Antiretroviral drugs have variable effects on bones and fat, lipoatrophy uncommon with modern regimens

This article is more than 12 years old. Click here for more recent articles on this topic

Different antiretroviral regimen components are linked to varying changes in bone mineral density and body fat, according to a sub-study of the US ACTG 5202 trial reported at the 17th Conference on Retroviruses and Opportunistic Infections (CROI) last week in San Francisco. Tenofovir/FTC (Truvada) led to greater bone loss, whilst boosted atazanavir (Reyataz) produced both more bone loss and greater limb and trunk fat gain.

ACTG 5202 was a phase IV clinical trial that compared four commonly used combination regimens. More than 1800 previously untreated participants were randomly assigned to receive either Truvada or abacavir/3TC (Kivexa or Epzicom) as a nucleoside/nucleotide reverse transcriptase inhibitor (NRTI) 'backbone'. They also received either ritonavir-boosted atazanavir or efavirenz (Sustiva or Stocrin) as a 'base' drug on an open-label basis.

The final results of the parent study were also presented at CROI. In brief, the various regimens produced comparable viral suppression in people with baseline viral load below 100,000 copies/ml, but showed some differences in side-effects.



Loss of body fat from specific areas of the body, especially from the face, arms, legs, and buttocks.

bone mineral density (BMD)

The higher your bone mineral content, the denser your bones are. And the denser your bones, the stronger they are and the less likely they are to break. A bone density test uses X-rays to measure how many grams of calcium and other bone minerals are packed into a segment of bone. The bones that are most commonly tested are in the spine, hip and sometimes the forearm. 

intent to treat analysis

All participants in a clinical trial are included in the final analysis, in the groups they were originally assigned to, whether or not they actually completed their course of treatment. This method provides a better estimate of the real-world effect of a treatment than an ‘on treatment’ analysis.


A clinical trial where both the researcher and participants know who is taking the experimental treatment. 

boosting agent

Booster drugs are used to ‘boost’ the effects of protease inhibitors and some other antiretrovirals. Adding a small dose of a booster drug to an antiretroviral makes the liver break down the primary drug more slowly, which means that it stays in the body for longer times or at higher levels. Without the boosting agent, the prescribed dose of the primary drug would be ineffective.

ACTG 5224s was a sub-study of the larger trial that looked at two types of side-effects, bone loss and body fat changes. Grace McComsey and colleagues used dual energy X-ray absorptiometry (DEXA) scans to measure bone mineral density changes in the lumbar spine (lower back) and hip bone, as well as changes in limb fat and trunk (primarily abdominal) fat.

They also looked at the number of bone fractures and the percentage of people who developed lipoatrophy, or fat loss in the limbs and face, after 96 weeks of therapy. For this study, lipoatrophy was defined as a 10% or greater loss of limb fat from baseline. Dr McComsey described this degree of fat loss as "mild", a change that might not be visually evident to patients or clinicians.

The sub-study included 269 participants, with similar numbers taking each of the four regimens. Demographic and HIV status were similar across all groups. Most participants (85%) were men, about half were white, the median age was 38 years, and the median body mass index (BMI) was 24.9, the top of the normal range.

The participants had relatively advanced HIV disease, with a baseline median CD4 cell count of 233 cells/mm3, and just over 40% having fewer than 200 cells/mm3; 41% had a high viral load of 100,000 copies/ml or more. None had diabetes or other conditions known to affect bone or body composition.

Bone density

All regimens led to an initial steep drop in bone mineral density soon after starting treatment, which then rose and stabilised at a somewhat higher level – though not back to the baseline level – after about a year. There was no evidence of any significant interactions between the NRTI backbones and the base drugs.

In the primary intent-to-treat analysis at week 96 comparing the NRTI backbones, people taking Truvada had significantly larger decreases from baseline in lumbar spine and hip bone density compared with Kivexa recipients (-1.3% vs -3.0% for spine; -2.6% vs -3.9% for hip).

Looking at the open-label base components, atazanavir/ritonavir was associated with more bone loss than efavirenz at both sites, but the difference was only significant for the lumbar spine (-1.7% vs -3.2% for spine; -3.1% vs -3.4% for hip).

Though the primary analysis was at week 96, Dr McComsey also showed longer follow-up data through week 192, at which point approximately 45% of the initial participants were still being followed. Bone density continued to rise with Kivexa (plus either efavirenz or atazanavir), but there was a downward turn with Truvada starting around week 144.

