Tobacco Documents Online

Research Reports on Tobacco and Health Vol. 5. No. 1 July 1962 IN THIS ISSUE Lung cancer origins, p.1 Adenovirus cancer, p.1 'Competing risks,' p.4 ~~PiMa tt ~rirvilY md ppdM~iar rp.a~ .~ a>~ Ya~` 1~ut1R, , ~ talaymalY '1f1 r.+ 1~a>y pvLisirair Lrt f.rr x atafs. 11 atwa w. (..wi.t. .ritei iile.l: il.t 4% rf~y ~~~~a~ia.lial. rq.aa aa.. .t ti._a. .rn.rt.ia ~ .~k.. _ . . ~ _ . ~ . . . .. < . :~:+-~.is..,...13c:r..h:.... ~a.. asr :..at ai~_... . . .w~..~ ._ . ... .. ._ .._.~.~l8x. Scientists Report ... Human Virus Induces Animal Lung Cancers Three Texas scientists report the induction of lung cancers in labora- tory animals with the use of a human adenovirus responsible for common respiratory disorders.' Adenovirus, type 12, was injected into the lungs of new-born hamsters, the researchers told the April meeting of the American Association for Can- cer Research. Within 5 to 15 weeks, 80% to 90% of the hamsters died of tumors of the thorax. Four animals also developed liver tumors. When transplanted into young adult hamsters the malignant cancers continued to grow, the researchers said. No tumors arose in animals in- jected with control tissue culture fluids. The scientists were Drs. John J. Trentin and Yoshiro Yabe, Baylor University College of Medicine, and Dr. Grant Taylor, M.D., Anderson Hospital, Houston, Texas. ( Continraed on page 3 ) '"Tumor induction in hamsters by human adenovirus." Proceedings o/ the American Association /or Cancer Research, April 1962. Primary Lung Cancers Found in Study of Nuns A survey of lung cancer in nuns disclosed "48 cases of primary carci- noma," two New Orleans scientists re- port.' This "would seem to indicate that lung cancer can occur in women with little or no smoking experience," the authors said. Dr. J. T. Nix and Carolyn Villar- rubia obtained questionnaire data from 483 Catholic general and surgi- (Continued on page 3) "Primary carcinoma of the lung in nuns." Linacre Quarterly, St. Louis, Mo., Novem- her 1961. New Research Work Raises Questions About Origins, Types of Lung Cancer Find Most Cancers Start In Periphery of Bronchial Tree, Not in Main Bronchi Lung cancers originate chiefly in the peripheral areas of the lungs, rather than in the main bronchi, ac- cording to a report by five San Fran- cisco scientists following an extensive stu dy. • Their findings contrast with many previous reports, the authors observed, saying: "It has frequently been re- ported that most of them (lung can- cers) arise in the main and lobar bronchi. "The precise locus or site of origin of bronchial cancer is of interest to students of the air pollution theory of lung carcinogenesis," they wrote. "The concentration of inhaled gaseous or particulate materials should be greatest in the trachea and, in healthy lungs, smallest in the terminal bron- chioles, with variable gradations in the intervening bronchi. "For some unknown reason," they said, "carcinoma of the tracheal epi- (Contituced on paae 2) •"The apparent sites of origin of carcino- mas of the lung." Radiology, January 1962. Drs. L H. Garland, R. L Beier, W. Coul- son, J. H. Heald, and R. L Stein. Study Shows Decline In Epidermoid Cancer Frequency in Males "The proportion of epidermoid car- cinomas in men is not increasing while the proportion of adenocarcinomas apparently is increasing," two scien- tists report after examining 956 pri- mary lung cancer necropsies taken over a 31-year period.' Their findings are the reverse of earlier reports by "numerous au- thors," who "concluded that the in- crease (in lung cancers) is manifested almost exclusively in the epidermoid and undifferentiated types of neo- plasms," they said. They noted that earlier reports indicated that "the fre- quency of adenocarcinoma has re- mained relatively constant." Previous investigators had sug- gested that while adenocarcinomas were of glandular or of unknown ori- (Continued on page 4) ' "Distribution of primary lung carcinomas in relation to time as determined by histo- chemical techniques:' lournal of the National Cancer Institute. December 1961. Dr. Doris L. Herman, pathologist, Los Angeles County Hospital and the University of Southern California Medical School, and Margaret Crittenden. bioatalistician. Nation- al Cancer Institute. Study of 163 U. S. Areas Shows ... Lung Cancer Death Rates Vary Widely Lung cancer death rates for white males in the U. S. vary widely from city to city, a recently published U. S. Public Health Service study reports.' Sharp differences in rates also exist between cities of compar- able size in the same geographic area, the study shows. Similar variations exist in the lung cancer death rates • "Lung cancer mortality, 1949-S1." Public Health Reports, April 1962. for white women. although the range is smaller. The study covered 163 metropol- itan areas in the U. S. It is a sup- plement to a 1957 Public Health Serv- ice report on 102 causes of death. That study listed cancers of the lung, bronchus and trachea as two separate series: (1) primary cancers and (2) cancers unspecified as to primary or (Continued on page 4) TIMN 0115593 T200076

