Background
To date clinical trials have been underpowered to assess the preferred antiretrovirals in therapy-naïve people with advanced HIV disease (PWAH). We investigated the efficacy and safety of an integrase inhibitor (INI) versus a boosted protease inhibitor (PI) containing regimen in PWAH.
Methods
In this open-label, randomised, multicentre, non-inferiority trial, therapy-naive people with HIV aged ≥18 years, a viral load >1000 copies/mL, and either AIDS at any CD4 cell count, severe bacterial infection (BI) with a CD4 cell count <200/µL, a CD4 cell count <100/µL, or currently being treated for opportunistic infections (OI), were randomised 1:1 to receive bictegravir or darunavir/cobicistat, each co-formulated with tenofovir alafenamide/emtricitabine for 48 weeks. The time to first occurrence of the primary outcome, which was a composite of specified virological failure (insufficient virological response was defined as HIV-1 RNA reduction <1 log 10 copies/mL at week 12, or viral load >50 HIV-1 RNA copies/mL at week 48) or clinical events plus the individual components were evaluated by Kaplan Meier and Cox regression analyses, with a non-inferiority margin of 1.606 in the hazard ratio.
Results
442 participants were randomised and well matched for baseline characteristics with a median age of 43 years, 81% male, and 62% white. 85.8% had CD4 T cell counts <100/µL and 44.6% viral loads >500,000 HIV-1 RNA copies/mL. In intention-to-treat analyses, the primary composite outcome event occurred in 49/220 patients in the INI group versus 70/222 patients in the PI group by 48 weeks (adjusted hazard ratio [aHR] 0.70; 95% confidence interval [CI] 0.48-1.00; p=0.05, noninferiority demonstrated). The per-protocol analysis gave a similar estimated aHR of 0.69 (95% CI 0.48-1.00). Kaplan-Meier estimates for virological failure was significantly lower in the INI versus PI group (aHR 0.54; 95% CI 0.33-0.88, p=0.013); however, clinical events were similar between groups (Figure 1). Incidence of drug-related AE (grade ≥2) was 13.7 in the INI group versus 21.7 per 100 person-years in the PI group (p=0.04). The incidence of grade 3-4 drug-related AEs (p=0.99), AE leading to study drugs interruption (p=0.92), and serious AEs (p=0.82) did not differ between groups.
Conclusions
In PWAH the bictegravir containing regimen was non-inferior to the darunavir containing regimen in terms of the composite outcome but had a better virologic response at week 48 and fewer overall AE.