Late 2017 HIV Conferences Focus on Antiretrovirals and Comorbidities

Bictegravir is an unboosted integrase inhibitor that is being coformulated as a once-daily agent with emtricitabine (Emtriva, FTC) and tenofovir alafenamide (TAF) in a single pill called B/F/TAF. It has a high in vitro barrier to resistance and a low potential for drug-drug interactions.

A Phase 3 open-label trial randomized HIV-positive people who had a viral load below 50 copies/mL while taking a regimen consisting of a boosted protease inhibitor (either darunavir [Prezista] or atazanavir [Reyataz]) plus two NRTIs to either 1) continue that regimen or 2) switch to B/F/TAF. Participants were then monitored for 48 weeks.

A Week 48 snapshot analysis of the 577 participants presented at IDWeek 2017 determined that 1.7% in each study arm had a viral load above 50 copies/mL (the primary endpoint). That result was enough to establish the noninferiority of switching from a suppressive protease inhibitor-based regimen to B/F/TAF (at a noninferiority margin of 4%).

The proportions of participants in the B/F/TAF arm and the control arm with a 48-week viral load below 50 copies/mL were 92.1% and 88.9%, respectively.

Drug-resistant HIV did not emerge in anyone who switched to B/F/TAF. Two participants discontinued B/F/TAF for adverse events, compared to one person on the control regimen.

Participants who switched to B/F/TAF had statistically significant improvements in triglycerides and total-to-high-density lipoprotein cholesterol ratio.

B/F/TAF appears to be an effective and well-tolerated switch option for patients taking a suppressive protease inhibitor-based regimen, the researchers concluded.

Fostemsavir is a novel HIV attachment inhibitor without in vitro cross-resistance to drugs in other antiretroviral classes.

In the U.S., Europe, and South America, the ongoing phase 3 BRIGHTE trial recruited participants taking a failing regimen who, although they had one or two antiretroviral classes they had not yet tried, were unable to construct a viable regimen with available agents.

Researchers randomized participants 3:1 to add either fostemsavir (600 mg twice daily) or placebo to their failing regimen for eight days. Then both groups received open-label fostemsavir plus an optimized background regimen (OBR). Meanwhile, a separate, nonrandomized set of patients with zero remaining fully active antiretrovirals who were taking a failing regimen received open-label fostemsavir plus OBR.

The results were presented at the European AIDS Conference. At Day 8 in the randomized group, fostemsavir proved superior to placebo, with a mean viral load drop of 0.79 log10 copies/mL versus 0.17 log10 copies/mL in an intention-to-treat-exposed analysis (difference -0.625; 95% confidence interval -0.810 to -0.441; P < 0.0001).

A Week 24 snapshot analysis determined that 54% of those taking fostemsavir in the randomized cohort had a viral load below 40 copies/mL, while 77% reached a viral load below 400 copies/mL. In the nonrandomized group, 36% had a viral load below 40 copies/mL at week 24, while 45% had a load below 400 copies/mL.

Through 24 weeks, CD4 count rose by an average 90 cells/mm3 in the randomized group and 41 cells/mm3 in the nonrandomized group.

Treatment-related serious adverse events affected 2% in the randomized group and 3% in the nonrandomized group. The proportion of participants experiencing an adverse event that led to study discontinuation were 4% and 9%, respectively. Seventeen people (5%) taking fostemsavir died during the study; 12 deaths were attributed to AIDS events, including immune reconstitution inflammatory syndrome.

BRIGHTE investigators believe these results support continued development of fostemsavir as a treatment option for patients with heavy antiretroviral experience.

A single tablet combining the cobicistat-boosted protease inhibitor darunavir with emtricitabine/tenofovir alafenamide -- a tablet currently referred to as D/C/F/TAF -- is licensed in the European Union under the brand name Symtuza and is under regulatory review in the U.S.

