Short summary

We will use liver stiffness measurements with transient elastography (TE) to screen 3000 patients from at-risk populations and 1000 participants from the general population for advanced liver fibrosis. At-risk is defined as either (A) a prior or current alcohol overuse for more than 5 years, or (B) presence of the metabolic syndrome with or without concomitant type 2 diabetes mellitus. Our goal is to evaluate the aptitude of TE as a screening tool for advanced liver fibrosis, based on analyses of benefit, harm, detection rate, technical applicability and prognostic potential. Secondly, we aim to compare novel serum markers of liver fibrosis as potential screening tools against TE.

---

Rationale

Liver cirrhosis is the 13th leading cause of years of life lost worldwide and the 7th in Denmark. It causes 1.2 million deaths annually and is one of the leading causes of hospital admissions. Cirrhosis represents a major worldwide health issue due to the high mortality rate in combination with a significant economic burden of the disease on health care systems and advanced social burden on patients and relatives.

Cirrhosis is the final consequence of liver disease, which may be caused by several etiological factors including alcohol, type 2 diabetes mellitus (T2DM), the metabolic syndrome (MetS), obesity or a combination of these factors. These factors are very common worldwide and chronic liver disease is therefore highly prevalent. The global prevalence of non-alcoholic fatty liver disease (NAFLD) is estimated to be 25.2%, and alcohol causes more than 50% of all deaths due to cirrhosis. It is expected that the overall burden of liver disease will increase due to the obesity epidemic and continued high rates of harmful alcohol drinking.

Liver fibrogenesis occurs over decades, but the disease is rarely diagnosed in the pre-cirrhotic stage or in compensated cirrhosis because it remains asymptomatic. Currently, 75% of patients are diagnosed with a decompensating event, most commonly ascites. At this stage of disease, the median survival is only three years.

Timely diagnosis would allow for identification of the etiological factors driving the fibrogenesis and allow specific targeted interventions and timely institution of preventive measures.

The gold standard for diagnosing non-alcoholic steatohepatitis (NASH), that can lead to liver fibrosis, and staging of liver fibrosis is liver biopsy, but liver biopsies have major limitations such as invasiveness, high costs and sampling variability. Given the high prevalence of ALD and NAFLD worldwide, liver biopsies cannot be proposed to all patients. As a consequence there has been a growing interest for alternative non-invasive strategies. Non-invasive methods rely on two approaches: Imaging techniques and serum biomarkers. The reference non-invasive method is transient elastography (TE; FibroScan). TE is based on the physical properties of the liver and measures liver stiffness to evaluate the amount of fibrosis in the liver. We have recently showed that liver stiffness correlates highly with liver fibrosis and has excellent accuracy for the diagnosis of significant alcoholic fibrosis and cirrhosis. Serum biomarkers can be divided in direct and indirect markers. The widely used "liver blood tests" - or indirect markers of liver fibrosis - reflect the liver function, hepatic inflammation, cholestasis or cholangiocyte damage. Direct markers of liver fibrosis reflects the extracellular matrix (ECM) generation and includes the Enhanced Liver Fibrosis test and neoepitope markers of ECM turnover. Elastography is probably superior to serum markers for detecting significant fibrosis and cirrhosis but combining elastography with indirect and/or direct serum markers may increase diagnostic accuracy and is recommended in European guidelines.

Current guidelines provide conflicting recommendations on screening in diabetic patients due to lack of data. An ALD and NAFLD screening program may be in accordance with Wilson and Jungners ten principles of early disease detection. Secondary prevention of ALD and NAFLD would require an assessment of benefit, harms and costs of a screening program of at-risk patients.

The identification of patients in early stages of chronic liver diseases faces two major challenges: (1) Is there a reliable methods for early diagnosis? (2) Can this method be used for screening in at-risk populations and (3) if so; does this improve the patients' prognosis?

We aim to evaluate the aptitude of TE as a screening tool for advanced liver fibrosis based on analyses of benefit, harm, detection rate, technical applicability and prognostic potential. We will use an 8.0 kPa cut-off value to rule out advanced fibrosis.