Chest specialist says . . . Basic Cancer Data Being Obscured 4 . i 4 I I , The "wide publicity given to the statistical surveys" associating smok- ing with lung cancer "has relegated to the backgruur.d many basic observa- tions" on the development of the dis- ease, says Dr. Milton B. Rosenblatt, associate professor of medicine at New York Medical College.' "A revival of investigations in this field is warranted," he said in a letter to the editor of the Jourrtal o/ the American Medical Association. He noted that Drs. Wynder and Day, in an article in the March 18, 1961, I.A.M.A., "issued another blast in the lung cancer-smoking contro- versy. "Their passionate communication paid great homage to statistical sur- veys, belittled the necessity for biolog- ical parameters, and concluded with an appeal to common sense," Dr. Rosenblatt wrote. "The latter is an ex- cellent suggestion and merits further elaboration. "During the past decade, epidemi- ologists have applied a relatively new discipline, biometry, to establish the etiology of lung cancer. In a remark- ably short period a problem which defied biological research for 150 years was readily solved by statistical data. The absence of unequivocal bio- logical confirmation was embarrassing but inconsequential. "The problem resolves itself into whether a statistical association be- tween 2 variables, smoking and can- cer, is sufficient to establish a causal relationship, particularly when the same data show a relationship between smoking and respiratory infections, coronary disease, peptic ulcer, liver cirrhosis, and prostatic cancer. Has the etiology of disease been simplified into one universal toxic agent? "The allegation that bronchogenic carcinoma was a rare disease before the era of widespread cigarette con- sumption is unfounded and based chiefly on the paucity of data," Dr. Rosenblatt said. "As bronchogenic car- cinoma becomes progressively more accessible to diagnosis, the rate of in- crease will stabilize as it has for other cancers. "Obsession with statistical data on smoking has led to rejection of nec- • Letter to editor. "Quantophrenia, lung cancer and common sense," 1.A.M.A., April 7. 1962. ropsy studies from European hospitals showing the same quantitative rela- tion between lung cancer and total can- cer in the 19th cecturv as exists today. The increase in necropsy in- cidence of lung cancer has been but a small fraction of the increase in clin- ically diagnosed cases. The 'pan- demic' of lung cancer is a tribute to progress in diagnostic acumen. Ex- foliative cytology, exploratory thora- cotomy, and greater utilization of ra- diology, bronchoscopy, and biopsy are the chief factors responsible. "Lung cancer is predominantly a disease of males, and the attempts to attribute this sexual favoritism to smoking are amusing. The sex ratio favored males before cigarette smok- ing became popular and it has con- tinued so, despite, the prodigious in- crease of smoking among women. Lung cancer also occurs in areas where cigarette smoking is virtually un- known." Peripheral Cancer Sites (Continued /rom page 1) thelium is quite rare. It has been re- ported that carcinoma of the bronchus occurs in descending order of fre- quency from the main to the subseg- mental divisions. What are the facts?" They said "the important point is that many tumors ostensibly primary in large bronchi have actually arisen in a small, more distant bronchus and involved the larger one by direct ex- tension." Reviewing 463 microscopically ver. ified lung cancer cases, the investi- gators found 150 suitable for "reason- able identification" of site of origin by two or more of these methods: radiologic, bronchoscopic, operative, or autopsy. Percentage distribution of site of tumor origin in the identifiable cases was: main bronchus (including inter- mediate), 11%; lobar bronchus, 29%; segmental bronchus, 29%; seg- mental area (i.e., peripheral tumor), 31%. The scientists made this observation on their study: "At first glance, 'an excursion to the bronchial forest' merely to locate the common site of origin of primary lung cancer would seem superfluous in the year 1961. Conversation with pathologists and surgeons, however, revealed a paucity of information on the subject, largely Tobacco and Health Published by THE TOBACCO