Improved darunavir genotypic mutation score predicting treatment response for patients infected with HIV-1 subtype B and non-subtype B receiving a salvage regimen.

OBJECTIVES: The objective of this study was to improve the prediction of the impact of HIV-1 protease mutations in different viral subtypes on virological response to darunavir. METHODS: Darunavir-containing treatment change episodes (TCEs) in patients previously failing PIs were selected from large European databases. HIV-1 subtype B-infected patients were used as the derivation dataset and HIV-1 non-B-infected patients were used as the validation dataset. The adjusted association of each mutation with week 8 HIV RNA change from baseline was analysed by linear regression. A prediction model was derived based on best subset least squares estimation with mutational weights corresponding to regression coefficients. Virological outcome prediction accuracy was compared with that from existing genotypic resistance interpretation systems (GISs) (ANRS 2013, Rega 9.1.0 and HIVdb 7.0). RESULTS: TCEs were selected from 681 subtype B-infected and 199 non-B-infected adults. Accompanying drugs were NRTIs in 87%, NNRTIs in 27% and raltegravir or maraviroc or enfuvirtide in 53%. The prediction model included weighted protease mutations, HIV RNA, CD4 and activity of accompanying drugs. The model's association with week 8 HIV RNA change in the subtype B (derivation) set was R(2) = 0.47 [average squared error (ASE) = 0.67, P < 10(-6)]; in the non-B (validation) set, ASE was 0.91. Accuracy investigated by means of area under the receiver operating characteristic curves with a binary response (above the threshold value of HIV RNA reduction) showed that our final model outperformed models with existing interpretation systems in both training and validation sets. CONCLUSIONS: A model with a new darunavir-weighted mutation score outperformed existing GISs in both B and non-B subtypes in predicting virological response to darunavir.

Patterns of drug resistance among newly diagnosed HIV-1 infected patients in Greece during the last decade: the crucial role of transmission networks.

INTRODUCTION: The prevalence of drug resistance is approximately 10% in Europe and North America among newly infected patients. We aim to investigate the temporal patterns of resistance among drug naive HIV-infected individuals in Greece and also to determine transmission networking among those with resistant strains. MATERIALS AND METHODS: Protease (PR) and partial reverse transcriptase (RT) sequences were determined from 2499 newly diagnosed HIV-1 patients, in Greece, during 2003-2013. Genotypic drug resistance was estimated using the HIVdb: Genotypic Resistance Interpretation Algorithm. We identified transmission clusters of resistant strains on the basis of a large collection of HIV-1 sequences from 4024 seropositives in Greece. Phylodynamic analysis was performed using a Bayesian method. RESULTS: We estimated drug resistance levels among naïve patients on the basis of all resistance mutations in PR and partial RT. The overall prevalence of resistance was 19.6% (490/2499). Resistance to NNRTIs was the most common (397/2499, 15.9%) followed by PIs (116/2499, 4.6%) and NRTIs (79/2499, 3.2%). We found a significant trend for decreasing resistance to NRTIs over time (6.7%-1.6%). There was no time trend for the overall PI and NNRTI resistance. The most frequently observed major resistant sites in PR were V82 (2.0%) and L90 (1.8%). In RT, we found E138 (58.6%), K103 (13.1%), V179 (8.4%) and T215 (7.1%), M41 (4.7%) associated with resistance to NNRTIs and NRTIs, respectively. The prevalence of K103N and E138Q were significantly increased during 2003-2013. Crucially, we found that both K103N, E138Q are associated with transmission networking within men having sex with men (MSM) and intravenous drug user (IDU) local networks. The K103N network included seropositives across Greece, while the latter only from the recent IDU outbreak in Athens metropolitan area (1). Phylodynamic analyses revealed that the exponential growth for K103N network started in 2009 (Figure 1) and for the E138Q in 2010. CONCLUSIONS: The overall resistance has been stable in Greece over time; however, specific NNRTI resistance patterns are increasing. Notably, they are associated with local transmission networking, thus suggesting that this is the cause for the increased patterns of NNRTI resistance and not multiple transmissions of resistant strains from different sources among treated individuals. Our study highlights the advance of molecular epidemiology for understanding the dynamics of resistance.

Effect of HIV type 1 subtype on virological and immunological response to combination antiretroviral therapy: evidence for a more rapid viral suppression for subtype A than subtype B-infected Greek individuals.

Whether response to combination antiretroviral therapy (cART) differs between those infected with HIV-1 subtype A or B remains unclear. We compared virological and immunological response to cART in individuals infected with subtype A or B in an ethnically homogeneous population. Data derived from the Athens Multicenter AIDS Cohort Study (AMACS) and analysis were restricted to those of Greek origin. Time to virological response (confirmed HIV-RNA <500 copies/ml) and time to failure (>500 copies/ml at any time or no response by month 6) were analyzed using survival models and CD4 changes after cART initiation using piecewise linear mixed effects models. Of the 571 subjects included in the analysis, 412 (72.2%) were infected with subtype B and 159 (27.8%) with subtype A. After adjusting for various prognostic factors, the rate of virological response was higher for those infected with subtype A versus B (adjusted HR: 1.35; 95% CI: 1.08-1.68; p=0.009). Subtype A was also marginally associated with a lower hazard of virological failure compared to subtype B (HR=0.73; 95% CI: 0.53-1.02; p=0.062). Further adjustment for treatment adherence did not substantially changed the main results. No significant differences were observed in the rates of CD4 increases by subtype. The overall median (95% CI) CD4 increase at 2 years of cART was 193 (175, 212) cells/µl. Our study, based on one of the largest homogeneous groups of subtype A and B infections in Europe, showed that individuals infected with subtype A had an improved virological but similar immunological response to cART compared to those infected with subtype B.

