CROI 2020 Abstract eBook

Abstract eBook

Oral Abstracts

inflammation and immune activation is not a determinant factor for SIV- related gut dysfunction. Our results also indicate that AGMs’ AIDS-resistance is independent of the CD4+ T cells. 121 INFERIORITY OF SHORT DURATION SOFOSBUVIR-VELPATASVIR FOR RECENT HCV (REACT STUDY) Gail Matthews 1 , Sanjay Bhagani 2 , Marc van der Valk 3 , Jürgen K. Rockstroh 4 , Christine Thurnheer 5 , Arthur Kim 6 , Jordan J. Feld 7 , Julie Bruneau 8 , Edward Gane 9 , Margaret Hellard 10 , Tanya Applegate 1 , Marianne Martinello 1 , Kathy Petoumenos 1 , Gregory J. Dore 1 , for the REACT Study Group 1 Kirby Institute, Sydney, NSW, Australia, 2 Royal Free Hospital, London, UK, 3 Academic Medical Center, Amsterdam, Netherlands, 4 University of Bonn, Bonn, Germany, 5 University Hospital of Bern, Bern, Switzerland, 6 Massachusetts General Hospital, Boston, MA, USA, 7 University Health Network, Toronto, ON, Canada, 8 Centre de Recherche du CHUM, Montreal, QC, Canada, 9 University of Auckland, Auckland, New Zealand, 10 Burnet Institute, Melbourne, VIC, Australia Background: Shortened duration therapy for acute and recently acquired HCV infection has been shown to be highly effective in several small non- randomised studies with direct-acting antiviral agents (DAAs), however guidelines remain conservative in their recommendations with no currently approved regimens for this indication. Methods: The REACT study was an NIH-funded multicentre international, open-label, randomised, phase 4 non-inferiority trial examining the efficacy of short course (6 weeks, Arm A) versus standard course (12 weeks, Arm B) therapy with sofosbuvir/velpatasvir for recently acquired HCV infection (estimated duration of infection <= 12 months). Randomisation was at week 6 and stratified by site and HIV status. The primary endpoint was sustained virologic response at 12 weeks post-treatment (SVR12) reported in the intention-to treat (ITT) population. A total of 250 participants were planned for enrolment and DSMB analysis was scheduled after the first 50 participants in each arm reached SVR12. Following the initial DSMB review, a second review was scheduled following 60 participants through SVR12 in each arm. Results: At second DSMB review, 185 participants were enrolled, 165 had been randomised and 127 were through SVR12; n=65 in Arm A and n=60 in Arm B. Of those randomised, 98%were male, and 72%HIV positive. The predominant genotype was 1 (65%) and median baseline viral load 5.6 log 10 IU/ml. 38%were reinfections. At second review (May 2019), the DSMB recommended study cessation for evidence of inferiority in the short arm (A). By ITT SVR12 was 78% (53/65) in Arm A (95%CI 70-90) versus 95% (57/60) in Arm B (95%CI 86-99) (p=0.021). Relapse occurred in 6/65 participants in Arm A (9%, 95%CI 3-19) versus 0 in Arm B (95% CI 0-6). Median baseline viral load in people with viral relapse was 6.4log 10 IU/ml (range 5.9-7.1 log 10 IU/ml). Two further participants in Arm A had virological failure at end of treatment, one had reinfection, two died, and four lost to follow-up. Three participants in Arm B were lost to follow-up. Conclusion: In this randomised study of treatment for recently acquired HCV infection, 6 weeks sofosbuvir/velpatasvir was inferior to 12 weeks. Final analysis of the full randomised dataset will be completed October 2019. 122 INDIVIDUAL AND POPULATION-LEVEL IMPACT OF HCV TREATMENT AMONG PEOPLE WHO INJECT DRUGS Javier Cepeda 1 , David L. Thomas 2 , Rachel E. Gicquelais 3 , Jacquie Astemborski 3 , Gregory D. Kirk 3 , Shruti H. Mehta 3 1 University of California San Diego, San Diego, CA, USA, 2 Johns Hopkins University School of Medicine, Baltimore, MD, USA, 3 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA Background: The availability of highly curative direct acting antivirals (DAA) has led to calls for elimination of HCV as a public health priority. Specifically, a key target is a 65% reduction in HCV-associated mortality by 2030. Mathematical models have predicted that dramatic scale-up of DAAs can achieve this ambitious target, however this has not been assessed empirically. We evaluate temporal trends in individual and population-level liver disease burden (as a proxy for HCV-related mortality) in a community-based cohort of people who inject drugs (PWID) in Baltimore in the context of expanding DAA use. Methods: From 2005-2018, we collected 11,471 liver stiffness measurements (LSM) using transient elastography from 1,668 PWID who were HCV antibody positive. At the individual-level, we estimated the impact of DAAs on LSM using generalized linear mixed modeling after adjusting for age, sex, race, time, alcohol use and HIV status. To assess population-level impact, we used segmented regression to determine changes in the proportion of the population

