Thank you for your interest in CCO content. As a guest, please complete the following information fields. These data help ensure our continued delivery of impactful education.
Become a member (or login)? Member benefits include accreditation certificates, downloadable slides, and decision support tools.
Let’s move on to the metabolic complications of ART.
Before continuing, please take a moment to answer the following question.
Let’s start with the COCOMO study, an ongoing observational, longitudinal, cohort study evaluating the prevalence and incidence of non-AIDS comorbidities among PWH in Denmark.11 An investigation of HIV-specific and traditional risk factors for insulin resistance in PWH from the COCOMO study was presented at EACS 2021 (N = 651).12 The median Homeostasis Model Assessment of Insulin Resistance was 1.86 (IQR: 1.23-3.14) mmol/L x mIU/mL. The high levels of insulin resistance were defined by that upper quartile at greater than 3.14.
This investigation was also a fine dissection of prior exposure to thymidine analogs (TA) and didanosine (ddI), more specifically the interaction between these agents and insulin resistance risk factors. This is different from prior studies looking at the classifications for protease inhibitors, integrase inhibitors, nucleos(t)ide reverse-transcriptase inhibitor (NRTIs). Now we are looking more critically at effect modifiers within this ARV category.
The factors significantly associated with insulin resistance were age, abdominal obesity, BMI ≥25 kg/m2, duration of combination ART, and previous use of TA/ddI.12 We know that as people get older, they can develop insulin resistance.
In this analysis, BMI >30 kg/m2 was the overwhelming risk factor for developing insulin resistance.12 These results are not surprising and are seen throughout the general population as well. Abdominal obesity created almost 5-fold increased risk. A finding of interest, especially for those with prior TA or ddI use, is the 88% increase in odds of developing insulin resistance. We see here that with the effect modifiers, particularly among participants with a background history of use of TA and ddI, it really accentuated that risk for those with either abdominal obesity or aging, even if not quite statistically significant.
We also see that for those patients who are aging through decades of treatment—many patients are living with HIV for 30 or 40 years since the genesis of the epidemic—prior TA and ddI, in addition to their known mitochondrial toxic properties, can be a factor in developing insulin resistance.
Data from studies 1489 and 1490, which were the licensing trials for bictegravir (BIC)/FTC/TAF, were updated at IDWeek 2021. These 2 randomized, double-blind, active-controlled, phase III trials evaluated BIC/FTC/TAF in treatment-naive PWH.13 The active controls were single-tablet regimen DTG/ABC/lamivudine (3TC) for Study 1489 and DTG plus FTC/TAF for Study 1490, so this study is able to isolate just the integrase inhibitors.
The objective of these studies is to get an overview of treatment-emergent metabolic changes with contemporary integrase-based regimens. The investigators compared metabolic parameters—specifically treatment-emergent diabetes, hypertension, BMI changes, and fasting glucose—more than 3 years in participants randomized to BIC/FTC/TAF, to DTG/ABC/3TC, and to DTG plus FTC/TAF.
The results showed no significant differences between arms at Week 144 for treatment-emergent diabetes, treatment-emergent hypertension, changes in BMI, or fasting glucose.13
Fasting lipid changes were small, and few participants initiated lipid-lowering therapy.
BMI increased overall in all groups but did not differ by treatment regimen in either study, and in study 1489 there was a greater short-term increase in BMI with BIC/FTC/TAF vs DTG/ABC/3TC through 96 weeks but no difference at Week 144. In Study 1490, there were no significant differences in BMI at any timepoint for either arm, including subgroup analyses by race and sex at birth.
Proportions of people with normal initial BMI dropped 7% to 15% through 144 weeks across treatment arms in both studies. Each trial saw a corresponding jump of 9% to 15% in the proportions of participants with overweight, obesity, and morbid obesity.
So again, isolating out TAF and comparing the integrase inhibitors DTG and BIC in Study 1490, we see there was not much difference in weight gain. The caveat is that the trial population was 90% male and had a much younger median age (33 years) than we would expect to enroll in the United States, where the HIV population is aging, and more than 50% of PWH are older than 50 years of age.
This 2-trial analysis offers reassuring data on metabolic changes through 144 weeks of treatment. It also confirms the BMI gains in the first years of treatment that we see with common, integrase inhibitor-based regimens that we are starting in our patients, but the BIC/FTC/TAF regimen does not seem to follow out to that degree of incline in the 2- to 3-year period.
