Background: Temporal changes in HIV-1 resistance both reflect and influence the clinical use of antiretrovirals (ARVs).Objective: To determine temporal trends in reduced susceptibility to ARVs and resistance mutations in routine clinical samples (RCS) from HIV infected patients.Study design: Calculated fold-changes (FC) for ARVs were determined for viral genotypes from RCS received between July 1998 and June 2007 using vircoTYPE HIV-1 (Version 4.2.01). The prevalence of isolates with clinically relevant reduced susceptibility and of resistance mutations were determined for consecutive 6-month periods.Results: 242,003 RCS were identified. The prevalence of RCS exhibiting reduced susceptibility to >= 1 drug among any of three ARV classes decreased from 82% to 66.5% and of these, to >= 1 drug in each of the three ARV classes and from 30.5% to 15.7% from December 98 to June 07 (p <= 0.0001). The prevalence of mutations associated with NRTI, NNRTI and PI resistance generally reflected evolving use of these drug classes. Among fully susceptible RCS, isolates that exhibited resistance mutations were rare.Conclusions: Clinically relevant reduced susceptibility to ARVs declined over the 9-year period. There was a general higher prevalence of reduced susceptibility to RT inhibitors than to Pis. (c) 2008 Elsevier B.V. All rights reserved.

Background In vitro phenotypic resistance studies suggest that the presence of the M184V mutation leads to a reduction in HIV-1 susceptibility to didanosine (ddI). The relevance of this to clinical outcomes remains unclear. In this study, we compared the virological response of ddI- and non-ddI-containing regimens in the presence or absence of the M184V mutation. Methods Data from an observational cohort study of all HIV-1 patients who had phenotypic resistance testing following the emergence of virological failure to an existing highly active antiretroviral therapy (HAART) regimen were analysed. A total of 586 patients entered the study and were followed-up over 48 weeks; 281 (48%) were switched to ddI-containing HAART, of whom 105 had the M184V mutation at baseline. Virological efficacy of combination therapy was studied by reference to average area under the curve of viral load (VL) response and the proportion of patients attaining an undetectable VL (<400 copies/ml). Baseline characteristics and univariate analysis of changes in VL were compared using the Wilcoxon rank sum test. Multivariate analyses were performed using the Van Elteren test. Additional variables included the number of baseline nucleoside reverse transcriptase inhibitor mutations and the number of active antiretroviral drugs given to each group as compared by ‘real phenotype’ resistance test results. Results Amongst patients on ddI-containing HAART, median fold changes in phenotypic susceptibility to ddI were greater in patients with the M184V mutation (fold changes of 2.2 vs 1.2, P<0.001). Nonetheless, the median change in VL and percentage of patients attaining an undetectable VL were similar in those taking ddI, irrespective of whether the M184V mutation was present at baseline. In the group of patients with the M184V mutation at baseline, the virological outcome was significantly better in those treated with ddI-containing HAART than in those on HAART without ddI ( P<0.05). Conclusions While the M184V did increase the fold resistance of HIV to ddI, these changes appeared to be lower than the clinically relevant threshold for phenotypic resistance for this drug.

BACKGROUND: The clinical utility of phenotypic resistance data can be enhanced by evaluation of the levels of resistance associated with reduced treatment response. To define clinical cutoffs (CCO) for multiple drugs in a consistent manner, we collected resistance, treatment, and virological response data for patients treated with combination regimens. Quantitative phenotypic resistance levels were predicted by VirtualPhenotype™ analysis of viral genotype. METHODS: Data from 11 clinical trials and two cohorts were utilized. Models of virological response as a function of baseline fold change in IC50 (FC) were constructed using linear (change in viral load from baseline), and logistic regression (treatment response rate). Phenotypic susceptibility scores (PSS) of the background regimens, baseline viral load, and FC were included in each model. RESULTS: Records for >13000 patients yielded ~3150 regimens with the required baseline and outcome variables, ranging from 60 regimens including unboosted saquinavir soft gel to 1546 including lamivudine. Median log baseline viral load ranged from 3.8 (regimens with tenofovir) to 4.7 (regimens with boosted indinavir). Median PSS of background regimens was 2 (range 0-7). In a preliminary linear regression analysis, FC values (95% confidence interval) associated with a 20% diminution of virological response at 8 weeks compared to the maximal response were: zidovudine 1.8 (1.5-2.5), lamivudine 1.1 (1.1-1.2), stavudine 1.3 (1.2-1.4), didanosine (extended release) 1.3 (1.2- 1.9), abacavir 1.6 (1.1-2.6), tenofovir 1.2 (1.1-1.5), indinavir 1.2 (1.1-1.9), indinavir/r