Triple nucleoside treatment with abacavir plus the lamivudine/zidovudine combination tablet (COM) compared to indinavir/COM in antiretroviral therapy-naïve adults: results of a 48-week open-label, equivalence trial (CNA3014)

Abstract

Objective: An equivalence (non-inferiority) trial comparing antiviral response, tolerability, and adherence with a triple nucleoside regimen containing abacavir 300 mg (ABC) plus a lamivudine 150-mg/zidovudine 300-mg combination tablet (COM) twice daily vs. a regimen containing the protease inhibitor indinavir (IDV) 800 mg three times daily plus COM twice daily (IDV/COM) in antiretroviral-naïve, HIV-infected patients.

Methods: Adult patients with plasma HIV-1 RNA levels > or = 5000 copies/mL and CD4+ cell counts > or = 100 cells/mm(3) were randomized to receive open-label ABC/COM (n = 169) or IDV/COM (n = 173) for 48 weeks. The intent-to-treat (ITT) population was the primary population evaluated. ITT: switch/missing equals failure (ITT: S/M = F) and as-treated (AT) analyses were used for assessing the proportion of patients achieving plasma HIV-1 RNA level < 400 and < 50 copies/mL at each clinic visit. In the ITT: S/M = F analysis, patients who switched treatment or had missing values were considered treatment failures; the AT analysis examined virologic data only while patients received study treatment. ABC/COM was considered equivalent (non-inferior) to IDV/COM if the lower limit of the 95% confidence intervals (CIs) about the difference in proportions of ABC/COM- vs. IDV/COM-treated patients attaining plasma HIV-1 RNA < 400 copies/mL exceeded -15% at week 48.

Results: The study population was diverse with respect to ethnicity (38% Asian, 27% Hispanic, 28% white, 3% black, 4% other) and gender (39% women, 61% men). Baseline median HIV-1 RNA was 4.80 log(10) copies/mL and CD4+ cell count was 315 cells/mm(3). ABC/COM met the criterion of equivalence to IDV/COM. In the ITT: S/M = F analysis at Week 48, a greater proportion of ABC/COM-treated patients achieved HIV-1 RNA < 400 copies/mL (66% [109/164] vs. 50% [82/165]; treatment difference 16.6%, 95% CI (6.0, 27.2), p = 0.002) and HIV-1 RNA < 50 copies/mL (60% [99/164] vs. 50% [83/165]; treatment difference 9.6%, 95% CI [-1.1, 20.2]), whereas the AT analysis showed similar proportions achieving these endpoints (< 400 copies/mL: 85 vs. 83%; < 50 copies/mL: 79 vs 81%). Comparable proportions of patients with screening HIV-1 RNA values > 100 000 copies/mL achieved HIV-1 RNA < 400 copies/mL (ABC/COM: 60% [35/58]; IDV/COM: 51% [33/65]; treatment difference 9.6%, 95% CI [-7.9, 27.1]; ITT: S/M = F analysis). A significantly greater proportion taking ABC/COM were > or = 95% adherent (72% [109/151] vs. 45% [70/154] with IDV/COM, p < 0.001). Median increases from baseline in CD4+ cell counts were similar in the two treatment groups (+148 vs. +152 cells/mm(3)). Significantly more patients on IDV/COM reported drug-related adverse events (87% [142/165] vs. 65% [108/164] with ABC/COM, p < 0.001), similar proportions discontinued treatment due to adverse events (13 vs. 10%), and a slightly greater proportion in the ABC/COM group reported serious adverse events (13 vs. 8%). About half of the latter comprised suspected ABC-related hypersensitivity reactions (overall rate, 6%). Most adverse events were gastrointestinal in nature in both treatment groups.

Conclusion: ABC/COM was at least equivalent to IDV/COM over 48 weeks in the treatment of antiretroviral-naïve patients. ABC/COM was associated with a significantly higher adherence rate and lower incidence of drug-related adverse events than IDV/COM. The study was limited in that it was not powered to determine equivalence of treatments within high vs. low viral load strata, adherence was not monitored electronically, and bias could not be ruled out due to the open-label study design.