDD, and DDE in food (postregulatory. 2.3 yg 13 (3) , :1980-1982) 0.002 Dieldrin in food (preregulatory, 1968-1969) 1.5 Ag 1.1 (4) 0.001 Dieldrin in food (postregulatory, 1980-1982) 1.1 µg . 1.1 (4) 0.004 EDB in treated apples (preregulatory) 4.1 µg 1.5 (5) 0.0004 EDB in grain products (prerngulatory, 1983) 0.42 µg 1.5 (5) 0.01 PC:Bs in food (preregulatory, 1971) 15 µg 1.7 (6) 0.0002 )'C:Bs in food (postregulatory, 1980-1982) 0.2 ug 1.7 (6) 0.003 Sodium saccharin in diet soda (1977-1978) 4.9 ng 2.100 (7) Natural carcinogens in foods and bevercges 0.003 Aflatoxins in peanuts and peanut butter (1977) 5.8 ng 0.0026 (8) <0.0001 Estragole in basil <3.8 µg 52 (9) 1.6 Ethyl alcohol in beer(1981) 10:2 g 9,100 (10) 0.4 Ethyl alcohol in wine (1981) 2.7 g 9,100 (l0) 1.3 Ethyl alcohol in hard liquor (1981) 8.1 g 9.100 (l0) 0.01 Hydrazines in mushrooms (1977) 0.16 g 20.000 (11) 0,001 DMN in cured meat and bacon (1980) 0.12 µg 0.16 (12) 0.002 DEN in cured meat and bacon (1980) 0.034 Icg 0.021 (12) 0.03 Aeibient air pollutm:ts &:nzene (Los Angeles, preregulatory. 1968) 1.0 mg 53 (l3) 0.0C19 &:nzene (Los Angeles, postregulatory, 1984) 0.32 mg 53 (13) 0.0005 Carbon tetrachloride (U.S. urban and surburban 48 lag 140 (14) areas, 1973-1974) 0.0004 Carbon tetsachloride (U.S. urban areas, 1980) 42 µg 140 (14) 0.0002 DDT (U.S. rural areas, przregulatory, 1972) 2.0 µg 13 (15) 0.00003 DDT (U.S. rural a