Weight gain on HIV treatment: more than a 'return to health' phenomenon?

Weight gain on antiretroviral treatment is not solely a ‘return to health’ effect, research presented earlier this month at the Conference on Retroviruses and Opportunistic Infections (CROI 2024) suggests.

Since the introduction of integrase inhibitor-based antiretroviral treatment, clinical trials and cohorts around the world have reported weight gain in people with HIV after starting treatment. In most people the weight gain is modest, but in a substantial minority, weight gain can approach 5-6kg within two years, and some people develop obesity.

Scientists studying weight gain on antiretroviral treatment have tended to favour one of two explanations. One is that weight gain is caused by antiretroviral drugs – the original assumption. The other is that weight gain is a ‘return to health’ phenomenon that occurs when antiretroviral drugs suppress HIV and the weight-sapping effect of fighting a chronic viral infection is ended. According to this model, the return to health effect is especially noticeable in people with low CD4 counts who have been living with HIV for some time. And if there are differences in weight gain between drug regimens, this is largely explained by the suppressive effect of two drugs – efavirenz and tenofovir disoproxil – on weight rather than any weight-promoting effect of specific drugs.

Studies have found plenty of reasons to question the role of antiretrovirals in weight gain, but less attention has been paid to the validity of the ‘return to health’ theory. Is the weight gain a normal recovery from chronic infection, or is it being amplified by medication?

At this month’s Conference on Retroviruses and Opportunistic Infections in Denver, several poster presentations raised questions about whether the ‘return to health’ effect provides a complete explanation of why many people gain weight after starting treatment. Some presentations also provided new information about biological mechanisms that might explain why differing patterns of weight gain are seen with various drugs.

Weight gain in people who start treatment soon after acquiring HIV

The CASCADE cohort collaboration set out to investigate whether the people least likely to show a ‘return to health’ effect – people who have recently acquired HIV – show any signs of weight gain after starting treatment.

The analysis used follow-up data on people who started antiretroviral treatment within 12 months of acquiring HIV and who had been on stable treatment for at least three months. Participants came from cohorts in Europe, Australia, Canada and Africa.

A total of 5,698 people were eligible for inclusion in the analysis. People included in the analysis started treatment a median of four months after their estimated date of HIV acquisition and had a median age of 34. The median CD4 count at the time of treatment initiation was 459 and participants were followed for just under two years.

Forty-one percent of those included in the analysis started treatment with an integrase inhibitor, 31% with a protease inhibitor and 19% with non-nucleoside reverse transcriptase inhibitor.

The investigators assessed changes in weight by antiretroviral regimen and baseline body mass index.

In those with the lowest body weight before starting treatment (body mass index below 18.5), people taking integrase inhibitors or protease inhibitors had gained substantially more weight six months and three years after starting treatment than people taking NNRTIs (4kg - 5kg vs 2.7kg).

A similar difference between drug classes was evident in the normal weight and overweight categories, although the amount of weight gained was less for all classes of antiretroviral. In the obese weight category, the difference between drug classes was modest and those taking integrase inhibitors or NNRTIs gained less weight than those taking these drugs in lower weight categories (+2.2kg and +1kg respectively) after three years.

Across all weight categories people taking tenofovir alafenamide (TAF) gained the greatest amount of weight, irrespective of which other agents were included in the combination. Weight gain associated with TAF treatment after three years ranged from just under 4kg in the obese category to 5.1kg in the overweight category.

The proportions of those in the normal weight category who gained substantial weight (+5% of body weight or +10% of body weight) was highest in those taking integrase inhibitors (53% and 31% respectively) and lowest in those taking NNRTIs (40% and 20%).

Within the integrase inhibitor class, regimens containing bictegravir and elvitegravir combined with TAF were associated with greater weight gain.

Participants were not switching from other regimens that might have suppressed weight and did not have advanced immunosuppression, both factors that might affect the extent of weight gain after starting the regimens associated with greater weight gain in this study.

However, the CASCADE study lacks a control group in the form of a matched sample of people without HIV. (Pantazis)

Weight gain in people with and without HIV compared

A study of people with HIV in the United States led by Dr Richard Elion did compare weight changes in 11,888 people with HIV on treatment and 56,768 people without HIV. Rather than analysing a relatively homogenous sample of people with HIV – early diagnosed and treated HIV – the study of people from clinics participating in the Trio Health HIV Research Network pulled in both treatment-experienced and newly treated people with supressed HIV. Treatment-experienced people had at least three years of follow-up, while newly treated people were eligible for inclusion if they had been on treatment for at least one year. The analysis then adjusted for differences between participants and matched a sub-group of people with HIV (n=1296) to people without HIV (n=4168).

