Tobacco Documents Online

The DDT : Example Dr William Hazeltine

For Presentation to a Seminar Sponsored by the International Center for Scientific Ecology, May 10, 1993, at Paris, France TITLE:Is the Concept of a Linear Relationship Between Dose and Effect Still a Valid Model for Assessing Risks Related to Low Doses of Carcinogens--the D. D. T. Example. William Hazeltine, Ph.D., B.C.E., Oroville, California, USA It is gratifying to think that a Scientific Organization would want to look at the old and new data on carcinogenicity of DDT, 20 years after the United States Environmental Protection Agency (EPA) Administrator announced that DDT was a carcinogen. The Administrator made this decision and overturned his hearing Officer's findings concerning the risks of allowing DDT's continued use in the United States. Background You have to understand that I am an Entomologist-Ecologist by training, who used DDT for crop protection, and for control of public health pests. I tried to understand the details of risk from such use, because I was the applicator on many occasions, and I was often exposed to higher doses that most others in the area being treated. In addition, I have continually practiced an approach to pest control which begins with a threshold determination, to decide when and if control is necessary. In many areas, this is called Integrated Pest Management (IPM). I began to look at DDT in far greater detail in the early 1960s, when Rachel Carson's book "Silent Spring" appeared, particularly because her narratives were not consistent with my oun experience. If I had been mistaken, I wanted to be the first to know. At the time, I was working on control methods for an aquatic midge in Clear Lake, California, and Ms. Carson had told about the earlier control efforts for this midge in Clear Lake, in infinite (and less than accurate) detail. A few years later, I was listening to a founder of the Environmental Defense Fund (EDF) who told our State Legislature that "anyone who used or recommended the use of DDT did not know his business ". As part of the EDF presentation, which include d an unequivocal statement that DDT causes cancer, I obtained the book of photocopied papers which the EDF witness had said were the "very best of the Literature on DDT". Thirty years later, I am still looking for convincing evidence to show that I was wrong in allowing the use of this pesticide for food production and health protection. In the scientific field dealing with pesticides, there is a need for more people to read the literature in detail, because often times, there exists a temptation, even among scientists, to use "adversary science" in place of objective science. There is also a tendency to rely on the abstract or summary, in place of reading the whole article. Adversary science is the art of telling only the facts which are favorable to one's own belief, and hoping that the opposition (or the media) will not be able to cite the data which was omitted. Right now, the U.S.Supreme Court is trying to decide what is acceptable scientific

2 evidence. This problem has come into focus, because we are allowing lawyers and judges to replace the older and more appropriate method of scientific debate. A real scientist will be the first to try to refute his own hypothesis. Part of my committment to look for the "facts" about DDT, and having seen apparent abuses of scientific integrity, led me to volunteer to help the lawyers for the U.S. Department of Agriculture and the manufacturer of DDT in the U. S., when they examined the witnesses from the Enviroamentalist's side, in the 6 month long public hearing on Cancellation of the uses of DDT. As some EDF members had bragged, they would be happy to submit to "the crucible of cross-examination ." This was an excellent learning experience, and it provided me with the exhibits and sworn transcripts of the interesting parts of the proceedings. The issue of carcinogenicity of DDT was one of the topics covered, and it is still under debate. Separating the Issues The specific topic of this seminar, "is there a linear relationship between dose (and time of exposure) and the development of cancers in animals," requires some ground rules, before the question can be answered for the chemical DDT. These separate topics need consideration. 1. Is each isomer and metabolite considered separately or are all of them considered as one chemical? The literature on testing seems to be based both on purified and "Technical" DDT, neither of which are expected to contain much if any of the metabolite DDE. Good protocol should specify the use of a purified single chemical of known molecular structure. 2. Are the doses limited to those which do not damage the homeostasis of the test animals? DDT is considered to be a relatively cumulative chemical in vertebrate animals, and lipid residue data for all test animals at sacrifice, correlated with presence and type of cancers could be a better measure of dose response than constant feeding at given doses in food or gavage of neonatal animals. 3. With DDT, it is imperative to decide whether the dose- response data should be consistent for all test animal species or hybrid classes, or only for one or more kind or species of test animal, or for larger taxonomic groups of test animals. 4. The time of exposure and the latent period before autopsy is important, as well as the value of multi-generational studies. 5. Is the interest in the cancers only in the test animals, or is the interest to provide predictions of risk for humans, or both? 6. Should tumors be counted as malignancies? As I read the Background Statement for this Symposium, it seems clear that the discussion should try to compare, or at least be aware of the conditions imposed in the U.S. by the Delaney Clause (no cancers in any test 'animals under good test conditions) as well as to the "no-effect dose level coupled with a good safety factor" system for setting allowable residue tolerances in raw agricultural commodities. A corollary requires 2501171304

