Rosiglitazone improves fat loss and insulin resistance in people with lipoatrophy

The diabetes drug rosiglitazone (Avandia) improved limb lipoatrophy in HIV-positive people taking antiretroviral therapy, researchers reported on Wednesday at the Sixteenth Conference on Retroviruses and Opportunistic Infections in Montreal. However, the researchers did not study the drug's effect on facial fat loss.

Many people with HIV develop peripheral lipoatrophy, or fat loss in the face, limbs or buttocks. Fat loss can be distressing because it changes an individual's appearance and is an obvious sign of HIV and AIDS.

Lipoatrophy has been linked to certain antiretroviral drugs, in particular the thymidine nucleoside reverse transcriptase inhibitors (NRTIs) d4T (stavudine, Zerit) and AZT (zidovudine, Retrovir). Due to their long-term side-effects, these drugs are no longer considered preferred therapies in Europe and the US, but they are still widely used in developing countries.

Rosiglitazone belongs to a class of drugs known as thiazolidinediones, which work by increasing the activity of a cell receptor protein called PPAR-gamma that plays a role in fat metabolism. Thymidine NRTIs strongly inhibit PPAR-gamma, so researchers explored whether rosiglitazone might counteract this effect.

This prospective study included 71 HIV-positive participants with fat wasting who had taken thymidine NRTIs for at least 12 months in the past, but had not done so for at least six months prior to enrolment. Patients were randomly assigned to receive 4mg twice-daily rosiglitazone or placebo for 48 weeks. Laboratory tests and dual energy X-ray absorptiometry (DEXA) scans were used to assess metabolic and limb fat changes.

Baseline characteristics were similar in the two groups, except that the placebo recipients had a higher average total cholesterol level. Most participants (85%) were men and the average age was about 50 years. The overall mean amount of limb fat was about 6500g. Close to 40% had pre-existing insulin resistance, but people with frank diabetes were excluded.

Patients had well-controlled HIV disease, with about 90% having viral load below 400 copies/ml and a mean CD4 count of approximately 600 to 700 cells/mm³. On average, participants had been off thymidine NRTIs for nearly four years.

By week 48, limb fat had increased in both groups, as would be expected when patients stop taking thymidine NRTIs (fat restoration appears to occur very slowly in the absence of thymidine NRTIS, even without a treatment intervention such as rosiglitazone). But the improvement was significantly greater in the rosiglitazone group than in the placebo group, with fat gains of 900g vs 300g, respectively. Rosiglitazone also produced greater improvement in the percentage of limb fat gained (15% vs 5%, respectively).

Furthermore, insulin resistance (assessed using the HOMA method) and insulin levels decreased in patients taking rosiglitazone, whilst rising slightly in those taking the placebo.

Mean increases in total cholesterol and non-HDL ("bad") cholesterol at 48 weeks were significantly greater in the rosiglitazone group, and triglycerides showed a non-significant trend in the same direction. HDL ("good") cholesterol, however, did not change in either group. Body mass index and bone mineral density also did not change significantly in either arm.

Rosiglitazone was generally well tolerated. Nine participants were lost to follow up (about half in each treatment arm), but only one person in the rosiglitazone group stopped early due to a possible drug-related adverse event (worsening of pre-existing coronary artery disease).

In previous studies of the HIV-negative general population, rosiglitazone has been linked to increased cardiovascular disease risk and decreased bone density leading to an elevated risk of fractures. To further explore this issue, the researchers are conducting additional analyses including measurements of carotid artery intima-media thickness (a marker for atherosclerosis), biomarkers of inflammation and bone biomarkers.

Given these findings, the investigators concluded that – in the absence of thymidine NRTIs – rosiglitazone "significantly improves peripheral lipoatrophy even in subjects without insulin resistance", and may be a "promising addition" for fighting lipoatrophy in HIV-positive people.

Some past studies of rosiglitazone, and the related drug pioglitazone (Actos), have not shown similar improvements in HIV-related lipoatrophy, but these often included participants who remained on thymidine NRTIs. For patients who experience fat wasting, switching to a different NRTI backbone – if appropriate alternative are available – would be a preferred first step for managing lipoatrophy before trying additional medications.

In response to a question from an advocate in the audience, presenter Marisa Tungsiripat acknowledged that the study did not measure changes in facial fat – often the biggest concern of patients – so it remains unclear whether these findings will translate into 'real-world' benefits that are noticeable to people with HIV.