Lipoatrophy Following Long-term Antiretroviral Therapy

Case Patient

A 45-year-old man with long-term HIV infection, likely acquired through intravenous (IV) heroin use.

The patient recalls being diagnosed with HIV in 1985 and reports a CD4+ cell count nadir of 90 cells/mcL in the early 1990s, although he denies any history of opportunistic infection or malignancy. The patient was not treated with antiretroviral therapy (ART) until the early 1990s, when he initiated stavudine and lamivudine. Despite ongoing IV drug use, the patient stated that he was very careful to take his ART. Upon further questioning, he did admit to missing occasional doses, but he claimed that this was a rare occurrence. Despite relatively good compliance, he had detectable HIV plasma viral loads since 1997. His most recent plasma viral load measurements were in the range of 10,000-15,000 copies/mL, and his CD4+ cell count remained stable in the range of 250-300 cells/mcL.

Other Medical History

Previous HIV clinicians who cared for the patient had tried to convince him to change to therapy that included a protease inhibitor (PI) and/or a nonnucleoside reverse transcriptase inhibitor (NNRTI), along with optimization of his nucleoside reverse transcriptase inhibitor (NRTI) backbone. The patient was repeatedly unwilling to do so, citing concern that his drug use could at some point affect his compliance and result in more resistant virus or drug interactions with his heroin. He therefore deferred any changes in therapy until he “got clean.” He recalls 2 months of abstinence from IV drugs in 1999, but has failed to achieve abstinence again for any significant duration of time. Despite stating that he was interested in discontinuing illicit drug use, the patient remained precontemplative in his formulation of a plan to stop IV use of heroin.

The patient claimed that he was happy with his medication regimen and denied any treatment-related adverse effects. His physical exam, however, was significant for severe facial and limb lipoatrophy, a side effect that he had not previously attributed to his ART. Though still unwilling to change therapy, he agreed to resistance testing to facilitate discussion of further options. A genotype consistent with resistance to most members of the NRTI class revealed multiple thymidine analogue-associated mutations (TAMs), M184V, but no K65R. His genotype did not detect any PI or NNRTI mutations, as expected given his lack of exposure to these classes of antiretroviral drugs.

Discussion

This case is illustrative of a NRTI-experienced patient with virologic failure and relative immunologic stability, yet who does not appear to be gleaning significant benefit from the current ART regimen and who is experiencing toxicity due to long-term thymidine NRTI use. One of the most challenging aspects of this case is the patient's concern about development of PI and NNRTI resistance related to potential noncompliance to ART because of illicit drug use. Clearly, discontinuation of IV drugs would be a landmark step in the overall clinical management of this individual, and intensive interventions should be offered to facilitate abstinence or at least harm reduction. Several studies have indicated that this stigmatized portion of the HIV-infected population often receives suboptimal access to ART, despite mixed evidence regarding compliance to medications.[1–5] Physicians' attitudes toward IV drug users and perceptions of noncompliance appear to have a significant impact on initiation of ART in this population.[6] Each case needs to be approached individually, and patients capable of or interested in initiating ART should be adequately supported to allow for successful administration of life-extending therapy.

From the perspective of optimizing this patient's HIV therapy, options include discontinuing all ART if he is unwilling to change regimens, using an NRTI-sparing regimen, or starting a salvage regimen that includes NRTI. The first option, though unappealing from the risk perspective, merits brief discussion.

The patient appears to be benefiting only slightly, if at all, from his current regimen, while suffering the cosmetic and potentially metabolic complications of long-term exposure to a thymidine NRTI (stavudine). Lipoatrophy is an increasingly recognized complication of antiretroviral therapy, with thymidine NRTIs such as stavudine frequently implicated in the theories of mitochondrial toxicity-related pathogenesis.[7–9] The severity of this patient's ART-related complication has to be carefully weighed against the possible viral fitness cost that this nonsuppressive regimen may be providing. The patient's CD4+ cell count nadir of 90 cells/mcL makes a complete discontinuation of ART unappealing, given the likelihood of CD4+ cell depletion caused by more fit virus that would probably emerge upon ART discontinuation. Nevertheless, partially effective ART in the setting of worsening antiretroviral drug-related complications may require discontinuation of therapy, despite the physician's responsibility to first do no harm with interventions.

