Dolutegravir-based first-line non-inferior to efavirenz-based ART but associated with substantial weight gain: results from the ADVANCE study

Polly Clayden, HIV i-Base

Dolutegravir (DTG)-based ART was non-inferior to efavirenz (EFV)-based treatment in a South African study comparing three first-line regimens. [1, 2] Participants receiving DTG experienced significant rises in weight and this was more pronounced among those also receiving tenofovir alafenamide (TAF). [3]

Week 48 safety and efficacy data from the ADVANCE study were presented at IAS 2019, alongside a simultaneous publication in the New England Journal of Medicine (NEJM). A pooled analysis with the NAMSAL trial, looking at body weight, was also shown at the conference.


ADVANCE is a 96-week, open-label, randomised trial conducted by the newly-hatched Ezintsha – a sub-syndicate of the Wits Reproductive Health and HIV Institute (Wits RHI) – and colleagues.

The study is currently underway in Johannesburg and compares TAF/emtricitabine (FTC)/DTG, tenofovir disoproxil fumarate (TDF)/FTC/DTG and TDF/FTC/EFV.

Participants aged 12 years and above, with no previous ART for more than 30 days, creatinine clearance greater than 60 mL/min (80 mL/min if less than 19 years) and viral load greater than 500 copies/mL were included. Pregnancy and tuberculosis (TB) were exclusion criteria.

Participants did not have baseline genotyping, in accordance with South African treatment guidelines. Background NNRTI resistance in South Africa is estimated to be 5–15%.

The primary endpoint was week 48 viral load less than 50 copies/mL, discontinuation or missing data (Intent-to-treat population, non-inferiority margin -10%, significance level p=0.017, adjusted for multiple comparisons).

Professor Francois Venter from Ezintsha presented the findings on behalf of the ADVANCE investigators.

A total of 1053 participants (351 per arm) were randomised between February 2017 and May 2018: 99% black, 60% women, mean age at baseline 32 years and CD4 336 cells/mm3. Approximately 20% had viral load over 100,000 copies/mL. Baseline median Body Mass Index (BMI) was 24.1 kg/m2.

The percentage of participants with viral load <50 copies/mL at week 48 were: 83.8% for TAF/FTC/DTG, 84.9% for TDF/FTC/DTG and 78.6% for TDF/FTC/EFV. In the on-treatment analysis the respective percentages were: 96% for TAF/FTC/DTG, 94% for TDF/FTC/DTG and 95% for TDF/FTC/EFV.

Professor Venter noted that social issues played a greater role in participants achieving virological suppression than regimen, particularly age and employment. Rates for young and unemployed people were about 60% compared to older employed people who achieved almost 100% suppression.

The study confirmed non-inferiority for both DTG arms versus the EFV arm.

The majority of participants with viral load >50 copies/mL re-suppressed after adherence counselling. Only 2/18 in the TAF/FTC/DTG arm, 3/19 for TDF/FTC/DTG and 7/16 for TDF/FTC/EFV did not re-suppress. There was no DTG-emergent resistance.

Adverse events and laboratory abnormalities were similar between treatment arms with the notable exception of weight gain.

Weight gain

In the 96 week NAMSAL trial, 613 treatment naive participants in Cameroun were randomised to TDF/3TC/DTG or TDF/3TC/EFV 400 mg. Body weight was measured at baseline and week 48.

In ADVANCE, body weight was measured at baseline and every 12 weeks. DEXA scans evaluated limb and trunk fat at baseline, weeks 48 and 96.

Dr Michelle Moorhouse from Ezintsha showed data from a pooled analysis of the two trials: changes in body weight and BMI were compared between arms. Changes in trunk fat were also compared between arms in ADVANCE.

Participants in NAMSAL were 66% women, with a median age of 37 years. At baseline, median BMI was 23 kg/m2, CD4 approximately 280 cells/mm2, 67% had viral load above 100,000 copies/mL and 30% above 500,000 copies/mL.

This analysis revealed that in NAMSAL mean weight rose by a mean of 5 kg (BMI 1.7 kg/m2) for TDF/3TC/DTG vs 3 kg (BMI 1.2 kg/m2) for TDF/3TC/EFV (both p<0.001). Treatment-emergent clinical obesity (BMI >30 kg/m2) was seen in 12% of participants on TDF/3TC/DTG vs 5% on TDF/3TC/EFV (p<0.01).

At week 48, 44% of women in the TDF/3TC/DTG arm had 10% or more change from baseline weight vs 34% in the TDF/3TC/EFV arm (p<0.05). For men, the difference was non-significant and the percentage with 10% or more change was about 20%.