The study found that after adjusting for variables that affect weight, there was no difference between people with HIV and people without HIV in weight change over three years. However, people with HIV were 30% more likely to shift upwards from one body mass index class to another (e.g. normal to overweight) than people without HIV. Similarly, a greater proportion of people with HIV gained at least 10% in body weight than people without HIV. (Elion)

CD4 count and weight-suppressive drugs

However, an analysis of three randomised trials conducted in Africa and India suggests that weight gain on treatment might be better thought of as an interaction between HIV and its treatment. In the ADVANCE and NAMSAL studies, treatment with dolutegravir was associated with greater weight gain, especially when combined with TAF. Dr Andrew Hill and trial investigators assessed weight gain according to antiretroviral regimen at various CD4 count strata in the 2202 participants in the three trials, who received one of three regimens. Of those with CD4 counts below 100 (11%), who might be expected to have the lowest body weight and the greatest weight gain in a ‘return to health’, weight gain was greatest in those who received both dolutegravir and TAF, and least in those who received both efavirenz and TDF. And the analysis also found that the proportion who developed obesity after 96 weeks on treatment was greatest in people with baseline CD4 counts below 100, suggesting that a prolonged period of immunosuppression may prime the body to respond more strongly to drug-induced metabolic changes – or that when weight increases, some drugs will blunt the weight gain. (Hill)

A retrospective analysis of 4194 people with HIV who started antiretroviral treatment between 2008 and 2022 at a large HIV clinic in Texas found that in the first three years on treatment, body mass index increased by a median of 1.8 kg/m2 (from 24.4kg/m2), an increase of 7.5%. Only two drugs were significantly associated with the extent of weight change: people taking TDF or efavirenz gained significantly less weight than people taking other antiretrovirals. Use of TAF or an integrase inhibitor did not affect the amount of weight gained. (Drechsler)

Mechanisms to explain differing drug effects on weight

During World War Two, researchers at the University of Minnesota carried out an experiment to investigate the effects of semi-starvation and subsequent re-feeding in healthy young men, to guide food relief programming for refugees. It showed that after a period of semi-starvation, increased food intake encouraged the body to prioritise the creation of new fat reserves, especially in those with the greatest fat mass prior to starvation. So, it is unsurprising that people with advanced HIV and CD4 counts below 100 should gain the most weight after starting treatment.

But what happens in people with less advanced HIV? An animal study presented at CROI suggests that even in people who are not obese and do not have advanced HIV, antiretroviral treatment has an effect akin to re-feeding after a period of starvation.

A group at the National Primate Research Center, Oregon Health and Sciences University investigated the metabolic impact of starting antiretroviral treatment after a short period of untreated infection in primates. The group fed 22 macaques one of two diets, either a diet that induced a lean body or another that induced obesity. The animals were then infected with simian immunodeficiency virus (SIV). Five weeks after infection they were started on antiretroviral treatment containing tenofovir DF, emtricitabine and dolutegravir. Essentially, what the experiments showed was that in lean animals, SIV infection without treatment followed by treatment induced a metabolic state typical of pre-diabetes– signalled by a decreased ratio of adiponectin to leptin. This state encourages the development of obesity through greater fat storage in adipocytes. (Sauter)

Several other poster presentations also explored the ways in which specific antiretroviral drugs might influence body weight. One study, by Dr Sangwon Kim and colleagues at Johns Hopkins University, looked at the action of dolutegravir on energy expenditure. The sex hormone oestrogen regulates energy expenditure through signals in the hypothalamus, the region of the brain that regulates many bodily processes. The study found that in female mice fed dolutegravir for two weeks, oestradiol activity was blocked by dolutegravir, and energy expenditure fell, even though food consumption remained stable. The investigators say that their findings may point to a mechanism by which integrase inhibitors lead to weight gain, especially in women. (Jung)

Finnish researchers investigated whether the site at which TDF is processed into useable drug might explain differences in weight gain between people taking tenofovir disoproxil fumarate (TDF) and tenofovir alafenamide (TAF). TDF, the prodrug of tenofovir, must be transformed into free tenofovir in enterocytes, cells that line the gut and absorb water, proteins, fats, sugars and other substances that have been broken down in the small intestine. The study sampled enterocytes from 12 people with HIV taking TDF and 12 taking TAF. None had gastrointestinal disease. The study found that people taking TDF had greater signs of damage to the villi, the tiny fingers that line the surface of the intestines. The study found a trend towards lower levels of some vitamins in people taking TDF, suggesting malabsorption, but the investigators say more research is needed to determine if the damage to villi affects absorption of nutrients or correlates with lower levels of lipids or body fat. (Kauppinen)