The Phase 3 open-label European/U.S. EMERALD trial recruited patients taking a suppressive regimen that consisted of a boosted protease inhibitor plus emtricitabine (FTC, Emtriva) and tenofovir disoproxil fumarate (TDF, Viread). Participants were randomized 2:1 to either (1) switch to once-daily D/C/F/TAF or (2) continue their current combination. The results were presented at IDWeek.

In the Week 48 primary endpoint analysis, 2.5% of the 763 people randomized to D/C/F/TAF and 2.1% of the 378 people randomized to continue their regimen had a confirmed virologic rebound. That was close enough to establish the noninferiority of D/C/F/TAF at a 4% noninferiority margin (P < 0.001).

Snapshot analysis determined that 94.9% of those assigned to D/C/F/TAF and 93.7% of those assigned to continue their regimen had a viral load below 50 copies/mL at Week 48.

No one stopped their regimen for virologic nonresponse. Among viral rebounders, 12 of 19 taking D/C/F/TAF and four of eight continuing their initial regimen returned to a viral load below 50 copies/mL by Week 48 without changing antiretrovirals.

There were three rebounders in the control arm and one in the D/C/F/TAF arm. No detectable darunavir, primary protease inhibitor, tenofovir, or emtricitabine resistance mutations emerged in any rebounder.

Low proportions of people in the D/C/F/TAF arm and the control arm had one or more serious adverse events (4.6% and 4.8%, respectively) or adverse events leading to discontinuation (1.4% and 1.3%, respectively).

EMERALD investigators stated, "D/C/F/TAF combines the safety advantages of TAF, with the known efficacy and higher genetic barrier to resistance of darunavir, in a single-tablet regimen."

D/C/F/TAF is not only being explored as an HIV treatment switch option; it is also under study for first-line therapy. AMBER, a Phase 3 double-blind European/U.S. trial, randomized antiretroviral-naive patients to receive either D/C/F/TAF or a regimen consisting of nearly the same drugs, but with TDF instead of TAF. All participants had genotypic susceptibility to darunavir, emtricitabine, and tenofovir.

A Week 48 snapshot analysis presented at the European AIDS Conference determined that 91.4% of people randomized to D/C/F/TAF and 88.4% of people randomized to the control combination had a viral load below 50 copies/mL. The result established the noninferiority of D/C/F/TAF to the control regimen at a noninferiority margin of 10% (P < 0.0001).

The proportions of participants with a viral load above 50 copies/mL at Week 48 were 4.4% in the D/C/F/TAF group and 3.3% in the control group. One participant (0.3%) in the D/C/F/TAF arm and nobody in the control arm stopped treatment for lack of efficacy. Among patients genotyped at failure, no one in either arm had darunavir or primary protease inhibitor resistance mutations.

Eight patients (2.2%) randomized to D/C/F/TAF and 16 (4.4%) randomized to the control regimen discontinued treatment because of an adverse event. Grade 3 or 4 adverse events affected 5.2% in the D/C/F/TAF arm and 6.1% in the control arm. No Grade 3 or 4 lab abnormalities crossed the 5% incidence threshold in either arm.

No one stopped treatment because of bone, renal, or central nervous system adverse events. Bone, renal, and lipid measures proved consistent with previous research on TAF and cobicistat. AMBER investigators concluded that D/C/F/TAF is an effective and safe regimen for people with previously untreated HIV infection.

In 2015 the FDA revised its noninferiority margin from 12% to 4% for antiretroviral switch trials. Researchers in Spain conducted an individual patient data meta-analysis of randomized controlled trials to determine whether dual therapy with a protease inhibitor plus lamivudine (PI/3TC) met that new standard when compared to continuing a protease inhibitor plus two NRTIs.

A systematic search of electronic databases (1990-March 2017) and major HIV meetings (2014-March 2017) yielded four trials testing a switch to PI/3TC versus continued protease inhibitor-based triple therapy. Results from the analysis were presented at the European AIDS Conference.

Together the four trials (SALT, ATLAS, DUAL, and OLE) included 525 patients randomized to PI/3TC dual therapy and 526 randomized to continue triple-drug therapy. Across all four trials, about 75% of participants were men, 22% had HCV infection, and median baseline time with a viral load below 50 copies/mL stood at about 30 months.

