CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

539LB Effect of Baseline Resistance-Associated Variants on SVRWith the 3D Regimen Plus RBV Mark Sulkowski 1 ; Preethi Krishnan 2 ; RakeshTripathi 2 ; Gretja Schnell 2 ;Yan Xie 2 ; Daniel Cohen 2 ; RogerTrinh 3 ; Nancy S. Shulman 3 ;Tami Pilot-Matias 2 ; Christine Collins 2 1 The Johns Hopkins Univ, Baltimore, MD, USA; 2 AbbVie, North Chicago, IL, USA; 3 AbbVie Inc, North Chicago, IL, USA Background: The 3D regimen (ombitasvir/paritaprevir/ritonavir [paritaprevir identified by AbbVie and Enanta] and dasabuvir) + RBV, which target HCV NS3, NS5A and NS5B, respectively, is highly effective and is approved in the US and EU for the treatment of HCV GT1 infection. To assess the prevalence of baseline resistance-associated variants (RAVs) and their impact on SVR rate for this regimen, we used next generation sequencing (NGS) to evaluate baseline samples from patients enrolled in the Phase 3 studies of GT1a treatment-experienced (SAPPHIRE-II) and cirrhotic (TURQUOISE-II) patients. Methods: NGS was employed using Illumina MiSeq on baseline samples from 332 GT1a patients (n=118 in TURQUOISE-II; n=214 in SAPPHIRE-II) all of whom received 3D + RBV. Patients who discontinued treatment early for non-virologic reasons were excluded. The prevalence and impact of baseline RAVs in NS3, NS5A, and NS5B were determined using a threshold of NGS detection ranging from 1% to 20%. The impact of baseline RAVs on treatment outcome was determined by comparing SVR rates in patients with or without detectable RAVs. Results: 320/332 (96%) patients in this sample set achieved durable SVR (SVR 48 ). Baseline variants at positions 80, 155, 156 and 168 in NS3; 28, 30, 58, and 93 in NS5A; and 316, 414, 448, 556, and 561 in NS5B were evaluated. Using a NGS threshold of 1%, NS5A RAVs were present in 22% of patients; at a sensitivity of 15% (comparable to population sequencing), ≥1 NS5A RAVs were present in 38 (11%) patients. Using the prevalence data at the 15% NGS threshold, similar SVR rates were seen in patients with or without ombitasvir-specific RAVs (95% vs. 97%). RAVs in NS3 and NS5B were rare (1 and 2%, respectively). All 3 patients with a baseline NS3 RAV achieved SVR, while an NS5B RAV was seen in 1 of the 12 virologic failure patients. The highly prevalent Q80K polymorphism in NS3 was assessed and had no impact on SVR rate (96%with Q80K vs. 97%without Q80K). Conclusions: As most current HCV therapies include an NS5A inhibitor, the high prevalence of baseline NS5A RAVs is of particular importance. Patients treated with the 3D regimen + RBV achieved similarly high SVR rates, regardless of presence or absence of baseline NS5A RAVs. 540 HCV IgG Antibody Avidity as a Biomarker to Estimate Population-Level Incidence Eshan U. Patel 1 ; Andrea L. Cox 2 ; Shruti H. Mehta 3 ; Denali Boon 3 ; Andrew D. Redd 4 ;William Osburn 2 ; Gregory D. Kirk 3 ; David L.Thomas 2 ;Thomas C. Quinn 2 ; Oliver Laeyendecker 1 1 NIH, Bethesda, MD, USA; 2 Johns Hopkins Univ Sch of Med, Baltimore, MD, USA; 3 Johns Hopkins Univ, Baltimore, MD, USA; 4 NIAID, NIH, Bethesda, MD, USA Background: Sensitive methods to estimate population-level incidence are needed to monitor the trajectory of the HCV epidemic. We demonstrate how HCV IgG antibody avidity, a biomarker associated with recent infection, can be used to estimate population level HCV incidence. Methods: From 568 people who inject drugs and are enrolled in prospective cohort studies in Baltimore, Maryland, we tested 997 serum or plasma samples [n=233 collected <2 years post-HCV seroconversion and n=764 collected >2 years post-HCV seroconversion]. HCV IgG antibody avidity was determined using a modified Ortho 3.0 HCV ELISA test system. Avidity-based testing algorithms were evaluated by their: (1) mean duration of recent infection (MDRI), the average time an individual is identified as recently infected by a given testing algorithm, (2) false recent rate, the proportion of samples collected >2 years post-HCV seroconversion misclassified as ‘recent’, (3) sample size needed to precisely estimate incidence in various simulated epidemics (relative standard error <30%), and (4) power to detect a 50% reduction in incidence between serial cross-sectional surveys. Results: A multi-assay algorithm consisting of an avidity index <30% followed by detection of HCV viremia (RNA>500 IU/mL) had an MDRI of 147 days (95%CI: 125-195), and a false recent rate of 0.7% (95%CI: 0.2-1.8) and 7.6% (95%CI: 4.2-12.3) among HIV negative and positive persons, respectively. In several simulated high-risk populations (HCV incidence >10%), this multi-assay algorithm required less than 1000 individuals to estimate incidence. This multi-assay algorithm also had >80% power to detect a 50% reduction in HCV incidence in various epidemic scenarios where baseline incidence was >20% (see Figure 1; N=1000-2000). Avidity-based algorithms consistently required smaller sample sizes than current approaches (viremic detection among seronegative individuals). Conclusions: Integrating HCV avidity-based testing algorithms into current surveillance systems may improve the capacity to accurately estimate incidence among high-risk populations. This methodology may serve as an effective tool to rapidly assess the impact of interventions. Algorithms to improve performance in HIV coinfected individuals should be explored.

Poster Abstracts

541 HCV Phylogenetic Analysis Complements HIV Contact Tracing in an HIV Outbreak Setting Romeo R. Galang 1 ; Jessica Gentry 2 ; Sumathi Ramachandran 1 ; Kathleen A. Backfish 2 ; Sara J. Blosser 2 ; Anupama Shankar 1 ; Pamela Pontones 2 ;William Switzer 1 ; Philip Peters 1 ;Yury Khudyakov 1 1 CDC, Atlanta, GA, USA; 2 Indiana State Dept of Hlth, Indianapolis, IN, USA Background: In January 2015, a cluster of HIV infections was detected in rural Indiana among persons who reported injecting the prescription opioid oxymorphone. We conducted molecular analyses of HCV sequences from a subset of patients co-infected with HIV/HCV in this outbreak investigation.

211

CROI 2016