A randomized comparative 96-week trial of boosted atazanavir versus continued boosted protease inhibitor in HIV-1 patients with abdominal adiposity

Background: Abdominal adiposity in HIV-1 patients initiating antiretroviral therapy may be part of a restoration-to-health phenomenon. Lipoatrophy is associated with long-term thymidine analogue therapy. Individual protease inhibitors (PIs) differ in their effects on lipids and insulin resistance. Methods: A randomized open-label multicentre 96-week trial compared changes in fat distribution in patients with suppressed HIV-1 RNA and abdominal adiposity, who either continued on their current twice-daily ritonavir-boosted PI (PI/r) or switched to once-daily boosted atazanavir (ATV/r). Treatment with two nucleoside reverse transcriptase inhibitors was unchanged. Body composition was assessed by dual-energy x-ray absorptiometry (DEXA) and abdominal computerized tomography (CT) scanning. Results: In total, 201 patients were randomized; 131 switched to ATV/r. Viral suppression (<50 copies/ml) was similarly maintained (93% ATV/r versus 89% PI/r). Mean changes from baseline in trunk-to-limb fat ratio were similar; difference estimates 0.03 (95% CI -0.06, 0.12; P=0.48 at week 48) and 0.02 (95% CI -0.10, 0.14; P=0.73 at week 96). More patients in the PI/r arm had a decrease of ≥20% in limb fat from baseline at week 96. Significantly greater reductions in proatherogenic lipids occurred following switch to ATV/r. Both treatment regimens were generally well-tolerated; the incidence of grade 3-4 treatment-related clinical adverse events was 34% among ATV/r recipients versus 4% of PI/r-treated patients. Conclusions: Switching to ATV/r had no demonstrable benefit on abdominal adiposity. Maintenance of efficacy, less limb fat loss and marked reduction in proatherogenic lipids was observed with ATV/r compared with continuing a PI/r regimen.