Abstract
BACKGROUND: Several long-acting antiretroviral treatment regimens contain second-generation non-nucleoside reverse transcriptase inhibitors (2ndGenNNRTI). As first-generation NNRTIs (1stGenNNRTI) exhibit some cross-resistance with 2ndGenNNRTI, we sought to evaluate the rate of acquired cross-resistance to 2ndGenNNRTI and its determinants at treatment failure in a typical low- and middle-income country (LMIC) such as Cameroon. PATIENTS AND METHODS: A facility-based cross-sectional study was conducted among patients failing first-/second-line regimens between 2019 and 2023 in Cameroon. HIV-1 Sanger sequencing was performed on plasma and resistance-associated mutations (RAMs) to etravirine, rilpivirine and doravirine were interpreted using HIVdb program v.9.5.0 (HIVdb penalty scores were, ≥60, high resistance; 15-59, intermediate resistance and <15, susceptible) and the IAS-USA 2022 list. RESULTS: Overall, 653 individuals previously exposed to 1stGenNNRTI were enrolled [median (IQR) age 39 (26-46) years and viraemia 59 370 (10 442-244 916) copies/mL]. Importantly, 361 participants were on 1stGenNNRTI-based first-line and 292 on protease inhibitor-based second-line regimen. NNRTIs RAMs were found in up to 90.64% of individuals, with 36.45% having more than three RAMs. Concerning 2ndGenNNRTIs, 77.18% of individuals harboured RAMs conferring high or intermediate-level resistance, with the predicted efficacy of etravirine, doravirine and rilpivirine being 47.17%, 33.23% and 32.31%, respectively. Major 2ndGenNNRTIs RAMs were driven by Y181C (23.74%), K101E (8.57%), Y188L (8.42%) and H221Y (8.42%), while minor RAMs were A98G (18.83%), G190A (18.68%) and P225H (14.70%). A higher prevalence of RAMs was observed in those failing first-line versus second line (81.71% versus 71.57%, respectively, P < 0.001), driven predominantly by the difference in doravirine-RAMs [first line (72.85%) versus second line (59.58%), P < 0.001]. CONCLUSIONS: Among patients failing treatment in Cameroon, there is a high-level of cross-resistance to 2ndGenNNRTI due to wide exposure to 1stGenNNRTI. Thus, in LMICs sharing similar programmatic features, the use of NNRTI-sparing regimens should be prioritized as a public health approach, while second-generation-NNRTI long-acting regimens should be guided by genotyping or for clients without previous exposure to NNRTIs.