517 Impact of Therapeutic Interruptions in Patients Treated during Primary Infection C. Goujard*1, C. Deveau2, M. Sinet3, C. Jacomet4, C. Lascoux5, T. Allegre6, H. Bazin7, C. Rouzioux8, J. F. Delfraissy1, L. Meyer2 1Bicêtre Hosp, Le Kremlin-Bicêtre, France; 2INSERM U569, Bicetre Hosp, Le Kremlin-Bicetre, France; 3INSERM E0109, Le Kremlin-Bicetre, France; 4Clermont-Ferrand, France; 5St Louis Hosp, Paris, France; 6Aix en Provence, France; 7Caen, France; and 8Necker Hosp, Paris, France
Background: Whether or not to initiate HAART during primary-infection is still debated. We describe the evolution of plasma HIV-RNA in 27 patients (pts) who interrupted treatment after being virologically controlled during at least 6 months.
Methods: Since 1996, the ongoing French PRIMO cohort has followed 305 pts recruited during primary infection (defined as an incomplete WB, or a positive p24 Ag with a negative ELISA, or a positive following negative ELISA in the previous 6 mos); 82% presented with symptomatic primary infection, median plasma viral load at enrolment was 5.1 log, 80% pts initiated HAART at enrolment. Among the 172 pts who initiated HAART at enrolment and were followed at least 6 months, 65 interrupted treatment during at least 14 days. The evolution of plasma viral load after interruption was described in the subset of the 27 pts in whom plasma viral load was controlled (< 200 copies/ml) since at least 6 mos at the time of treatment interruption. The cut-off date for analysis was January 2002.
Results:The median follow-up after interruption and duration of interruption were 15.6 mos (range 4.4- 55) and 9.9 mos (1.3-55), respectively; the median of maximum plasma HIV RNA during interruption was 4.12 log; 33% of pts reinitiated HAART afterwards. Pts whose viral load was 4.12 log or more after treatment interruption tended to be older (38 vs 30 yrs, p = 0.11), had higher viral load and lower CD4 at enrolment (5.7 vs 3.6 log, p = 0.01 and 475 vs 767 cells/muL, p = 0.004) than those in whom viral load remained < 4.12 log. Treatment characteristics (treatment duration: 22 vs 20 mos, duration of HIV-RNA < 200 copies/ml during treatment: 21 vs 19 mos, CD4 increase during treatment: +309 cells vs +278) were similar between the two groups. Interestingly, plasma viral load remained spontaneously under 200 in 4 pts during an interruption period varying from 18 to 55 mos.
Conclusions: Our results suggest that subject characteristics at the time of treatment initiation during primary infection, but not the duration of the treatment or the duration of the virological control, are associated with plasma viral load levels during treatment interruption. Although median viral load after treatment interruption was similar to that observed during natural history, there was a huge heterogeneity: 4 pts controlled their viremia after an interruption of more than 18 months.