The objectives of this study are therefore: (1) to screen 3000 patients at risk of ALD or NAFLD and 1000 randomly selected participants from the general population for advanced fibrosis with TE; (2) to investigate benefits and harms of screening for advanced fibrosis in a population-level screening programme, compared to unscreened control cohorts from the Inter99 study and the Copenhagen and Odense alcohol rehabilitation cohorts; (3) in patients with a positive screening test, to investigate the accuracy, prognostic potential and applicability of imaging and serum non-invasive markers to diagnose advanced fibrosis, using liver biopsy as reference.

---

Description of the cohort

Prospective cohort study including 1000 participants at risk of ALD, 2000 participants at risk of NAFLD and 1000 randomly selected participants.

Patients are eligible for screening if the following inclusion criteria are fulfilled:

• Age 30-75 years (except the general population, which should be aged 45-75)
• Informed consent to study investigations
• Ability to read and write Danish

AND (only at-risk patients)

• Prior or current alcohol overuse, defined as an average intake of more than 24 grams/day (14 units/week) for women and more than 36 grams/day (21 units/week) for men, for at least 5 years; OR
• Presence of the metabolic syndrome defined by central obesity plus any two of the following four metabolic risk factors: (a) raised triglycerides, (b) reduced HDL cholesterol, (c) raised blood pressure and (d) raised fasting plasma glucose; OR
• Type 2 diabetes mellitus defined by either fasting plasma glucose more than 7 mmol/L, HbA1c  more than 48 m mmol/mol, a random plasma glucose more than 11.1 mmol/L in the presence of classic diabetes or an oral glucose tolerance test with fasting plasma glucose more than 7.0 mmol/L and/or 2 hour plasma glucose more than 11.1 mmol/L.

We will recruit patients from the Region of Southern Denmark. Potential participants can sign up for the trial at a secure website administered by Odense Patient Data Network at Odense University Hospital. Here they will leave name and contact information and a project worker will contact the patients with further information (for text see appendix). It is also possible to contact the project doctor by telephone or e-mail.

Alcoholic liver disease

• Municipal alcohol rehabilitation centers
• Private alcohol rehabilitation centers
• Family physicians
• Web and paper adverts at sundhed.dk, Google.dk, news outlets, local press, libraries, community centers, train stations and social medias (see appendix for advert text).
• Hospital Emergency Departments
• Referrals with a diagnosis of alcohol overuse, alcoholic liver disease or alcohol withdrawal symptoms at departments of gastroenterology and hepatology, internal medicine, surgery and psychiatry.

Non-alcoholic fatty liver disease

• Family physicians
• Web and paper adverts at sundhed.dk, Google.dk, news outlets, local press, libraries, community centers, train stations and social medias (see appendix for advert text).
• Hospital Emergency Departments
• Patient associations
• Referrals with a diagnosis of type 2 diabetes, the metabolic syndrome, arterial hypertension, obesity, dyslipidemia or hypercholosterolemia to departments of gastroenterology and hepatology, endocrinology, internal medicine, or infertility clinics
• The DD2 database, which contains data on newly diagnosed type 2 diabetes patients in Region of Southern Denmark. Patients in the DD2 database have agreed to receive proposals for participation in research studies (see appendix for the letter).

General population

We will recruit participants from the general population by random-drawing of 5000 personal identification numbers.

Data and biological material

The biobank will contain material from study participants from visit 0, 1 and 2. The material at visit 0 and 2 is 110 ml. blood, 9 ml. urine, 2 ml. saliva and 20 grams of feces. The material from visit 1 is 110 ml blood and 10 mm x 0.4 mm liver tissue.

Questionaires: Food questionaires, Quality of life, AUDIT, Consequences of screening

We will assess clinical outcomes 1, 5 and 10 years after inclusion. Clinical outcomes will be investigated from patient files and national registers (CPR registry, Landspatientregisteret, Cancerregisteret, Dødsårsagsregisteret, Den Nationale Receptdatabase, receptudstedelser i Fælles medicinkort (FMK) og i Sundhedsdatastyrelsens blodprøveregister).

We define clinical outcomes as:

• All cause mortality
• Disease specific mortality: liver related and cardio-vascular
• All cause hospitalizations
• Disease specific hospitalizations: liver related and cardio-vascular
• Infections
• Progression to liver-related complications:
• Ascites
• Hepatic encephalopathy, covert or overt
• Spontaneous bacterial peritonitis
• Variceal bleeding
• Hepatocellular carcinoma
• MELD-Na score more than 15

We will also address progression of liver disease by evidence of clinically significant portal hypertension.

