Combined analysis of two-year follow-up from two open-label randomized trials comparing efficacy of three nucleoside reverse transcriptase inhibitor backbones for previously untreated HIV-1 infection: Ozcombo 1 and 2

Purpose: To compare inhibition of HIV replication, improvements in CD4+ T-cell counts, metabolic parameters, and body shape changes after 2 years of assigned therapy in OzCombo patients. Method: Study participants were those who were recruited into the open-label OzCombo 1 (1996/1997) and OzCombo 2 (1997/1998) trials. Patients in OzCombo 1 were randomized to receive indinavir in combination with zidovudine+lamivudine (AZT+3TC; n = 35), stavudine (d4T)+3TC (n = 34), or d4T+didanosine (ddl) (n = 37). OzCombo 2 patients were randomized to the same nucleoside reverse transcriptase inhibitor (NRTI) backbones with nevirapine (n = 20, 22, 23, respectively). The mean time-weighted changes from baseline in CD4 T-cell count/mL, HIV RNA (log copies/mL plasma), and proportions with detectable viral load (>500 copies plasma HIV RNA/mL) between NRTI arms over 2 years were compared by formal meta-analysis. A cross-sectional study of metabolic and body shape complications was also undertaken. Results: For the comparison of d4T+3TC and d4T+ddl to AZT+3TC, mean differences in time-weighted change from baseline in CD4 T-cell count/μL and log copies HIV RNA/mL adjusted for baseline CD4+ T-cell and HIV RNA counts were: -44 (p = .08) and -14 (p = .56) cells/μL and -0.1 (p = .40) and -0.1 (p = .6) copies/mL. Odds ratios for detectable viral load in the last study quarter were 0.6 (p = .44) and 1.0 (p = .95). The mean percent leg fat was lower in the d4T+3TC and d4T+ddl than the AZT+3TC arm (mean difference 5.1% [p = .07] and 7.6% [p = .02], respectively). Conclusion: For all regimens, virological control and immunological response were maintained over 2 years. Regimens containing d4T and particularly d4T+ddl were significantly associated with increased peripheral fat loss compared with, AZT+3TC.