Observational studies suggest that male circumcision is protective against HIV and STD acquisition. In Rakai, Uganda, we observed an HIV incidence of 0.9/100 Py in circumcised and 1.8/100 py in uncircumcised men (adjusted RR=0.53). This is biologically plausible: the inner preputial mucosa is rich in HIV target cells and is retracted during intercourse, exposing a large surface area to female genital fluids. The foreskin is also vulnerable to trauma and ulceration, and the moist subpreputial space may prolong survival of HIV or STDs and delay healing of GUD. However, the observed protection may also reflect behavioral confounding since circumcision is performed for religious/cultural reasons or for medical indications. For example, Muslims who are circumcised before puberty may be at reduced risk of HIV due to lower risk behaviors and cultural practices such as ritual post-coital genital washing. Conversely, non-Muslims are more likely to be circumcised for medical indications which may be sequelae of prior STDs associated with higher risk behaviors. Observational data are inadequate to determine policy and randomized trials are needed before circumcision programs can be promoted. We propose to conduct a two stage randomized controlled trial to evaluate adult male circumcision for HIV prevention among HIV-negative men in Rakai, Uganda. In the first stage, we propose a pilot study of male circumcision safety-accessibility in HIV-neg men. Following DSMD approval, we will conduct a two arm individually randomized trial to test the primary hypothesis that: (i) circumcision is safe and acceptable, and (ii) circumcision reduces male HIV incidence compared to uncircumcised controls. Secondary endpoints include the effects of circumcision on male STDs and risk behaviors, and the effects of pre-existing pathology on circumcision safety. Consenting HIV-negative men aged 15-49, resident in 72 communities, will be enrolled, and individually randomized into a Circumcision Intervention Arm aqnd a Control Arm (n=2,500 per arm). Randomization of men who accept circumcision will avoid self-selection bias. Participants will be followed annually for 2 years. The study is powered to detect a RR=0.53 for male HIV incidence, adjusting for cross-over. HIV incidence in the circumcision arm relative to the control arm will be assessed by an intent-to-treat analysis using multivariate Poisson regression. Testing and counseling (VCT), condoms, HIV/STD prevention education, STD treatment and selected medical care will be provided. Control arm men will be offered circumcision as a service in Years 4-5, contingent on DSMB approval.
Showing the most recent 10 out of 44 publications
