icon-folder.gif   Conference Reports for NATAP  
  9th Conference on Retroviruses and Opportunistic Infections
Seattle, Washington, February, 2002
Back grey_arrow_rt.gif
Rosiglitazone In The Treatment of HAART-Associated Lipodystrophy: a randomized, double-blinded, placebo controlled study
Written by Jules Levin
  Author: Jussi Sutinen, Helsinki University Central Hospital, Helsinki, Finland
The negative results from this study have created quite a bit of discussion and disappointment. Here are the details of this study presented at the oral late breaker at the recent Retrovirus Conference.
Background offered by authors. Lipodystrophy in HIV is associated with insulin resistance & impaired glucose tolerance, loss of subcutaneous fat, and intra-abdominal fat accumulation. Glitazones (PPAR-gamma agonists) are new insulin sensitizing anti-diabetic drugs. These drugs stimulate adipocyte (fat cell) differentiation (making more fat cells) in vitro (test tube). It reverses the block in adipose tissue differentiation induced by HAART in vitro. They increase insulin sensitivity in type-2 diabetics, and increase truncal subcutaneous fat by 34% in HIV-negative patients with lipodystrophy. A number of studies suggest that HIV-infected patients with fat redistribution demonstrate fasting hyperinsulinemia, with normal fasting glucose levels. Decreased glucose uptake has been shown on clamp studies, indicating insulin resistance.
In vitro and in HIV-negative study volunteers have shown that protease inhibitors may inhibit glucose transport. Recent in vivo studies conducted in non HIV-infected patients demonstrated that exposure to indinavir over 4 weeks increased insulin resistance as assessed by euglycemic, hyperinsulinemic clamp. It has also been suggested that changes in the immune system, perhaps due to HIV or HAART-induced immune reconstituion may be related to body changes and insulin resistance. As well, body changes themselves may lead to insulin resistance. It remains uncertain whether insulin resistance contributes to body changes or if body changes cause insulin resistance. To read more about this go to this link to read a review by Steve Grinspoon:
And the NATAP website Lipodystrophy section:
Back to the Finnish study.
The aim of this small pilot study was to determine whether rosiglitazone as compared to placebo has favorable effects on HAART associated lipodystrophy (body composition, features of insulin resistance). And whether rosiglitazone is safe in patients using HAART. The study was 24 weeks in duration, and 8 mg per day vs placebo was used. Visits to the study site were at weeks 0, 2, 6, 12, 18, and 24. Patients had self-reported and investigator confirmed lipodystrophy. Patients were excluded if they had diabetes, ALT 3 times greater than the upper limit of normal, were pregnant, or had cardiac insufficiency.
30 patients were randomized to placebo or rosiglitazone (5 female), CD4s were 572. 70% had <50 copies/ml HIV RNA. 73% were currently on a PI. All patients were HCV antibody and HBV s-antigen negative.
There were 15 patients randomized to placebo and 15 to rosiglitazone. 35 HIV-negative patients were used as controls. Ages were similar across all 3 groups (42-44). BMI (kg/m2) was similar (23.6 to 23.7). Waist to hip ratio was 0.99 in the placebo group, 0.98 in the rosiglitazone group, and 0.94 in the HIV negative group. Apparently, the HIV positive patients had greater waist to hip ratios. Serum-insulin (mU/l) (pmol/l) was 10 (60) in the placebo group, 13 (78) in the rosiglitazone group, and 6 (36) in the HIV-negative group. Apparently, serum insulin was lower in the HIV-negative group. Serum triglycerides (mmol/l) (mg/dl) were 3.2 (282) in the placebo group, 3.5 (308) in the rosiglitazone group, and 1.1 (97) in the HIV-negative group.
Visceral fat (deep in the belly) and abdominal subcutaneous fat (surface fat that is supposed to be the same as the fat loss in the face & periphery) were measured using MRI (16 slices). Liver fat was determined by proton spectroscopy. Serum leptin, which is synthesized supposedly exclusively in adipose tissue and is said to correlate closely with body fat mass, was determined as an additional measure of fat mass. Skinfold thickness was determined. Bioimpedance was analyzed. And self-assessment was evaluated.
There was no change in weight in either the placebo or rosiglitazone groups.
There was no change in body fat mass in either the placebo or rosiglitazone groups.
Interestingly, improvements in self-assessment were reported from week 0 to week 24 for both fat loss and fat accumulation in the both the placebo and rosiglitazone groups, but these differences between weeks 0 and 24 were statistically significant only in the placebo group.
There was no change from week 0 to week 24 in subcutaneous fat (as measured by MRI) and in intra-abdominal fat (also measured by BMI) in either the placebo or rosiglitazone arms. At week 0 the placebo group appeared to have more subcutaneous fat than the rosiglitazone group. (editorial note: perhaps, these patients had progressed too far for rosiglitazone to work. Maybe early in the stage of developing fat loss rosiglitazone might be effective).
Serum leptin concentrations remained unchanged.
The sum of skinfolds done at several sites (I think he said 6) was unchanged.
The fasting serum insulin and liver fat content were improved in the rosiglitazone groups but worsened in the placebo groups. The mean changes in insulin resistance and liver fat between the two groups were statistically significant.
There were statistically significant increases in fasting serum cholesterol and fs triglycerides in the patients receiving rosiglitazone compared to the placebo patients. The peak increases in triglycerides occurred at week 6 and slowly declined. The cholesterol increases also peaked at week 6 but remained stable. The week 6 changes were statistically significant, and the subsequent differences between the placebo & rosiglitazone regarding cholesterol were significant (p<0.01). The difference in triglycerides between the placebo & rosiglitazone groups were significant at week 6 only.
Anemia is a specific adverse event for glitazones. In keeping with this finding from previous data, hemoglobin declined from 150 to 139 from week 0 to week 24 in the rosiglitazone group (statistically significant), but not in the placebo group.
There were no changes in HIV viral load, CD4 count, PI concentration (baseline vs 6 weeks), and in lactate.
Liver enzymes were measured every 6 weeks and no liver toxicity was found. On the contrary, serum ALT declined significantly in the rosiglitazone group but not in the placebo group. The author suggested the decrease in ALT may be due to the decreased liver fat content (watch your diet if you have hepatitis). One patient dropped out at week 12 due to serum triglycerides of 32.5 mmol/l (2860 mg/dl) in the rosiglitazone group. I think the author said other commonly observed side effects or toxicities seen with glitazones were not observed in this study (such as edema).
The author summarized that although rosiglitazone seemed to improve insulin resistance, it did not change the amount of subcutaneous or intra-abdominal fat. It induced marked hyperlipidemia in several patients. The author discussed his thoughts on why they saw these negative results. He said dose was not the issue because they used the highest dose of rosiglitazone recommended. He doubts if the study were longer if better results would have been seen as substantial fat gains have been seen in type-2 diabetics and HIV-negative persons with lipodystrophy after only 12-24 weeks of treatment. He said we do not know if different baseline characteristics, patients with proven insulin resistance, would have changed the results. When patients were divided up by having either higher or lower fasting serum insulin at baseline there was no difference in the study results (body changes). Although he felt convinced by the results he said the small study size might obscure some benefit to some patients.
Regarding a potential explanation for the negative results, the author suggested that perhaps the target molecule of glitazones, PPAR gamma, is either absent or nonfunctioning in the subcutaneous adipose tissue of these patients. He referred to a new paper about to be published in the Lancet by a French research group demonstrating a drastic decline in the amount of PPAR gamma messenger RNA and I think he said protein levels in subcutaneous fat samples in patients with HAART associated lipodystrophy.