Document 44GJRDO0wDKpejZDLOo3XK70a

Thk Mocnt Sinai Joi'knm.ok Mkdicink Vwl. W, No. 6. June 15'S(> Printed in L'.S A. Neoplastic Asbestos-Induced Disease W. T. Elliott McCaughey, M.D. Recognition of the carcinogenic effects of as bestos has evolved slowly over the past 50 years. Lung cancer first came under suspicion in the 1930s and 1940s, and by the 1950s it was well established in Europe as a frequent complication of asbestosis (1, 2). Strong evidence of an associa tion between exposure to asbestos and mesothe lioma appeared in 1960 (3), and subsequently the link has been confirmed by many studies. Fi nally, from 1964 onward evidence has been growing that asbestos may cause cancer in a number of other sites (Table). Cumulatively, these cancers of other sites add significantly to the mortality of asbestos workers (4). In 1964 and 1965 several important papers concerning asbestos-related cancer came from the Mount Sinai group which included Drs. J. Churg, Selikoff, and Hammond (5-7). These reports de scribed a striking excess of lung cancer and meso thelioma among American insulation workers and a probable excess of gastrointestinal cancer. It thus became obvious that asbestos-induced cancer was an important occupational disease in the United States and was not, as had sometimes been inferred, a disease mainly afflicting Euro peans and South Africans. Subsequently, it was shown that many other categories of American asbestos workers were at risk and that paraoccupational neighborhood and domestic exposures to asbestos might also lead to cancer. Moreover, as awareness increased of the widespread contami nation of the general environment by asbestos, it became apparent that there was at least a poten tial risk of asbestos-induced cancer for a large group of people in the general population. From the Canadian Tumour Reference Centre, Univer sity of Ottawa, Ottawa, Ontario, Canada. Address reprint re quests to the author there. Presented at a scientific meeting in honor of Drs. Jacob Churg and Lotte Strauss, "Selected Topics in Pulmonary, Vascular, and Pediatric Diseases," at the New York Academy of Medicine, June 14,1985. Dr. Churg's special contribution to this un folding story has come through the leadership and direction he has given to the evolution of ob jective criteria for the diagnosis of asbestos-re lated disease, especially mesothelioma. For many years he and his close friend and colleague, the late Dr. Milton Kannerstein, ran the U.S. Meso thelioma Panel and, in a number of important studies during that period, made us all more fa miliar with the characteristics of mesothelial tumors. The result has been that diagnosis jjf these controversial neoplasms is now much more objective, and related epidemiologic and clinical observations have become more secure. Lung Cancer Though it has been estimated that asbestos may cause some 5% of present-day lung cancers (8), the extent of the contribution of asbestos to the development of lung cancer in an individual patient may be difficult to quantify precisely, especially when the patient has been a cigarette smoker or the evidence of excessive exposure to asbestos is not clear cut. Initially asbestos was thought to cause lung cancer by multiplying the effects of tobacco smoke (9), but subsequently it was observed that asbestos caused lung cancer in nonsmokers (10-12). There is recent evidence that the relative risk of lung cancer in non smoking asbestos workers may be as high as or higher than in smoking asbestos workers ill. 12). All commercial forms of asbestos have caused lung cancer, but some reports suggest that chrysotile, which accounts for more than 90% of the asbestos used in recent decades, is less potent than the two other widely used forms (amosite and crocidolite). However, in one setting (a chrysotile asbestos textile plant) the standard mortality ratio for lung cancer at different 'levels of exposure has been among the highest reported 416 Viti. No. NKOPLASTIC ASHKSTOS-INDl'CKP l)ISIiASK -- MCCAl'GHKY 117 TABLE Xcop/astic Effects of Asbestos