Three commonly used first-line HIV medications -- boosted atazanavir (Reyataz), boosted darunavir (Prezista) and raltegravir (Isentress) -- all deserve their place among the list of preferred drugs in official U.S. treatment guidelines, according to results from a major head-to-head study presented at CROI 2014. But relative to one another, atazanavir was not as well tolerated, and raltegravir had the best combined efficacy and tolerability profile, the study found.
It is an increasingly rare sight in HIV research: large, well-constructed studies directly comparing HIV antiretrovirals that are already in widespread use. (And it's likely to become even rarer, given recent changes to federally funded HIV research in the U.S.) But the AIDS Clinical Trials Group study known as A5257 was designed to do precisely that.
Among the first-line regimens currently recommended within HIV treatment guidelines published by the U.S. Department of Health and Human Services (HHS), three are:
tenofovir/emtricitabine (Truvada) plus ritonavir (Norvir)-boosted atazanavir, a protease inhibitor
tenofovir/emtricitabine plus ritonavir-boosted darunavir, a protease inhibitor
tenofovir/emtricitabine plus raltegravir, an integrase inhibitor
Notably missing from this list, of course, is any regimen containing efavirenz (Sustiva, Stocrin), a key component of the still-tremendously popular fixed-dose combination drug Atripla. But with efavirenz a less-than-optimal drug for many patients -- including those with psychiatric concerns, NNRTI resistance or pregnancy plans -- the relative viability of other first-line options has been a key question.
Raphael Landovitz, M.D., and colleagues crafted A5257 to answer this question. Their hypothesis entering the 96-week study was that all three drugs would perform similarly in terms of efficacy and tolerability. As the data he presented at CROI 2014 show, that hypothesis proved only partially true.
A5257 Study Construction
A5257 enrolled treatment-naive adults with HIV from throughout the U.S., each with a viral load of 1,000 copies/mL or higher. They were randomized 1:1:1 to receive one of the three regimens listed above and monitored for at least 96 weeks. Of the 1,809 eligible participants who began the study, about 92% were still in it after 96 weeks.
Baseline characteristics were similar across the three study arms. Average age was 37; roughly 24% of volunteers were female (a solid percentage for an HIV study). This was also a racially diverse cohort: 34% was white, 42% was black and 22% was Hispanic.
This was not precisely the healthiest group of volunteers, immunologically speaking -- which may, in a way, make it more reflective of the actual HIV-positive population. Median CD4 count at baseline was only 308; 30% of volunteers had a CD4 count below 200, and 30% had a viral load of 100,000 or higher (7% were over 500,000).
Assessing Failures on Atazanavir, Darunavir and Raltegravir
As with many head-to-head trials of this ilk, comparisons focused on the cynical realities: the odds of treatment failure. A5257's endpoints were the time it took for patients to experience virologic failure (which was primarily defined as having a viral load over 200 after 24 weeks of therapy) or tolerability failure (defined as discontinuing the study drug because of toxicity). Since there was a risk that some patients might experience both unfortunate endpoints, the researchers also had a "composite endpoint" that counted whichever of the two occurred first.
Overall, all three regimens were quite successful; here are the failure percentages for each arm after 96 weeks:
| Study Drug | Virologic Failure % | Tolerability Failure % |
| Atazanavir (boosted) | 13% | 14% |
| Darunavir (boosted) | 15% | 5% |
| Raltegravir | 10% | 1% |
A few things on the data:
All three arms had overlapping confidence intervals for virologic failure, meaning that none was significantly better than another. Although it may appear that raltegravir trended better (especially compared to atazanavir), in the context of this study, the drugs were considered equivalent virologically.
In terms of tolerability, atazanavir was a hands-down loser, with the additional failures caused almost entirely by jaundice/hyperbilirubinemia (47 discontinuations, usually grade 3 or 4) and gastrointestinal complications (25 discontinuations, usually grade 3).
Darunavir and raltegravir were considered to have equivalent tolerability within the context of this study.
Darunavir tolerability failures were primarily due to gastrointestinal problems, but occurred at a less frequent rate (14 discontinuations) relative to atazanavir.
The composite endpoint analysis yielded a clear sequence of superiority: raltegravir, then boosted darunavir, then boosted atazanavir. This was driven largely by the tolerability differences.
An additional analysis of patients achieving an undetectable viral load (less than 50 copies/mL) yielded excellent success rates across the board, with raltegravir again appearing to trend a little better. At 96 weeks, 88% of atazanavir patients, 89% of darunavir patients and 94% of raltegravir patients had an undetectable viral load in an intent-to-treat analysis that did not take into account antiretroviral therapy (ART) changes. (A unique feature to this study was that patients were permitted to change to other regimens due to toxicity, while still remaining in the study.)
By contrast, in an "off-ART = failure" snapshot analysis that considered patients who switched regimens to be failures, however, differences emerged; raltegravir was noticeably best (80% success at 96 weeks), followed by darunavir (73%) and atazanavir (63%).
Another analysis focusing on CD4 change from baseline found nearly identical benefits at week 96 for atazanavir (+284 cells/mm³) and raltegravir (+288 cells/mm³), with darunavir slipping slightly behind (256 cells/mm³). "These modest differences do achieve statistical significance compared to the other two arms," Landovitz said in his presentation, but "their clinical relevance is questionable."
As expected, raltegravir had a gentler lipid profile than the two protease inhibitors, with patients on atazanavir or darunavir experiencing a statistically significant level of LDL cholesterol and triglyceride increases relative to patients on raltegravir. Landovitz noted that additional analyses are being done to explore a wide range of other variables, including the drugs' relative effects on cardiovascular risk, inflammatory biomarkers and adherence.
Drug resistance was rare to emerge among patients who failed any drug, ranging from 2% in the darunavir arm to 2.8% in the atazanavir arm and 3.3% in the raltegravir arm -- the one area in this study in which raltegravir appeared to fare a bit more poorly.
Since this study began, raltegravir has ceased to be the only integrase inhibitor recommended by the HHS treatment guidelines, having been joined by elvitegravir (part of Stribild) and dolutegravir (Tivicay) in October 2013. "Is the moral of this story that raltegravir is a good way to start, or that integrase inhibitors is a good way to start?" Joel Gallant, M.D., the associate medical director of specialty services at Southwest CARE Center in Santa Fe, N.M., asked rhetorically. "It may not be fair to extrapolate these data to those other integrase inhibitors, but I suspect that's what will happen in clinical practice, where there's already widespread use of Stribild and Tivicay."
Nonetheless, Gallant suggested that the results of A5257 dovetailed with other trends seen in first-line antiretroviral therapy within resource-rich regions. "There's still a role for protease inhibitors, but it's shrinking," he said, adding: "A completely unintended message of this study is that we're moving away from efavirenz. It's been a gold standard for many years, but we now have a number of better-tolerated options."
Myles Helfand is the editorial director of TheBody.com and TheBodyPRO.com.
Follow Myles on Twitter: @MylesatTheBody.