Background:  Appendicular fat loss continues to be one the most troubling side effects of long term ART regimens. Therapeutic options for lipoatrophy remain limited. Nucleoside reverse transcriptase inhibitor (NRTI)-sparing regimens are attractive potential alternatives to avoid toxicities associated with NRTI therapy.

Methods:  A5125s, the metabolic sub-study of A5116, examined the effects of NRTI-sparing and protease inhibitor (PI)-sparing regimens on fat distribution and bone mineral density; using whole body and regional DXA, and metabolic parameters in subjects with advanced HIV (nadir CD4 ≤ 200 cells/mm³ or HIV RNA ≥ 80,000 copies/mL) and undetectable HIV RNA ≤ 200 copies/mL after at least 18 months of ART, who were randomized to switch their initial successful antiretroviral regimen to open-label ritonavir-boosted lopinovir (LPV/r) (533/133 mg twice daily) + efavirenz (EFV) (600 mg once daily) vs EFV+2 NRTI.

Results:  We enrolled 62 subjects (85% men, 60% white). Median age was 42 years and median baseline CD4 was 432 cells/mm³ (p = ns). Median baseline appendicular fat was 6 kg. At week 48, median change in appendicular fat in the LPV/r+EFV arm was +562 g (IQR –218 to 1186 g) vs a loss of –246 g (–631 to 459 g) in the NRTI-containing arm (p = 0.097). At the time of last observation (n = 46, median 104 weeks) a median gain of 782 g (+10%, IQR –380 to 1168 g) of appendicular fat was noted in the non-NRTI arm vs a loss of –900 g (–15%, IQR –1301 to –466 g) in the NRTI-containing arm (p = 0.002). The study was not powered to detect differences between specific NRTI. At week 48, the LPV/r+EFV arm had greater increases in triglycerides (85 vs 11 mg/mL, p = 0.01) and in total cholesterol (+19 vs –7 mg/mL; p = 0.0008) compared with the NRTI arm. HDL cholesterol changes were similar (+6 mg/mL for LPV/r+EFV vs. +2 mg/ml for the NRTI- arm). There were no significant changes within each arm or differences between arms in fasting glucose, insulin and HOMA-IR levels. Trunk fat and bone mineral density remained stable.

Conclusions:  The switch to a non-NRTI containing combination of LPV/r+EFV was associated with significant improvement in appendicular fat, increases in serum lipids, and stable glucose metabolism and regional bone mineral density. These findings support the observations that LPV/r has minimal effects on glucose metabolism, but is associated with greater increases in triglycerides and cholesterol than a NRTI-containing regimen. These results provide additional evidence that NRTI are important in progressive appendicular fat loss that characterizes HIV-lipoatrophy. The switch to a NRTI-sparing regimen represents a therapeutic option for patients with lipoatrophy.

Keywords: lipoatrophy; nucleoside reverse transcriptase inhibitors; switch study