903    

Background:  Evidence indicates that protease inhibitors (PI) might have a specific anti–Kaposi’s sarcoma (KS) activity. This suggests that a PI-containing HAART regimen may be more effective in preventing KS than other HAART regimens. We studied the incidence rates of KS in 4 calendar periods of time characterized by different ART availability and studied the factors associated with KS.

Methods:  Patients were selected from the French Hospital Database on HIV, a large prospective hospital cohort. Hazards of KS was estimated by Cox proportional hazards models adjusting for age, nadir of CD4, HIV exposure category, prior diagnose of AIDS, initiation of a mono, and dual or HAART regimens. Several HAART regimens were distinguished according to whether they contained PI, non-nucleoside analogue (NNRTI), both, or only NRTI. All treatment variables were included as time dependent variables.

Results:  A total of 1634 patients with KS were identified from the 54,999 patients included in the study (182,752 person-years of follow-up). The incidence rate of KS decrease over time from 32 of 1000 person-years in 1993–1994 to 3 of 1000 person-years after 1999. In the most recent period, KS occurred at higher median CD4 cells counts (134 vs 24 cells/mm³, p < 0.0001). In the multivariate analysis, age at entry into the database, nadir of CD4 cells count, previous AIDS diagnosis, homosexual transmission were associated with an increased risk of KS. Prescription of HAART containing PI was associated with a significant decreased risk of KS (0.82; CI 0.72 to 0.93). HAART containing NNRTI was associated with a similar reduced risk although nonsignificant (0.83; 0.67 to 1.03). When restricting the analysis to homosexual patients the reduction associated with the use of HAART with PI and HAART with NNRTI was similar (0.68 and 0.71) and both significant. Use of HAART with NRTI was associated with a reduced risk of KS (0.83 and 0.88) in both populations although not significant.

Conclusions:  There has been a dramatic reduction of incidence of KS since the introduction of HAART. Similar reduction of KS incidence was observed in patients taking HAART including PI and HAART including NNRTI indicating that PI and NNRTI might be equally effective in preventing KS. The same tendencies were observed with NRTI-containing HAART, but larger treated-populations are required to confirm these results.

Keywords: Kaposi's Sarcoma; Protease inhibitor; Non nucleoside reverse transcriptase inhibitor