INSTITUTE. INC. Address correspondence to: Editorial Office Tobacco and Health 150 East 42nd Street New York 17. N. Y. because of the very nature of bron- chial cancer itself. It is usually de- tected only after the evidence of the ex- act site of origin has been destroyed." They quoted from a personal com- munication from Dr. A. A. Liebow of Yale Medical School who, discuss- ing his analysis of 153 lungs.or lobes resected for cancer, said: "Approxi- mately one-third of the peripheral tumors in this series were adenocarci- nomas. It appears to us that over half of all epidermoid tumors begin be- yond a segmental bronchus." Dr. Leo Rigler was quoted as telling the American Association for Tho- racic Surgery in 1957: "A majority of all cases (of lung cancer). arise in a branch bronchus rather than in a ma- jor bronchus ... and many eventually grow centrally to invade a major one . . ." Study Questions Accuracy Of Death Certificates Death certificate diagnoses "are in- su(liciently accurate to permit their use as a reliable indication of the inci- dence of cancer," says a report by two Canadian investigators.' Drs. T. H. Crawford Barclay and A. J. Phillips studied death certificates in Saskatchewan in the period 1950- 1956, when 7,146 deaths were ascribed to cancer of various sites. All deaths in the province in the period were checked, they said, and all relevant data were obtained from physicians, hospitals and records. They said overdiagnosis by site of the disease was shown by the fact that primary cancer of the lung, trachea and bronchus was not confirmed in 96 casea. or 13.6% of 708 recorded cases, compared to 22.7% overdiag- nosis for cancer of the pancreas, 15% for cancer of the stomach and 14.5% for cancer of the large intestine. •"The accuracy of cancer diagnosis on death certificates." Cancer. January-Febru- ary 1962. 2 TIMN 0115594 T200077

Report by Little Cites New Tests, Human Type Tumors in Animals The Scientific Director of the To- bacco Industrv Research Committee cited "several significant develop- ments" in his recently published 1961 Annual Report on research progress. Dr. Clarence Cook Little reviewed scientific findings and discussed these developments: 1. "The means now seem to be at hand" for designing tests to evaluate what influence, if any, each of many external and internal factors may have in inducing lung cancer in animals. Several "activity scales" have been developed for testing the relative can- cer-causing potencies of various sub- stances. 2. By exposing animals to certain viruses along with chemical agents, scientists have induced animal lung cancers similar to those found in hu- man beings. Such results were not ob- tained when either the viruses or the chemicals were used alone. This is an- other step in the "rapid development of methods for study of squamous cell carcinoma in animal lungs." 3. Continuing tests of tobacco smoke indicate that "the smoke is much too feeble to play a role in the etiology of lung cancer as a direct contact car- cinogen. This does not necessarily ex- clude tobacco from having any role at all, nor is any such role established. More investigations are necessary on the suggestions of possible indirect, contributory roles for tobacco." Dr. Little said that the experimental systems discussed in the report "pro- vide a basis for evaluating the rela- tive effects of many extrinsic and in- trinsic factors on the rate and level of incidence" of lung cancer in ani- mals. "The fortunate aspect is," he con- tinued, "that, as the contributions of the various suspects are evaluated, methods of nullifying or blocking some of them may become appar- New Report Available Copies of the 1961 Annual Re- port of the Scientific Director of the Tobacco Industry Research Committee are available upon re- quest to: Tobacco Industry Re- search Committee, 150 East 42nd St., New York 17, N. Y. ' ent . . . Hopefully, effective methods of reducing the toll of human lung cancer may become evident from this type of work." The present status of research, he said, "emphasizes the truth of what we and others have said repeatedly. The origin of lung cancer is complex and still obscure." The Annual Report noted that through 1961 the Scientific Advisory Board, which consists of nine scien- tists responsible for the T.I.R.C.'s re- search program, had made grants to more than 120 independent scientists in over 80 hospitals. universities and research institutions. The T.I.R.C. has so far appropriated $5,450,000 for such research. The 72-page report includes ab- stracts of 60 research reports pub- lished in 1961 by scientists who re- ceived grants from T.I.R.C. A total of 222 such reports was published by grant recipients through 1961. Study Finds Cancers Arise in Lung Scars A "close relationship" between lung scars and the development of lung can- cers is suggested in a study of seven cases of peripheral lung cancers re- ported by two Chicago scientists.' The lung cancers arose in scars, the researchers said, and the histological characteristics suggested a close rela- tionship between the two. "It was felt that [diagnoses of 1 car- cinomas of bronchogenic origin have been too frequently made on the as- sumption that most lung carcinomas arise in the larger bronchi, a situation which seems to require re-evaluation," they said. "In our series the nature of the scars could not be stated with cer- tainty ...