Metabolic and Fat Complications

Hopkins HIV Report 2006 Mar; 18(2):8-9,11

Joseph Cofrancesco, Jr., M.D., M.P.H., F.A.C.P.
Johns Hopkins

A number of important studies of metabolic and fat complications were presented at the 13th CROI in Denver.

Fat Loss and Fat Gain Are Not Linked

Phyllis Tien reported an analysis of 338 HIV-infected women in the FRAM study, all of whom underwent rigorous metabolic and body shape measurement [Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 749]. The authors found that the duration of stavudine (d4T) use was associated with less subcutaneous fat in all compartments but not with increases in visceral fat, and that indinavir (IDV) use was associated with more upper trunk fat, confirming that lipoatrophy and fat accumulation are not linked phenomena and have different risk factors. This study also found that in women, the effects of age and physical activity were stronger than drug effects.

Cardiovascular Disease (CVD) Risk

Everyone agrees there is an increase in CVD risk in HIV-infected patients, but there is controversy about whether the risk is due to antiretroviral therapy or HIV infection itself. Fueling the controversy were several studies presented at CROI. The SMART study (reported in this issue by Joel Gallant) randomized over 5,000 subjects with CD4 cell counts above 350 cells/mm³ to continue HAART or to discontinue, resuming when the CD4 count fell below 250 cells/mm³. In the interruption arm, in addition to more HIV progression, the relative risk (RR) for a serious complication or death was 2.2, and for a nonfatal cardiovascular event was 1.5 [El Sadr W, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 106LB]. It has been hypothesized that the increased risk following discontinuation of therapy may be due to inflammation resulting from rebound of HIV viremia. These results should give us pause before we consider stopping HAART to correct metabolic abnormalities.

The D:A:D group continued to report on its 23,400 patients in 11 cohorts from Europe, Australia and Europe [Friis-Moller, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 144]. After adjusting for other risk factors and other drug class use, the authors found that for myocardial infarction (MI), the RR increase was 1.16 per year of PI use (p=0.0001) and 1.05 per year of NNRTI (p=0.17). PI-specific data were not presented. Even after adjusting for lipids, the mechanism by which PIs presumably increase risk, the risk associated with PI use was maintained, suggesting additional medicationrelated factors independent of lipid effects. Importantly, the authors found fewer MIs in more recent calendar years, in large part explained by increased use of cholesterol-lowering agents.

However, Judith Currier reported on ACTG 5078, a smaller study of 134 subjects in 45 “triads” (the triads were: HIV-infected patients taking PIs, HIV-infected patients not taking PIs, and HIV-uninfected patients, all matched on age, race, sex, BP, smoking status, and menopausal status) [Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 145]. The authors found that neither HIV infection nor PI use predicted the 3-year change in carotid intima thickness, a predictor of heart disease. Models in the HIV-infected subjects did suggest that ritonavir (RTV) use, longer PI use, elevated fasting glucose, and high C-reactive protein were suggestive of more progression.

As with several other studies, the HOPS cohort of over 78,000 patients found that traditional risk factors were most predictive of CVD [Lichtenstein K, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 735]. Risk factors from multivariate analyses included age over 40 (AOR 3.3), hypertension (AOR 2.2), diabetes (AOR 2.4), and low HDL cholesterol (AOR 2.6). In their model, CD4 cell count at the initiation of HAART, time on HAART, switching HAART, and specific agents were not significant predictors. In those with hypertension or hypercholesterolemia, the use of antihypertensive and cholesterol lowering agents decreased risk. These studies remind us to be aware of CVD risk, which likely comes from a combination of antiretroviral therapy, HIV itself, and traditional risk factors. This is becomes increasingly important as our patients age. A large proportion of the risk can be reduced by focusing on traditional risk factors: stop smoking, control hypertension, lower lipids, and add aspirin where appropriate.