with cirrhosis (LSM≥ 12.3 kPa) within and between two periods (before and after 2015) based on presumed scale-up of DAAs in the community. Results: Overall, 69%were male, mean age was 48, 33%were HIV co- infected, and 14% had cirrhosis at baseline. Only 2% reported ever receiving HCV treatment in 2014 which increased to 41% by 2018. After adjusting for confounders, PWID who reported receipt of DAA had significantly lower LSM (-0.88 kPa 95%CI: -0.91 kPa, -0.86 kPa, p<0.0001) compared to PWID who had not received DAAs. HIV/HCV co-infected PWID had significantly higher LSM, especially among those who not HIV virally suppressed (1.14 kPa, 95%CI: 1.08 kPa, 1.20 kPa, p<0.0001) compared to PWID who were HIV negative. At the population-level, we observed a significant rise in the proportion with cirrhosis within the cohort from 2005-2014 (mean log odds of cirrhosis increased each year by a factor of 0.13, p=0.006); however from 2015-2018, the proportion with cirrhosis declined (mean log odds decreased each year by a factor of 0.53, p=0.003, see figure). Conclusion: Expansion of DAAs in the PWID community appears to have led to declines in liver disease progression at the individual and population level. Increasing access to all HCV infected persons, particularly those at greatest risk of liver disease progression will be critical to achieving HCV elimination targets.

Oral Abstracts

123LB LARGE HIV OUTBREAK AMONG PEOPLE WHO INJECT DRUGS, WEST VIRGINIA, 2018–2019 Robert P. McClung 1 , Nivedha Panneer 2 , Amy Atkins 3 , Sheryl Lyss 2 , Senad Handanagic 2 , Mi Chen 2 , Michael Kilkenny 4 , Kyle T.Bernstein 2 , Vicki Hogan 3 , Carolyn Wright 2 , Erica Thomasson 2 , Kara Willenburg 5 , David Wills 3 , Alexandra M. Oster 2 , for the West Virginia HIV Response Team 1 Centers for Disease Control and Prevention, Atlanta, GA, USA, 2 CDC, Atlanta, GA, USA, 3 West Virginia Department of Health and Human Resources, Charleston, WV, USA, 4 Cabell-Huntington Health Department, Huntington, WV, USA, 5 Marshall University, Huntington, WV, USA Background: In January 2019, the West Virginia Bureau for Public Health (BPH) detected an increase in HIV diagnoses among people who inject drugs (PWID) in Cabell County, which has experienced high rates of substance use disorder in recent years. Responding to HIV clusters and outbreaks is one of four pillars of the federal Ending the HIV Epidemic initiative and can be used to guide activities supporting the other pillars (diagnose, treat, prevent). BPH, Cabell-Huntington Health Department, and CDC collaborated to conduct a robust investigation and response. Methods: We analyzed surveillance data, including HIV-1 polymerase data, reported to BPH through November 2019; links were identified at ≤0.005 nucleotide substitutions/site. Outbreak cases were defined as HIV diagnoses during January 1, 2018–October 9, 2019 among 1) PWID linked to Cabell County, 2) their sex or injecting partners, or 3) people with linked sequences. We estimated transmission rate and timing of infections via molecular clock phylogenetic analysis and identified suspected recent infections based on initial viral load and CD4+ cell count, report of last negative HIV test, or presence in a molecular cluster. State, federal, and local partners implemented a comprehensive response. Results: We identified 81 cases, a 2,285% increase above the 2015–2017 annual average of 2 cases. Most people were male (58%), aged 20-39 years (74%), and white (91%). Almost all (99%) were PWID; many (73%) reported unstable housing. In all, 69 (85%) had ≥1 measure of recent HIV infection. Among 45 people with an available HIV-1 sequence, 41 (91%) were in a large molecular cluster with 35/41 (85%) inferred transmissions occurring after January 1, 2018. Estimated transmission rate in the molecular cluster was 78 per 100 person-years. A comprehensive response featured activities from all four pillars (figure). Conclusion: Evidence of rapid transmission in this outbreak—the largest relative increase over baseline in the United States since the large 2015 outbreak in rural Scott County, Indiana—galvanized robust collaboration among

40

CROI 2020