519
Randomized Intensification of a Triple Nucleoside Regimen with Efavirenz or Tenofovir in ACTG 5095
Roy Gulick*1, C Lalama2, C Shikuma3, H Ribaudo2, B Schackman1, W Meyer4, K Squires5, E Acosta6, K Klingman7, D Kuritzkes8, and the ACTG 5095 Study Team
1Weill-Cornell Med Coll, New York, NY, US; 2Harvard Sch of Publ Hlth, Boston, MA, US; 3Univ of Hawaii, Honolulu, US; 4Quest Diagnostics, Baltimore, MD, US; 5David Geffen Sch of Med, Univ of California, Los Angeles Med Ctr, US; 6Univ of Alabama at Birmingham, US; 7Division of AIDS, NIAID, NIH, Bethesda, MD, US; and 8Brigham and Women's Hosp, Boston, MA, US
Background: Quadruple nucleoside (4-nuc) combinations
have potential as active, convenient, well-tolerated regimens with few drug
interactions, but comparative data with standard regimens are limited.
Methods: ART-naïve subjects randomized originally on ACTG
5095 to receive zidovudine/lamivudine/abacavir (3-nuc)
who suppressed HIV RNA to <200 copies/mL were
re-randomized to intensify with either efavirenz
(EFV-based regimen) or tenofovir (4-nuc regimen). Primary
study endpoint was time to treatment failure, defined as minimum time to either
virologic failure (2 successive HIV RNA >200 copies/mL) or treatment discontinuation (allowing within-class
substitutions for treatment-limiting toxicity). Analyses were intent-to-treat.
The study was designed to have 64 to 80% power (depending on enrollment) to
detect a hazard ratio of treatment failure (HR) of 0.57 with 5% type I error.
Results: Originally we randomized 382 subjects to
3-nuc in A5095 and 208 had HIV RNA <200 copies/mL
after a median 36 weeks of treatment. Of them, 25 chose to continue 3 nuc off study, 13 changed or discontinued treatment, and
170 (21% women; 56% non-white) chose to enter the current study. At baseline,
73% had HIV RNA <50 copies/mL and median CD4 count
was 453 cells/mm3. Over a median follow-up of 79 weeks, 163 (96%)
completed the study, 5 (3%) discontinued early, and 2 (1%) died. Overall, treatment
failure occurred in 32 subjects: 13
(15%) on EFV-based treatment vs 19 (22%) on 4 nuc (p = 0.28 log-rank), however the time-failure curves crossed. There
was a significant change in treatment effect over time (p = 0.03), with more early treatment failures (and none after week
40) on EFV-based treatment (e.g., HR 1.95; 95%CI, 0.62, 6.12 at week 8) and more later treatment failures on 4 nuc
(e.g., HR 0.15; 95%CI 0.03, 0.75 at week 48). Overall, HIV RNA was suppressed
through 48 weeks in ≥89% of subjects to <200 c/ml and in ≥79%
to <50 copies/mL, without significant differences
by treatment arm (p >0.1). There
were no significant differences between arms in CD4 increases (median +55
cells/mm3 at week 72), time to new grade 3 or 4 adverse events
(occurring in 31% and 12% of subjects, respectively), or adherence rates (≥77%
reported not missing a dose over 4 days at weeks 4, 24, and 48).
Conclusions: In subjects with virologic
suppression on 3 nuc, intensifying with efavirenz was not different to intensifying with tenofovir in safety and efficacy over a median 1.5 years of
follow-up. The relative effect of treatment between the 2 regimens changed over
time.
|