3 a consideration of whether or not there is a true "no-effect" level for a carcinogen; this could involve use of epidemiological considerations dealing with large areas and human or other animal populations exposed to higher or lower doses of DDT for prolonged periods of time. Extrapolation From Laboratory Animals to Humans During the DDT hearing testimony of Dr Marvin Schneiderman, a statictician with the National Cancer Institute, it suddenly occurred to me that aDT, as a 25 or 502 "Tracking Powder" was very effective in controlling mice and bats, but it did not control rats. I understood the difference to be that rat livers were like human livers in their response to DDT, but mouse livers could essentially not be induced with DDT. Dr. Schneiderman testified, as I recall, that the protocol for laboratory cancer research required that two species of test animals be used, and that both of these animals must not have a known physiological system difference, if the data was to be used to extrapolate the dose response information to humans. The following exchange took place between Dr. Schneiderman and a lawyer at the Public Hearing (Jan. 12, 1972, page 7142): Q. If the purposes of these studies, Doctor, is to some day draw some extrapolation to man, would you consider it a drawback that the metabolism is different in mice and man? A. If, as you say, it has been demonstrated, then clearly one should not use an animal that clearly has a different metabolism than man as an experimental animal.-- Later in his testimony, Dr Schaeiderman was asked if transplantation had any relevance to carcinogenitic studies, because there was concern about how to accurately determine what was a cancer. His answer was: One of the criteria for malignancy is the transplantability of a tumor. If a piece of tissue does not transplant and grow up into, another tumor, it's assumed that it may not be malignant. Dr Schneiderman admitted that transplanting was not done in the Bionetics study which he called a "minor defect" in that study. Between mid February and November of 1972, 1 exchanged letters with Dr. Schneiderman. The first references to species differences in liver inducibility were (1) J.W.Gillett, 1970. in Biological Impact of Pesticides in the Environment. Oregon State Univ., Corvallis Oregon, which deals with Japanese Quail and (2) Hart and Fouts, 1965. Arch. f. Exp. Path. and Pharmak. 249:486-500, which considers sleep time chances with barbiturates when test animals are dosed with DDT and other chlorinated hydrocarbon pesticides. My last letter, which was not answered, included a copy of a paper by J. R. Fouts, 1972. in Environmental Health Perspectives, October 1972, 55-66, that contains the following: ---. Thus, at doses and schedules of doses and at times after last dose of the inducer that clearly established hepatic microsomal enzyaae induction by DDT or beazpyrene in rats and other species including monkeys, the mouse did not show such induction or showed it only

q marginally. (references omitted, emphasis in text) This article gives some evidence for variation in inducibility for rats and mice, but the differences in tolerated DDT doses between mice and rats in the major papers usually relied upon for determining carcinogenisis, clearly points to physiological differences, with the mouse being aberrant, compared to rats and humans in its response to DDT treatment. This article also has a long list of references concerning indwcibility. A later paper by L. Tomatis et al., 1973 (Int. J. Cancer 12: 1-20) reports that tumors were found in Rat livers following doses of DDT. ?he only reference to rats was a paper by Fitzhugh and Nelson, 1947 ( J.Pharm.& Exp. Therap. 89:19-30). No other organs besides the liver were involved, and only tumors were found, according to Tomatis. In 1986, EPA published "Guidelines for Carcinogen Risk Assessment. (1986 Vol 51, of the U.S. Federal Register, pages 33992-34003 and reprinted in EPA/ 625/3-90/017, Sept. 1989). Some of the pertinent statements are: --- The strength of the evidence supporting a potential human carcinogenicity judgment is developed in a veight-of-evidence stratification scheme.(p33994) The weight of evidence that an agent is potentially carcinogenic for humans increases---:(3) with the occurrence of clear-cut dose response relationships as well as a high level statistical significance of the increased tumor incidence in treated compared to control groups; (4) when there is a dose-related shortening of the time- to-tumor occurrence or time to death with tumor; and (5) when there is a dose-related increase in the proportion of tumors that are nalignant.(p33995) --. A statistically significant excess of tumors of all types in the aggregate, in the absence of a statistically significant increase of any individual tumor.type, should be regarded as minimal evidence of carcinogenic action, unless there are persuasive reasons to the contrary (p33995) In the absence of appropriate human studies, data from species that respond most.like humans should be used, if information to this effect exists.---(p33997) Major Studies The "Bionetics" Study (Ianes et il. 1969. J Nat. Cancer Inst 42:1101-14) was an ambitious project which was intended to screen a 120 chemicals on mice at doses which were close to that which would cause toxic symptoms. For DDrT, the dose was 46.4 mg/Rg. by stomach tube from day 7 through day 28 of age, and the dose in the food was then 140 ppm for the remainder of an 18 month test period. The mice used were two essentially new crosses, using two different male stocks and the same mother stock for both crosses. The test began with 18 mice of each gender and of each cross (72.aice per chemical). All mice tested for each chemical were given the same dose of chemical, apparently for the purpose of answering only the