At Week 48, a pooled 4% of patients on dual therapy and 3% on triple therapy had a viral load >50 copies/mL. The 95% confidence interval (CI) for the difference fell within the 4% noninferiority margin, confirming the noninferiority of switching to PI/3TC.

A secondary analysis found that the rates of people with a viral load below copies/mL at Week 48 were 84.7% among patients who switched to dual therapy and 83.2% among those who continued triple therapy, a result that only confirmed the noninferiority of switching to PI/3TC using the older, 12% noninferiority margin (difference 1.47%; 95% CI -2.9% to 5.8%).

In both analyses, the difference in virologic response between switching to dual therapy and continuing triple therapy was not affected by sex, HCV status, or whether the specific protease inhibitor used was atazanavir, darunavir, or lopinavir.

Total cholesterol, low-density lipoprotein cholesterol, and triglycerides all rose in the dual-therapy group while falling among those who continued triple therapy, and between-group differences were statistically significant. Glomerular filtration rate rose (i.e., improved) with dual therapy and fell with continued triple therapy, also a significant difference (3.32 mL/min vs. -1.89 mL/min).

Clinicians pondering the potential merits of switching to PI/3TC must consider the relative convenience and resistance barrier of a two- or three-pill boosted-PI/3TC regimen versus available one-pill multi-drug combinations, the researchers suggested.

ABX464, a novel HIV drug called a Rev inhibitor that modulates RNA splicing, has been found to induce dose-dependent drops in viral load when given alone to antiretroviral-naive patients.

A randomized, double-blind, placebo-controlled trial spanning Belgium, France, and Spain aimed to test the efficacy and safety of ABX464 when added to darunavir monotherapy for 28 days.

To gauge the impact of ABX464 on HIV reservoirs, researchers measured total and integrated HIV DNA in peripheral blood mononuclear cells (PBMCs) collected on days 0 and 28.

The trial randomized 30 participants: 22 added ABX464 (50 or 150 mg daily) to boosted darunavir, while the remaining eight added a matched-dose placebo. All participants had a viral load below 50 copies/mL on darunavir monotherapy. All treatment was then stopped after 28 days, to be resumed when viral load rebounded above 1,000 copies/mL. Study results were presented at the European AIDS Conference.

A single Grade 3 adverse event, fatigue, affected a participant taking 150 mg of ABX464. One person stopped ABX464 because of Grade 1 epigastric pain and Grade 2 abdominal pain. Another participant stopped ABX464 because of viral rebound before Day 28.

Eight of 15 participants (53%) with evaluable HIV DNA were "responders," defined as experiencing a minimum HIV DNA reduction of 50 copies and a more than 25% decline in total HIV DNA copies.

Among evaluable participants, HIV DNA rose by a mean of 58 copies per million PBMCs in four placebo recipients, rose by a mean of 67 copies per million PBMCs in seven ABX464 nonresponders, and fell by a mean of 186 copies per million PBMCs (38%) in eight ABX464 responders. Integrated HIV DNA declined by a mean of 131 copies per million PBMCs (55%) in ABX464 responders and rose by a mean of 5 copies per million PBMCs in ABX464 nonresponders.

Time to viral rebound after all treatment stopped averaged 13 days with ABX464 and 14 days with placebo, a finding indicating that recorded declines in HIV DNA were not large enough to delay viral rebound.

The researchers wrote that they believe this is "the first time a signal has been observed demonstrating the reduction in HIV reservoirs in patients with a therapeutic candidate."

The use of integrase inhibitors often leads to rapid drops in HIV viral load and gains in CD4 cells. But those beneficial effects are also linked to the potential emergence of immune reconstitution inflammatory syndrome (IRIS), an excessive response against antigens of opportunistic infections.

