Document mBn13wx1Mmkj9vjG53XmXp0n0

216 MARGARET R. BECKLAKE human carcinogen, or a cocarcinogen, and/or potentiator of cigarette smoke and/or other factors. The Influence of Fiber Type and the Nature of Exposure on Biological Response After the 1964 New York Conference (5), the UJCC working group on cancer urged study of "the relationship of dust dosage (including concentration and duration of exposure) and the composition and physical state of the dust to the incidence of asbestosis, carcinoma of the lung, mesothelioma and other cancers." In other words, 2 areas were identified for urgent future research: (7) to establish whether dose-response relationships exist between exposure and bio logic response, and (2) to establish whether the composition and physical state of the dust affects these dose-response relationships. In the ensuing 11 years, health scientists throughout the world have gathered data in sup port of the first hypothesis, namely, that a doseresponse relationship exists for all of the re sponses listed (asbestosis, carcinoma of the lung, other cancers, and, probably, mesothelioma). Furthermore, this holds for all types of fiber and for all types of exposure investigated. The evi dence has been already summarized in table 4. It has proved more difficult to investigate the second hypothesis, that composition and physi cal state of the dust influence the responses, be cause this requires the comparison of dose-re sponse curves for different fibers, or for the same fiber under different exposure conditions, e.g., mining and milling compared to manufac turing. Even animal studies in which exposure can be relatively well controlled are inconclu sive about the relative fibrogenicity of different fiber types (62, 224) although differences in car cinogenic potential have been shown (224). Epidemiologic studies in man in different oc cupationally exposed groups suggest differences between fibers, and between exposures, in terms of their carcinogenic potential both for lung cancer (table 6) and for mesothelioma (224): however, it is usually impossible to es tablish to what extent these can be explained by differences in exposure levels and associated fac tors, such as cigarettes and other co-cardno gens. Furthermore, exposures to one fiber type only are rare (usually in mining), and most production workers have mixed exposures (224). For instance, table 8 summarizes data collected in Quebec asbestos workers, exposed only to ehrysotile fibers. Even in these results, which show a dose effect for all responses, there ate between-area differences, particularly for ra diologic changes, for men of the same stock and working the same geologic deposit, and with ex posure calculated using the same type of index. Civcn these between-area differences, to what ex tent. if any, arc these dose-response curses ap plicable to other working populations? Criteria for comparability require that both the dose and the response be measured in a similar fashion and that the populations com pared have equivalent susceptibility. Enough has already been said to indicate that even if the first 2 criteria were met, virtually nothing is known about the factors accounting for suscep tibility, much less whether it is possible to mea sure them in practice. The only comparison at tempted was incondusive, because mortality was assessed by fundamentally different techniques (figure 5). Thus, it is not possible to deduce from this comparison whether, indeed, there is greater risk for production workers compared to miners (figure 5, upper panel) attributable to differences in the nature of the exposure, an interpretation for which there is little scientific justification, or whether the risk is comparable for both groups, but the scaling (figure 5, lower panel) requires appropriate adjustment. Despite the difficulties in making betweenstudy comparisons, a consensus lias emerged, out lined in a carefully reasoned paper by Kleinfetd (224). and in the conclusion of the Advisory Committee on Asbestos Cancers (7. p. 341). It is believed that there are gradients in the meso thelioma-producing potential, related to fiber type (greatest with crocidolite, less with amosite and with ehrysotile, least with amhophyllite) and to occupation (e.g., for amosite, greatest in insulation workers compared to mi ners). Gradients in fibrogenic capability of the different fibers, less clear, may also be present, with crocidolite leading ehrysotile, whereas gra dients in lung cancer risk may be more closely related to the nature of the exposure, with production lending mining, at least for chrysotile. To date, the best explanation of these gradi ents in biologic potential is that developed by Timbrell (7, p. 295), namely, that biologic ac tivity relates to the degree of penetration and deposition in the Jung (see also figure 6). Thus, the greater mesothelioma potential of crocidoliie compared to amosite, authophyllite, and ehrysotile, could be due to its smaller fiber size and its other aerodynamic properties, which per mit greater penetration and deposition. Similar-