Overall, 5.6% of participants had at least one bone fracture, all of them traumatic (due to injury). The researchers also reviewed data on fractures in the ACTG 5202 parent study. In this larger group, the fracture frequency was 4.3% (a rate of 1.7 per 100 person years); 12.7% of these were non-traumatic (spontaneous). Despite variations in bone mineral density, there were no significant differences in fracture frequency between the NRTI arms or between the base drugs in either the sub-study or parent study.

Body fat

Turning to fat, Dr McComsey reminded listeners that fat gain in the arms and legs is considered good, relative to the lipoatrophy seen in the past. (People often gain a modest amount of weight after starting antiretroviral treatment.)

Trunk or central fat gain is often unfavourable, but could also signal a "return to health" in people with wasting due to advanced HIV disease.

Overall, 16% of participants had a 10% or greater limb fat loss (protocol-defined lipoatrophy) at week 96, and there were no statistically significant differences between either the NRTI backbones or the base drugs (intent-to-treat analysis). The frequency of lipoatrophy was 14.3% for Truvada plus efavirenz, 15.6% for Truvada plus atazanavir/ritonavir, 18.9% for Kivexa plus efavirenz and 16.3% for Kivexa plus atazanavir/ritonavir.

Dr McComsey told a subsequent press conference that the measure of lipoatrophy used in this study was very strict, and that for lipoatrophy to be noticeable to patient and doctor, limb fat loss of 40 to 50% would need to occur. In this population limb fat loss of 10% represents a total loss of around 1kg (2.2lbs) over two years.

The researchers also added a post-hoc (unplanned) analysis of the percentages of people who experienced a 20% or greater limb fat loss. This showed more variation: 8.9%, 0%, 3.8% and 6.1%, respectively.

By week 96, the average amount of limb fat had increased for both Truvada and Kivexa recipients, with average gains of 1.1 vs 1.7 kg, respectively, in the intent-to-treat analysis, not a significant difference. Percentage changes were also statistically similar. After 96 weeks, limb fat continued to rise with both NRTI backbones.

In an analysis of patients as treated (that is, taking into account treatment discontinuations and switches), limb fat gains were 1.2 kg with Truvada vs 2.1 kg with Kivexa, which did reach statistical significance. The difference in percentage change, however, remained non-significant. After 96 weeks, limb fat continued to rise with Truvada but fell with Kivexa.

With regard to the open-label base drugs, the amount of limb fat rose more with atazanavir/ritonavir than with efavirenz by week 96 (1.9 vs 1.0 kg, respectively), and the percentage increase was about twice as large with atazanavir/ritonavir (roughly 15% vs 30%, respectively); both were significant in an intent-to-treat analysis. With longer follow-up, efavirenz appeared to catch up by week 192.

In an unplanned analysis of trunk fat, average amounts gained (about 1.5 to 2.0 kg) and percentage changes (about 25%) were similar in the Truvada and Kivexa groups. As with limb fat, these changes were modest.

Trunk fat rose significantly more with atazanavir/ritonavir than with efavirenz (about 1.2 vs 2.5 kg, or roughly 20% vs nearly 40%, respectively).

Summing up, the researchers found that all regimens appeared to produce an initial bone loss, with subsequent stabilisation after week 48. Truvada led to greater bone mineral density loss in the lumbar spine and hip than Kivexa, and atazanavir/ritonavir led to greater bone loss in the lumbar spine (but not hip) than efavirenz. Fractures were similarly distributed amongst the study arms.

Further, regimens containing Truvada or Kivexa increased both limb fat and trunk fat, and were not significantly different in an intent-to-treat analysis. Boosted atazanavir led to greater gain in limb fat and trunk fat than efavirenz.

The investigators concluded that lipoatrophy – even the mild protocol-defined form – occurred in about 16% of participants, and was not significantly different between Truvada and Kivexa or between atazanavir/ritonavir and efavirenz.

Speaking at an accompanying press conference, McComsey said that regimens containing Truvada "did worse" than those with Kivexa, keeping in mind that bone loss is bad and limb fat gain is good.

These findings are important, she added, because they can help clinicians tailor regimens for people who have a higher risk of bone loss, such as post-menopausal women.

With regard to limb fat, she said the lipoatrophy rates were "very low", which offers "very encouraging news" that these modern regimens do not cause some of the problems seen with some of the older drugs.

Further information

You can view the abstract on the official conference website.

You can also view a webcast and slides of this session on the official conference website.


McComsey G et al. Bone and limb fat outcomes of ACTG A5224s, a substudy of ACTG A5202: a prospective, randomized, partially blinded phase III trial of ABC/3TC or TDF/FTC with EFV or ATV/r for initial treatment of HIV-1 infection. Seventeenth Conference on Retroviruses and Opportunistic Infections, abstract 106LB, San Francisco, 2010.