CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

in progressors versus non-progressors were: CPT score 6 versus 5 (p=0.379); MELD score 9 versus 8 (p=0.032); total bilirubin 1.33 mg/dL versus 0.8 mg/dL (p=0.006); INR 1.25 versus 1.09 (p=0.004); albumin 3.5 g/dL versus 4 g/dL (p=0.002) and platelets 68000/L versus 121500/L (p=0.001). Only one out of 44 (2%) patient with >100000 platelets/L and total bilirubin <1 mg/dL at baseline was progressor, being the negative predictive value of the combination of these two parameters for progression 0.98. Conclusions: Although most cirrhotics on DAA-based therapy experience an improvement in liver function tests in the short term, MELD and/or CPT scores worsen in around one third of them. Those with a poorer baseline liver function are more prone to suffer from impairment, and should be carefully monitored while on therapy. The combination of plasma bilirubin and platelet count may accurately predict those who will not have a worsening in liver function scores.

604 Liver Cancer After Hep C Cure: Less Cirrhosis and Less Liver Fat Than Expected Andrea Branch ; Chiara Rocha; M. I. Fiel; Erin Doyle; Nicholas Goosens;Yujin Hoshida;Thomas Schiano; Myron Schwartz Icahn Sch of Med at Mount Sinai, New York, NY, USA Background: Patients who achieve a sustained virological response (SVR) to treatment for chronic hepatitis C (HCV) remain at risk for hepatocellular carcinoma (HCC). Aims: To characterize patients who developed HCC more than 12 mo after achieving SVR. Methods: Demographic, laboratory, and HCC stage and treatment data on 41 cases were obtained frommedical records (1/2010-12/2014). Histology of H&E stained non-tumor tissue was evaluated by blinded review using the Knodell system for necroinflammation (HAI scale, 0-18) and the Scheuer system for stage (scale, 0-4); the specimens were from 17 patients treated with resection or liver transplantation. Patients with HCC detected through surveillance were compared to those diagnosed when symptomatic. Results: HCC was diagnosed a median of 6 yr post-SVR when patients were a median age of 58 yr. Only 71% of the 41 cases had cirrhosis, and only 56% had AFP >10 ng/mL, 83% were male, and 88%were non-Asian. Comorbidities included diabetes (29%), HBV (2%), and HIV (12%). Median tumor size was 2.8 cm (range, 0.8-18.2), 83% had a single lesion on imaging and 51% had vascular invasion. Median laboratory values at the time of HCC diagnosis indicated that liver function was generally well-preserved: albumin 4.2 g/dL (2.1- 5.0), platelets 148 x 10 3 cells/μL (48-446), and total bilirubin 1.3 mg/dL (0.2-8.8). HCC was diagnosed via surveillance (imaging and/or AFP) in 27 patients. Among the surveillance group, 85%were within Milan criteria versus 23% in the symptomatic group (p<0.01). Survival at 1 and 3 yr were 91% and 60% in the surveillance group and 62% and 62% in the symptomatic group (p=0.3). The histology study yielded important insights: Only 10/17 (59%) of the (non-tumor) tissues had stage 4 cirrhosis, whereas 6/17 (35%) had stage 0-2 fibrosis. Median HAI was 4 (range, 0-8), indicating that most had mild necroinflammation. Periportal hepatitis was present in 59%, only two had (mild) steatosis and three had steatohepatitis (5/17 = 29%). Conclusions: Many patients lacked traditional HCC risk factors: 44% had AFP < 10 ng/mL and 40% of 17 histological specimens were non-cirrhotic and 70% lacked liver fat (steatosis/steatohepatitis). Future research should seek novel indicators of persistent HCC risk after HCV cure, e.g., somatic DNA mutations and epigenetic changes. This study did not show a survival benefit to screening; however, a higher percentage of patients diagnosed through surveillance were within Milan criteria. 605 Residual Risk of Disease Progression After Hepatitis C Cure in HIV-HCV Patients Dominique Salmon-Ceron 1 ; Camille Gilbert 2 ; Philippe Sogni 1 ; Francois Dabis 3 ; Julie Bottero 4 ; Patrick Miailhes 5 ; Lionel Piroth 6 ; Firouzé Bani-Sadr 7 ; LindaWittkop 8 ; for the ANRS CO13 Hepavih Study Group 1 INSERM-APHP, Hosp Cochin, Paris, France; 2 INSERM, Bordeaux, France; 3 INSERM U897, ISPED, Univ de Bordeaux, Bordeaux, France; 4 Hôpitaux de Paris, Hôpitaux Universitaires Paris-Sud, Le Kremlin-Bicêtre, France; 5 CHU Lyon, Lyon, France; 6 CHU Dijon, Dijon, France; 7 CHU Reims, Reims, France; 8 CHU de Bordeaux, Bordeaux, France Background: Incidence rates of hepatic events and mortality after cure of HCV in HIV-HCV co-infected patients are not yet documented in real-life settings. Methods: We included HIV-HCV co-infected patients with a sustained virological response (SVR24 – negative HCV-RNA ≥6 months after anti-HCV treatment completion) within the French nationwide ANRS CO13 HEPAVIH cohort. Hepatic events were defined as ascites, digestive hemorrhage, bilirubin level >2.05 mg/mL, hepatic encephalopathy and hepatocellular carcinoma (HCC). All events were validated by a medical committee. We estimated incidence rates (95% confidence intervals [CI]) and mortality rates overall, and separately in patients with severe (F3/F4) and mild/no fibrosis (F0/F1/F2). Crude incidence risk ratios (IRR) were assessed using Poisson regression. Results: We included 324 patients (36%) with SVR24 treatment with peg Interferon + Ribavirin (n=283) or triple therapy with Boceprevir/Telaprevir (n=41) and a median follow-up of 3.6 years. At SVR, median age was 45 years, median CD4 cell count was 450 (IQR 302-624)/mm 3 , median BMI was 22 kg/m² (IQR 20-24), 7% had an excessive alcohol consumption (>3 glasses of alcohol/day for men and >2 for women), 25% had a severe fibrosis (F3 n=31/F4 n=44) and 45% had a HCV genotype 1. Incidence rate of a first hepatic event was 4.2 per 1,000 person years (PY) (CI: 0.8-7.5) occurring 7 months in median after SVR. Patients with severe fibrosis tended to have a higher risk of first hepatic event compared to patients with mild/no fibrosis (IRR: 5.0 CI: 0.7; 26.3, P =0.10). Hepatic events were: HCC (n=2) and decompensation (n=4). At the time of event, median fibrosis score was 10.3 kPa (IQR 9.0; 16.1) and 60% of these patients had an elastometry value >9.5 kPa. Death occurred 5.3 years in median after SVR. Overall mortality rate was 4.8 per 1,000 PY (CI: 1.2-8.3), 4.0 (CI: 0.2-7.2) and 5.0 (CI: 0.6-9.4) per 1,000 PY in patients with severe and mild fibrosis, respectively ( P =0.63). Causes of death were: HCC (n=2), infection (n=2), cardio-vascular disease (n=2) and unknown (n=1). Conclusions: Both hepatic events and death after cure were rare after short/mid-term follow-up. A residual risk of hepatic events after hepatitis C cure in HIV-HCV co-infected patients persisted after cure and patients with severe fibrosis tended to have a higher risk for a hepatic event. Our early findings underline the need for long-term follow-up and are in favor of an early access to anti-HCV treatment.

Poster Abstracts

244

CROI 2016