Tobacco Documents Online

i C:t<r<<<r-.. The Effects of Cigarett_e Smoking U2on Dogs Research program supported by the American Cancer Society and Veterans Administration and presented by E. Cuyler Hammond and Oscar Auerbach at special scient;_fic meeting of the society at Waldorf Astori3 Hotel, New York City, February 51, 1970. In its pronouncement that "for the first time, scientists have produced lung cancer in a significantly larger experimental animal as a result of heavy cigarette smoking", the American Cancer Society acknowledged that all previous pub- iicizad claims were false. The evidence now presented and the conclusions drawn from these experiments are also subject to considerable criticism. Design of Experiment The experiment started with 97 dogs certified healthy but 3 of the dogs were quickly eliminated because of brain tumor, bronchopneamonia, and pulmonary infarction, respectively. The remaining 94 dogs were subjected to tracheostomies kept patent my means oi hollow Teflon tubes. There were 8 dogs used as controls. The rest of the dogs were divided into groups accord- ing to whether they smoked filter or non-fr ilter cigaret-tes. The latter category was subdivided into light smokers and heavy smokers. The heavy smoking dogs were further segregrated into two groups accordin; to ;~eight. It was the inter.tion of the

-2- experimenters not to sacrifice the dogs weighing the most but to keep them smoking until all eventually died. All the heavy smoking dogs were exposed to 9 cigarettes daily f or a period of approximately two and a half years. (9 cigarettes daily for dogs weighing 20-39 pounds would correspond to approximately 60-80 cigarettes daily when related to human male exposure) Results The number of dogs examined pathologically was 68 which included 12 in the heavy weighing group not intentionally sacrificed. There were tumors found in 36 of the dogs, some with multiple tumors. In 24 dogs, the tumors were non-invasive of the bronchioloalveolar type. ("Many so small they would probably not have been found by routine autopsy") Invasive bronchiolo-alveolar tumors were found in 12 of the heavy smokers of non-filter cigarettes; 2 of these dogs also had early squamous cell carcinomas. C omment The findings are purported to confirm the cigarette- cancer link previously suggested by the statistical studies. The histologic findings are of 3 types, namely, non-invasive bronchioloalveolar tumors, invasive bronchioloalveolar tumors, and early squamous cell carcinoma. The results in each category will be discussed separately.

Non-invasive bronchioloalveolar tumors are, in actu- ality forms of benign epithelial proliferation of the terminal bronchioles. They occur commonly in all types of acute and chronic lung infections and calling them tumors is a question of semantics. The dogs were subjected to tracheostomies and the smoking dogs to repeated exposure of the lungs to the external environment. It is surprising that only 24 of the 36 dogs shoc•red epithelial pro- liferative changes. It is also of interest that in Auerbach's group H (lighter weighing dogs smoking non-filter cigarettes) there were 19 non-invasive tumors in 24 dogs whereas in gxoup h (heavier dogs smoking non-filter cigarettes) there were only 4 non-invasive tumors in 12 dogs. The attempt to relate cigarettes to the etiology of lung cancer on the basis of these findings is highly imaginative and. borders on experimental desperation. Benign epithelial prolifera- tion was described as early as 1876 by Friedlander (1) and a spate of reports(2-13)~ has since appeared in the medical literature re- lating the condition to bronchitis', pneumonia, pulmonary infarction, bronchiectasis, pulmonary fibrosis, and thromboembolism. Benign epithelial proliferation (fiuerb ach's non-invasive bronchioloalveolar tumors) has been commonly observed at autopsy since its first description a century ago. It is found in children and in adults, in smokers and non-smokers. It is irresponsible to associate N O this finding with lung cancer because it occurred in smoking dog,s.U1 O Auerbach, himself, found it in 25 per cent of his non-smoking ~ controls. .J O

Invasive bronchiolo--alveolar tumors were found in 12 (33.3 per cent) of 36 heavy smoking dogs exposed to non-filter cigarettes. If these resul.ts were extrapolated to experience in humans there would be at least 20,000,000 cases of bronchiolo- alveolar carcinoma inithe United States. Equally paradoxical was the proportion of invasive bronchiolo-alveola:r tumors to the total number of malignant tumors in the dogs. There were 17 tumors (15 invasive bronchiolo-alveolar and 2 squamous cell carcinoma in the 12 dogs) so that the incidence of invasive bronchiolo-alveolar tumors among all the lung cancers was almost 90 per cent. In human lung cancer the vast majority of cases are squamous cell carcinomas and the bronchiolo-alveolar cancers are an almost negligble quantity. In a study of 4,068 cases of primary lung,cancer at the Memorial Hospital in the period 1926-1960, Watson (14) found the incidence of bronchiolo-a:lveolar carcinoma to be 6.5 per cent. Inasmuch as the resemblance of primary bronchiolo-alveolar car- cinoma to metastases from extrathoracic glandular carcinomas was not fully appreciated by pathologists in the early years of the study, it is probable that the true incidence was even lower (15-21). Auerbach explained the disparity between the large pro- portion of invasive brcnchiol.o-alveolar tumors found in the dogs and the small percentage found in hunans by rationalizing that some bronchiolo-alveolar tumors are reported under the term, adenocarcinoLqa. While this may be true it offers a very tenuous

