Predictive factors of treatment interruption duration in a cohort of HIV-1 infected patients with CD4 count greater than 350 cells per mm3.

Objective: To determine predictive factors of treatment interruption (TI) duration within a cohort of HIV-1 infected patients having stopped their treatment with CD4 above 350 cells per mm(3).

Design: Data were collected from computerized medical records. Patients were selected if they were HIV-1 positive, 18 years of age or older, and had stopped their treatment between January 1st, 1999 and July 1st, 2003, with CD4 count above 350 cells per mm(3). The study period was censored on October 1st, 2003. Patients were assessed every 3 months from inclusion to censure. A survival analysis using the Cox proportional hazard model was performed.

Results: One hundred eighty-five patients were included. The median duration of TI was 43 weeks. Sixty-three patients remained off-treatment at censure. In the multivariate analysis, TI duration was shorter if CD4 nadir was below 250 cells per mm(3) before TI (relative hazard, 2.10), age superior to 40 (relative hazard, 1.72), viral load higher than 2.3 log.copies per ml (relative hazard, 1.52), and CDC class C (relative hazard, 1.78) at TI. Neither CD4 cell count at TI, numbers of treatments, nor duration of treatment and infection before TI were independent predictive factors of early treatment resumption (TR).

Conclusion: Some clinical and biological data may be used as predictive factors of early TR. Our results can have implications on future therapeutic strategies, in which the goal of therapy is to maintain CD4 cell count above a predetermined threshold using cycles of therapy followed by prolonged interruption according to CD4 count.