Researchers working with the Netherlands ATHENA cohort conducted a study to determine whether first-line integrase inhibitor use is associated with emergence of IRIS, as defined by French and colleagues or by clinical diagnosis. Phase 3 integrase inhibitor trials are unable to answer this question because they included few late presenters, the researchers asserted.

The analysis focused on ATHENA cohort members at high risk for IRIS who either: (1) started antiretroviral therapy after March 2009 and had a pretreatment CD4 count below 200 cells/mm3, (2) developed an opportunistic infection before or after starting HIV treatment, (3) used corticosteroids within 12 months after starting antiretroviral therapy, or (4) died within 12 months after starting therapy. Results were presented at the European AIDS Conference.

The 155 patients who began an integrase inhibitor and the 517 who began other antiretrovirals had the same median pretreatment viral load (5.5 log10 copies/mL) and similarly low median baseline CD4 counts (39 vs. 33 cells/mm3, respectively).

Adjusted logistic regression determined that starting an integrase inhibitor nearly doubled the odds of IRIS by the French criteria (odds ratio [OR] 1.95, 95% confidence interval [CI] 1.14 to 3.34) or combined French and clinical criteria (OR 1.87, 95% CI 1.22 to 2.86), independently of risk factors such as CD4 count, viral load, prior opportunistic infections, or corticosteroid use before IRIS diagnosis.

Cox regression analysis determined that raltegravir use tripled IRIS risk (hazard ratio 3.2, 95% CI 2.0 to 5.0), but use of dolutegravir or elvitegravir was not associated with IRIS risk.

The ATHENA team concluded that "use of integrase inhibitors is a risk factor for development of IRIS in HIV-1 late presenters," but this association was limited to patients starting raltegravir in their analysis.

Cardiovascular disease risk runs higher in women with HIV than in other women, especially as they lose the cardioprotective effect of estrogen. That said, little is known about HIV-related cardiovascular risk in premenopausal women, leading researchers to conduct a cross-sectional national-sample analysis, the results of which were presented at IDWeek.

The investigators used National Health and Nutrition Examination Survey data sets from 1999 through 2014 to identify premenopausal women who had no history of myocardial infarction and for whom sufficient data were available to calculate their 10-year Framingham cardiovascular risk score. Among 9,639 eligible women, 25 (0.3%) were HIV positive, 9,610 were HIV negative, and HIV status could not be determined for four.

Women with HIV were marginally older than HIV-negative women (mean 39.4 versus 33.9 years, P = 0.147). The HIV-positive group included a higher proportion of blacks (84% versus 19.9%) and a lower proportion of whites (4% versus 41.9%) (P < 0.001). The two groups had similarly low proportions of women who currently smoked (0% with HIV vs. 8.9% without HIV, P = 0.242).

Among Framingham risk score variables other than age and smoking, only two differed meaningfully between women with versus without HIV: high-density lipoprotein cholesterol (mean 45.4 mg/dL with HIV vs. 57.4 mg/dL without HIV, P = 0.050) and antihypertensive use (20% of women with HIV versus 9.7% of women without HIV, P = 0.082).

Mean 10-year Framingham risk score was significantly higher in women with HIV than without HIV: 2.12 versus 0.95 (P = 0.010).

The clinicians treating these women were more likely to tell a woman with HIV to exercise than a woman without HIV (28% versus 13%, P = 0.027) or to modify their diet (28% versus 13.3%, P = 0.031). However, overall rates of exercise were similar between the two groups (24% with HIV and 19% without HIV, P = 0.52), as were rates of diet change (32% vs. 24.5%, P = 0.385).

Because this study appears to confirm that HIV itself is a cardiovascular risk factor in premenopausal women, the researchers emphasize the importance of modifying traditional risk factors like smoking, hypertension, diabetes, and dyslipidemia.

The investigators caution, however, that their study findings are limited by the low number of women with HIV.

Statins lower atherosclerotic cardiovascular disease (ASCVD) risk in the general population, and people with HIV have higher rates of ASCVD. But previous research found low statin use in HIV-positive people, at least according to older (Adult Treatment Panel III) guidelines.

