569    

Background:  Many patients with chronic HIV infection with long-term controlled viremia and high CD4⁺ count asked to stop HAART (group I). Many patients on HAART would not have been started on therapy based on today’s more conservative guidelines (group 2). We present a 6-month follow-up of PBMC DNA genotype on these pts in whom HAART was discontinued.

Methods:  This prospective pilot study included 57 patients. At the time of structured treatment interruption (M0), 28 patients had HIV RNA <50 copies/mL since 38 (30; 48) months and 841 (644 to 1063)/µL CD4⁺ (group I); 29 had 3.3 (2.6; 4) HIV RNA copies/mL, 630 (537 to 748)/µL CD4⁺ with a median pre-HAART CD4⁺ cell count of 416 (336; 532)/µL (group 2). Proviral DNA extracted from patients PBMC at M0 and M6 after structured treatment interruption was quantified by real-time PCR and analyzed for reverse transcriptase (RT) and protease (PR) genotypes.

Results:  Considering the entire population at M0, median CD4⁺ count and plasma HIV RNA were 739 (584 to 962) cells/µL and 1.9 (1.7 to 3.3) log₁₀ copies/mL, respectively. A 6-months structured treatment interruption was associated with a median decrease of 233 (127 to 346) CD4⁺/µL and increase of 1.9 (1.2 to 2.8) log₁₀ HIV RNA copies/mL. Changes in HIV RNA and CD4⁺ count were significantly higher in group 1 than in group 2 (p = 0.04 and <10^-4). At M6, no patient had to restart treatment. Median number RT+PR mutations in DNA sequences at M0 were 1 (1 to 3) and 4 (3 to 6) in group I and II, respectively. When considering M0 mutated sequences, changes at M6 were as follows: for RT sequences: 9.3% lost 3 to 4 mutations, 48.8% lost 1 to 2 mutations, 35% remained unchanged, finally 50% shifted to wild-type; for PR sequences: 23.7% lost 1 to 2 mutations, 2.6% lost 6 mutations, 60% remained unchanged. Plasma genotyping in group 2 patients showed a shift from mutated to WT in 64% sequences. To assess the more reliable assay to detect mutations related to previous regimens and thus help guide decisions about subsequent therapy, we compared M0 genotyping in plasma RNA and in PBMC DNA. M0 RT plasma genotypes were identical to those in PBMC at M0 in 63%, but were able to detect more resistance mutations in 33%. When HIV RNA<50 at M0 of structured treatment interruption, genotyping in PBMC DNA at M0 is able to reflect archival mutations.

Conclusions:  Despite the repopulation of sensitive virus in plasma and the loss of some mutations in PBMC within 6 months of structured treatment interruption, resistant HIV-1 still persist in PBMC in >50% patients. To help guide decisions about subsequent therapy after structured treatment interruption, M0 genotyping in plasma RNA is the best indicator.

Keywords: PBMC DNA genotype; STI; high CD4 cell count