Mesothelioma Cases - Chrysotile Asbestos in Drinking-Water
7.1.3 Other malignant diseases
Results of cohort studies of workers almost exclusively exposed to chrysotile asbestos and considered by the Task Group to be most relevant to this evaluation are summarised in Table 23 and described in section 7.1.3.1. Studies that contribute less to our understanding of the effects of chrysotile, due primarily to concomitant exposure to amphiboles or to limitations of design and reporting, are presented in section 7.1.3.2.
7.1.3.1 Critical occupational cohort studies involving chrysotile
There has been considerable unresolved controversy regarding the possible carcinogenic effect of asbestos on the larynx, kidney and gastrointestinal tract. Moreover, there is little evidence that permits an assessment of chrysotile, in particular, as a risk factor for these cancers. In four of the cohorts exposed almost exclusively to chrysotile, data were presented on SMRs for laryngeal cancer (Hughes et al., 1987; Piolatto et al., 1990; McDonald et al., 1993; Dement et al., 1994). Non-significant excesses were observed in some of the studies. It is not possible to draw conclusions about the association with laryngeal cancer because the data are too sparse and because confounding may play an important role in creating associations. Where examined, laryngeal cancer was strongly associated with cigarette smoking (McDonald et al., 1993) and alcohol consumption (Piolatto et al., 1990).
Owing to the rarity of kidney cancer, cohort studies have limited statistical power to detect even moderate increases of kidney cancer. There was no overall excess of kidney cancer in the cohort of miners and millers followed by McDonald et al. (1993), although some increases occurred in subgroups stratified by mine and exposure; however, the number of cases precludes meaningful interpretation. In the study in asbestos-cement production workers, in which the SMR for kidney cancer in plant 1 (chrysotile) was 2.25, based on only four cases, the SMR for lung cancer was 1.17 (Hughes et al., 1987). No other data on kidney cancer risks were presented for the other cohorts of chrysotile workers.
In predominantly "chrysotile"-exposed cohorts, there is no consistent evidence of excess mortality from stomach or colorectal cancer. In the analysis of mortality in the Quebec cohort up to 1989 (McDonald et al., 1993), the SMR for gastric cancer was elevated in the highest exposure category (SMR = 1.39); the corresponding SMR for lung cancer was 1.85. Overall, there was no systematic relationship with exposure.
7.1.3.2 Other relevant studies
Most case-control studies have investigated the association between exposure to unspecified or several forms of "asbestos" and various cancers (see, for example, Bravo et al., 1988; Parnes, 1990; Jakobsson et al., 1994). In the multisite case-control study conducted in Montreal (see section 7.1.2.3d), 177 cases of kidney cancer were included (Siemiatycki, 1991). The OR of any exposure to chrysotile was 1.2 (90% CI=0.9-1.7; 31 exposed cases), and that of substantial exposure was 1.8 (90% CI=0.9-3.7; 6 cases). Corresponding ORs of exposure to amphiboles were 0.7 (8 cases) and 0.8 (1 case).
In this study, a total of 251 stomach, 497 colon and 257 rectal cancer cases were included (Siemiatycki, 1991). The ORs for any and substantial exposure to chrysotile were 1.3 and 0.7 for stomach cancer, 1.0 and 1.6 (90% CI=1.0-2.5) for colon cancer, and 0.7 and 0.5 for rectal cancer. Exposure to amphiboles was not associated with a significant increase in risk of any of these cancers.
7.2 Non-occupational exposure
Data available on incidence or mortality in populations exposed in the vicinity of sources of chrysotile since Environmental Health Criteria 53 was published have not been identified. In studies reviewed at that time, increases in lung cancer were not observed in four limited ecological epidemiological studies of populations in the vicinity of natural or anthropogenic sources of chrysotile (including the chrysotile mines and mills in Quebec) (IPCS, 1986). Data available on incidence or mortality in household contacts of asbestos workers were reviewed in Environmental Health Criteria 53. In several case-control studies reviewed therein, there were more mesothelioma cases with household exposure than in controls, after exclusion of occupation. However for most of these investigations, it is not possible to distinguish the form of asbestos to which household contacts were exposed on the basis of information included in the published reports. Available data on effects of exposure to chrysotile asbestos (specifically) in the general environment are restricted to those in populations exposed to relatively high concentrations of chrysotile asbestos in drinking-water, particularly from serpentine deposits or asbestos-cement pipe. These include ecological studies of populations in Connecticut, Florida, California, Utah and Quebec, and a case-control study in Puget Sound, Washington, USA, reviewed in Environmental Health Criteria 53. On the basis of these studies, it was concluded that there was little convincing evidence of an association between asbestos in public water supplies and cancer induction. More recent identified studies do not contribute additionally to our understanding of health risks associated with exposure to chrysotile in drinking-water.