Cancer risk among men with, or at risk of, HIV infection in southern Europe

Objective: To evaluate the cancer risk in southern European men with, or at risk of, HIV infection. Design: An analysis of longitudinal data to assess time-dependent rare events. Methods: Data from a cohort of HIV seroconverters, and from two hospital-ba sed HIV seroprevalent cohorts were combined and analysed. The number of can cer cases observed was compared with the expected number, obtained from can cer incidence rates among men in the general population. Age-standardized i ncidence ratios (SIR) and their 95% confidence intervals (CI) were computed . Results: A total of 19 609 person-years of observation were accumulated amo ng HIV-positive men, and 7957 person-years among HIV-negative men. Among HI V-positive men, statistically significant increased SIR were seen for Hodgk in's disease (HD) (SIR = 8.7), liver cancer (SIR = 11.0), and cancer of the salivary glands (SIR = 33.6). An excess of lung cancer was seen among intr avenous drug users (IDU), but not among homosexual men. When the risk of al l non-AIDS-defining cancers was considered, HIV-positive men had a nearly t wofold excess (95% CI: 1.2-2.8). A risk of similar magnitude emerged among HIV-negative IDU (95% CI: 1.0-4.5), largely attributable to lung cancer and HD. Conclusion: These findings confirm that HIV infection increases the risk of HD, whereas they suggest that the risk of hepatocellular carcinoma may als o be enhanced by HIV infection. The observation of an elevated risk of lung cancer in both HIV-positive and HIV-negative IDU points to personal behavi ours unrelated to HIV infection. (C) 2000 Lippincott Williams & Wilkins.