Peripheral and central fat changes in subjects randomized to abacavir-lamivudine or tenofovir-emtricitabine with atazanavir-ritonavir or efavirenz: ACTG Study A5224s

Abstract

Background: We compare the effect of 4 different antiretroviral regimens on limb and visceral fat.

Methods: A5224s was a substudy of A5202, a trial of human immunodeficiency virus type 1 (HIV-1)-infected, treatment-naive subjects randomized to blinded abacavir-lamivudine (ABC-3TC) or tenofovir DF-emtricitabine (TDF-FTC) with open-label efavirenz (EFV) or atazanavir-ritonavir (ATV-r). The primary endpoint was the presence of lipoatrophy (≥ 10% loss of limb fat) at week 96 by intent-to-treat (ITT) analysis. Secondary endpoints included changes in limb and visceral fat. Statistical tests included linear regression, binomial, two-sample t test, and Fisher's exact test.

Results: A5224s enrolled 269 subjects; 85% were male, and 47% were white non-Hispanic. The subjects had a median baseline HIV-1 RNA level of 4.6 log(10) copies/mL, a median age of 38 years, a median CD4+ cell count of 233 cells/μL, median limb fat of 7.4 kg, median visceral adipose tissue (VAT) of 84.1 cm(2), and VAT: total adipose tissue (TAT) ratio of 0.31. At week 96, estimated prevalence of lipoatrophy (upper 95% confidence interval [CI]) was 18% (25%) for ABC-3TC and 15% (22%) for TDF-FTC (P = .70); this was not significantly less than the hypothesized 15% for both (P ≥ .55 for both). The secondary as-treated (AT) analysis showed similar results. At week 96, the estimated mean percentage change from baseline in VAT was higher for the ATV-r group than for the EFV group (26.6% vs 12.4%; P = .090 in ITT analysis and 30.0% vs 14.5%; P = .10 in AT analysis); however, the percentage change in VAT:TAT was similar by ITT and AT analysis (P ≥ .60 for both). Results were similar for absolute changes in VAT and VAT:TAT.

Conclusions: ABC-3TC- and TDF-FTC-based regimens increased limb and visceral fat at week 96, with a similar prevalence of lipoatrophy. Compared to the EFV group, subjects assigned to ATV-r had a trend towards higher mean percentage increase in VAT.

Clinical trials registration: NCT00118898.

Figure 1.

Details of disposition and outcome of study subjects. Subjects were to remain in follow-up regardless of having modified antiretroviral therapy. Nucleoside reverse-transcriptase inhibitors (NRTIs) were double-blinded through 25 February 2008 for those with screening human immunodeficiency virus (HIV) RNA levels of ≥100,000 copies/mL and until final visits starting 1 July 2009 for those with screening HIV RNA levels <100,000 copies/mL. Reasons for study discontinuation are split into no. of subjects with and no. of subjects without week 96 limb fat primary endpoint. Site closure was censored for premature study and treatment discontinuation. Death was censored for premature study discontinuation and counted as the reason for treatment discontinuation if there was no prior modification. Reasons for the first treatment modification are split into no. of subjects before, no. of subjects after, and no. of subjects without week 96 limb fat primary endpoint. Subjects in the as-treated sample continued to receive randomized treatment through the week 96 whole-body dual-energy X-ray absorptiometry.

Figure 2.

Estimated mean percent changes in limb fat and trunk fat by study week by NRTI components (combining third drugs) and NNRTI/PI components (combining NRTIs) by ITT and AT analyses. Vertical bars are 95% confidence intervals on the mean change.