The authors reviewed the epidemiologic studies on exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and cancer risk, published since the last full-scale review made by the International Agency for Research on Cancer Monographs program in 1997. The update of a cohort of US herbicide producers generated negative results overall; the internal analysis provided evidence of an increased "all-cancer" risk in the highest exposure category, with a statistically significant exposure-response association in some of the many analyses performed. The update of a similar Dutch cohort did not confirm the previously observed association with TCDD exposure. The updated surveillance of the Seveso population provided evidence of increased all-cancer mortality 15-20 years after exposure among those living in the most contaminated area but might also reflect random variation, as overall excesses in the most recent follow-up were not observed. Corresponding data on cancer incidence offer little support to the mortality results. Updated results from cohort studies of Vietnam veterans potentially exposed to TCDD did not consistently suggest an increased risk of cancer. Results of additional, smaller studies of other occupational groups potentially exposed to TCDD, and of community-based case-control studies, did not provide consistent evidence of an increased cancer risk. In conclusion, recent epidemiological evidence falls far short of conclusively demonstrating a causal link between TCDD exposure and cancer risk in humans. The emphasis on results for overall cancer risk-rather than risk for specific neoplasms-is not justified on epidemiologic grounds and is not a reason for ignoring the weaknesses of the available evidence.