Objective: Determine whether male circumcision would be effective in reducing HIV transmission among men who have sex with men (MSM).
Design: Retrospective analysis of the VAXGen VAX004 HIV vaccine clinical trial data.
Methods: Survival analysis was used to associate time to HIV infection with multiple predictors. Unprotected insertive and receptive anal sex predictors were highly correlated, thus separate models were run.
Results: Four thousand eight hundred and eighty-nine participants were included in this reanalysis; 86.1% were circumcised. Three hundred and forty-two (7.0%) men became infected during the study; 87.4% were circumcised. Controlling for demographic characteristics and risk behaviors, in the model that included unprotected insertive anal sex, being uncircumcised was not associated with incident HIV infection [adjusted hazards ratio (AHR) = 0.97, confidence interval (CI) = 0.56-1.68]. Furthermore, while having unprotected insertive (AHR = 2.25, CI = 1.72-2.93) or receptive (AHR = 3.45, CI = 2.58-4.61) anal sex with an HIV-positive partner were associated with HIV infection, the associations between HIV incidence and the interaction between being uncircumcised and reporting unprotected insertive (AHR = 1.78, CI = 0.90-3.53) or receptive (AHR = 1.26, CI = 0.62-2.57) anal sex with an HIV-positive partner were not statistically significant. Of the study visits when a participant reported unprotected insertive anal sex with an HIV-positive partner, HIV infection among circumcised men was reported in 3.16% of the visits (80/2532) and among uncircumcised men in 3.93% of the visits (14/356) [relative risk (RR) = 0.80, CI = 0.46-1.39].
Conclusions: Among men who reported unprotected insertive anal sex with HIV-positive partners, being uncircumcised did not confer a statistically significant increase in HIV infection risk. Additional studies with more incident HIV infections or that include a larger proportion of uncircumcised men may provide a more definitive result.