Although these analyses provide valuable information on the comparability of risks and of modifying factors in different cohorts, the results for the A-bomb survivor cohort itself generally confirm the findings reported earlier in the chapter, and they are not discussed further here. Biologically based models have also been applied to the A-bomb survivor data (Kai and others 1997; Pierce and Mendelsohn 1999).
Because the exposure of A-bomb survivors was whole-body exposure, studies of the LSS cohort afford the opportunity to compare cancer risks by site. Inferences for site-specific cancers are based on smaller numbers than those for all solid cancers and involve smaller ERRs than leukemia. This often means that there is considerable uncertainty in quantifying risk, in evaluating modifying factors, and even in determining whether or not there is a dose-response relationship. Although it is likely that radiosensitivity varies across sites, it is often not possible to separate true differences from chance fluctuations. Cancers at some sites may fail to exhibit associations because of small numbers of cases and diagnostic misclassification, which is more problematic for mortality data than for incidence data.
Preston and colleagues (2003) used common models for expressing risks for cancers at different sites. Specifically, 15 sites were analyzed with parameters expressing the modifying effects of age at exposure and attained age set equal to those for all solid cancers. Results of these analyses are summarized in Figure 6-3, which shows the ERR/Sv for exposure at age 30 and attained age 70. Except for sex-specific cancers, estimates are averaged for the two sexes. Preston and colleagues (2003) note that the variability in this plot is generally consistent with what would be expected if the true site-specific ERRs were all equal to that for all solid cancers. More detailed analyses of the five most common types of solid cancer (stomach, colon, liver, lung, and female breast) were conducted. With ERR models, the age-time patterns were similar for these sites, although the decrease in risk with attained age was more rapid for colon cancer. With EAR models, statistically significant departures from the solid cancer temporal model were found for lung cancer, which increased more rapidly with attained age than other solid cancers, and breast cancer, which decreased more rapidly with age at exposure than other solid cancers.
Data from the Hiroshima and Nagasaki tumor registries are preferable to mortality data for evaluating site-specific risks. These data have the major advantages of including