Longer-term effectiveness of protease-inhibitor-based second line antiretroviral therapy in four large sub-Saharan African clinics

Summary Objectives Data on the longer-term effectiveness of second line combination antiretroviral therapy (ART) in sub-Saharan Africa (SSA) are lacking. We sought to assess the probability and determinants of 2nd line ART failure in SSA. Methods A retrospective, multi-center study of 2nd line ART initiated between 2005 and 2017 at four ART centers in Ethiopia, Ghana and Uganda. Main outcome measure was virologic failure (VF) defined as VL>1000 copies/ml after >6 months on 2nd line therapy. Predictors of VF and virologic re-suppression on 2nd line were evaluated using Cox Proportional Hazards and multivariable logistic regression models, respectively. Results 2191 subjects started 2nd line therapy, 61.5% females. Switching from 1st line (56.4% NVP-based, 70.3% including thymidine-analogues) to 2nd line therapy occurred after mean of 4.1 years. 98.9% of patients started boosted PI with NRTI backbone (TDF+3TC/FTC 67.3%, AZT+3TC 18.5%, others 14.2%). There were 267 (12.0%) VF with a 5-year estimated probability of 15.0% (95% CI 13.2–16.9). Key determinants of VF were concomitant rifampicin use (aHR 2.50 [95% CI 1.54–4.05]) and clinical/immunological failure versus virologic failure as reason for switching therapy (aHR, 0.53 [0.33–0.86]). 138 of 267 (51.7%) subsequently achieved virologic re-suppression and predictors included HIV RNA levels at 2nd-line failure: +1 log higher aOR 0.59 [0.43–0.80], experiencing change within 2nd line ART before VF: aOR 0.17 [0.05–0.56], and more recent calendar year of 2nd line initiation: aOR 0.85 [0.75–0.94]. Conclusions The effectiveness of current 2nd line ART regimens in SSA is good but challenged by interactions with TB therapy.