ACTG 5142 was a large randomised study that compared a nucleoside-sparing regimen of Kaletra and efavirenz against Kaletra or efavirenz paired with lamivudine (3TC) plus either d4T, AZT or tenofovir. The study was designed to test whether avoidance of nucleoside analogue drugs, some of which can cause lipoatrophy, was effective and safe.
Virological results of the study, presented at the International AIDS Conference in Toronto in August 2006 showed that patients randomised to Kaletra plus two NRTIs experienced virological rebound significantly more quickly than patients randomised to efavirenz plus two NRTIs. By intent-to-treat analysis, 89% of the participants receiving efavirenz-based triple therapy had viral loads below 50 copies/ml after 96 weeks, compared with 77% receiving Kaletra-based triple therapy (p=0.003). Kaplan-Meier analysis showed that recipients of Kaletra-based triple therapy had a significantly faster time to virological failure than the efavirenz triple therapy arm (P=0.006)
The study recruited 753 patients in the United States, 64% non-Caucasian and 20% female. The median baseline CD4 cell count was 191 cells/mm3 and the median baseline limb fat as measured by DEXA scan was 7.1kg.
In the two-nucleoside analogue-containing arms, 42% received AZT, 24% d4T and 34% tenofovir as their second nucleoside analogue.
After 96 weeks those in the nucleoside-sparing arm had experienced an average 18% gain in limb fat (around 1kg), compared to a 9.8% gain in the Kaletra + 2 NRTIs group and a gain of 1.4% in the efavirenz group.
Lipoatrophy, defined as a 20% loss of limb fat at week 96, was experienced by 32% of the efavirenz + 2NRTI group, 17% of the Kaletra + two NRTI group, and 9% of the Kaletra/efavirenz group.
When the incidence of lipoatrophy was analysed according to the second nucleoside analogue used, it was most commonly seen in patients who received d4T – 42% compared to 27% in the AZT group. This difference was statistically significant. Nine per cent of tenofovir-treated patients developed lipoatrophy, a significantly lower proportion than the AZT recipients.
When lipoatrophy incidence was analysed according to pairings of drugs, it was evident that tenofovir recipients who received efavirenz were more likely to develop lipoatrophy (12%) than those who received Kaletra (6%). Similarly, AZT recipients who received efavirenz were also at greater risk of lipoatrophy (40% vs 16% for Kaletra recipients). The difference was less pronounced for d4T recipients (51% for efavirenz, 33% for Kaletra).
Efavirenz recipients had a 2.7-fold higher risk of lipoatrophy compared to Kaletra recipients (p<0.001), d4T recipients had a 1.9-fold higher risk of lipoatrophy compared AZT recipients (p=0.029) and tenofovir recipients had a significantly lower risk of lipoatrophy when compared to AZT recipients (OR 0.24, p=0.001).
Total cholesterol and triglyceride increases were significantly higher in the nucleoside-sparing arm than in the Kaletra arm (+57mg/dl vs 32mg/dl, p=0.001; +62mg/dl vs 46mg/dl, p=0.03)); there was no significant difference in cholesterol elevation between the Kaletra and efavirenz arms. The increase in HDL cholesterol was significantly greater in the nucleoside-sparing arm.
Presenting the results, Richard Haubrich of the University of California, San Diego, said that revision of treatment guidelines would need to take into account the effects of different regimens on body fat distribution.