Deep-Sequencing Analysis of the Dynamics of HIV-1 Quasiespecies in Naive Patients during a Short Exposure to Maraviroc

ABSTRACT In this study, we have characterized quasispecies dynamics and the evolution of viral tropism in naive HIV-1-infected patients treated with a short course of maraviroc monotherapy (ClinicalTrials.gov registration no. NCT01060618) independently of the tropism of the infecting virus. We randomly selected 20 patients infected with viruses displaying different basal tropisms—10 carrying R5 and 10 carrying dual/mixed X4 (DM/X4) viruses—at recruitment as determined by phenotypic assay (Trofile). Evolution of viral quasiespecies at the end of treatment was determined by ultradeep sequencing of the V3 region using a 454 Life Sciences Platform and geno2pheno (g2p) algorithm for viral tropism prediction. The false-positive rate (FPR) that defines the probability of classifying an R5 virus falsely as X4 was set at 10%. X4-specific HIV-1 viral load (VL) was calculated from sequences with an FPR of <3.75%. Virological response as defined as >1-log 10 copies/ml reduction in VL was detected in 70% of patients independently of the basal tropism of the infecting virus. Viral tropism remained stable, and nonsignificant differences in FPR values before and after treatment were found for the majority of patients in both tropism groups. Only three patients (one with R5 and two with DM/X4 viruses) showed an increased (>1 log) X4 VL, and one patient harboring a DM/X4-tropic virus displayed a significant reduction in FPR values at the end of treatment. Fast changes in the composition of viral populations were observed in all patients after 10 days of maraviroc (MVC) monotherapy treatment, and a complete replacement of viral quasiespecies was found in 3/10 patients carrying R5-using viruses and 4/10 patients carrying DM/X4-using viruses. IMPORTANCE Initiation of treatment with maraviroc requires previous determination of viral tropism by genotypic or phenotypic methods because of the risk of treatment failure and selection of DM/X4-tropic variants. In this study, we confirm previous work showing that the virologic response to maraviroc is independent of basal tropism. By deep-sequencing analysis, we determined that fast changes in viral populations were due to the emergence of minority variants in some patients whereas in others generation of new strains was detected. The risk of DM/X4 selection was very low as FPR values remained stable, and only one patient showed a detrimental switch to DM/X4 variants. Our data show that some DM/X4 viruses are sensitive to maraviroc treatment probably because only a low proportion of DM/X4 viruses use preferentially the X4 receptor and contain authentically maraviroc-resistant viruses that are not accurately detected by current assays.