Abstract
OBJECTIVES: Women are under-represented in clinical trials and must often commit to using contraception to enroll. We sought to determine the incidence and predictors of pregnancy in women participating in HIV treatment trials. DESIGN: Individual participant data meta-analysis. METHODS: We included data from multicountry HIV treatment trials conducted during the period 2005-2019 by the AIDS Clinical Trials Group that included females with HIV who were of reproductive potential, did not intend to become pregnant, and agreed to use effective contraception during study treatment. We extracted data from all female participants of age 18-55 years, including occurrence and dates of pregnancy on-study; however, only a few incident pregnancy predictor variables were available for analysis. RESULTS: One thousand six hundred twenty-six women from 4 trials were included. Over a median of 28 months (6461 person-years) of follow-up, 143 (9%) women became pregnant, for an overall incidence of 2.2 pregnancies/100 person-years (range 0.5-3/100 person-years, by study). In multivariable analysis including baseline age, type of regimen, and country as predictor variables, younger age remained the strongest predictor of incident pregnancy ( P < 0.0001 adjusted for country and antiretroviral treatment regimen). CD4 and HIV-1 RNA were not associated with pregnancy incidence. CONCLUSIONS: Pregnancy incidence was 2.2/100 person-years in female participants of HIV treatment trials. Rather than leading to exclusion of young women from trials, this finding should prompt appropriate adaptations in study design and analysis for earlier generation of pregnancy safety information for drugs.