If the patient is willing to accept a new regimen, the challenge of its design revolves around the decision to include or exclude the NRTI antiretroviral class. Inclusion of these agents in the setting of likely class-wide resistance might decrease viral fitness by maintaining resistance mutations associated with attenuated virulence. Both TAMs and M184V have been associated with in vitro impact on replication capacity, with still inconclusive clinical correlation in vivo.[10–12] Selection of NRTI drugs in this case would need to be made in the context of the patient's severe lipoatrophy. Several medication “switch” studies using abacavir or tenofovir have demonstrated improvement in lipoatrophic subjects previously treated with thymidine NRTI-containing ART regimens.[13–16] These data can be extrapolated to the present clinical situation, which favors an NRTI backbone that includes lamivudine along with abacavir or tenofovir in combination with further optimized ART.

The second option that does not include an investigational agent is to use a NRTI-sparing regimen. In this patient, that could include a PI in combination with a NNRTI, or a dual-PI regimen. NNRTI options would include efavirenz or nevirapine along with a PI (eg, lopinavir, atazanavir, or fosamprenavir) boosted with low-dose ritonavir. PI dosage adjustments would likely be required given decreased NNRTI levels that occur during coadministration of some combinations of NNRTIs and PIs.

Dual-PI therapy, though an option, is less appealing in this situation given likely activity of NNRTIs, as indicated by the patient's genotype and lack of previous exposure to this drug class. Limited clinical experience with NRTI-sparing regimens has been reported. The NEKA study compared ritonavir-boosted lopinavir in combination with nevirapine to nevirapine in combination with 2 NRTIs. After 48 weeks of follow-up, the subjects enrolled in both arms maintained viral suppression.[17] ACTG study 5110 randomized 77 subjects with undetectable viral loads on thymidine NRTI-containing regimens to switch from stavudine or zidovudine to abacavir, or to start a NRTI-sparing regimen of lopinavir/ritonavir with nevirapine. Preliminary results at 24 weeks demonstrated increase in limb fat of 8% in subjects on the NRTI-sparing group while no change in limb fat was observed in the abacavir switch group. CD4+ cell counts of those on the NRTI-sparing regimen increased by 8 cells/mcL compared with the NRTI group (P = .03), and 93% of those on the NRTI-sparing arm maintained virologic suppression, a proportion similar to that of subjects on the NRTI-containing regimen (92%).[16] Similar data on fat gain were demonstrated in ACTG trial 5116 and its substudy 5125s. In this trial, individuals with advanced HIV disease (CD4+ cell count nadir < 200 cells/mcL or pretreatment plasma viral load of > 80,000 copies/mL) were treated with standard ART for at least 18 months. If plasma viral load was maintained at < 200 copies/mcL at this time point, then patients were switched to either an NRTI-sparing regimen of lopinavir/ritonavir with efavirenz, or efavirenz with 2 NRTIs. Despite a demonstrated gain in peripheral fat, the NRTI-sparing regimen trended toward more toxicity-related discontinuations (17% compared with 5%), which was mostly attributed to elevated triglycerides. Furthermore, although it did not reach statistical significance, there appeared to be a trend toward increased virologic failure in the NRTI-sparing regimen compared with standard therapy.[18,19]

Case Resolution

The patient ultimately decided to accept switch in his regimen following several more discussions about his therapeutic options.

Given his naivet to the NRTI and PI class, he was started on dose-adjusted lopinavir/ritonavir with efavirenz. Because of mixed data on the virologic potency of NRTI-sparing regimens, he was also started on tenofovir and lamivudine in an attempt to exert a positive effect on viral fitness. The decision to avoid the thymidine NRTIs (stavudine and zidovudine) in his salvage regimen was based on the lipoatrophy he had developed while on stavudine, as well as the presence of that drug in his previous nonsuppressive ART regimen.

The patient ultimately suppressed his viral load to < 50 copies/mL 8 weeks after starting his new regimen. Though not formally measured through nuclear studies or CT scans, no clear change in facial or limb atrophy was noted on initial follow-up. The patient's fasting triglyceride level was 200 mg/dL prior to initiating his new regimen, while a subsequent value was 250 mg/dL; he was referred to a nutritionist to optimize his diet. To date he has not developed any overt toxicities or intolerances to his regimen outside of transient vivid dreams from efavirenz.

Readers are encouraged to respond to George Lundberg, MD, Editor of MedGenMed, for the editor's eye only or for possible publication via email: glundberg@medscape.net