Effectiveness of the dolutegravir transition in Uganda: Disco cohort week-24 results

Background: The fixed-dose combination of tenofovir (TDF), lamivudine (3TC), and dolutegravir (TLD) is now preferred first-line antiretroviral therapy (ART) for most adults with HIV in Sub-Saharan Africa. Yet, concerns remain about durability of TLD with high circulating resistance to 3TC and TDF and metabolic abnormalities observed in clinical trials. Limited programmatic data are available to describe the success of the TLD transition in the region. Methods: We established the DISCO cohort to quantify viral suppression and regimen tolerability during the TLD transition. We prospectively enrolled adults from public clinics in Uganda and South Africa who had been on non-nucleoside reverse transcriptase inhibitor-based ART for ≥6 months and were programmatically switched to TLD. We obtained demographics, medical history data, and plasma specimens at enrollment and week 24. We conducted retrospective HIV-1 RNA viral load (VL) testing using the Cepheid GeneXpert platform. Though both sites were interrupted by COVID-19, here we report complete week 24 results for the Uganda cohort. Results: We enrolled 500 participants (41% female) in Uganda. Median age was 47 years (IQR 40-53). Median ART duration was 8.8 years (IQR 5.7-12.2). The most common regimens prior to TLD switch were 3TC/TDF/efavirenz (44%) and 3TC/zidovudine/nevirapine (39%). Retrospective VL testing demonstrated that 95% (475/499) had VL 1,000 copies/mL at enrollment. 90% (448/500) completed week 24 visits, with 50 additional visits delayed during COVID-19, 1 disenrollment, and 1 death. By week 24, 1% (6/448) discontinued TLD due to side effects or clinician discretion. At week 24, 96% (432/448) had VL 1,000 copies/mL. Of those with week 24 VL >50 copies/mL, 31% (5/16) had detectable VL >50 copies/mL at enrollment, versus 3% (15/431) in those with suppressed VL at week 24 (χ2 p-value<0.001). Conclusion: The great majority of participants transitioned to TLD with an undetectable VL. Overall, we documented 86% suppression at week 24 after TLD switch in the midst of the COVID-19 pandemic and 96% suppression in those completing a week 24 visit. These data support early tolerability and efficacy of TLD transition in the public sector. However, detectable VL at switch predicted detectable VL at 24 weeks. Vigilance and programmatic monitoring are needed to ensure long-term durability of TLD.