Four-drug single-pill antiretroviral equivalent to Atripla at 48 weeks

Quad – the coformulated antiretroviral combining the integrase inhibitor elvitegravir, the booster cobicistat, tenofovir, and emtricitabine-proved virologically equivalent to efavirenz plus tenofovir/emtricitabine (coformulated as Atripla) after 48 weeks in previously untreated people. [1]

In a separate 48-week analysis, atazanavir/cobicistat yielded virologic response rates similar to atazanavir/ritonavir.

These two phase 2 studies enrolled antiretroviral-naive people with a viral load at or above 5000 copies, a CD4 count above 50, and no resistance to NRTIs, NNRTIs, or PIs. Both trials were double-blind and active controlled, the first comparing Quad with Atripla, the second comparing atazanavir/cobicistat with atazanavir/ritonavir in people also taking tenofovir/FTC.

Researchers randomised 48 people to Quad, 23 to Atripla, 56 to atazanavir/cobicistat, and 29 to atazanavir/ritonavir. Ages averaged about 35, and about 90% of study participants were men. Median pretreatment viral loads were 4.6 or 4.7 log in all four treatment groups (about 40,000 to 50,000 copies). Median pretreatment CD4 counts were 354 in the Quad group, 436 in the Atripla group, 341 in the atazanavir/cobicistat group, and 367 in the atazanavir/ritonavir group. AIDS rates were 16% or lower in all treatment arms. Six people in the atazanavir/cobicistat group never received an antiretroviral dose and were not included in the analysis.

Three people (13%) discontinued Quad, none because of adverse events; 3 people quit Atripla, 1 because of an adverse event (suicidal ideation); 5 people (10%) quit the cobicistat group, 2 because of side events (vomiting and rash); and 3 people (10%) quit the ritonavir group, 1 because of an adverse effect (ocular icterus). Rates of drug-related grade 1 to 4 adverse events were 46% with Quad, 57% with Atripla, 36% with atazanavir/cobicistat, and 48% with atazanavir/ritonavir. Rates of grade 3 or 4 adverse events were 4% with Quad, 9% with Atripla, 4% with cobicistat, and 0% with ritonavir.

In a missing-data-equal-failure analysis, proportions with a week-48 viral load below 50 copies were 90% with Quad versus 83% with Atripla, and 82% with atazanavir/cobicistat versus and 86% with atazanavir/ritonavir. The first result meant Quad was noninferior to Atripla in 48-week efficacy. In a missing-data-excluded analysis, 48-week sub-50-copy rates were 96% with Quad versus 95% with Atripla, and 91% with cobicistat versus 96% with ritonavir. CD4 count gains through 48 weeks averaged 240 with Quad, 162 with Atripla, 230 with atazanavir/cobicistat, and 206 with atazanavir/ritonavir.

Mean percent changes in glomerular filtration rate estimated by the Cockroft-Gault method (eGFR) were -14% with Quad versus -4% with Atripla and -12% with atazanavir/cobicistat versus -11% with atazanavir/ritonavir. Changes in eGFR appeared early in the trial and remained stable through 48 weeks. Concerns over renal toxicity with cobicistat arose at week 24 in this trial, when eGFR and serum creatinine were worse with Quad than with Atripla. [2]

In the cobicistat-ritonavir comparison, rates of jaundice were low (3% to 4%) and similar in the two treatment arms. Rates of ocular icterus were higher but equivalent in the two arms12% in the cobicistat group and 14% in the ritonavir group.

comment

The change in eGFR is reported as a calculation error rather than evidence of toxicity but is clearly a difficulty that will require a new management algorithm.