Abacavir/lamivudine versus tenofovir DF/emtricitabine as part of combination regimens for initial treatment of HIV: final results

Abstract

Background: AIDS Clinical Trials Group A5202 compared blinded abacavir/lamivudine (ABC/3TC) to tenofovir DF/emtricitabine (TDF/FTC) with efavirenz (EFV) or atazanavir/ritonavir (ATV/r) in human immunodeficiency virus (HIV)-infected treatment-naive patients, stratified by screening HIV RNA (< or ≥ 10(5) copies/mL). Due to higher virologic failure with ABC/3TC in the high HIV RNA stratum, blinded treatment was stopped in this group, but study follow-up continued for all patients.

Methods: Primary endpoints were times to virologic failure, regimen modification, and safety event.

Results: In the low HIV RNA stratum, time to virologic failure was similar for ABC/3TC vs TDF/FTC with ATV/r (hazard ratio [HR] 1.25, 95% confidence interval [CI] 0.76, 2.05) or EFV (HR 1.23, 95% CI 0.77, 1.96), with significantly shorter times to regimen modification for ABC/3TC with EFV or ATV/r and to safety events with EFV. Prior to stopping blinded treatment in the high stratum, higher virologic failure rates were seen with ABC/3TC with EFV (HR 2.46, 95% CI 1.20, 5.05) or ATV/r (HR 2.22, 95% CI 1.19, 4.14).

Conclusions: In the low HIV RNA stratum, times to virologic failure for ABC/3TC or TDF/FTC were not different with EFV or ATV/r. In the high stratum, virologic failure rate was significantly higher for ABC/3TC than for TDF/FTC when given with either EFV or ATV/r.

Figure 1.

Enrollment, randomization, and disposition of patients with screening HIV RNA <10⁵ copies/mL. Patients were to remain in follow-up regardless of whether antiretroviral therapy was modified; therefore, study follow-up and treatment modification disposition are both presented. Reasons for treatment modification are split into (number before, number after, number without protocol-defined virologic failure) to summarize the amount of censoring of primary efficacy endpoints in as-treated analyses. * Nucleoside reverse transcriptase inhibitors were blinded through 25 February 2008 for persons with HIV-1 RNA levels of 100 000 copies/mL or more at screening and until final visits starting 1 July 2009 for those with HIV-1 RNA levels less than 100 000 copies/mL at screening. † Death was censored for premature study discontinuation and counted as a reason for treatment discontinuation if there was no previous regimen modification. Site closure was censored for premature study and treatment discontinuation. ‡ Site-declared virologic failure was by clinical determination of the site investigator, whereas protocol-defined virologic failure was determined strictly by the quantitative definition set forth in the protocol. Numbers may differ because not all patients who had protocol-defined virologic failure modified the regimen, or the drug modification may have been attributed to another reason, such as “nonadherent with medications or visits.”

Figure 2.

A, Time to protocol-defined virologic failure in those with screening plasma HIV RNA<10⁵ copies/mL. B, Proportion of patients in low HIV RNA stratum with HIV RNA level <50 copies/mL and 95% binomial confidence intervals at each study week where analysis includes patients with available data, regardless of whether they had previously switched therapy or met criteria for virologic failure. C, Time to regimen completion (time to the first occurrence of either confirmed virologic failure or discontinuation of initially randomized regimen) in low HIV RNA stratum. D, Time to protocol-defined virologic failure in the high HIV RNA stratum.

Figure 3.

In low HIV RNA stratum, (A) Time to tolerability endpoint, defined as first change in regimen; and (B) time to first primary safety endpoint, defined as first grade 3 or 4 sign, symptom, or laboratory abnormality while on initial randomized treatment that was at least 1 grade higher than baseline, excluding hyperbilirubinemia and elevation in the creatine kinase level.