Increasing prevalence of HIV-1 subtype A in Greece: estimating epidemic history and origin.

BACKGROUND: In North America and Europe, human immunodeficiency virus (HIV)-1 infection has typically been dominated by subtype B transmission. More recently, however, non-B subtypes have been increasingly reported in Europe. METHODS: We analyzed 1158 HIV-1-infected individuals in Greece by DNA sequencing and phylogenetic analyses of protease and partial reverse-transcriptase regions. RESULTS: We found that the prevalence of non-B subtypes has increased over time and that this significant trend can be mainly attributed to subtype A, which eventually surpassed subtype B in prevalence in 2004 (42% and 33%, respectively). Multivariate analysis revealed that the year of HIV diagnosis was independently associated with subtype A infection (odds ratio for being infected with subtype A for a 10-year increase in the time period of diagnosis, 2.09 [95% confidence interval, 1.36-3.24]; P<.001). Phylogenetic analysis revealed that the subtype A epidemic in Greece is the result of a single founder event. The date of the most recent common ancestor of the subtype A in Greece was estimated to be 1977.9 (95% highest posterior density interval, 1973.7-1981.9). CONCLUSIONS: Subtype A circulates among the long-term residents of Greece. This is in contrast to the situation in most European countries, in which infection with non-B genetic forms is associated either with being an immigrant or heterosexual or with intravenous drug use.

Evaluation of the clinical sensitivity for the quantification of human immunodeficiency virus type 1 RNA in plasma: Comparison of the new COBAS TaqMan HIV-1 with three current HIV-RNA assays--LCx HIV RNA quantitative, VERSANT HIV-1 RNA 3.0 (bDNA) and COBAS AMPLICOR HIV-1 Monitor v1.5.

The COBAS TaqMan HIV-1 test (Roche Diagnostics) was compared with the LCx HIV RNA quantitative assay (Abbott Laboratories), the Versant HIV-1 RNA 3.0 (bDNA) assay (Bayer) and the COBAS Amplicor HIV-1 Monitor v1.5 test (Roche Diagnostics), using plasma samples of various viral load levels from HIV-1-infected individuals. In the comparison of TaqMan with LCx, TaqMan identified as positive 77.5% of the 240 samples versus 72.1% identified by LCx assay, while their overall agreement was 94.6% and the quantitative results of samples that were positive by both methods were strongly correlated (r=0.91). Similarly, in the comparison of TaqMan with bDNA 3.0, both methods identified 76.3% of the 177 samples as positive, while their overall agreement was 95.5% and the quantitative results of samples that were positive by both methods were strongly correlated (r=0.95). Finally, in the comparison of TaqMan with Monitor v1.5, TaqMan identified 79.5% of the 156 samples as positive versus 80.1% identified by Monitor v1.5, while their overall agreement was 95.5% and the quantitative results of samples that were positive by both methods were strongly correlated (r=0.96). In conclusion, the new COBAS TaqMan HIV-1 test showed excellent agreement with other widely used commercially available tests for the quantitation of HIV-1 viral load.

Comparison of three current viral load assays for the quantitation of human immunodeficiency virus type 1 RNA in plasma.

The LCx HIV RNA quantitative assay (Abbott Laboratories, Delkenheim, Germany) was compared with the Versant HIV-1 RNA 3.0 (bDNA) assay (Bayer, Tarrytown, NY) and the COBAS Amplicor HIV-1 Monitor v1.5 test (Roche Diagnostics, Branchburg, NJ), using plasma samples of various viral load levels from HIV-1-infected patients. Considering the lower limit of the linear range of 50 copies/ml of both assays, the detection range of the LCx was 127/151 (84.1%) versus the 131/151 (86.8%) of the bDNA 3.0 assay, while overall agreement between the two assays was 93.4% (141/151). LCx and bDNA 3.0 results were found to be strongly correlated (r = 0.96). The fitted regression line was described by the equation log10(LCx copies/ml) = 0.05 + 1.06 x log10(bDNA 3.0 copies/ml) with 95% CI for the estimated slope and intercept at 1.01, 1.12 and -0.16, 0.26, respectively. Similarly, the detection range of the LCx was 115/148 (77.7%) versus the 128/148 (86.5%) of the Monitor v1.5 test. A 91.2% concordance (135/148) was observed between these two assays at a cut-off of 50 copies/ml. LCx and Monitor v1.5 results were highly correlated (r = 0.96). The fitted regression line was described by the equation log10(LCx copies/ml) = 0.06 + 1.03 x log(10)(Monitor v1.5 copies/ml) with 95% CI for the estimated slope and intercept at 0.97, 1.09 and -0.16, 0.28, respectively.