CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

543 Tracing the Origin of HCV NS3 Q80K Among HIV-Infected MSM in the Netherlands

Astrid M. Newsum 1 ; Cynthia K. Ho 2 ; Faydra I. Lieveld 3 ;Thijs J. van de Laar 4 ; JanT. van der Meer 5 ; Anne M.Wensing 3 ; Greet J. Boland 3 ; Joop E. Arends 3 ; Maria Prins 1 ; Janke Schinkel 2 1 PH Service of Amsterdam, Amsterdam, Netherlands; 2 Academic Med Cntr, Amsterdam, Netherlands; 3 Univ Med Cntr Utrecht, Utrecht, Netherlands; 4 Sanquin Blood Supply Fndn, Amsterdam, Netherlands; 5 Cntr of Infectious Diseases and Immunology Amsterdam, Academic Med Cntr, Amsterdam, Netherlands Background: The naturally occurring Q80K polymorphism in the nonstructural protein 3 (NS3) of hepatitis C virus (HCV) has been associated with a reduced response to simeprevir (a ‘second wave’ first generation HCV protease inhibitor) / pegylated-interferon/ribavirin triple therapy. This polymorphism is transmissible between hosts, and its prevalence varies geographically and per risk group. We describe the prevalence of Q80K among the major HCV risk groups in the Netherlands, e.g. HIV-1 coinfected men who have sex with men (MSM) and people who inject drugs (PWID). Using phylogenetic analysis, we examined whether the presence of Q80K was linked to specific HCV transmission networks. Methods: Stored blood samples from 150 HCV genotype 1a infected patients attending two large Dutch medical centers, AMC Amsterdam and UMC Utrecht, were used for this study. Data on transmission route and HIV-status were extracted from patient records. At the time of sampling, all patients were treatment-naïve for NS3/4A protease inhibitors. A 611 bp fragment of the NS3 genomic region including the Q80 amino acid position was amplified and sequenced. The NS3 maximum likelihood (ML) phylogeny was reconstructed using MEGA v6 with the HKY+G substitution model and 1000 bootstrap replicates. Results: Of the 150 patients, 45%was coinfected with HIV-1, 39%was MSM (all HIV-1 coinfected), 17% PWID, 14% had other risk factors including blood transfusion and for 30% the route of transmission was unknown. The Q80K polymorphismwas present in 35% of patients and throughout the time span of sample collection (2000-2015). Q80K was most prevalent among MSM (52%), followed by persons with other or unknown risk factors (30%) and PWID (8%). Robust clustering in the ML phylogenetic tree (figure 1) was only observed for MSM; 5 clusters supported by high bootstrap values were identified. Q80K was present in 100% of sequences in 3 out of these 5 clusters. The largest cluster included 17 patients. Interestingly, sequences did not cluster according to treatment center. Conclusions: The Q80K polymorphism naturally occurs in 35% of our study population and has persisted over at least 15 years. Among HIV-1 coinfected MSM, the prevalence of Q80K was highest and distinct transmission networks with and without Q80K were identified. This suggests a founder effect, with the introduction and expansion of Q80K variants in this key population which potentially jeopardizes future treatment with simeprevir of HIV-1 coinfected MSM.

Poster Abstracts

213

CROI 2016