Background: The emergence of resistance mutations is one of the major safety concerns of intermittent HAART in chronic HIV infection. We monitored the occurrence of mutations in the ISS PART study, an ongoing randomized multicenter clinical trial comparing continuous (arm A) versus intermittent (arm B) HAART in 273 subjects with chronic HIV infection on their first line HAART (and no previous mono or dual therapy)

Methods: Study population consisted of 136 patients of arm B, followed up during 4 cycles on/off  (4 structured treatment interruptions of 1, 1, 2, and 2 months, respectively, separated by 3 months of treatment). At enrolment, all had CD4 >350/mm³ and viremia <400 copies/mL. HIV-1 genotype was analyzed on plasma samples during viral rebound and on baseline PBMC (when available), using the TrueGene assay.

Results: Overall, mutations were found in 39/136 patients (28.6%). The proportion of subjects with mutations slightly increased over structured treatment interruptions.  The most frequent individual mutations were: M184V (16.2% of all 3TC-treated subjects); K103N (7% of NNRTI-treated), T215Y (2.9% of NRTI-treated), L90M (10.3% of subjects receiving SQV or NFV). The rate of virologic failure (HIV RNA >400 copies/mL at the end of treatment phase) was 30.8% in patients with mutations and 12.4% in patients without mutations (chi-square p = 0.011, 95%CI:  2.3 to 34.6).For 27/39 subjects with mutations HIV DNA genotype was performed on a baseline PBMC sample. Mutations were detected in 11/27(40.7%). On the other hand, baseline mutations were present in 1/22 (4.5%) only of  22 subjects not developing mutations during structured treatment interruptions. In general, the baseline samples did not show the PI- and NNRTI-related mutations seen during structured treatment interruptions (K103N:and L90M); 2/12 subjects with M184V during interruption had this same mutation in the baseline PBMC sample. Most of the other mutations were already detectable at study entry. At univariate analysis, the use of a PI-contaning regimen appeared to be the only baseline factor associated with the emergence of resistance during interruptions.

Conclusions: Our findings show that a remarkable proportion of subjects undergoing structured treatment interruptions develops mutations that may have a negative impact on virologic response to therapy reinstitution.

Such mutations are more likely to occur  in patients harbouring mutations in PBMC at baseline

as well as in subjects receiving PI-containing regimen (this latter finding should  be confirmed in a multivariate analysis including adherence).

Keywords: STI; resitences