In order to examine whether biomarkers of cytogenetic damage and susceptibility, such as spontaneous and mitomycin C-induced sister chromatid exchange (SCE) can predict cancer development, a nested case-control study was performed in a blackfoot endemic area with known high cancer risk. A cohort of 686 residents was recruited from three villages in the arseniasis area. Personal characteristics were collected and venous blood was drawn for lymphocyte culture and stored in a refrigerator. The vital status and cancer development was followed using the National Death Registry, Cancer Registry, and Blackfoot Disease Registry. The follow up period was from August 1991 to July 1997. During this 6-year-period, 55 residents developed various types of cancer. Blood culture samples from 23 of these subjects were unsuitable for spontaneous SCE experiments and 45 of these subjects were unsuitable for mitomycin C-induced SCE experiments due to improper storage. Finally, a total of 32 cancer cases had cytogenetic samples that could be analyzed. About 32 control subjects were selected from those who did not develop cancer in the study period and these subjects were matched to cases by sex, age, smoking habits, and residential area. The results showed that there was no significant difference in the frequencies of spontaneous and mitomycin C-induced SCE between the case and control groups. There was also no significant difference in the net difference of spontaneous and mitomycin C-induced SCE between the case and control groups. These results suggest that SCEs, either spontaneous or mitomycin C-induced, might not be good markers to predict cancer risk.