The Effect of Number of Mutations and of Drug-Class Sparing on Virological Response to Salvage Genotype-Guided Antiretroviral Therapy

Abstract

To assess on longitudinal data the impact of number of drug-associated mutations at genotype resistance testing (GRT) and history of previous exposure to antiretrovirals on the virological response to genotype-guided antiretroviral therapy. Subjects that failed HAART who underwent GRT between June 1999 and March 2002 were enrolled. GRT was performed by Viroseq-2 with expert advice offered to physicians. Main outcome was reaching undetectable (< 80 copies/ml) HIV-1 RNA level after GRT and maintaining undetectable viraemia for at least 6 months. The number of mutations conferring resistance to each class of antiretrovirals was categorized and their effect on virological outcome investigated. Mutations considered in the analysis were those reported by the IAS-USA in 2002. Multivariate analysis was performed by Cox proportional hazard model. Four-hundred-and-seventy consecutive subjects were enrolled and followed-up for a median of 14 (IQR 9-19) months after GRT. Sustained undetectable viraemia was reached by 80 of 449 subjects (18%). Using as end-point reaching and maintaining for at least 6 months < 400 copies/ml after GRT, 103 out of 447 subjects (23%) reached the outcome. For each single protease inhibitor (PI)-, nucleoside reverse transcriptase inhibitor (NRTI)-and non-nucleoside reverse transcriptase inhibitor (NNRTI)-associated mutation, there was a reduction of, respectively, 11% (P = 0.008), 12% (P = 0.001) and 39% (P = 0.005) in the likelihood of reaching virological outcome. Subjects carrying > or = 6 mutations to NRTIs, > or = 7 mutations to PIs and > or = 2 mutations to NNRTIs were less likely to reach the virological success compared with those carrying 0-1 (NRTI and PI) or 0 (NNRTI) mutations [HR = 0.25 (95% CI: 0.10-0.65); HR = 0.33 (950% CI: 0.16-0.67); HR = 0.33 (95% CI: 0.14-0.77)], respectively. However, at multivariate analysis the probability of reaching a favourable virological outcome in patients with > or = 7 mutations to PIs, if naive for NNRTIs [HR = 1.74 (0.69-4.36)], and in subjects with > or = 2 mutations for NNRTIs if naive for PIs [HR = 1.23 (0.22-6.80)], was comparable to those observed in patients with none or one mutation. Our data showed a non-linear association between resistance-conferring mutations and virological outcome. GRT-guided therapy still provided remarkable chances of durable virological success even in subjects with > or = 7 mutations to PIs and in subjects with > or = 2 mutations to NNRTIs, when the subjects did not have a three-class exposure or if GRT showed no evidence of mutations for a drug class. GRT and as-long-as-possible sparing of a drug class could be a convenient strategy for long-term management of drug-failing patients.