5 single question of could the administered chemical induce tumors or cance rs. Dr Hans Falk, one of the Co-authors of the Innes et al. paper presented a speech which was later printed as an article (Nat'l. Pest Control Operators News, Octobe r 1971 p4-6 and 29-31).. In this presentation, Dr Falk provided comments about the test methods and the implications of the test results. He also commented about the way the test mice were bred and the fact that the suitability of these crosses for cancer testing was unknown. There apparently was no history of cancer susceptibility of these crosses. Dr Falk told about various parts of the test, then he said: We came up with positive results, for instance, a significant number of idepatomas in mice on 140 ppm DDT in the diet.---. But as we decrease the DDT level to that which we [humans] are exposed to now, most likely we would have to conclude that the hepatoma incidence would be 0 X.(emphasis added) One paragraph of the Falk article is long but important to our understanding. It says, refering to the Bionetics Study: The lack of malignancy of the liver and lung tumors produced a controversy which while quite artificial was, nevertheless, quite heated. The Delaney amendment compels the Food and Drug Administration to ban any chemical for use on or in food if it can produce cancer at any dose leve3l and by any mode of administration. It specifically states that the tumor has to be a cancer. So the question arose, did these tumors have a tendency to become cancers? Three types of tumors were observed in these studies. They were either hepatomas, and by definition a hepatoma is a benign tumor, i.e. it does not invade surrounding tissue or metastasize to other organs, in other words it grows confined to the organ of origin. With time the tumor may become invasive, in which case its cells attack the surrounding tissues, but it may as yet not spread to distant tissue like lung or lymph nodes. At that stage, it would be malignant. This behavior is quite characteristic for mouse tumors. They do not 9uite behave like human cancers. Whereas a hepatoma is a benign tumor, it is generally conceded that given another six months, it would most likely have become malignant and killed its host. But you can appreciate that the FDA has a difficult decision to make because at the moment of sacrifice at 1$ months, most of the tumors were benign. Lung tumors, the second type of tumors, similarly were benign tumors which probably would have turned malignant with time. The only neoplasm that was recognized as a malignant tumor was the reticulum cell sarcoma (a type of lymphoid tumor.) (Emphasis added) The data published for the Bionetics Study (Innes et al.,cited above) shows "lymphomas" as occurring at a 0.05 confidence level for all mice tested, with females of each hybrid type having "lymphomas" at the 0.01 confidence level. These levels of confidence do not suggest clear and convincing evidence of carcinogenicity, unless one wants to enter into the speculation or "educated guess" type of logic which Dr Fa llc referred to. Dr Falk then goes on to explain that if the tests were allowed to go to 2