( Four suggested ) a primary tuberculous complex ... It is our im- pression that the earlier the surgical excision of lung cancer is performed, the more frequently peripheral cancer of the lung may be diagnosed." * "Peripheral lung cancers arising in scars." Drs. Hidejiro Yokoo and Earl E Suckow, Departments of Pathology, Veter- ans Administration Research Hospital and Northwestern University Medical School. Cancer, Nov.-Dec. 1961. Lung Cancers Found In Non-Smoking Nuns (Continued f ront page 1) cal hospitals in the United States, Can- ada and Puerto Rico. Of the 48 cases, 44 were proven pathologically and four were diagnosed by x-ray, the au- thors said. "As the majority of lung malignan- cies covered by this survey were re- ported in elderly female religious, and as smoking was uncommon in women at the time of their entry into the novi- tiate, the results obtained may well represent lung carcinoma data from the last large group of female reli- gious with no prior smoking experi- ence," they said. The investigators said that in Cath- olic religious orders for women the use of cigarettes is prohibited and "ex- posure to smoke is largely limited to that of kitchen ranges and vigil lights." Tumors Are of Many Types The lung malignancies were of a variety of types, they said: 14 cases were labeled bronchogenic carcinoma, 9 adenocarcinoma, 8 undifferentiated, 7 alveolar, 2 epidermoid, and 8 un- known. Ages ranged from 27 years to 86 years, with 21 being between 60 and 70 years. They noted that "Scientific studies of occupational habits and environ- ments in many fields of industry have shown that individuals and groups can acquire predispositions to some dis- eases, and relative immunity to others." The authors said their study was supported by the Committee on Medi- cal Care of Clergy and Religious of the National Federation of Catholic Physicians' Guilds and the Catholic Hospital Association, and by the Louisiana Division of the American Cancer Society. Hamster Virus Cancers (Continued f rom page 1) Houston, Texas. The American Cancer Society, which helped support the research, said in a news release: "The human findings add a new dimension to the possible role of viruses in human can- cers. While the hamster cancers by no means constitute proof that human cancers are virus caused, they stimu- late once more the conjecture that some human cancers could be caused by viruses." 3 TIMN 0115595 T200078

Epidermoid Cancers Decline in Frequency (Continued from page 1) gin, epidermoid lung cancers might be due to external causes. The scientists noted that in the past. "Several inves• tigators suggested that exogenous fac- tors, espeeially cigarette smoking, are of greater importance in the produc- tion of epidermoid and undifferenti- ated carcinomas than of adenocarcin- omas." The new study showed that the pro• portion of epidermoid carcinomas for 1945-57 was well under that for 1927- 44; while the proportion of adenocar• cinomas was higher for 1945•57 com• pared to 1927•44. The authors also said a large pro- portion of the tissues that appeared to be epidermoid cancers when viewed under a microscope were actually adenocarcinomas, as determined by chemical tests which showed a glandu• lar origin for the tumors. Their study was divided in two parts: the classification of lung tumors by the size, appearance and distribu- tion of cancer cells in the tissues exam- ined under a microscope; and a study of the chemical nature of the tissue sections. The investigators reviewed 1.037 primary lung cancer necropsies from the Los Angeles County Hospital, taken from 1927 to 1957. Of this num• ber, 956 were found satisfactory for complete histochemical study. They reclassified the tumors accord- ing to two systems: one based on the system used in the U. S., which was devised by Dr. A. A. Liebow of Yale University; the other system used by Dr. Leiv Kreyberg of Norway in 1952 and since adopted by the World Health Organization. Dr. Kreyberg's view has been that epidermoid lung cancers