INTEGRATIVE HEPATIC TRANSCRIPTOME ANALYSIS IDENTIFIES BIOMARKERS OF NON-ALCOHOLIC STEATOHEPATITIS ACTIVITY AND FIBROSIS
PROGRESSION
Olivier Govaere1, Simon J Cockell2, Dina Tiniakos1,3, Ramy Younes1,4, Michele Vacca5, Federico Ravaioli1,6, Jeremy Palmer1, Salvatore Petta7, Jerome Boursier8, Chiara Rosso4, Katherine Johnson1, Christopher P. Day1, Rebecca Darlay9, Heather J Cordell9, Fabio Marra10, Antonio Vidal-Puig5, Pierre Bedossa1,11, Jörn Schattenberg12, Karine Clement13, Mike
Allison14, Elisabetta Bugianesi4, Vlad Ratziu15, Ann Daly1 and Quentin M. Anstee1
(1)Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, United Kingdom, (2) Bioinformatics Support Unit, Newcastle University, Newcastle upon Tyne, United Kingdom, (3)Department of Pathology, Aretaieion Hospital, National & Kapodistrian University of Athens, Greece, (4)Department of Medical Sciences, University of Turin, Turin, Italy, (5)University of Cambridge Metabolic Research Laboratories, Wellcome-MRC Institute of Metabolic Science, Addenbrooke’s Hospital, Cambridge, United Kingdom, (6)Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy, (7)Sezione Di Gastroenterologia, Dipartimento Biomedico Di Medicina Interna e Specialistica, Università Di Palermo, Palermo, Italy,
(8)Hepatology Department, Angers University Hospital, Angers, France, (9)Institute of Genetic Medicine, Newcastle University, Newcastle upon Tyne, United Kingdom, (10) Dipartimento Di Medicina Sperimentale e Clinica, University of Florence, Florence, Italy, (11)Assistance Publique-Hôpitaux De Paris, Hôpital Beaujon, University Paris-Diderot, Paris, France, (12)Department of Medicine, University Hospital Mainz, Mainz, Germany, (13)Institute of Cardiometabolism and Nutrition, Pitié-Salpêtrière Hospital, Paris, France, (14) Liver Unit, Department of Medicine, Cambridge Biomedical Research Centre, Cambridge University NHS Foundation Trust, United Kingdom, (15)Hôpital Pitié Salpêtrière, Paris, France.
Background: Non-alcoholic fatty liver disease (NAFLD) is a spectrum including ‘simple’ steatosis (non-alcoholic fatty liver, NAFL) and non-alcoholic steatohepatitis (NASH) with varying degrees of fibrosis. The mechanisms that drive disease progression and the pathogenic transitions that occur remain poorly understood This multicentre study aimed to unravel the pathogenesis of NAFLD using an integrative transcriptomics approach.
Methods: 436 liver biopsies comprising 414 NAFLD samples and 22 controls were included in this study Histological semi-quantitative NAS and SAF scores were applied to grade and stage disease A discovery cohort comprising 206 NAFLD and 10 control samples underwent high-throughput RNA sequencing Results were validated in a further 175 NAFLD and 12 control cohort using Nanostring and 33 NAFLD using immunohistochemistry A proof-of-principal analysis was conducted using logistic regression models and tested in serum samples to explore translation to putative biomarker.
Results: Sequencing revealed that NAFLD was characterised by concomitant dysregulation of several pathogenic pathways occurring early in NAFL, including lipid metabolism (e g ELOVL1, FADS2, SCD, SREBF1), insulin signalling (IGF1, IGFBP1, GCK), bile secretion (CYP7A1), apoptosis (e g FOS, IKBKG/NEMO, TRAF2), senescence (e g CDKN1A/p21) and autophagy (e.g. SQSTM1/p62, ATG9A) Using NAFL and NASH F0/1 as baselines, we identified 25 genes
consistently differentially expressed compared to NASH F2, F3 or F4 Predictive modelling identified increased levels of ANKRD29, CLIC6 and STMN2 expression together with T2DM as independent predictors for advanced fibrosis in both cohorts. HSD17B14 was an independent protective variable for NAS≥4, while GDF15 expression independently predicted SAF activity≥2.
Patients with a SAF activity score ≥2 showed a 1.37 fold increase in GDF15 serum levels compared to patients with a score < 2 (p<0 001).
Conclusion: The current study provides important insights into the evolution of NAFLD and progression from NAFL to NASH/fibrosis and provides proof of principal that transcriptomic changes may correlate to potentially tractable biomarkers.
