Resistance to at least one PI was observed by RNA but not DNA genotyping in 50% of patients (78 of 156) and by DNA but not RNA genotyping in 7% of patients (11 of 156). The median (IQR) GSS was 1.5 (1.0, 1.5) based on RNA genotyping and 2.0 (1.5, 3.0) based on DNA genotyping (P < 0.001). The GSS was 2 or more in 18% and 58% of patients based on RNA and DNA genotyping, respectively,

suggesting that 20% of patients had at least two active drugs in their regimen (excluding enfuvirtide) based on previous RNA genotyping, compared with 58% of patients based on current DNA genotyping (Table 1). RNA genotyping showed that 160 (95%) of the 169 patients harboured triple-class-resistant viruses, compared with 42 (35%) of Afatinib ic50 the 121 patients with assessable DNA genotypes. Among age, gender, Centers for Disease Control and Prevention (CDC) status, nadir click here and baseline CD4 cell counts, and total duration of antiretroviral treatment, only a lower nadir CD4 cell count was significantly associated with triple-class resistance based on DNA genotyping vs. no resistance to at least one of the 3-class (26 vs. 88 cells/μL; P = 0.001). The efficacy of switch drugs for patients receiving a virologically effective regimen depends mainly

on the number and nature of resistance mutations archived in the proviral reservoir following previous Nintedanib (BIBF 1120) therapeutic failures [2-7, 10, 11]. Viraemia is suppressed by the antiretroviral regimen in these patients, so resistance genotyping must be performed only on cell-associated HIV-1 DNA. Here we compared DNA genotyping results at randomization among 169 patients on successful highly active antiretroviral therapy (HAART) enrolled in the ANRS 138-EASIER switch trial [8] with the results of historical RNA genotyping in the same patients. We found that DNA genotyping

detected significantly fewer resistance mutations in the RT and PR genes than previous RNA genotyping. Indeed, mutations conferring resistance to at least one antiretroviral drug were detected exclusively by RNA genotyping or exclusively by DNA genotyping in 63% and 13% of patients for NRTIs, 47% and 1% of patients for NNRTIs and 50% and 7% of patients for PIs, respectively. Despite frequently suboptimal therapy in the past, only 35% of patients harboured triple-class-resistant archived viral DNA, a situation associated with a lower CD4 cell nadir. Our study confirmed the findings of a recent study showing, in a large number of patients with undetectable or low level viral load under an antiretroviral regimen, that the concordance between DNA and RNA was 46.7% for NRTI mutations, 26.3% for NNRTI mutations and 43.7% for PI mutations [12].