To that end, John Gerber presented data suggesting that fish oil (3 g twice daily, each gram containing 500 mg eicosapentaenoic acid and 310 mg docosahexaenoic acid) could help those with high TG levels [Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 146]. Approximately 100 subjects entered step 1, randomly receiving open-label fish oil or fenofibrate for 8 weeks; those in both groups with TG>200 at week 8 went onto step 2, in which both fish oil and fenofibrate were administered. At baseline, participants in both groups were similar, with a mean age of 43 years and TG of 660-690; 90% were male and 56% were white. Enrollment criteria included having an LDL <160 mg/dL; diabetics and those on other lipid-lowering medications were excluded. A small percentage of subjects achieved TG <200 mg/dL after step 1 (8.5% for fish oil and 16.7% for fenofibrate); the remainder were offered step 2. By ITT analysis, 17/75 (22.7%) achieved TG <200 mg/dL, and even for those who did not, there was an average decrease in TG from 377 mg/dL to 279 mg/dL. LDL cholesterol increased in each group, as is known to occur with fish oil, (22, 37, and 14 mg/dL for the fish oil, fenofibrate, and combined groups, respectively), although it is not clear if this was clinically significant. There were no negative effects on a variety of immune stimulation parameters measured, and there were few dropouts, suggesting good tolerability of the medications. Fish oil had no effect on lopinavir trough concentrations, suggesting no significant drug interactions. This study suggests a role for high dose fish oil in patients with high TG levels; in many cases it may need to be used in combination with fibrates and/or statins.

Malan reported 48-week data of a 96-week trial of the efficacy and safety of atazanavir (ATV) in 199 ART-naïve subjects, beginning to answer the question of whether low-dose RTV increases lipids. [Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 107LB, also discussed in this issue by Gallant]. Subjects were randomized to receive open-label ATV 400 mg daily or ATV 300 mg plus RTV 100mg daily in combination with d4T and lamivudine (3TC). There was more jaundice (grade 2-4) (3% vs. <1%), a slightly larger increase in total cholesterol (15% vs. 6 % , P<0.01), and a larger increase in TG (26% vs. -3%, p<0.01) in the boosted arm. LDL cholesterol increased (23% vs. 16%) and HDL cholesterol decreased (-30% vs. -9%) in both arms. The study demonstrates that 100 mg of RTV does cause statistically significant changes in cholesterol, though they were relative small and may not be clinically significant. Patients with untreated HIV infection tend to have low cholesterol levels, and any treatment can raise them to what would have been the patient’s “healthy” baseline. Also, all subjects in this study received d4T, which can negatively affect cholesterol levels. This study supports the characterization of boosted and unboosted ATV as a “lipid friendly” PI.

Fat Gain/Loss

The large HOPS cohort studied 2,304 subjects who had at least 2 visits to a one of multiple HIV clinics from 1996-2005 and found that patients starting and remaining on HAART at higher CD4 cell counts were significantly less likely to experience toxicity [Lichenstein K, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 769]. In a multivariate analysis the adjusted odds ratio for lipoatrophy suggested a benefit from starting HAART with a CD4 count of 200-349 cells/mm³ (AOR 0.5) or CD4 >350 cells/mm³ (AOR 0.4). Tenofovir use was associated with less lipoatrophy (OR 0.6), whereas white race (2.7) and use of d4T (1.8), didanosine (ddI) (1.6), and any PI (2.1) were associated with more lipoatrophy. Lipoatrophy is not generally associated with PI use; however, subjects were not randomized in this observational study, and there may be residual confounding even after adjustment. Providers may have used PIs in patients with more advanced disease, or PI use may be a marker for longer and/or more ART use.