The TANGO study is an ongoing, phase III non-inferiority study evaluating the efficacy and safety of switching from a stable 3- or 4-drug TAF-based regimen to DTG/3TC.14 PWH receiving a stable TAF-based regimen were randomized 1:1 to either continue their TAF-based ART or switch to the 2-drug regimen of DTG/3TC. Secondary and exploratory endpoints at Week 144 included changes from baseline in weight, metabolic biomarkers, and lipids.
These were illuminating data. The changes in weight were similar in the 2 groups, with a mean 2.2-kg increase in the group that switched to DTG/3TC vs a mean 1.7-kg increase in the group that continued TAF-based ART.15 This is comparable to what we would expect in the general population, with a weight gain of approximately 0.5-1.0 kg/year.16
The proportion of weight increases of 10% or more was similar in the DTG/3TC and TAF-based regimen groups (13% and 12%, respectively), suggesting that TAF does not play a significant role in weight gain for patients on stable ART.
The authors noted that in the subgroup of participants with obesity at baseline, a higher proportion of those on TAF-based ART vs DTG/3TC were referred for weight management interventions.
Changes in fasting lipids favored the DTG/3TC group, whereas changes in other metabolic health parameters were similar between groups.17 This parallels numerous studies showing lipid elevations in TAF with no significant change in the total cholesterol–to–high density lipoprotein (HDL) ratio,18-20 which may prove to be a useful index for cardiovascular risk.
DRIVE-FORWARD is a multicenter, randomized, double-blind phase III study comparing treatment with 2 NRTIs and either DOR or boosted DRV in treatment-naive PWH.21 At Week 48, the DOR-based regimen was noninferior to DRV-based therapy.21 High virologic efficacy of DOR plus NRTIs was maintained through 96 weeks with a higher Week-96 rate of virologic suppression vs DRV/RTV, as defined by HIV-1 RNA <50 copies/mL by FDA Snapshot analysis (treatment difference: 7.1%; 95% CI: 0.5%-13.7%).22
Patients who completed the double-blind phase of the study and met the inclusion criteria could receive another 96 weeks of DOR plus 2 NRTIs in an open-label extension.23 Virologic suppression was maintained through Week 192 in 81.1% of patients who continued DOR-based ART and 80.7% of those who switched from DRV/RTV to DOR-based ART.
There were minimal increases in fasting lipids during the 192-week study in participants who continued DOR plus 2 NRTIs.23 For those who switched from DRV/RTV to DOR after Week 96, substantial reductions in all other lipid parameters were observed without significant changes in HDL cholesterol, so DOR exhibited a favorable safety profile in comparison with DRV.
For patients who continued DOR plus 2 NTRIs, there was a minimal median weight gain of 1 kg after Week 96 and a modest increase (median: 1.9 kg) from baseline through Week 192.
For the patients who switched from DRV to DOR after 96 weeks, there was a median 1.5-kg weight increase after Week 96, which appears to affirm the comparatively greater weight gain in the base study with DOR vs DRV/RTV (1.8 kg vs 0.7 kg, respectively). These findings suggest that DOR is not neutral when it comes to weight gain while on ART for HIV.
The next study, from the university hospital in Bonn, Germany and presented at EACS 2021, evaluated the influence of integrase inhibitors on hepatic steatosis in PWH.24 Nonalcoholic fatty liver disease (NAFLD) has emerged as a growing threat to liver health both in PWH and uninfected individuals. Obesity and metabolic syndrome are known risk factors for NAFLD, so investigators for this study examined a potential further association between NAFLD and INSTIs in PWH.
The study included adult PWH receiving INSTIs between 2013 and 2020.24 Participants were assessed for hepatic steatosis at baseline and annually.
Controlled attenuation parameter (CAP) was measured annually via transient elastography and compared with a baseline mean CAP value of 257.5 ± 61.4 dB/m, below the cutoff of 260 for any grade of significant steatosis.24
Comparing mean CAP values at the start and end of the exposure period, there was a nonsignificant increase in CAP values indicating a trend toward higher steatosis for those on elvitegravir. Only 4.5% of participants were receiving elvitegravir, so the number of participants is small.
Mean CAP value decreased among participants on BIC, but again, this representing a small proportion of patients within the cohort (13.6%, or 15 patients).
