Background:  DART (Development of Anti-Retroviral Therapy in Africa) is a randomized trial of monitoring strategies and planned interruptions (after 24 weeks) in 3300 symptomatic ART-naive adults with CD4 < 200 cells/mm³ from 3 clinical sites (2 in Uganda, 1 in Zimbabwe). Of 3263 patients enrolled to  October 1, 2004, 2466 (76%) have received zidovudine (ZDV)+lamivudine (3TC)+ tenofovir DF (TDF) first-line. As there are few viral load data on this regimen, we conducted a 24-week virology sub-study (n = 300).

Methods:  Plasma HIV-1 RNA was retrospectively assayed at 0, 4, 12, and 24 weeks in 100 patients at each site (50 with baseline CD4 of 1 to 99 and 50 of 100 to 199 cells/mm³) using the Roche Amplicor assay. All assays have been performed in Africa under cross-site quality assurance. Here we present results from Ugandan patients, based on available samples (intention to treat).

Results:  Of the 200 Ugandan adults, 68% were women, median age was 37 years, and median baseline CD4 100 cells/mm³ (IQR 36 to 144). Median HIV RNA was 333,600 copies/mL (IQR 106,000 to 662,300 copies/mL), and mean 5.4 log₁₀ copies/mL (SD 0.7). At week 24 (n = 188), 54% achieved < 50 copies/mL and 72% < 400 copies/mL (intent to treat M = F 51% and 68%, respectively). Corresponding proportions were 14% and 40% at week 4 (n = 193), and 49% and 85% at week 12 (n =1 87). Mean decreases from baseline were 2.5, 3.6, and 3.5 log₁₀ copies/mL at weeks 4, 12, and 24, respectively, and median CD4 increases 83 and 88 cells/mm³ at weeks 12 and 24. Six patients died before 24 weeks:  2 died before 4 weeks, the other 4 had final HIV RNA before death of < 1000 copies/mL. At 12 and 24 weeks, 10% and 19% had HIV RNA ≥ 1000 copies/mL (6% and 15% ≥ 2000 copies/mL). In the preceding 12 weeks, 28 of 37 patients with HIV RNA ≥ 1000 copies/mL at week 24 had been off ART for > 1 week (n = 4), had incomplete adherence (n = 23), became pregnant (n = 3), or had severe adverse events, grade 3/4 adverse events, or other ART-modifying adverse events (n = 6) or malaria (n = 12). Preliminary 24 week data from 100 patients in Zimbabwe are similar.

Conclusions:  Triple nucleoside regimens are highly relevant in resource limited settings:  27% of DART patients have a prior diagnosis of pulmonary or extra-pulmonary tuberculosis, the latter also being the third most frequently reported WHO grade 4 event after baseline. ZDV+3TC+TDF has good virological efficacy in advanced HIV disease, comparable to that reported after HAART introduction in equivalent industrialised populations, but against a background of intercurrent illnesses. Evaluation of genotypes in those with HIV RNA ≥ 1000 copies/mL at 24 weeks, and response to 48 weeks are ongoing.

Keywords: HIV-1 RNA response; triple NRTI; Africa