explanation for the paradox of invasive bronchiolo-alveolar tumors constituting 90 per cent of the lung cancers produced. Adenocarcinomas provide a small percentage of all lung cancers in humans. In a series of 676 lung cancer autopsies at the Memorial Hospital Watson (14) foundthat the number of adenocarcinomas and'bronchiolar carcinomas, combined, were less than one third of the total lung cancers. Liebow (22) estimated that approximately 15 per cent of pulmonary cancers were adenocarcinomas. Haenzel et al (23) found the incidence of adenocarcinoma and alveolar cell carcinoma, combined, to be 21 per cent of all lung cancers. As further justification for his bizarre findings, Auerbach referred to an epidemiologic study (23) showing that the standardized mortality ratio for adenocarcinomaa of the lung was 26 times as high among male smokers of more than 1 pack daily as among male non-smokers. Without discussing the validity of these findings it is very apparent that they have no relevancy, wha.tsbever, to Auerbach's experiments which produced bronchiolo- alveolar tumors. It should be further noted that Auerbach culled from the above report only those findings which gave him comfort and omitted to state that the same study found only 59 alveolar cell carcinomas out of a total of 2,381 lung cancers. Actually only 27'of the 59 alveolar cell carcinomas were confirmed by autopsy. But even if all had been confirmed this would give an incidence of 2.5 per cent of alveolar cell carcinomas among all the lung cancers contrasting considerably with Auerbach's f inding of 90 per cent in the dogs. Whatever was produced in these exper•i-- mznts bears little relationship to lung cancer in humans.

-6- A clue to the Auerbach findings may be obtained from studies of spontaneous lung tumors in dogs. Nielson (24) found that most of the spontaneous primary tumors of the lungs were located in the peripheral areas of the lung and were classifiable as bronchiolo-alveolar carcinomas; it is significant that 2 of his 11 spontaneous lung cancers in dogs were squamous cell car- cinomas. In another veterinary study by Cohen, et al (25) there were 22 spontaneous primary cancers diagnosed of which 9(4Gper cent) were of the lungs and bronchi. It was also noted that the risk of acquiring a malignant tumor increased with the age of tre* dog (Auerbach's dogs had doubled their age at the end of the experiment). Dahme (26) found that adenocarcinomas were the most common spontaneous lung tumors in domesticated animals but his illustration of adenocarcinoma closely resembled the histologic features of bronchiolo-alveolar carcinoma:. Animal experimentation has shown that when lung tumors were produced in beagles, the tumors were usually bronchiolo- alveolar carcinomas. Park, et al (27) using,plutonium aerosols via masks produced 6 primary lung cancers in 42 exposed dogs; all were bronchiolo-alveolar carcinoma and in 5 of the dogs the N tumors originated in more than one lobe. In another experiment, ~ UI Andersen and Guttr.ran (28) expcsedbeagles to whole body radiation Q and produced 7 lung cancers with histologic features typical of ~ bronchiolo-alveolar carcinoma. ~ W It would therefore appear that when lung cancers occur i'n,dog:, whsther spontaneous or experimentally produced, the tumors

histologically are bronchiolo-alveolar carcinomas in the vast majority of cases. This is an inherant biologic feature of lung, cancer in animals. The pro.duction of these tumors in dogs by cigarette, or other exposure, has little relevancy to lung can- cer in humans in whom bronchiolo-alveolar carcinomas occur in-- frequently and are often found to be metastatic from extrathoracic cancers erroneously diagnosed. (.howhere in the Auerbach pre- sentation is it stated that complete autopsies were done to exclude , primary cancers elsewhere metastasizing to the lungs). There were 2"Early squamous cell carcinomas" found in Auerbach's experiment each in a heavy smoking dog. These same dogs also showed invasive bronchiolo-alveolar tumors. It was stated'that the early squamous cell carcinomas" were indistin- guishable from "early invasive squamous cell carcinomas" found in humans who smoke cigarettes. Evaluation of these findings must await publication of the photomicrographs. At present, one can only comment that Auerbach has been more fortunate than most pathologists in f inding early squamous carcinomas. In a previous N study (29) of humans he found them slightly more frequent in the O proximal parts of the main bronchial tubes, an area in which lung N UT cancer seldom develops. In this experiment, one of the early ~ squamous carcinomas was found in a dog's left main bronchus, also ~ a proximal part of the mainn bronchial tubes. ~ The only significant conclusion from the Auerbach experi- mPnts is that 2 early squamous carcinomas were found. Whether the tumors occurred spontaneously as the dogs grew older o~.~