In ACTG 5082, 105 subjects were randomized to receive metformin 1000 mg bid, rosiglitazone 4 mg qd, both drugs, or no drugs in a placebo controlled randomized trial [Mulligan K, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 147]. Subjects had an increased waist-to-hip ratio (>0.95 in males or >0.85 in females) or waist circumference >100 cm, indicative of fat accumulation, and evidence of impaired glucose tolerance but not diabetes; 66% percent were male, 65% were white, 65% were taking PIs, 43% were taking NNRTIs, and 96% were taking NRTIs. Many metformin subjects dropped out (12/26), primarily due to diarrhea, which limited the analysis. However, the study failed to show significant decreases in abdominal visceral fat or subcutaneous adipose tissue over 16 weeks in any of the treatment arms. Not surprisingly there was a decrease in the insulin area under the curve concentration (AUC), indicating improved insulin sensitivity, but there were no changes in 2-hour glucose by oral glucose tolerance test. Limb fat also tended to decrease with metformin, a potentially undesirable effect for patients with lipoatrophy, but increased significantly with rosiglitazone (+4.8% vs. -8.3% in the placebo group, p=0.034). Combining drugs ameliorated this effect on fat. There was also a decrease in HDL cholesterol with rosiglitazone. There was no significant increase in lactate.

Similarly, Kohli and colleagues found no benefit with respect to fat accumulation or lipid levels in 25 subjects receiving 1,500 mg metformin daily for 24 weeks compared to 23 receiving placebo [Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 148]. There was a trend in the metformin arm toward reduction in appendicular fat mass. Metformin may be a less viable option then previously thought for the treatment of fat accumulation, especially in patients who also have lipoatrophy. This study says nothing about patients with diabetes, however, and the literature in HIV-uninfected patients suggests that metformin is a good choice for diabetic patients who are obese.

There are conflicting studies on the efficacy of rosiglitazone to improve lipoatrophy. Some show a small benefit (see above), while others have demonstrated no benefit. No study has demonstrated a “normalization” of fat. At CROI, we heard results of another “glitazone” study, in which 130 subjects were randomized to receive pioglitazone 30 mg qd or placebo for 48 weeks [Slama L, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 151LB]. Approximately 80% of the subjects were male, 27% had AIDS, and 25-30% remained on d4T. Overall, there was improvement in limb fat (+0.38 kg, p=0.051) without a change in visceral fat. However, it is important to look at the sub-analysis. Improvement only occurred in patients not currently receiving d4T (+0.45 kg compared to +0.04 kg with placebo, p=0.013). Pioglitazone patients also experienced improved limb circumference and triceps skin fold measurement as well as small but statistically significant increases in HDL cholesterol. However, patients perceived no improvement in their lipoatrophy. Here we see a small but real increase in subcutaneous fat, but we are reminded that we must also remove offending agents (e.g., d4T) and that the changes may not be noticed by patients, at least in the short-term. It is possible that the differences will become more noticeable if fat continues to increase over time.

The jury is still out on the use of these compounds. Metformin didn’t decrease visceral fat and tended to worsen lipoatrophy; rosiglitazone and pioglitazone caused small increases in limb fat, but the changes were not dramatic. We still need to find better ways to deal with fat accumulation and lipoatrophy. The best approach is still to select medicines that don’t cause the problems in the first place.

Testosterone and anabolic steroids can be useful in rebuilding muscle mass in patients with HIV wasting. However, they have also been used by some clinicians to treat fat accumulation. Shikuma reported a study of 75 male subjects with mild hypogonadism (total testosterone 125-400 ng/dL) and a waist-to-hip ratio >0.95 or waist circumference >100 cm, who were randomized to receive testosterone gel 10 mg/day or placebo for 24 weeks in a double blinded fashion [Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 149]. The study had mixed results: In the treatment group, total body fat by DEXA decreased (-7.9% vs. +4.5% , p<0.001), extremity fat decreased (-10.1% vs. 3.8%, p=0.001), and lean body mass (muscle) increased (+1.3% vs. -0.3% p=0.02). The loss of limb fat was also seen on CT scan, with no affect on visceral fat. The results of lipids, glucose and hormone levels were not presented. The men in this study were only mildly hypogonadal and received twice the standard replacement dose. Muscle mass improved, but limb fat decreased and there was no decrease in visceral fat. It may be that the usual 5 mg daily dose would have had different effects. Once we see the hormone level data we may be able to extrapolate further. Clinicians must remember there are often benefits with respect to mood, sexual function, energy, anemia, bone density, and overall quality of life when a frankly hypogonadal man receives testosterone at replacement doses, so the decision to treat cannot be made based solely upon this study. But as other studies in this report, clinicians should not assume that testosterone, and by extension other anabolic steroids, will decrease visceral fat. In addition, use of these agents may exacerbate pre-existing lipoatrophy.