occur mostly in males, are rising in frequency, and result mainly from cigarette smoking; while adenocarcinomas of the lung show an equal sex distribution, are not rising in frequency, and result from unknown or hormonal factors. The new report said : "Independent of criteria used in our series, the proportion of epidermoid carcinomas in men is not increasing, while the proportion of adenocarcino- mas apparently is increasing. "Our ratios of epidermoid and un- differentiated carcinomas to adeno- carcinomas are decreasing, while ra- tios reported by Kreyberg and others are increasing." The authors said the difference in results "is in need of explanation. Cer- tainly, such environmental factors as geographical location, variations in quality and quantity of air pollutants, and socioeconomic factors must be considered . . ." 'Competing Risks' . . . Diseases Have More Than One Cause The consensus today is that all dis- eases have more than a single cause, says Dr. George L. Saiger of Colum- bia University in an article on the uses of epidemiology.' An example is lung cancer, he said, and several host and environmental factors have been implicated, includ- ing age, sex, heredity, nutrition, chronic infections resulting in meta- plasia and scarring, viruses, hor• mones, tobacco smoke, occupational exposure to nickel, chromates, asbes- tos, inorganic arsenic and radiation and air pollution. "In spite of this, some people still believe that tobacco smoke is the only cause of lung cancer," Dr. Saiger's article said. • "Ten uses of epidemiology." Canadian Medical Association Journal. Oct. 28, 1961. "These facts cannot be disputed: tobacco smoke is not a proven cause; and persons who never smoke also de- velop this disease. For reasons such as these, it must follow that the prob- ability that an individual with a cer- tain characteristic will develop a dis- ease is not necessarily the probability that he will develop the disease as a result of being exposed to a substance identified with that characteristic un- less it is known unequivocally that no other substance or environmental fac- tor is a cause of the disease in the same species. The above wording has been carefully chosen to preserve the notion of competing risks in disease etiology." Dr. Saiger's article discussed the present status of epidemiology and said one use of it is to aid in the search for causes of disease. Lung Cancer Death Rates (Continued from paae 1) secondary loriginating elsewhere and moving to the lung). Combining primary and unspeci- fied cancers- the report said, has "not heen of maximum value because in practice reported deaths from lung cancer unspecified include varying proportions of deaths from primary lung cancer in different (U.S.) areas." The 1957 study said: "The reason for excesses of mortality in particular areas is often obvious. Sometimes they are due to the concentration of cer- tain occupational groups; in other cases. persons with existing disahili- ties have tended to migrate to certain areas. Where the reason is not imme- diately identifiable, the excesses pro- vide an area for research on the in- fluence of environmental and other factors." The body of the new study consists of a table giving the ratios of age- adjusted lung cancer mortality in the 163 areas to the total U. S. white pop- ulation, by sex, 1949•51. The tabula- tion notes: White males in N'ew Orleans showed twice the lung cancer mortality com- pared to the entire U. S. Charleston, S. C., showed 70% more, and New York City, Buffalo. Baltimore and vIo- bile showed 60c o more. For white females, Shreveport, La., showed 90% more than the national average. and Charleston. S. C., 70oJo more. The cities most below the national average for white males were Lima, 0., and Green Bay, Wis., both 60~Jo below. For white females. Sioux Falls, S. D., and Ogden, Utah- were 80% below the national rate. The tabulation also showed some marked differences in rates between cities of comparable size and in the same state or area. For white males. the rate in Dayton, 0., was 10% below the national average. while Toledo was 10c,'ii+ above. Fresno. Calif.- had an average lung cancer death rate. while the rate for Sacramento was 40% above. In Worcester, Mass., the rate was 20% below the average. while Bridgeport, Conn., was 30~~c above. In addition, the rates for the two sexes in the same city also showed some sharp variations. In the Duluth, Minn.-Superior. Wis., area. the male death rate was 20% below the national average while the female rate was 30~Jo above. Conversely, the male rate for Washington. D. C.- was 40% above the national average while the female rate was 10~fo below. TIMN 0115596 4 T200079 PR/NTEO IN U.S.A.