Appropriateness of hydroxyethylcellulose gel as a placebo control in vaginal microbicide trials: a comparison of the two control arms of HPTN 035

Abstract

Objective: To compare the 2 control arms of HPTN 035 [a hydroxyethylcellulose (HEC) gel control arm and a no-gel control arm] to assess the behavioral effects associated with gel use and direct causal effects of the HEC gel on sexually transmitted infections (STIs), pregnancy, and genital safety.

Design: Randomized trial with 1 blinded (HEC gel) and 1 open-label (no-gel) control arms.

Methods: HIV-uninfected, sexually active women were randomized into the HEC gel arm (n = 771) and into the no-gel arm (n = 772) in 5 countries. Participants in the HEC gel arm were instructed to insert the study gel intravaginally <1 hour before each vaginal sex act. Data on sexual behavior, adherence, safety, pregnancy, and STIs were collected quarterly for 12-30 months of follow-up.

Results: During follow-up, mean reported condom use in the past week was significantly higher in the no-gel arm (81% versus 70%, P < 0.001). There were no significant differences, after adjusting for this differential condom use, between the 2 arms in the rates of genital safety events, pregnancy outcomes, or STIs, including HIV-1.

Conclusions: In this large randomized trial, we found no significant differences between the no-gel and HEC gel arms in the rates of genital safety events, pregnancy outcomes, or STIs. These results aid interpretation of the results of previous vaginal microbicide trials that used the HEC gel as a control. The HEC gel is suitable as a control for ongoing and future vaginal microbicide studies.

Trial registration: ClinicalTrials.gov NCT00074425.

Figures

Figure 1. Reported Percent Condom Use by Arm Over Time Since Enrollment

Dark solid line is the mean for the placebo group, light solid lines are 95% upper and lower confidence interval limits for the placebo group. Dark dashed line is the mean for the no gel arm, light dashed lines are 95% upper and lower confidence interval limits for the no gel arm.