OBJECTIVE

To compare a quadruple-nucleoside with an efavirenz-containing regimen for treatment of HIV-1 infection.

DESIGN

A randomized, open-label study of the AIDS Clinical Trials Group (ACTG).

METHODS

Subjects receiving zidovudine/lamivudine/abacavir on ACTG 5095 with HIV-1 RNA less than 200 copies/ml were randomly assigned to intensify either with tenofovir or efavirenz. Subjects were followed for time to treatment failure, defined as either virological failure or treatment discontinuation. Analyses were intent-to-treat.

RESULTS

One hundred and seventy subjects (21% women; 56% non-white) entered the study. At baseline, 95 and 73% had HIV-1-RNA levels less than 200 and 50 copies/ml, respectively; the median CD4 cell count was 453 cells/microl. Over a median 79 weeks follow-up, 165 (97%) completed the study, three (2%) discontinued, and two (1%) died. Treatment failure occurred in 31 subjects: 18 (21%) (quadruple nucleosides) and 13 (15%) (efavirenz-containing regimen); however the failure-time curves crossed and demonstrated a non-constant treatment effect over time, characterized by more early treatment failures on the efavirenz-containing regimen and more late treatment failures on the four-nucleoside regimen. HIV-1 RNA remained suppressed in more than 88% of subjects to less than 200 copies/ml and in more than 78% to less than 50 copies/ml at weeks 24, 48, and 72, without differences by treatment arm. There were no significant differences between the regimens in CD4 cell increases, time to new grade 3/4 adverse events, or adherence.

CONCLUSION

The safety, tolerability, and efficacy of the four-nucleoside regimen were not significantly different from the efavirenz-containing regimen. These pilot data support further investigation of the quadruple-nucleoside regimen.