U.S. treatment guidelines equally recommend first-line treatment regimens that include either a non-nucleoside reverse transcriptase inhibitor (NNRTI), efavirenz (Sustiva, Stocrin); one of two ritonavir (Norvir)-boosted protease inhibitors (PIs), atazanavir (Reyataz) or darunavir (Prezista); or one of three integrase inhibitors, raltegravir (Isentress), elvitegravir/cobicistat or dolutegravir (Tivicay, DTG). Meanwhile, recent drug-company-sponsored studies have concluded that integrase inhibitors are superior to other drug classes.
At CROI 2014, the AIDS Clinical Trials Group (ACTG) did it again, and presented data of their massive, 1,800-patient randomized trial ACTG 5257, known as the ARDENT study. In this open-label study, participants received tenofovir/emtricitabine (Truvada) and were randomized to receive either raltegravir (twice daily), atazanavir or darunavir.
It's notable that the group used the well-known ACTG endpoints in ARDENT, which will raise discussions during 2014. Virologic failure was defined as a viral load above 1,000 copies/mL between weeks 16-24, or a viral load above 200 copies/mL after week 24. Treatment failure was defined as discontinuation of randomized treatment components. This could have more relevance than initially thought, since we could be misclassifying cases of slow initial viral load decreases as virologic failures.
After 96 weeks, 80% of raltegravir participants had viral loads less than 50 copies/mL versus 63% in the atazanavir group and 73% in the darunavir group. Looking at virologic failure, darunavir was statistically inferior to raltegravir, based within the predefined confidence intervals.
With regard to tolerability failure, raltegravir and darunavir were both superior to atazanavir. And when investigators analyzed the data on cumulative incidence of either virologic or tolerability failure, raltegravir was the winner -- superior to both atazanavir and darunavir, while darunavir was superior to atazanavir.
Much of the differences in this study were driven by an unanticipated higher rate of discontinuations due to toxicity (elevated bilirubin and gastrointestinal toxicity) in the atazanavir arm (16%). This rate was higher than in previous studies of atazanavir such as the CASTLE (3%) and QUAD 103 studies (6%). We suspect that part of this difference was because of the open-label nature of ARDENT compared to the blinded design of CASTLE and 103.
Furthermore, raltegravir had significant advantages in lipid profile and effects on bone mineral density as assessed by DXA scans. Taken together, ARDENT and other studies are signaling a shift in the standards of care for first-line treatment, with the ascension of integrase inhibitors and the fading (perhaps to "alternate" status) of boosted PIs.
Indeed, one of us (Ben Young) now rarely initiates patients on either NNRTI- or PI-based treatment, using integrase inhibitor-based treatment for nearly all first-line treatment. Moreover, this is the first time ever that we have a randomized study that shows that a twice-daily regimen is superior to two preferred once-daily regimens, a finding we will need some weeks to digest.
Which other studies presented at CROI 2014 will have lasting impact? Read more of Dr. Llibre and Dr. Young's top picks.