Exposure in Humans the lung and the occupational history. Because discrepancies in the two forms of evidence are Definite Lung sometimes marked, both must be taken into ac count. carcinoma Pleura and peritoneum diffuse mesothelioma Gastrointestinal tract carcinoma Probable Ovary carcinoma Larynx Malignant Mesothelioma Association with Asbestos. Recently, the long controversy over the status and the diagnosis of mesothelioma has been augmented because of certain aspects of its association with asbestos. Though this tumor has undoubtedly been in carcinoma Pericardium and tunica vaginalis diffuse mesothelioma Probable'possible Reticuloendothelial system malignant lymphoma creasing in frequency, it remains rare. Sixteen hundred cases are estimated to have occurred in the United States in 1980 (18). However, the tumor is probably still underdiagnosed in those not uncommon situations in which the behavior Breast, kidney, and pancreas carcinoma or histopathology is not typical or adequate mate rial is not available. When epidemiologic and lung-asbestos- burden studies are taken into account, it appears that asbestos has caused some 60% to 80% of for lung cancer in asbestos workers (13, 14). This pleural and peritoneal mesotheliomas. However, may be due to the fact that textile processing pro the reported extent of the association varies duces a greater proportion of the form of asbestos markedly in epidemiologic studies. Thus, in some fiber (long and thin) that is particularly carcino- reports over 80% of patients give an occupational _ genic. Experimental studies have also shown that exposure history, whereas in other reports the fine chrysotile fiber is just as potent as other figure is less than 30% (see ref. 19 for references). forms of asbestos in causing lung cancer (15). One can only speculate on the extent to which The nature of the dose-response relationship these variations reflect varying occupational and between asbestos and lung cancer is particularly environmental conditions in different parts of the important in assessing the effects of low-level ex world and the enthusiasm and knowledge of the posures. If, as occupational studies suggest, the history-taker. Only a. small percentage of women relationship is linear or almost so (13, 14), very with mesothelioma have been occupationally ex light exposures may proportionately increase the posed (20, 21). risk. No threshold level of asbestos exposure has Quite possibly some people develop mesothe been recognized below which no lung cancer risk lioma as a result of very low level nonoccupa- exists. It is also apparent from dose-response pro tional exposure to asbestos. Supporting this pos jections that exposures much less than those re sibility* are extrapolations based on the dose-re quired to produce asbestosis may cause lung sponse relationship between asbestos and cancer. Asbestos appears to act as a classic pro mesothelioma in occupational studies (22) and moter in the development of lung cancer, though the fact that in both humans and experimental the pathogenetic mechanisms of asbestos-induced animals, mesotheliomas have been associated lung cancer are unclear (16). Because asbestos with brief exposures (15, 23). It is also known causes lung cancer in nonsmokers and experi that nonasbestos forms of mineral fiber with sim mental animals, it appears that the fiber may ilar physical characteristics (aspect ratio) to also have an initiating influence. those of asbestos fibers may induce mesothelioma The frequency of the main cell types in as in experimental animals and humans. In bestos-induced lung carcinoma appears to be humans, erionite, a fibrous form of zeolite not re about the same as in the general population. lated to asbestos, has caused mesotheliomas in There is evidence, however, that asbestos lung Turkey (24). cancers are more likely to be located in the lower Much controversy in recent years centers on ( lobes than lung cancers in the general population whether chrysotile asbestos induces mesothe (17). The pathologist's assessment of