Collaborating researchers and departments

Department of Gastroenterology and hepathology, Odense University Hospital

• Aleksander Krag, MD, PhD, professor of Hepatology
• Maja Thiele, MD, PhD, associate professor
• Linda Møller, MD, PhD, postdoctoral fellow
• Ditlev Rasmussen, MD, PhD student
• Bjørn Stæhr Madsen, MD, PhD student
• Camilla Dalby Hansen, MD-PhD student
• Suganya Ganesalingam, MD, PhD student
• Mads Israelsen, MD, PhD student
• Katrine Thorhauge, pregraduate research fellow
• Thor Lars Hansen, pregraduate research fellow

Unit for Clinical Alcohol Research, Institute for Clinical Research, University of Southern Denmark

• Anette Søgaard Nielsen, MD, PhD, associate professor of psychiatry

Emergency Department (FAM), Odense University Hospital

• Annmarie Touborg Lassen, MD, PhD and professor of emergency medicine

Department of endocrinology, Odense University hospital

• Henning Beck Nielsen, MD, professor of endocrinology
• Kurt Højlund, MD, professor of endocrinology

Department of Clinical Biochemistry and Pharmacology, Odense University Hospital

•  Lars Melholt, professor of clinical biochemistry

Department of Pathology, Odense University Hospital

• Sönke Detlefsen, MD, associate professor

Department of Clinical Biochemistry, Odense University Hospital Svendborg

• Steen Antonsen, chief consultant

Department of Internal Medicine, Esbjerg University Hospital of South-West Jutland

• Mette Munk Lauridsen, MD, PhD, postdoctoral fellow
• Torben Knudsen, MD, chief consultant

The Copenhagen Alcohol Cohort, National Institute of Public Health, University of Southern Denmark

• Ulrich Becker, professor

The Inter99 study, NNF Center for Metabolic Research, University of Copenhagen

• Torben Hansen, professor

The LiverScreen study, Liver Unit, Hospital Clinic, IDIBAPS, University of Barcelona, and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (Ciberehd), Spain

• Pere Ginés, professor

Nordic Bioscience A/S, Denmark

• Morten Karsdal, professor, CEO

VLV Bio, Peviva AB, Sweden

•  Richard Hermann, managing director

Manatee APS, Denmark

•  Kristian Nordahl, CEO

Siemens Healthcare A/S, Denmark

•  Merete Askestad

Department of Pathology, University Paris-Diderot, France

• Pierre Bedossa, professor

The Royal Free Sheila Sherlock Liver Centre, Royal Free Hospital and University College London Institute for Liver and Digestive Health, London, UK

•  Emmanuel Tsochatzis, MD, associate professor

Novo Nordisk Center for Proteine Research and Max Planck Institute of Biochemistry, Department of Proteomics and Signal Transduction, München, Germany

•  Matthias Mann, professor

Steno Diabetes Center, Gentofte, Denmark

• Peter Rossing, professor

Novo Nordisk Center for Basic Metabolic Research, University of Copenhagen

•  Torben Hansen, MD, PhD, professor

Department of Biochemistry and Molecular Biology, University of Southern Denmark

•  Susanne Mandrup, professor

Biomedical Research Foundation, Academy of Athens, Greece

•  Ema Anastasiadou

European Molecular Biology Laboratory, EMBL, University of Heidelberg, Germany 

• Michael Kuhn, postdoctoral fellow

Department for Visceral Surgery and Medicine, University Hospital Bern, Switzerland

•  Reiner Wiest, professor

Laboratory for Fibrosis and Portal Hypertension, Department of Internal Medicine 1, University of Bonn, Germany

•  Jonel Trebicka, adjunct professor

Department of Health Management and Health Economics, University of Oslo, Norway

•  Hans Melberg, associate professor

Department of Health Economics, Universitet Pompeu Fabra, Barcelona, Spain

•  Miquel Serra, Professor

NNF Center for Proteine Research, Disease Systems Biology Program

•  Lars Juhl Jensen, professor

NNF Center for Basic Metabolic Research, University of Copenhagen

• Manimozhian Arumugam, associate professor