Baseline Demographic Characteristics Subgroup Analysis of Metabolic Parameters using Paired t-test Tx: Treatment; IFN: Interferon; *** p<0.001; ** p<0.01; * p<0.05 1016 All-Oral 12-Week Combination Treatment With Daclatasvir (DCV) and Sofosbuvir (SOF) in Treatment-Experienced Patients Infected With HCV Genotype (GT) 3: A Subanalysis of the Ally-3 Phase 3 Study David Nelson, David Bernstein, Bradley Freilich, Eric Lawitz, Trevor Hawkins, Paul J. Pockros, Paul J. Thuluvath, Ziad H. Younes, Michael Bennett, Reem Ghalib, Peter J. Ruane, Myron J. Tong, Rafia Bhore, Philip D. Yin, Stephanie Noviello, Khurram Rana Background: The phase 3 ALLY-3 study evaluated the all-oral, ribavirin (RBV)-free combina- tion of daclatasvir (DCV; pangenotypic NS5A inhibitor) and sofosbuvir (SOF; NS5B polymer- ase inhibitor) in patients with GT3 infection. After 12 weeks of treatment, sustained virologic response at posttreatment Week 12 (SVR12) was achieved by 90% and 86% of treatment- naive and -experienced patients, respectively. Methods: Treatment-naive (N=101) and expe- rienced (N=51) patients received open-label DCV 60 mg + SOF 400 mg once daily for 12 weeks. This subanalysis provides further details of efficacy and safety outcomes in the S-1003 AASLD Abstracts experienced cohort. Results: Treatment-experienced patients were predominantly male (63%), white (88%) and non-cirrhotic (67%); 75% had baseline HCV RNA ≥800K IU/mL, and 61% had non-CC IL28B genotypes. Patients had previously received IFN-based (n=42), SOF-containing (n=7, including 7 treated with SOF/RBV and 1 who was retreated with SOF/peg/RBV) and alisporivir-containing (n=2) regimens. Prior responses included relapse (n=31), null response (n=7), partial response (n=2), and other forms of nonresponse or IFN intolerance (n=11). All patients completed 12 weeks of study treatment. SVR12 was achieved in 44 patients (86%); all prior null and partial responders, and IFN-intolerant patients achieved SVR12. SVR12 rates were higher in patients without cirrhosis and in those with IL28B CC genotype (Table). Treatment failure (relapse) was experienced by 7 patients, including 5 prior IFN/RBV recipients (prior response: relapsers, n=4; HCV RNA never undetectable, n=1) and 2 patients who relapsed after prior treatment with SOF/RBV. Of the two prior SOF relapsers, one had cirrhosis with grade 1 steatosis, and one had Fibrotest F3 with grade 2 steatosis, and a baseline NS5A-Y93 resistance-associated variant. There were no serious AEs or AEs leading to discontinuation. Grade 3/4 AEs (a single report of arthralgia) and grade 3/4 lab abnormalities (platelets, n=1; lipase, n=1) were uncommon. The most frequent AEs (any grade) were fatigue (26%), headache (20%), nausea (14%) and arthralgia (12%). Safety parameters were similar in those with or without cirrhosis. Conclusion: This all-oral, 12-week combination of DCV+SOF achieved high SVR12 rates in GT3 patients previously treated with all-oral DAA or IFN-containing regimens. DCV+SOF was well toler- ated. Sa1000 Natural History and Risk Factors of Carcinogenesis in the Iso-Enhanced Lesions on the Postvascular-Phase Sonograms in Chronic Liver Diseases Takayuki Kondo, Hitoshi Maruyama, Soichiro Kiyono, Tadashi Sekimoto, Osamu Yokosuka Aims: Postvascular-phase iso-enhanced hepatic lesions (PILs) on contrast-enhanced ultra- sound (CEUS) comprise several kinds of tumors, regenerative nodule, dysplastic nodule and well-differentiated hepatocellular carcinoma (HCC). Although care should be taken in the management of such lesions in chronic liver diseases (CLD) because of the malignant potential, performing