whether a lioma. Most observers feel that it is less potent given lung cancer is related to asbestos exposure than amosite and crocidolite in this regard. Par must obviously depend on the asbestos content of ticularly notable is the fact that peritoneal meso 418 T1IK MOIW'T SLNAI -JOCKNAL OF MFDK'LNT: .linn- l`<N> theliomas have rarely been observed in persons exposed to chrysotile. Experimentally, some forms of chrysotile have at least the same poten tial to induce pleural mesothelioma as other forms of asbestos. The role of chrysotile in initi ating mesothelioma will have to be further evalu ated in light of evidence that the tremolite form of asbestos, which is associated with chrysotile ore, could be a factor in the causation of pleural mesothelioma in chrysotile miners (25). Other Factors. It is likely that factors or agents other than mineral fiber may be involved in the causation of some mesotheliomas (see ref. 19 for references). Thus, it has been suggested that ra diation and chronic inflammation or scarring may have caused occasional mesotheliomas in humans. Genetic predisposition to cancer, as well as shared exposure to asbestos, may operate in the relatively frequent instances of familial oc currence of the neoplasm (26, 27). The list of agents that have caused mesothelioma in experi mental animals is a long one (19) and further en courages a critical attitude on the etiology of me sothelioma. Diagnosis. The pathologist can often diagnose mesothelioma in the early stages given experi ence, good material, and appropriate laboratory facilities. In several situations, however, consid erable potential for misdiagnosis exists. Cytologists are familiar with the fact that exuberant mesothelial hyperplasia can be readily confused with mesothelioma. Less well appre ciated is the extent to which mesothelial hyper plasia may mimic tumor grossly and in biopsies (28). Focal mesothelial hyperplasias on occasion may be so pronounced as to form grossly visible nodules. Mild, and even moderate, cellular atypia may also be present, and--rarely--hyperplastic mesothelium may assume a papillary disposition. Occasionally, hyperplastic mesothelium may in vade immediately underlying normal tissue, but necrosis and marked cytologic atypia do not occur in mesothelial proliferations proved to be benign on follow-up (28). Dr. Churg has pointed out a small group of peritoneal mesotheliomas which run a very slow course and which often begin with what looks like hyperplasia (29). We have encountered occa sional patients in whom marked mesothelial hy perplasia has been associated with chronic effu sions and after several years of recurrent effu sion, the cause of the mesothelial hyperplasia was not clear (28). In these situations the possi bility that the hyperplasia is precancerous must be kept in mind. Another difficult diagnostic problem in small serosal biopsies concerns differentiation of scattissue from desmoplastic diffuse mesothelioma. This entity was delineated by Kannerstein and Churg some years ago (30). The tumor may con tain extensive areas indistinguishable from banal scar tissue which may easily mislead the pathologist in small biopsies (30, 31). Further complicating the interpretation of fibrous serosal lesions is the fact that pleural fibrosis may some times be quite cellular and thus resemble fibrous tumor. The degree of nuclear atypia is usually the most critical factor in determining whether the tissue is benign or malignant. As in the case of mesothelial hyperplasia, necrosis favors malig nancy (31). A well-developed tissue pattern--storiform or gyriform--also points to mesothelioma (28). The most common diagnostic problem for the pathologist in biopsy material is differentiation of mesothelioma from metastatic carcinoma. When the tumor has a glandular or tubulopapillary pattern, the differential diagnosis is with ad enocarcinoma. Positive staining for mucin with mucicarmine or periodic acid Schiff after diastaseis usually considered to be decisive evidence of adenocarcinoma,' although I have occasionally