Modulation of environmental exposures by host genetic factors may explain interindividual variation in susceptibility to carcinogenesis. One determinant of susceptibility is mutagen sensitivity measured by the frequency of bleomycin-induced breaks in an in vitro lymphocyte assay. Mutagen sensitivity is a significant predictor of aerodigestive tract cancer risk. In this case-control study of lung-cancer susceptibility markers, 54% of 132 lung-cancer cases had mutagen-sensitivity scores greater than or equal to 1 break/cell, compared with only 22% of 232 controls. The mean breaks/cell value (+/-SE) for the 88 African-American cases was 1.11 (+/-0.60), compared with 0.82 (+/-0.49) for the 121 controls (P < 0.001). For the 44 Mexican-American cases and 111 controls, the comparable values were 1.11 (+/-0.52) and 0.76 (+/-0.38), respectively. The overall odds ratio (OR) for mutagen sensitivity (dichotomized at > or = 1 break/cell), after adjusting for ethnicity and smoking status, was 3.62 (95% confidence limits [CL] = 2.2, 5.9). For current smokers the adjusted risk associated with mutagen sensitivity was 2.52 (1.2, 5.3). For former smokers, the comparable OR (95% CL) was 6.19 (2.7, 14.1). The risk estimate for those under 61 years of age was 4.85 (2.3, 10.4), compared with 2.85 (1.5, 5.6) for older subjects. The risk also appeared to be higher for lighter smokers (< 20 cigarettes daily) than heavier smokers (ORs = 5.72 and 3.20, respectively). The ethnicity-adjusted ORs by quartile of breaks/cell were 1.0, 1.40, 2.46, and 4.80; the trend test was significant at P < 0.001. The joint effects of mutagen sensitivity and former smoking, current smoking, or heavy smoking were greater than additive, although the interaction terms were not statistically significant in the logistic model. Mutagen sensitivity may therefore be a useful member of a panel of susceptibility markers for defining high-risk subgroups for chemoprevention trials.