biopsy for all of them is totally impractical. This study examined the natural history of PILs on CEUS to determine the potential risk and predictive factors for developing HCC in CLD. Methods: This prospective study included 94 PILs (hepatic lesions < 30 mm in size and up to three nodules per patient) on CEUS (postvascular-phase: 10 min post-injection of perflubutane microbubbles) in 79 patients with CLD (48 males and 31 females; age, 65.0 ± 10.8 years; PIL diameter, 12.3 ± 4.1 mm [range, 5-26.5]). The following PILs were excluded from this study: PILs diagnosed as typical hemangioma or focal nodular hyperplasia based on imaging findings (CEUS/ computed tomography [CT]/ magnetic resonance imaging [MRI]) and PILs diagnosed as HCC by CT, MRI or biopsy at the time of enrolment. The PILs were followed-up by ultrasound/CEUS, CT, or MRI at 3 to 6 months intervals. The primary endpoint was imaging-based detection of HCC derived from PILs or other area within the liver. The median observation period was 22.0 months (range, 3.3-53.1) Results: 1. Development of HCC: Twenty patients developed HCCs during the study period; a single lesion was detected in 11 patients, two lesions in 2 patients, and three or more lesions in 7 patients. The mean diameter of HCC at the time of detection/ diagnosis was 15.3 ± 4.1 mm (range, 10.0-28.6). The cumulative risk of HCC occurrence was 7.4% at 1 year and 33.6% at 3 years. The presence of coexistent HCC (hazard ratio [HR], 5.056; 95% confidence interval [CI], 1.959-13.047; P = 0.001), and alpha-fetoprotein > 20 ng/ml (HR, 5.219; 95% CI, 1.791-15.207; P = 0.003) were significant factors for the risk of HCC occurrence. The median observation time between HCC detection and the last imaging session was 4.0 months (range, 2.3-6.3). 2. Natural course of PILs: Fourteen lesions were diagnosed as HCCs that developed from PILs. Cumulative HCC occurrence rates from PIL > 14 mm which was the best cut-off value for HCC occurrence by receiver operating characteristic curve analysis was 23.5% at 1 year and 46.3% at 3 years. Cox regression analysis showed that PIL > 14 mm (HR, 6.563; 95% CI, 2.013-21.397; P = 0.002), and alpha-fetoprotein > 20 ng/ml (HR, 4.615; 95% CI, 1.491-14.281; P = 0.008) were statistically significant factors for HCC occurrence. Conclusions: Patients with coexistent HCC, alpha- fetoprotein > 20 ng/ml, or PIL > 14 mm should be carefully monitored in patients with CLD accompanied by PILs because of the high potential for HCC occurrence. Sa1001 Magnetic Resonance Elastography Is Superior to Clinical Prediction Models for Determination of Advanced Fibrosis in Patients With Non-Alcoholic Fatty Liver Disease: A Prospective Study Jeffrey Y. Cui, Tanya Wolfson, Brandon Ang, William M. Haufe, Carolyn Hernandez, Elizabeth C. Verna, Claude Sirlin, Rohit Loomba Background and aims: Two-dimensional magnetic resonance elastography (2D-MRE), a novel magnetic resonance method for assessing liver stiffness, has been shown to have high diagnostic accuracy for predicting advanced fibrosis (stage 3 and 4 fibrosis) in patients with non-alcoholic fatty liver disease (NAFLD). However, head-to-head comparisons between 2D-MRE and clinical prediction rules using laboratory and demographic data have not yet been performed in a prospective cohort of patients with biopsy-proven NAFLD. The aim of this study was to compare the diagnostic accuracy of 2D-MRE with that of 8 previously defined clinical prediction rules (AST:ALT ratio, APRI, BARD, FIB-4, NAFLD Fibrosis Score, Bonacini cirrhosis discriminant score, Lok Index, and NASH CRN model) for predicting AASLD Abstracts