seen otherwise typical mesotheliomas with small numbers of tumor cells containing diastase-resis tant PAS-positive granules or globules in their cytoplasm (32). Because a small proportion of ad enocarcinomas arising in the lung and elsewhere do not secrete mucin, a negative reaction does not help significantly in distinguishing between me sothelioma and adenocarcinoma. Immunostaining for carcinoembryonic antigen is often helpful. Many adenocarcinomas immunostain strongly, but mesotheliomas do not, in our expe rience. However, mesotheliomas may occasion ally react weakly. Though it has been claimed that positive immunostaining for human milk fat globule antigen distinguishes adenocarcinoma from mesothelioma (33), others have reported no discriminatory value (34). Electron microscopy has been increasingly useful to us in distinguishing between mesothe lioma and poorly differentiated carcinoma. In this situation the ultrastructural features (long slender microvilli, cytoplasmic cystemae, distri bution of intermediate filaments) that help dis tinguish classical tubulopapillary mesothelioma and well-differentiated adenocarcinoma (35, 36) are often not helpful. In a recent study of poorly differentiated diffuse epithelial mesotheliomas, we have found that although the classical char acteristics of well-differentiated neoplastic me sothelial cells are markedly modified, cytologic \ (>1. No. (i NEOPLASTIC ASBESTOS-INDUCED DISEASE -- MCCAUGHEY n y details remain which are useful in distinguishing mesothelioma from carcinoma. These include the presence of cytoplasmic processes lying parallel to the adjacent plasma membranes, abundant cy toplasm with a polar disposition of organelles, and disaggregation of nuclear chromatin (37). Other Cancers It seems certain that asbestos causes other cancers (Tabled, though the strength of the evi dence supporting a relationship with individual types of cancer varies. The strongest evidence of an association is with gastrointestinal carci nomas (38) and ovarian cancer (39, 40). In view of the relationship with ovarian cancer, it is of in terest that the common ovarian epithelial tumors are believed to be of mesothelial origin and that tumors histologically identical to ovarian serous carcinomas may arise from the peritoneum of fe males (41, 42). Because asbestos is rapidly trans mitted through the wall of the intestine into the lymphatic circulation, there is obviously the po tential to initiate cancer in virtually any part of the body. References 1. Annual Report of the Chief Inspector of Factories for the Year 1947. London: H.M. Stationary Office, 1949:79 81. 2. Doll R. Mortality from lung cancer in asbestos workers. Br J Indust Med 1955; 12:81-85. 3. Wagner JC, Sleggs CA, Marchand P. Diffuse pleural me sothelioma and asbestos exposure in the North Western Cape province. Br J Indust Med 1960; 17:260 271. 4. SeiikofTIJ, Hammond EC, Seidman H. Mortality experi ence of insulation workers in the United States and Canada. Ann NY Acad Sci 1979; 330:91-116. 5. Selikoff IJ, Churg J, Hammond EC. Asbestos exposure and neoplasia. JAMA 1964; 188:22-26. 6. Hammond EC, Selikoff U, Churg J. Neoplasia among in sulation workers in the United States with special ref erence to intra-abdominal neoplasia. Ann NY Acad Sci 1965; 132:519-525. 7. Selikoff IJ, Churg J, Hammond EC. Relation between ex posure to asbestos and mesothelioma. N Engl J Med 1965; 272:560-565. 8. Doll R, Peto R. The causes of cancer. New York: Oxford University Press, 1981. 9. Selikoff U. Hammond EC, Churg J. Asbestos exposure, smoking, and neoplasia. JAMA 1968; 204:106-112. 10. Hammond EC, Selikoff U, Seidman H. Asbestos expo sure, cigarette smoking and death rates. Ann NY Acad Sci 1979; 330:473-490. 11. Enterline PE. Attributality in the face of uncertainty. Chest 1980; 78:377-379. 12. Berry G, Newhouse ML, Antonis P. Combined effect of asbestos and smoking on mortality from lung cancer and mesothelioma in factory workers. Br J Indust Med 1985; 42:12-18. 