The difference in treatment-emergent obesity between arms was greater in men than women. In the TDF/3TC/DTG vs TDF/3TC/EFV respectively: 14% vs 2% (p<0.01) compared with 12% vs 7% (NS).

ADVANCE bodyweight changes overall are shown in Table 1. There were highly significant differences in weight change between arms (p<0.001). Treatment-emergent clinical obesity was higher in the TAF/FTC/DTG arm than other two arms (p<0.01).

Table 1: ADVANCE changes in body weight

Mean change in weight
Week 48 +6 kg +3 kg +1 kg
Week 96 +8 kg +5 kg +2 kg
Treatment-emergent overweight
Week 48 23% 14% 9%
Week 96 25% 13% 11%
Treatment-emergent obese
Week 48 14% 7% 6%
Week 96 19% 8% 4%

At the time of analysis, not all participants had reached week 96.

In men, there were greater mean increases in weight in the two DTG arms compared to the EFV arm and these were higher in the TAF arm than the TDF-containing arms. Up to week 48, changes in weight were progressive in the DTG arms, after which the rate of increase appeared to plateau.

DEXA scanning, found changes in trunk and limb fat to be higher with DTG-based treatment: +5.4 kg, +4.3 kg and + 0.5 kg, in the TAF/FTC/DTG, TDF/FTC/DTG and TDF/FTC/EFV arms respectively at 96 weeks.

Dr Moorehouse noted that most of the weight gain in the DTG arms is fat gain (both trunk and limb). In the EFV arm, although both limb and trunk fat increased at weeks 48 and 96, at week 96 there was a loss of lean trunk mass.

As weight changes were slightly lower with DEXA, she also explained that participants were weighed at each visit and, although the intention was to perform DEXAs on everyone, this was not always possible as the very obese might have been too big for the scanner so the DXA data might underestimate mean weight change compared to the main analysis.

Similarly, for women, there were greater percentage weight changes in the two DTG arms compared to the EFV one. And this was more pronounced in the TAF than the TDF-containing arm. Women receiving TAF/FTC/DTGexperienced an average 16% weight increase over 96 weeks.

But, unlike men, the women’s percentage weight change increases were progressive and linear in the DTG arms up to week 96 and do not appear to be reaching a plateau.

DEXA scans, also found changes in trunk and limb fat to be higher with DTG-based treatment: +9.2 kg, +5.4 kg and +2.8 kg, in the TAF/FTC/DTG, TDF/FTC/DTG and TDF/FTC/EFV arms respectively at 96 weeks.

In multivariate analysis (adjusted for socio-demographics, baseline factors, disease history and adverse events and concomitant medications), TAF/FTC/DTG, baseline CD4 count, baseline viral load, and baseline BMI were predictive of treatment-emergent obesity. Excluding baseline BMI, female sex, South African nationality, and employment were also significant.

TAF/FTC/DTG, baseline CD4 count, baseline viral load, female sex, age, and baseline weight were predictive of a 10% or more increase in body weight.


ADVANCE is ongoing, further analyses are underway as all participants reach 96 weeks and the study will continue to monitor people to see what happens in the long-term. Ideally the plan is to continue follow up for at least another two years (week 192). The study team are actively pursuing funding to do this.

The huge strength of these data is the inclusion of 60% women (and almost all black African participants) in an low- and middle-income setting. In ADVANCE, there are also DEXA scans, which, as Professor Venter pointed out, were in place to look at bone density in the three treatment arms but turned out to be very useful for the weight changes analysis.  

Regular HTB readers will have heard this over and over again, but once again, these findings are a rallying cry for greater diversity in clinical trials rather than the typical 80–85% white men (still) seen in registrational trials for new drugs. Having data across populations that are treated is a vast improvement on guessing things will be OK as antiretrovirals are rolled out to millions of people.

Polly Clayden is on the scientific committee of ADVANCE and co-author of the NEJM paper.


  1. Venter WDF et al. The ADVANCE trial: Phase 3, randomised comparison of TAF/FTC/DTG, TDF/FTC/DTG or TDF/FTC/EFV for first-line treatment of HIV-1 infection.10th IAS Conference on HIV Science. Mexico City, Mexico. 21–24 July 2019. Oral abstract WEAB0405LB.
  2. Venter WDF et al. Dolutegravir plus two different prodrugs of tenofovir to treat HIV. New England Journal of Medicine. Online ahead of print. 24 July 2019.
  3. Hill A et al. Progressive rises in weight and clinical obesity for TAF/FTC/DTG and TDF/FTC/DTG versus TDF/FTC/EFV: ADVANCE and NAMSAL trials. 10th IAS Conference on HIV Science. Mexico City, Mexico. 21–24 July 2019. Oral abstract MOAX0102LB.

Links to other websites are current at date of posting but not maintained.