Abstract
Clinical trials demonstrated intermittent preventive treatment in pregnancy with mefloquine (MQ) reduced malaria rates among pregnant women, yet an unexpected higher risk of mother-to-child transmission (MTCT) of HIV among HIV-positive women receiving MQ has also been observed. To determine if interactions between antiretroviral drugs (ARVs) and MQ could contribute to the increased MTCT observed in women receiving MQ, we performed a retrospective cross-sectional analysis of ARV plasma concentrations in peripheral blood (maternal plasma) and cord blood (cord plasma) collected at delivery from 186 mothers participating in a randomized clinical trial of MQ (n = 102) compared with placebo (n = 84) in Kenya. Plasma zidovudine (AZT), lamivudine (3TC), and nevirapine (NVP) concentrations were measured by high-performance liquid chromatography-tandem mass spectrometry. Although only 4% (7/186) reported not using these ARVs, AZT, 3TC, and NVP were all below the limit of detection in 44% of maternal plasma and 42% of cord plasma samples, and proportions were similar between the two study arms. Median concentrations of AZT and 3TC were not significantly lower in the MQ arm compared with the placebo arm for maternal plasma and cord plasma (p > .05). However, median NVP concentrations were significantly lower in the MQ study arm compared with the placebo study arm in both maternal plasma (1,597 ng/mL vs. 2,353 ng/mL, Mann-Whitney Rank Sum, p = .023) and cord plasma (2,038 ng/mL vs. 2,434 ng/mL, p = .048). Reduced NVP concentrations in maternal and cord plasma of women receiving MQ suggest MQ may affect NVP metabolism for both mother and infant. These results highlight the need to evaluate potential drug-drug interactions between candidate antimalarials and ARVs for use in pregnant women.