Switching Due to Lipoatrophy: Tenofovir Bests Abacavir in Tolerability, Lipid Profile

Abacavir (ABC, Ziagen) or tenofovir (TDF, Viread): Which is a better bet when attempting to help alleviate the lipoatrophy experienced by patients who are taking stavudine (d4T, Zerit) or zidovudine (AZT, Retrovir)? The latest RAVE study results show that, although both abacavir and tenofovir can modestly improve lipoatrophy after a switch, tenofovir is better tolerated than abacavir, and also has a more favorable impact on lipid levels -- particularly when patients switch to it from stavudine.

RAVE is a randomized, open-label, 48-week British study of 105 patients who were taking a thymidine analog (stavudine or zidovudine) along with another nucleoside analog and either a protease inhibitor or non-nucleoside agent. These patients were on stable antiretroviral therapy for more than 24 weeks and had a viral load below 50 copies/mL, but were experiencing moderate or severe lipoatrophy. After enrollment, all trial participants were randomized to change the thymidine analog to either tenofovir (n = 52) or abacavir (n = 53). (All participants were naive to both tenofovir and abacavir.)

Graeme Moyle presented 48-week data from this study at the 12th Conference on Retroviruses and Opportunistic Infections (CROI) this past February. Those results suggested that switching from a thymidine analog to tenofovir or abacavir resulted in significant improvements in limb fat, and noted that tenofovir led to fewer treatment discontinuations, and somewhat greater improvements in lipid profiles, than abacavir. The CROI results were not stratified by baseline thymidine analog, however; this presentation rectifies that omission, and updates the data presented at that conference.

But first, a brief review of the data we knew coming into this presentation. As noted in the RAVE results presented at CROI, patients switching to tenofovir or abacavir had been receiving antiretroviral therapy for 5.7 and 4.9 years, respectively. A disproportionate number of the patients randomized to receive tenofovir had been on stavudine at baseline (77%, versus 59% of those who switched to abacavir). The baseline CD4+ cell count was 522 cells/mm3 for patients who switched to tenofovir and 478 cells/mm3 for those who switched to abacavir.

During the 48-week study, more patients discontinued abacavir (8) than tenofovir (3); there were three cases of abacavir hypersensitivity (at the expected rate of 6%). By week 48, median limb fat increased by 393 g in patients who switched to tenofovir, versus 316 g in patients who switched to abacavir. Fasting levels of total and low-density lipoprotein (LDL) cholesterol decreased more significantly in trial participants who switched to tenofovir.

New data presented at this conference were stratified by baseline thymidine analog; select findings are outlined in the chart below. The data show that the effect on total and LDL cholesterol of switching to tenofovir was more marked in patients previously receiving stavudine than zidovudine, likely reflecting a more important role for stavudine in the pathogenesis of dyslipidemia. Abacavir had no appreciable effect on cholesterol levels, irrespective of the type of prior therapy.

Changes in Mean Fasting Lipid Levels (mmol/L)

 d4T or AZT => ABCd4T or AZT => TDFd4T => ABCd4T => TDFAZT => ABCAZT => TDF
Change in total cholesterol+0.2-0.50.0-0.6+0.4-0.1
Change in LDL cholesterol+0.1-0.3-0.1-0.3+0.4-0.1
Change in triglycerides+0.1-0.3+0.4-0.4-0.30.0

Among patients on stavudine at baseline, a decrease in triglyceride levels was only seen in those who switched to tenofovir. Interestingly, a switch from stavudine to abacavir led to a 0.4 mmol/L increase in triglyceride levels, while a switch from zidovudine to abacavir appeared beneficial (-0.3 mmol/L). These data are difficult to explain, in light of the general lack of effect of a switch to abacavir on lipid profiles, and may be artifactual. Finally, at week 48, there were fewer trial participants on tenofovir with dyslipidemia by National Cholesterol Education Program criteria (25% vs. 39%), despite the fact that a higher proportion in that group had such abnormalities at baseline (38% vs. 28%).

When we treat patients who are on long-term stavudine (and possibly also zidovudine) therapy, there is a tendency to want to change this agent to another nucleoside analog, so that the patients may avoid or reduce long-term drug-associated toxicity. A single-drug switch can be safely undertaken in the setting of maximal virologic suppression, and this is commonly done in clinical practice. Such a switch could even be contemplated in patients receiving an agent associated with dyslipidemia or lipodystrophy before any abnormalities develop, a strategy that may actually be of greater benefit to the patient in the long run.

As most patients are already on lamivudine (3TC, Epivir) or emtricitabine (FTC, Emtriva), the most common agents considered for this switch are tenofovir and abacavir. This carefully done RAVE study shows both strategies to be equivalent with respect to reversal of lipoatrophy, and gives us a good idea of the magnitude of improvement that can be expected over the course of a year. Beyond this equivalence, however, tenofovir is better tolerated and leads to a more favorable lipid profile. Based on the data presented in the RAVE study, as well as long-term experience with its efficacy and tolerability in clinical practice, tenofovir must be considered the agent of choice for this indication of therapeutic substitution.