Objective: To compare the long-term prognostic significance of different definitions of immunologic and virologic responses to highly active antiretroviral therapy (HAART) at 6 months.
Methods: This was a prospective study conducted in 68 French hospitals. HAART was initiated in 2236 protease inhibitor-naive patients included in the French Hospital Database on HIV. Multivariate Cox proportional hazard models measuring time from 6 months after starting HAART were used to compare the strength of the association between different definitions of immunologic and virologic responses at 6 months and subsequent progression to AIDS or death. The Akaike's Information Criteria were used to identify the most appropriate model.
Results: During a median follow-up of 58 months, 325 patients experienced an AIDS-defining event or died. The model that fitted best was the model in which the CD4 cell count and plasma HIV-1 RNA values attained at 6 months were considered. The risk of clinical progression at 5 years ranged from 7% (95% confidence interval [CI]: 4-10) in patients whose CD4 cell count at 6 months was >or=350 cells/microL and whose HIV-1 RNA concentration was <3 log10 copies/mL to 63% (95% CI: 52-75) in patients whose CD4 cell count at 6 months was <100 cells/microL and whose HIV-1 RNA concentration was >or=5 log10.
Conclusions: Plasma HIV-1 RNA concentration and CD4 cell count should be taken into account independently when evaluating early response to treatment. The persistent impact of early response on clinical progression at 5 years emphasizes the major importance of the success of first-line HAART.