In spite of the extensive use of cytogenetic analysis of human peripheral blood lymphocytes in the biomonitoring of exposure to various mutagens and carcinogens, the long-term effects of an increased frequency of chromosomal aberrations in individuals are still uncertain. Few epidemiologic studies have addressed this issue, and a moderate risk of cancer in individuals with an elevated frequency of chromosomal aberrations has been observed. In the present study, we analyzed data on 1323 cytogenetic assays and 225 subjects examined because of occupational exposures to radon (range of exposure from 1.7 to 662.3 working level month (WLM)). Seventy-five subjects were non-smokers. We found 36 cases of cancer in this cohort. Chromatid breaks were the most frequently observed type of aberrations (mean frequency 1.2 per 100 cells), which statistically significantly correlated with radon exposure (Spearman's correlation coefficient R=0.22, P<0.001). Also, the frequency of aberrant cells (median of 2.5%) correlated with radon exposure (Spearman's correlation coefficient R=0.16, P<0.02). Smoking and silicosis were not associated with results of cytogenetic analyses. The Cox regression models, which accounted for the age at time of first cytogenetic assay, radon exposure, and smoking showed strong and statistically significant associations between cancer incidence and frequency of chromatid breaks and frequency of aberrant cells, respectively. A 1% increase in the frequency of aberrant cells was paralleled by a 62% increase in risk of cancer (P<0.000). An increase in frequency of chromatid breaks by 1 per 100 cells was followed by a 99% increase in risk of cancer (P<0.000). We obtained similar results when we analyzed the incidence of lung cancer and the incidence other than lung cancer separately. Contrary to frequency of chromatid breaks and frequency of aberrant cells, the frequency of chromatid exchanges, and chromosome-type aberrations were not predictive of cancer.