609    

Background:  The main determinant of CD4⁺ cell count increase in patients treated with HAART is the persistence of virologic response. Some studies suggested that the increase of CD4 reaches a plateau after 2 or 3 years of complete virological response. We searched for subgroups of patients in whom increase of CD4 would continue after the first years of HAART.

Methods:  The APROCO cohort study enrolled 1281 patients at the initiation of a protease inhibitor-containing regimen in 1997to 1999. Patients are followed every 4 months. In the patients having a plasma HIV RNA < 500 copies/mL at month 4, we studied the slopes of CD4 expressed in cells/mm³/month using longitudinal mixed models (SAS version 8.2, PROC MIXED). Data were censored at the time of occurrence of a rebound of plasma HIV RNA > 500 copies/mL.

Results:  We selected for the analyses 870 patients having a plasma HIV RNA < 500 copies/mL at month 4. During a median follow-up of 57 months, rebound of plasma HIV RNA occurred in 438 patients (50.3%). Median follow-up before rebound was 33 months. In patients having a plasma HIV RNA persistently < 500 copies/m after 5 years of HAART, the proportion of pts having  CD4 > 200/mm³ was 100% in those who initiated HAART with CD4 > 200/mm³ and 85% in those who initiated HAART while having CD4 < 100/mm³. We observed 2 significant inflexions of the slopes of CD4:  mean estimated increase of CD4 was of 29.9 cells/mm³/month before month 4, of 6.4 cells/mm³/month between month 4 and month 36 and of 0.6 cells/mm³/month after month 36 (not significantly different from 0). The switch from the protease inhibitor to a non-nucleoside reverse transcritptase inhibitor or abacavir was not associated with an inflexion of the slopes of CD4. In unadjusted analysis, 3 factors were associated with a significantly (p < 0.05) positive slope of CD4 after month 36:  male gender (+0.9), naivete of antiretroviral therapy at baseline (+2.0), and baseline CD4 < 100/mm³ (+2.6) but none of these factors was associated with a significantly positive slope of CD4 after month 36 when they were adjusted for each other.

Conclusions:  Our data confirm that the increase of CD4⁺ cell counts reaches a plateau after 3 years of complete virological response. Even if some patients may still have an increase of CD4⁺ cell counts after 3 years, this increase appears small and is not sufficient to overcome the initial immune deficit in all patients who initiate HAART at a very low CD4⁺ cell count.

Keywords: Long-term; CD4 response; HAART