Hair Zidovudine Concentrations Predict Virologic Outcomes Among People Living with HIV/AIDS in China

BACKGROUND: Hair antiretroviral concentrations are an objective and non-invasive measure of adherence to long-term antiretroviral therapy (ART) and can further predict virologic outcomes among people living with HIV/AIDS (PLWH). Zidovudine, one of the mainstream antiretrovirals in China, has been verified to have high reliability in adherence assessment, especially for its hair concentrations. However, data are limited in its predicting virologic outcomes. Therefore, this study aimed to characterize whether hair zidovudine concentrations can predict virologic suppression among Chinese PLWH compared with hair lamivudine concentrations and two self-reported measures, the overall frequency of adherence behaviors and percentage adherence. METHODS: This cross-sectional study randomly recruited 564 PLWH currently treated with zidovudine, lamivudine, and other ART agents (efavirenz, nevirapine, or lopinavir/ritonavir) in Guangxi, China. Hair antiretroviral concentrations were determined using the LC-ESI(+)-MS/MS method. Receiver operating characteristic (ROC) curves were used to estimate the optimal classification thresholds of hair concentrations of zidovudine and lamivudine, and the two self-reported measures. Based on those optimal classification thresholds, logistic regression was used to examine whether those four adherence measures can predict virologic suppression (HIV-1 RNA <200 copies/mL). RESULTS: ROC curves demonstrated good classification performance for association with virologic suppression of zidovudine with the optimal threshold at 58 pg/mg and lamivudine at 255 pg/mg but no self-reported measures. PLWH with hair zidovudine concentrations >58 pg/mg had an adjusted odds ratio (aOR) of 43.191 (95% confidence interval (CI) = 10.171‒183.418, p < 0.001) for virologic suppression. Hair lamivudine concentrations were also associated with virologic suppression (aOR = 10.656, 95% CI = 3.670‒30.943, p < 0.001). However, two self-reported measures did not predict virologic suppression (aORs = 1.157 and 2.488, ps >0.149). CONCLUSION: Hair zidovudine concentrations can be served as an alternative tool for clinically predicting virologic suppression among PLWH in China.