years, there would have been natural mortality, with cannibalism or decay "when no one was looking." In early 1974, following the cancellation in the U.S. of most uses of DDT about 2 years earlier, an Appropriations Subcommittee of the U.S. Congress asked EPA "To Initiate a Complete and Thorough Review Based on Scientific Evidence of the Decision Banning The Use of DDT." Testimony by Leonard Axelrod, Ph.D., the EPA person who was doing the review is recorded in the transcript of hearing for April 4, 1974. It reads in part: 1. There is at the present time, no evidence that DDT is carcinogenic (or tumorogenic) in any animal species when administered at levels less than two orders of magnitude higher than the maximum dose attainable by plant manufacturers and workers over a lifetime of exposure.--- 6. There is at present no evidence from experimentation that DDT is either teratogenic or mutagenic. 7. There is no evidence from human monitoring studies over a 5-year period that DDT as occurring in the environment has caused an increase of any known physiological dyscrasia A copy of a later draft of this Report, dated June 5, 1974 was obtained from a student who was allowed to work at EPA in Washington, D.C. during the time this study was in process. In the part dealing with "Carcinogenesis and Tumoragenesis"(p.21-22), Dr Axelrod lists the published reports available since January 1972 which he had thoroughly reviewed. The list and conclusions with Dr. Axelrod's hand written corrections, suggest a reasonable review of the pertinent literature available at that time. Discussion and evaluation of the papers used by the environmental "experts " in the record of the public hearing are included by reference by Dr. Axelrod, in this draft. One of the conclusions in the 6/5/74 draft appears to be based on the studies referred to as the Lyon and the Milan Studies (Turusov et al.1973 (J. Nat'1. Cancer Inst. 11:983-97) and Terracini et al.1973 (Int'i. J. Cancer 11:747-64) respectively. It is important enough to be quoted here: ---. If all tumors were considered carcinomas then an extrapolation on a linear-dose-response•basis would correspond to one or two additional cases of liver cancer in 200 million humans--. If mice have a predictive value for human risk of cancers, which appears to be in serious disagreement with the EPA Guidelines (Protocols?), this data would suggest a of 1 to 100 to 1 to 200 million added risk value, or 100 to 200 times what our EPA has proposed as a de minimus cancer risk level for residues of pesticides in food. The conclusion of Dr Axelrod in June of 1974 was changed in December 2, 1974 Working Paper. In it he says: '--. One can conclude that available evidence indicates that the iikelihood that DDT is a human carcinogen is very low.

Note: Dr Axelrod died suddenly, shortly after the preparation of the 12/2/74 draft. The final Report by EPA is dated J uly 1975, and it does not read anything like the Axeirod writings. The final Report appears to have been sanitiz ed, possibly to sustain the EPA Administrator's earlier conclusion that DDT was carcinogenic. Dr. Samuel Epstein, a witness called by the EPA and the EDF at the Public Hearings on January 14, 1972 has a different perspective of human riskthan Dr. Axelrod. Epstein testified, on page 7370: Q. Based on your experience and upon your survey and analysis of the literature, would you give us your professional opinion, if you have one, whether the presence of DDT in the human environment represents a significant carcinogenic hazard to man? A. I can only answer that qualitatively, and the answer is yes,--. Q. In your professional opinion, can a man be safely exposed to any leve 1 of DDT? A. the answer is no. During his testimony, Dr. Epstein who said he was on several advisory committees to the group which produced the document known as "the Mrak Report", introduced 4 pages of a paper referred to as "the unpublished Fitzhugh Report," which Dr. Epstein characterized as one of only 4 studies up to that time, "-- which provide definitive evidence of carcinogenicity." This manuscript (O.G.Fitzhugh et al.,undated, titled "A Summary of a Carcinogenic Study of DDT in Mice- Unpublished data from the Bureau of Science, Food and Drug Administration), along with 2 attached memoranda (dated 1/30/69 and 9/15/69) were only made available after this witness was excused, and no longer available for testimony . The total document shows some possible reasons why this study was not published, and does not invoke a feeling of confidence in the predictive value of this study. Page 33 of Attachment number 2 contains information to show: Unusually high mortality in some male mice during the second year of the 2 year study resulting in relatively few animals at risk when they were most likely to exhibit tumors. Report of a tumor of the testis in a female mouse. Report of an ovariaa tumor in an animal reported lost from the animal quarters earlier. A problem with the DDT dose occurred in the 67th week, where the 100 PPM dose was increased to 300 PPM, resulting in increased mortality.(page 4 of Attachment 2, dated September 15, 1969) Note: The Report shows 38 of 100 male, and 22 of roo female mice died between weeks 62 and 66.) ' The unpublished Fitzhugh Study used only one dose of DDT, so no dose- =esponse information ca n be drawn from it.

