Abacavir Substitution for Nucleoside Analogs in Patients With HIV Lipoatrophy A Randomized Trial

CONTEXT: Peripheral lipoatrophy may complicate antiretroviral therapy of human immunodeficiency virus (HIV) infection, often related to duration and type of nucleoside analog therapy, and may have a mitochondrial pathogenesis. No proven therapy exists for lipoatrophy, but abacavir is a nucleoside analog that may be less toxic to mitochondria. OBJECTIVE: To determine if substitution of stavudine or zidovudine with abacavir improves HIV lipoatrophy without affecting control of HIV replication. DESIGN: Randomized, open-label 24-week study. SETTING: Seventeen hospital HIV outpatient clinics and primary care centers in Australia and England, with randomization from June 2000 through January 2001. PARTICIPANTS: A total of 111 adults (109 men) with moderate or severe lipoatrophy who were receiving stavudine (n = 85) or zidovudine (n = 26) and had stable plasma HIV RNA levels below 400 copies/mL and no prior abacavir therapy. INTERVENTION: Patients were randomly assigned to switch from stavudine or zidovudine to abacavir, 300 mg twice per day, while continuing all other antiretroviral therapy (n = 54) or to continue all antiretroviral therapy (n = 57). MAIN OUTCOME MEASURES: The primary end point was limb fat mass, measured by dual-energy x-ray absorptiometry; key secondary end points were plasma HIV RNA levels, adverse events, physician-assessed (via subjective measures) lipodystrophy severity, total and central fat mass, and fasting metabolic (lipid, glycemic, and lactate) levels. RESULTS: There was a significant increase in limb fat in the abacavir group relative to the stavudine/zidovudine group (0.39 vs 0.08 kg; mean difference, 0.31; 95% confidence interval [CI], 0.06-0.57 kg), as well as significant relative increases in subcutaneous thigh (P =.01), arm (P<.001), and abdominal (P =.001) fat areas on computed tomography. Switching had no significant effect on secondary end points, including plasma HIV RNA (for unadjusted comparison between groups at week 24, odds ratio, 1.38; 95% CI, 0.48-3.96). Change in limb fat mass at week 24 did not correlate with change in subjectively determined perceived lipoatrophy severity (r = -0.06; P =.53 by Spearman correlation). Hypersensitivity to abacavir was seen in 5 patients (10%). CONCLUSIONS: In this sample of lipoatrophic HIV-infected adults, switching from stavudine or zidovudine to abacavir for 24 weeks led to significant, albeit modest, objectively measured increases in limb fat. Clinical lipoatrophy, as assessed subjectively, did not resolve, however, and at the rate of increase observed may take years to resolve with use of this strategy. Longer-term follow-up is needed.