Metabolic Effects in Children

Although not the focus of this report, there were a number of presentations confirming that the metabolic effects of antiretroviral therapy we see in adults also occur in children and adolescents. These include studies demonstrating better lipid outcomes with nevirapine (NVP) compared to efavirenz (EFV) [Huhtamaki T, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 688], that use of PIs, especially RTV-boosted PIs, but not NNRTIs, increases total cholesterol and triglycerides, [Ramos JT, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 690], and that there are no significant lipid changes with ATV other than mild increases when boosted [Samson P, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 689]. HIV-infected children have worse cardiovascular markers and carotid intima media thickness [McComsey G, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 691] and changes in fat depots [Vigano A, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 692] compared to HIV-uninfected children. In children, switching from d4T to tenofovir DF (TDF) can lead to significant increases in total and limb fat, but without restoration to baseline levels [Vigano A, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 693].

Conclusions

CROI was a helpful meeting for those interested in fat changes, metabolic toxicity, and cardiovascular risk. Well beyond the scope of this report but quite promising is the movement toward using various genetic markers to predict who may or may not be at risk for specific complications [Ranade K, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 763; Arnedo M, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 764; Owen A, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 765; De Luca A, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 766; Cossarizza A, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 767; Gometz ED, et al. Conf Retroviruses Opportunistic Infect 2006 Feb 5-8;13:abstract no. 768]. We also saw confirmation that changes occur not only in adults but also in children. There appears to be agreement on the increased risk of cardiovascular disease in our patients, but the relative contributions of antiretroviral therapy, specific antiretroviral agents, and HIV itself remain unclear. Clinicians are reminded that controlling traditional risk factors is still the best approach. At this meeting, we learned that interrupting therapy may be unwise, and that there is evidence supporting earlier initiation of therapy. Fat loss and fat accumulation, while often occurring together, are distinct phenomena with separate risk factors. We are also seeing that some promising therapies, such as metformin or testosterone for fat accumulation, are ineffective and may have unwanted side effects such as exacerbation of lipoatrophy. Even those with evidence of some benefit, such as “glitazones” for lipoatrophy, only yield a mild improvement. The task of the clinician remains complex: to select treatment regimens that will maximize the likelihood of a durable virologic response while minimizing the potential for side effects and longterm toxicity.

2006-03-10
HHR-2006-03-05

Copyright © 2006 - The Johns Hopkins University AIDS Service, Division of Infectious Diseases. Permission to use and reproduce portions of this newsletter is hereby granted provided that author and publication are fully credited and both copyright and permission notice appear with reprinted material. Inquiries may be directed to Sharon McAvinue, Managing Editor. Website: Johns Hopkins AIDS Service.

AEGiS is a 501(c)3, not-for-profit, tax-exempt, educational corporation. AEGiS is made possible through unrestricted funding from Boehringer Ingelheim, Bridgestone/Firestone Charitable Trust, Bristol-Myers Squibb Company, Elton John AIDS Foundation, the National Library of Medicine, Roche and Trimeris, and donations from users like you. Always watch for outdated information. This article first appeared in 2006. This material is designed to support, not replace, the relationship that exists between you and your doctor.

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 2006. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content.