HIV-Associated Lipoatrophy: What Are the Kinder, Gentler Agents?

Received 12 September 2005; accepted 14 September 2005; electronically published 5 December 2005. Reprints or correspondence: Dr. Michael P. Dubé, Wishard Memorial Hospital, 1001 W. 10th St., Ste. OPW-430, Indianapolis, IN 46202 (mpdube@iupui.edu). Clinical Infectious Diseases 2006; 42:281–2 2005 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2006/4202-0020$15.00 Lipoatrophy associated with antiretroviral treatment of HIV infection is a common and serious problem that is associated with significant aesthetic and metabolic derangements [1]. Clearly, therapy with thiazolidenediones or switching of therapy to nucleoside reverse-transcriptase inhibitors is suboptimal for addressing severe established lipoatrophy, so when possible, it is preferable to avoid or minimize this complication altogether by choosing the kinder, gentler antiretroviral agents. Because it is not clear whether fat accumulation, such as abdominal obesity, breast enlargement, and buffalo hump, is a complication associated with the use of particular drugs, I will limit my comments to the more definitively antiretroviral treatment–related morphologic complication of lipoatrophy. Multiple lines of evidence from observational and randomized studies implicate nucleoside reverse-transcriptase inhibitors, particularly stavudine, in lipoatrophy. A recent randomized trial that compared stavudine versus tenofovir (either given in combination with lamivudine and efavirenz) reported greater limb fat levels (as determined by dual-energy x-ray absorptiometry [DEXA]) among tenofovir recipients at weeks 96 and 144 [2]. However, this study did not include baseline DEXA findings and thus could not evaluate the longitudinal patterns of fat changes over time or compare results during treatment with pretreatment status. Thus, it was not clear whether the overall pattern was one of gain or loss over time in this study, but the inferiority of stavudine was clear. In a substudy of a prospective, randomized trial that compared stavudine plus didanosine with zidovudine plus lamivudine (both given in combination with nelfinavir and/or efavirenz), subjects assigned to receive zidovudine and lamivudine had an increase in limb fat level from baseline (as determined by DEXA) of 4% at 64 weeks, whereas subjects assigned to receive stavudine and didanosine had a decrease of nearly 17% ( , by between-groups P ! .001 comparison for overall change from baseline) [3]. Although the slight increase in limb fat level from the baseline level with zidovudine and lamivudine assignment at 64 weeks in this study is encouraging, longer-term data are needed to assess longer-term outcomes [4]. Regardless, differences between nucleoside reverse-transcriptase inhibitors are becoming clearer. Studies that compare tenofovir-, zidovudine-, and abacavir-based combination regimens are anxiously awaited. Existing clinical data are inadequate to speculate whether there are differences between those agents with regards to lipoatrophy. HIV-1 protease inhibitors, when given in combination with nucleoside reversetranscriptase inhibitors, may also be capable of contributing to lipoatrophy. There are conflicting data from observational studies with regard to the independent contribution of protease inhibitors to the development of lipoatrophy [4, 5]. Substitution of nonnucleoside reverse-transcriptase inhibitors for protease inhibitors has not been effective for treatment of established lipoatrophy. However, one randomized prospective study revealed greater limb fat loss with nelfinavir treatment, compared with efavirenz treatment, over 64 weeks [3]. However, the magnitude of the protease inhibitor effect in this study was smaller than the effect of nucleoside assignment. In this issue of Clinical Infectious Diseases, in a substudy of a randomized trial of subjects with a CD4 cell count of 100 cells/mm whose initial antiretroviral therapy consisted of zidovudine and lamivudine plus either efavirenz or atazanavir [6], Jemsek et al. [7] report no evidence of lipoatrophy at 48 weeks of treatment. With both combinations, there was a small increase from the baseline level in both abdominal subcutaneous and limb fat levels that was not statistically significant, and