1078 ENHANCED LIVER FIBROSIS TEST ACCURATELY IDENTIFIES LIVER FIBROSIS AND PREDICTS CLINICAL OUTCOMES IN ALCOHOLIC LIVER DISEASE

Abstract

Introduction Alcohol is the major cause of chronic liver disease (CLD), the fifth commonest cause of death in the UK. Traditionally, liver fibrosis has been assessed using liver biopsy. The enhanced liver fibrosis (ELF) test is a validated panel of biomarkers of matrix biology. We report the performance of the ELF test in assessing liver fibrosis and predicting liver related clinical outcomes (LRO) (morbidity and mortality attributable to cirrhosis) 7 years after ascertainment in subjects with alcoholic liver disease (ALD). Methods Serum samples were obtained at the time of liver biopsy from 81 patients with a clinical diagnosis of ALD. The ELF test was performed at a central laboratory and liver biopsies were staged by one histopathologist using the Scheuer classification. Clinical outcomes were assessed 7 years (median) after biopsy in all patients by reviewing clinical notes, routine data sources and by contacting primary care physicians. Diagnostic performance of the ELF test for detection of histological stages of liver fibrosis, and in predicting LRO was assessed by calculating the area under the receiver operator characteristic curves (AUROC). Results Median age was 47 years. Biopsies stages were; F0: 14 subjects (17%); F1: 17 (21%); F2: 5 (6%); F3: 17 (21%); F4: 28 (35%). The ELF test demonstrated good performance in identifying fibrosis at all stages (Abstract PWE-273 table 1). ELF predicted LRO at 7 years, AUROC=0.81 (95% CI 0.71 to 0.90). An ELF score ≥9.5 was better at predicting LRO than cirrhosis on biopsy (p=0.002), correctly predicting outcomes in 84% of patients with LRO compared to 55% predicted by biopsy. Conclusion In ALD, the ELF test correlates closely with histological staging conducted by an expert liver pathologist at all stages of fibrosis and performs better than biopsy staging in predicting LRO at 7 years. Competing interests P Trembling: None declared, J Parkes: None declared, S Tanwar: None declared, A Burt: None declared, W Rosenberg Grant/Research Support from: Siemens Healthcare Diagnostics.