-8- whether they were producedby the exposure of the dogs to the equivalent of 18 years of human smoking of 60 - 80 cigarettes daily are questions which may be resolved by duplication of these experiments by other investigators. It also must be verified by non-committed pathologists that the lesions found are compatible with the diagnosis of squamous cell. carcinoma. With respect to Hammond's presentation, all it demon- strates is that if dogs are subjected to irritation of the respiratory tract in a way that interferes with their defenses against pulmonary infection, they will develop varying degrees of pneumonitis andsequelae such as compensatory emphysema, fibrosis, and enlargement of the right heart. Apparently the nature of the experiment also interfered with the ability of the dogs to swallow properly inasmuch as 2 of the dogs died from asphyxiation fron aspiration of food. It is impossible to state how many of the other dogs may have aspirated food into the lungs, at various times, resulting in repeateod pulmonary infections. Hammond's colossal ignorance of the subject of patho- logic changes in the lungs is emphasized in his statement "Both in human beings and in dogs, emphysema (rupturing or destruction ~7 O of alveolar septa), fibrosis ..............seld=occur, or occur ~ to only a slight degxee, in the lungs of non-smokers (except non- 0 ~Pb smokers with occupational exposure to certain dusts and vapors)." ~ N Emphysema andfibrosis were foundin the lungs of non-smokers for Clt hundreds of year s( 30) before cigars ettes were manufactured. There is a vast literature on emphysema in the 19Lh century proving that it was a common pulmonary ailment readily detected by clinicians and pathologists.

REFERENCES 1. Friedla.nder, C.: Experimental Untersuchungen ueber chronische Pneumoni') und Lungenschwindsucht, Virchow's Arch. f. path. Anat., 68:325, 1876. 2. W'olf, K.: Der primare Lungenkrebs, Fortschr. d. Med. 13:725, 1895. 3. Kitamura, S.: Ueber second'are Veranderungen der Bronchien undEinige Bemerkungen uber die Frage der Metaplasie, Virchow's Arch. f. path. Ana:t. 190:163, 1907. 4. McKenzie, L.: Epithelmetaplasie bei Bronchopneumonie, Vircho<<r's Arch. f. path. Anat., 190:350, 1907. 5. Menetrier, P.: Cemcer.primitif du Poumon, Prog. M6d. 3:436, 1886. 6. Ribbert, H.: Bemerkungen zu einen Falle von primaren Lungen- carcinom, Deutsch. Med. Wchnschr. 22:165, 18016. 7. Haythorne, S.R.: On the metaplasia of bronchial epithelium, J. Med. Research 24:523, 1912. 8. Tyzzer, E.E.: A series of spontaneous tumors in mice with observations on the influence of heredity on the fre- quency of their occ urrence, J. Med. Research, 21:479, ~ O 1909. N ~ 9. Slye:, M., Holmes, H.F., and~Wells, H.G.: The primary spon- p 4.1h taneous tvmors of the lungs in mice, J. Med. Research M 30:417 1914 N . , 0 10. Winternit--, M.C., Wason, I.M. and; Mc1?amara, F.P.: The Haven, Conn: Yale University Pathology of Influenza, New Press, 1920.

REFF;REI3CES 11. Bell, E.T.: Hyperplasia of the pulmonary alveolar epithelium 12. in disease, Am. J. Path 19:901, 1943. Balo, J., JuhAsz, E., and Temes, J.: Pulmonary infarcts 13. and pulmonary carcinoma, Cancer 9:918, 1956• Berkheiser, S.: Bronchiolar proliferation and metaplasia associated with bronchiectasis, pulmonary infarcts, and anthracosis, Cancer 12:499, 1959. 14. Watson, W'.L.: Lung Cancer, St. Louis: C. V. r7osby, 1968, p:-387. 15. Herbert, P.A.: Alveolar cell:tumor of the lung, Arch. Path. 41:1'75, 1946. 16. 'Overholt, et al.: Favorable bronchiolar carcinoma, Dis. . Chest 27:403, 1955. 17. Eck, H.: Uber den sogennante Alveolarzellk+e'js (Lungen- adenematose), Ztschr. fiir Krebsforsch. 60:433, 1955. 18. Jojzansen, C. and Olsen, S.: Alveolar cell carcinoma, Acta Path. Microbiol. Scand., 41:187, 1957. 19. Rossman, P. and Vortel, V.: Pulmonary metastases imitating alveolar-cell carcinoma, J. Path. Bacteriol. 81:313, 1961. 20. Hewer, IL". F.: The metastatic origin of alveolar-cell tumor of the lung, Ji. Path, Bacteriol. 81:323, 1961. 21. Rosenblatt, M.B., Lisa, J.R. and Collier, F.: Primary and ~ metastatic bronchiolo-alveolar carcinoma, Dis. Chest ~ 52:147, 1967. CA O ~ N