To determine whether statin prescription practices in an HIV clinic population reflect updated 2013 guidelines from the American College of Cardiology (ACC) and American Heart Association (AHA), researchers at Washington University in St. Louis conducted a study, the results of which were presented at IDWeek.

The analysis involved 1,085 HIV-positive people at least 40 years old who visited the clinic during 2015.

The research team classified participants into four statin benefit groups as specified in ACC/AHA guidelines: (1) clinical ASCVD; (2) primary hyperlipidemia (LDL-C ≥ 190 mg/dL); (3) 40 to 75 years old with diabetes and LDL-C 70 to 189 mg/dL without ASCVD; or (4) 40 to 75 years old without ASCVD or diabetes, with LDL-C 70 to 189 mg/dL, and with a 10-year ASCVD risk of ≥ 7.5%.

The study group averaged 51.9 years in age; 71% of participants were men, 68% were black, and 98% were on antiretroviral therapy, with 86% having an undetectable viral load.

Researchers found that among 1,085 participants, 450 (41%) had a statin indication by ACC/AHA criteria. However, only 161 of those 450 people (36%) had a statin prescription, including 36% in group 1, 44% in group 2, 49% in group 3, and 29% in group 4.

Among participants with a statin prescription, the proportion of people prescribed high-dose statins were 30% in group 1, 75% in group 2, 29% in group 3, and 21% in group 4.

Compared with patients not prescribed statins, those prescribed a statin had a higher median CD4 count (630 versus 573 cells/mm3, P = 0.050) and a marginally higher rate of viral suppression (93% versus 88%, P = 0.073). In indication group 4, a significantly higher proportion of people with a statin prescription had viral suppression (95% versus 87%, P = 0.031).

Statin users did not differ significantly from nonusers in the following: proportion of people taking five or more antiretrovirals; median number of total medications; or proportion of people taking an HIV agent that may interact with statins (namely ritonavir [Norvir], cobicistat, or efavirenz [Sustiva]).

The researchers conclude that "two thirds of our patients were not prescribed statins despite a strong indication ... based on the new [ACC/AHA] guidelines." The investigators surmised that HIV providers may not routinely consult ACC/AHA guidelines.

Plentiful research has confirmed a high prevalence of low bone mineral density (BMD) in people living with HIV, hepatitis C (HCV), or HIV/HCV coinfection. Therefore, University of Texas and Veterans Affairs researchers conducted a cross-sectional study to assess the impact of dietary calcium on BMD in male veterans with HIV, HCV, HIV/HCV, or neither infection. The results were presented at IDWeek.

All men with HIV or HIV/HCV had achieved HIV suppression on antiretroviral therapy, while no one with HCV had received treatment for that virus.

The researchers used the Hertzler-Frary dietary calcium questionnaire to assess dietary calcium intake. They used DXA scans to measure BMD at the lumbar spine, femoral neck, and total hip.

The study included 141 men with HIV, 99 with HCV, 46 with HIV/HCV, and 150 with neither virus. Across the four groups, mean age ranged from 54 to 56. Calcium intake did not differ across the four groups (it ranged from 606 to 788 mg per day), but lay well below the recommended daily intake of 1,000 mg for men of this age.

The researchers found that high proportions of men in all four groups (including uninfected controls) had osteopenia (ranging from 49% to 65%) or osteoporosis (ranging from 11% to 15%). In the general population, 38% of men in the same age group have osteopenia and 4% have osteoporosis. Among the people studied, there was no significant association between type of infection and BMD.

In adjusted analyses, median calcium intake did not differ significantly between men with normal BMD versus men with osteoporosis either in the HIV group or the HIV/HCV group. But HCV-monoinfected participants with osteoporosis had significantly lower calcium intake than those with normal BMD (mean difference 409 mg, P = 0.027).

Foods significantly associated with higher BMD were milk, yogurt, spinach, and almonds. Thus, the researchers proposed that "consumption of milk and yogurt along with a healthy diet might protect bone health in these patient populations."