Background: CD4+ lymphocyte counts and plasma HIV-1 RNA levels predict progression of HIV-related disease, but the relative importance of these and other virological factors in defining response to antiretroviral therapy is not yet clear.
Objective: To determine the short-term variability of plasma HIV-1 RNA level during stable therapy; the relative importance of pretreatment values and early changes in CD4+ count, HIV-1 RNA levels, and infectious HIV-1 titers in mononuclear cells of peripheral blood and pretreatment syncytium-inducing phenotype of an HIV-1 isolate for prediction of disease progression and decline in CD4+ counts during therapy.
Design: Data were collected prospectively in a randomized, clinical trial comparing two combination regimens (ACTG [AIDS Clinical Trials Group] Protocol 241) and pooled across treatments.
Setting: 8 AIDS Clinical Trials Units.
Patients: 198 adults with HIV-1 infection and no more than 350 CD4+ lymphocytes/mm3 who had received at least 6 months of nucleoside therapy.
Interventions: All patients received zidovudine and didanosine; 100 received nevirapine and 98 received placebo.
Measurements: CD4+ lymphocyte counts, plasma HIV-1 RNA levels, and infectious HIV-1 titers in cells were measured before and 8 and 48 weeks after study treatment. Assay for the syncytium-inducing viral phenotype was done at baseline. Progression was defined as occurrence of opportunistic infection, malignancy, or death during the 48 weeks after treatment began.
Results: The difference between two measurements of HIV-1 RNA levels at baseline was within ± 0.39 log10 copies/mL (2.5-fold) for 90% of 167 patients receiving stable therapy. In a multivariate model, risk for disease progression was reduced by 56% (95% CI, 8% to 79% [P = 0.028]) for every 10-fold lower HIV-1 RNA level at baseline, by 52% (CI, 6% increase to 79% reduction [P = 0.071]) for every 10-fold reduction in HIV-1 RNA level at 8 weeks after treatment initiation, and by 67% (CI, 42% to 81% [P < 0.001]) for every 2-fold higher CD4+ count at baseline. These risk factors and syncytium-inducing viral phenotype at baseline, but not infectious HIV-1 titers in circulating cells, were associated with change in CD4+ counts over 48 weeks.
Conclusions: For an individual patient, a change in plasma HIV-1 RNA level of 2.5-fold or more probably indicates a true biological change. Monitoring HIV-1 RNA levels and CD4+ lymphocytes before a change in antiretroviral treatment and monitoring HIV-1 RNA levels shortly thereafter improves prediction of disease progression and decline in CD4+ counts for 1 year compared with monitoring CD4+ counts or HIV-1 RNA levels alone. Additional monitoring of infectious HIV-1 titers in mononuclear cells of peripheral blood is not useful.