13. Dement JM. Harris *RL, Symons,MJ. Shy CM. Exposure and mortality among chrysotile asbestos workers. Part II: Mortality. Am J Indusl Med 19S3; 4:421-433. 14. McDonald AD. Fry JS, Wooley AJ, McDonald J. Dust ex posure and mortality in an American chrysotile textile plant. Br J Indust Med 1983;40:361-367. 15. Wagner JC, Berry* G, Skidmore J\V, Timbrell V. The ef fects of inhalation of asbestos in rats. Br J Cancer 1974; 29:252-269. 16. Craighead JE, Mossman BT. The pathogenesis of as bestos-associated diseases. N Engl J Med 1982; 306:1446-1455. 17. Kannerstein M, Churg J. Pathology of carcinoma of the lung associated with asbestos exposure. Cancer 1972; 30:14-21. . 18. Asbestiform fibers. Nonoccupational health risks. Wash ington DC: National Academy Press, 1984. 19. Peterson JT, Greenberg SD, Buffler PA. Non-asbestos-re lated malignant mesothelioma: a review. Cancer 1984; 54:951-960. 20. McDonald AD, Harper A, El Attar OA, McDonald JC. Ep idemiology of primary malignant mesotneliai tumors in Canada. Cancer 1970; 26:914-919. . 21. McDonald AD, McDonald JC. Malignant mesothelioma in North America. Cancer 1980: 46:1650-1656. 22. Report of the Chronic Hazard Advisory Panel on As bestos. Washington DC: U.S. Consumer Report Safety Commission, 1983. 23. Seidman H. Selikoff IJ, Hammond EC. Short term as bestos work exposure and long term observation. Ann NY Acad Sci 1979; 330:61-89. 24. Baris YI. Artvinli M. Sahin AA. Environmental mesothe lioma in Turkey. Ann NY Acad Sci 1979; 330:423-432. 25. Churg A, Wiggs B, Depaoli L, Kampe B. Stevens B. Lung asbestos content in chrysotile workers with mesothe lioma. Am Rev Respir Dis 1984; 130:1042-1045. 26. Browne K. Asbestos-reiated mesothelioma: epidemiolo gical evidence for asbestos as a promoter. Arch Environ Hllh 19S3; 38:261-266. 27. Vianna N'J, Polan AK. Non-occupationai exposure to as bestos and malignant mesothelioma in females. Lancet 1978; 1:1061-1063. 28. McCaughey WTE, Al-Jabi M. Differentiation of mesothe- lial hyperplasia and neoplasia in biopsies. Pathol Ann (in press). 29. Churg J. Peritoneal mesothelioma. Environ Hhh Per- spect 1974;9:317-318. 30. Kannerstein M, Churg J. Desmoplastic diffuse malignant mesothelioma. In: Fenoglio CM. Wolff M, eds. Progress in surgical pathology, Vol. 11. New York: Masson Pub lishing USA Inc, 1980:19-29. 31. Cantin R. Al-Jabi M. McCaughey WTE. Desmoplastic dif fuse mesothelioma. Am J Surg Pathol 1982; 6:215 222. 32. McCaughey WTE. Kannerstein M. Churg J. Tumors and pseudotumors of the serous membranes. Atlas of Tumor Pathology. Fascicle 20, Second Series. Wash ington DC: Armed Forces Institute of Pathology. 1955. 33. Battifora H, Kopinski MI. Distinction of mesothelioma from adenocarcinoma. An immunohistochemical ap proach. Cancer 1985; 55:1679-1685. 34. Marshall RJ, Herbert A, Braye SF, Jones DB. Use of anti bodies to carcinoembryonic antigen and human milk fat globule to distinguish carcinoma, mesothelioma and reactive mesothelium- J Clin Pathol 1984: 37:1215 1221. 420 Till'; .\k>i\\tsi.\\\i .hh'kx.a]. oF.\n:run\K June )!'><> 35. Suzuki Y, Churg J. Kannerstein M. L'ltrastruclure of human malignant diffuse mesothelioma. Am J Pathol 1976: 85:241-262. 36. Warhol MJ. Corson JM. An ultraslructural comparison of mesotheliomas and adenocarcinomas of the lung and breast. Hum Pathol 1985: 16:50-55. 37. Dardick I, Al-Jabi M. McCaughey WTE, et al. b'ltrastructure of poorly differentiated diffuse epithelial mesothe lioma. Ultrastruct Pathol 19S4; 7:151-160. 38. Miller AB. Asbestos fiber dust and gastrointestinal ma lignancies: review of literature with regard to a cause/ effect relationship.. J Chron Dis 1978; 31:23-33. 39. Wipnall BK. Fox AJ. Mortality of female gas mask as semblers. Br J Indust Med 19S2; 39:34-3S. 40. Cramer DW, Welch WR, Scully RE, Wojciechowski CA. Ovarian cancer and talc. Cancer 1982; 50:372-376. 41. Kannerslcin M, Churp J, McCaughey WTE, Hill DP. Pap illary tumors of the peritoneum in women: mesothe lioma or papillary carcinoma. Am J Obstet Gynecol 1977; 127:306-314. 42. Foyle A, Al-Jabi M, McCaughey WTE. Papillary perito neal tumors in women. Am J Surg Pathol 1981: 5:241 249.