CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

548 Liver Disease Progression in a Community-Based Sample of HCV-Infected PWID Javier A. Cepeda 1 ; David L.Thomas 2 ; Jacquie Astemborski 1 ; Xiangrong Kong 3 ; Gregory D. Kirk 3 ; Shruti H. Mehta 3 1 Johns Hopkins Bloomberg Sch of PH, Baltimore, MD, USA; 2 Johns Hopkins Univ Sch of Med, Baltimore, MD, USA; 3 Johns Hopkins Univ, Baltimore, MD, USA

Background: Highly efficacious all oral direct-acting antiviral (DAA) therapy is curative against HCV infection. However, because of cost, DAA is denied to many individuals without cirrhosis based on the assumption that liver fibrosis progression can be safely monitored until cirrhosis is detected. We tested that assumption by examining the natural history of HCV infection in a large, largely untreated cohort in whom fibrosis was uniformly monitored. Methods: 1278 HCV-infected people who inject drugs (PWID) were followed from 2006-2014 as part of the ALIVE study in Baltimore, MD. Liver fibrosis was ascertained semi- annually by transient elastography (FibroScan) using previously validated cutoffs (≤8.0: no/mild fibrosis, 8.0-12.3: moderate fibrosis; > 12.3: severe fibrosis/cirrhosis). Cox regression was used to determine the association between fibrosis and mortality. Semi-parametric growth mixture modeling was used to characterize long-term patterns of fibrosis among those with >5 FibroScan scores (N=743). Persons with patterns consistent with progression to cirrhosis were subsequently compared to those with low/stable scores. Results: Median age was 49; 68%were male. Persons with cirrhosis had the highest mortality risk (6.25 per 100 PY); however mortality in persons with moderate fibrosis (3.51 per 100 PY) at baseline was also significantly higher compared to those with no fibrosis (2.04 per 100 PY; p=<0.001 and p=0.004, respectively). After adjustment for age, HIV, alcohol use, and other comorbidities, those with baseline severe fibrosis/cirrhosis (mortality rate ratio [MRR]: 2.04; 95% confidence interval [CI]: 1.43, 2.92) and moderate fibrosis (MRR: 1.39; 95% CI 0.97, 1.99) had higher mortality than those with no fibrosis. Over 8 years, out of the total number of people without cirrhosis, 14.9% had a pattern consistent with progression to cirrhosis. Only high HCV viral load (>6 log 10 IU/mL) differentiated those who progressed versus those with low/stable FibroScan scores across follow-up (OR 5.14, 95% CI 1.79 – 14.77) compared to undetectable HCV viral load. However, high HCV viral load had low prognostic accuracy (ROC=0.63). Conclusions: Over the course of nearly 8 years, liver staging was generally stable, with 10% progressing to cirrhosis. Our data do not support restricting access to HCV treatment based on disease stage because mortality is increased even at moderate stages and progression to cirrhosis cannot be reliably predicted.

549 Protective Effect of Coffee Intake on Mortality of French HIV/HCV-Infected Patients Patrizia Carrieri 1 ; Camelia Protopopescu 1 ; Philippe Sogni 2 ; LindaWittkop 3 ; David Zucman 4 ; Francois Dabis 5 ; Bruno Spire 6 ; Dominique Salmon-Ceron 2 ; for the ANRS HEPAVIH CO13 Study Group 1 INSERM, Marseille, France; 2 INSERM-APHP, Hosp Cochin, Paris, France; 3 CHU de Bordeaux, Bordeaux, France; 4 Hosp Foch, Suresnes, France; 5 INSERM U897, ISPED, Univ de Bordeaux, Bordeaux, France; 6 INSERM UMR 912, Marseille, France Background: HIV-HCV co-infected patients are particularly concerned by liver disease due to immune activation/inflammation, exposure to antiretroviral agents and evolution of HCV co-infection. Polyphenols contained in coffee have several hepato-protective properties. Elevated coffee consumption has already been associated with a reduced risk of insulin resistance and lower levels of liver enzymes in co-infected patients (Carrieri et al. CID 2014, Carrieri et al. JHEPAT 2013). However its association with overall mortality is unknown. This study aimed to investigate the effect of coffee consumption and other behaviors on mortality risk in HIV-HCV co-infected patients. Methods: The ANRS CO13 HEPAVIH cohort is a French nationwide prospective cohort of HIV-HCV co-infected patients with medical and psycho-social/behavioral data collection using self-administered questionnaires at enrolment (M0) and every 12 months thereafter until M60. The present study’s outcome was all-cause mortality reported between M0 and M60. We used a Cox proportional hazards model to study the effect of coffee consumption on mortality. Results: Over a median [IQR] follow-up of 5.0 [3.9-5.8] years, 77 deaths occurred among 1,035 eligible patients, corresponding to a mortality rate [95% CI] of 1.64 [1.31-2.05]/100 person-years. Leading causes of death were: related to HCV (including hepatocellular carcinoma, n=33, 43%), to cancers (unrelated to AIDS or HCV) (n=9, 12%), and to AIDS (n=8, 10%). Elevated coffee consumption (≥3 cups/day) was reported by 26.3% of the patients at M0. It was, as a time-varying covariate, significantly associated with a 50% reduced risk of mortality (HR [95%CI] = 0.5 [0.3-1.0], p=0.045), after adjustment for gender and other time-varying factors as follows: precarious housing, having a steady partner, alcohol and tobacco use, HIV stage, CD4+ cell count ≤200/mm 3 , HCV treatment status (ongoing treatment: HR [95%CI] = 0.9 [0.4-2.1]; treated but not cured: 0.6 [0.3-1.3]; treated and cured: 0.2 [0.1-0.5]; treatment naive: reference). Conclusions: This study indicates a possible protective effect of elevated coffee intake on mortality in HIV-HCV co-infected patients. This association is independent of HIV immunological status and HCV clearance. As this effect may be mediated by coffee compounds having anti-inflammatory and anti-fibrotic properties, these results underline the need of evaluating the benefits of coffee extracts and supplementing dietary intake of other anti-inflammatory compounds in this population. 550 Statin Use and Cirrhosis Progression in an HIV/HCV Coinfected Population Nora T. Oliver 1 ; Christine Hartman 2 ; Jennifer R. Kramer 2 ; Elizabeth Chiao 1 1 Baylor Coll of Med, Houston, TX, USA; 2 Michael E. DeBakey VA Med Cntr, Houston, TX, USA Background: Metabolic risk factors such as diabetes and obesity have previously been described in patients with liver disease who progress to cirrhosis and hepatocellular carcinoma (HCC). Much less is known about the role of hypertension and lipid abnormalities on liver fibrosis. In addition, the role of statins in preventing cirrhosis in HCV has been

Poster Abstracts

216

CROI 2016