19 Dr. Epstein characterizes the results of the research reported by Hayes et al. 1971(Arch, Env. Health 22:119), Laws et al.1967(Arch. Env. Health 15:766), and Laws 1971 (Manuscript submitted to EPA May 13, 1970) as "inappropriate", yet the first two of these papers report on work done in humans, and the third showed apparent inhibition of the "take" of transplanted cancer cells in mice, when the mice had been treated with DDT.(see discussion under Benefits) There are tests reported where DDT was administered to dogs (Lehman, 1952. Q. Bull Assoc. F4D Officials of US.16:47, and later Ottoboni 1977 (Arch. Environm. Cont. & Tox. 6:83-101), Hansters (Agthe et al. 1970, Proc. Soc. Exp. Biol. & Med. 134:113) and Rhesus Monkeys (Durham et al. 1963, (Arch. Int. Pharmacodyn. 141:111). All of these tests, except for Ottoboni's work are questionable because of the conditions of the tests. Ottoboni tested purebred Beagle dogs continuously dosed for 3 generations at 10 ppmDDT in their feed, with sacrifice at age 2 1/3 years. No adverse effects due to the treatment were found. She reported (personal communication) that there was observable beneficial effects in the treated dogs in this multigenerational study; she also said that the tech nicians began to refer to the untreated control dogs as "the DDT deficient animals." Appended to this paper is a copy of some tables presented by Carrol Weil who was with the Mellon Institute in Pittsburg, Pennsylvania. These tables were part of a presentation to the Entomological Society of America annual meeting in 1975, to show some of the problems associated with trying to interpret the kind of data that can come from animal assays for carcinogenicity in mice. In the Table marked "A", if by chance the male group number 5 had been the control group, all the other untreared groups would have been called positive. "B " shows the potential effect of diet type and "C" shows the difference in tumors when mice are allowed to eat all they want, compared to some limitation on the amount of food they receive. The study reported by R. Kimbrough et al., 1964 (J Nat' 1. Cancer Inst.,33:215-25) confirms that Rats also can show adve rse effects (leukemia) due to use of a purified diet, and not due to the DDT used in their tests. Conclusions About Cancer Studies The most reasonable conclusion I can draw, based on the review of the references I have looked at and the other materials available to me concerning DDT and cancers in test animals, is that there is not enough data to get any kind of clear picture about dose/respoase for cancers with DDT in mice, if such a relationship exists. Furthermore, there will probably be few if any new studies because DDT is essentially gone from the market-place in the {I. S. , which is the source of most of the funding for massive cancer research programs. If one accepts the idea that mouse livers are essentially non-inducible with DDT, then there seems to be two alternative and perhaps connected hypotheses to work with. One is that the mouse studies are really looking at animals whose physiological systems are overwhelmed by a foreign chemical and that adverse results occur because of this disruption of homeostasis in the test animals, and not due to a site or system specific carcinogen. The second hypothesis is that there is no dose/response correlation for cancers or hepatomas, when DDT is L11 ~ E.. ~ -.] W O

9 administered. Both of these hypotheses neglect the question of whether or not DDT is carcinogenic in the animals tested so far, and the most reasonable answer to this question is that the evidence available to date does not support the idea that DDT is a carcinogen. Two studies with humans have been reported, and both are negative for definitive adverse effects. These were the studies which Dr. Epstein dismissed as inappropriate. Hayes et al.1971 (cited earlier) fed DDT volunteers doses of DtJT at rates of up to 35 mg. per day for 21.5 Months, and some of the people were observed for 5 years from the initial feeding period. The dose was calculated at 535 times the average daily intake in the U.S., and no adverse effects were observed. Laws et al. 1967, (cited earlier) followed 35 men with 11 to 19 years of high exposure at a facility where D17T was manufactured, and these men showed no adverse effects attributed to exposure to DDT. Based on storage and excretion levels, the average daily intake was estimated at 3 to 18 mg. per man per day. This compared to .04 mg per day for the general population. Fat residues for these workers was 39 to 128 times the level in the general population. Even though the evidence is negative, it seems to be of more value in assessing the risks of DDT exposure to humans than all of the mouse data accumulated over the years. Even with the use of DDT beginning with World War II, and extending up to 1972, any evidence of increased human liver cancer rates appears to be absent. This may be analogous to a hypothetical cancer researcher saying "We must study reality to see if it works out in theory." Beneficial Effects of DDT in Animal Tests It is interesting to look at the tests which do not cause some "harm", but rather show some measurable results which seem to be beneficial to the test animals. Once in a while, there may be a surprise and the expected results may not occur. We all need to be open to "letting the data tell us what it has to say. M. A. Ottoboni, 1984 (The dose makes the poison, Vincente Books, Berkeley, California, 222 pp.) provides an understandable discussion of the types of chemical carcinogens. The list of types include: Primary Carcinogens-They directly start the cancer process. (example-itadiomimetic drugs) Procarcinogens-Not carcinogenic, but may be converted to a carcinoge n (example-Senzpyrene ) Cocarcinogens (-promoters)-Enhance the carcinogenic action of another chemical. They may change the rate of metabolism of carcinogens, alter biochemical pathways or interfere with the repair mechanisms that would otherwise reduce the effect of the carcinogen. Secondary Carcinoge ns-Not carcino gens, but they indirectly cause tissue damage. (example-Oxalic acid may produce bladder stones which over time can irritate the bladder and cause cancer from the irritation) N Ln ~ ~ ~ W ~ ~