Looking in more granular detail, none of the 15 BIC patients without steatosis at baseline developed steatosis during the study period, as compared with 36% of those receiving DTG, 33% receiving elvitegravir, and 66% receiving RAL who developed steatosis. Owing to the limited sample size, these differences did not reach statistical significance.
Clinically significant weight gain was observed in patients who received DTG and RAL (DTG: 77.5 ± 11.3 kg vs 80.3 ± 11.8 kg, P = .005; RAL: 83.5 ± 12.6 kg vs. 87.9 ± 16.6 kg, P = .007), but it was not associated with any further significant increase in steatosis in patients receiving DTG or RAL.
Again, the smaller patient numbers must be acknowledged, but the results show that there is a lot of variability in many of these studies when it comes to whether DTG is influencing weight gain and whether RAL is influencing weight gain to a lesser degree.
In the NA-ACCORD study that I mentioned previously, RAL was associated with emergent diabetes, but here there is no association in those who gained weight on RAL or DTG with significant increases in steatosis that one may consider within the metabolic syndrome continuum. You can see how we are building a picture as the data roll in, but it is not yet a clear picture. It will take larger numbers. The authors did conclude that larger studies need to be performed to see if the favorable effects of BIC vs the impact of these other INSTIs holds up the development of steatosis within the context of larger, adjusted multivariable analyses in larger cohorts.
Previously, I mentioned that the ADVANCE trial was one of the studies that really brought attention to the weight-gain phenomenon, particularly in women as well as with integrase inhibitors and then TAF.
Looking back at CROI 2020, the ADVANCE study was conducted in South Africa and was an open-label, phase III study that compared 3 regimens in treatment-naive patients.25,26 Patients were equally randomized to either DTG plus FTC/TAF or DTG plus FTC/TDF or combined NNRTI EFV/FTC/TDF.
What emerged and what caught people’s interest was the weight changes. Body weight was measured at baseline and every 12 weeks, and then dual-energy x-ray absorptiometry (DXA) scans were used to evaluate limb and trunk fat at baseline and at Weeks 48 and 96.25
The 96-week data jumpstarted some of our conversations about what contributes to this weight gain in our patients, and that the weight gain results were notable, particularly for women receiving DTG plus FTC/TAF, and this contributed fuel to the debate of the integrase inhibitors and TAF as contributory factors for weight gain.
There were progressive linear rises in body weight at Week 96 for women treated with DTG plus FTC/TAF and DTG plus FTC//TDF, and in men, mean body weight rose in the DTG arms at Week 48 but then stabilized to Week 96.
In addition, DTG plus FTC/TAF was associated with significantly higher risk of rises in visceral and subcutaneous adipose tissue as measured by DXA, as well as treatment-emergent obesity and treatment-emergent metabolic syndrome, recognized as the International Diabetes Federation definition as central obesity (BMI >30 kg/m2), and any 2 of the following conditions: increased triglycerides, reduced HDL cholesterol, increased blood pressure, or increased fasting glucose.25 The increase in prevalence of metabolic syndrome from baseline to Week 96 in the DTG plus FTC/TAF arm was statistically significant compared with EFV/FTC/TDF.
The ADVANCE investigators also reported data on bone mineral density (BMD) and renal function in HIV-positive patients receiving TDF- vs TAF-containing regimens.27 Because TDF has been associated with decreases in both BMD and renal function in prior studies, this 192-week open-label randomized trial compared BMD scores obtained on DTG plus FTC/TAF, DTG plus FTC/TDF, or EFV/FTC/TDF.
Significant baseline to Week 48 decreases in hip and whole-body BMD were observed in all groups, and BMD stabilized at Week 48. Greater decreases in hip and whole-body BMD were seen in the TDF groups as compared with the TAF group. Similarly, the decrease in spine BMD was greater in the TDF groups than in the TAF group. However, all 3 arms showed recovery in spine BMD from Week 96 to Week 144, with the TAF group returning to near-baseline levels compared with the TDF group. This is an interesting and favorable finding, and one of the reasons why TAF now has superseded TDF for both renal and bone safety.
The investigators also evaluated changes in hip BMD by weight quartiles at Week 144 and found that being in the highest weight quartile was associated with less reduction in hip BMD. So weight gain appears to confer some protection against BMD loss. I’m not going to call it a silver lining for weight gain, but it is consistent with what we see in the general population: less reduction in BMD in those experiencing more weight gain.