Intro: Hepatocellular carcinoma (HCC) is the 6th leading cause of cancer worldwide. In France, more than 10,000 new cases are identified each year. The latter occur in 85% of cases in cirrhosis, the most frequent causes of which are excessive alcohol consumption, metabolic syndrome or HBV/HCV infection. Patients with cirrhosis justify being included in monitoring programs involving the performance of a semi-annual liver ultrasound (US) in order to detect HCC eligible for curative treatment (liver resection or percutaneous ablation). This practice is considered to be cost-effective in the event of an annual incidence of HCC> 1.5%. US in this context has a low sensitivity for the detection of HCC at the very early stage and the following observations have been made in the last 20 years:

• The rate of patients detected at early stage BCLC 0 is around 30% by ultrasound
• The rate of patients included in surveillance programs detected with advanced HCC eligible for palliative treatment is around 20%
• Reducing the periodicity of liver ultrasounds from 6 to 3 months does not improve these results.

In parallel, liver MRI has been evaluated as a tool for the early detection of HCC. Its performance for the detection of HCC at the very early stage exceeds 80%. However, due to the higher cost compared to US, it was estimated that its use in screening context would only be cost effective in the event of an annual incidence> 3%. In addition, the practice of these expensive and long-lasting MRIs (30 to 45 minutes) can be optimized by carrying out abbreviated MRI protocols" or Fast-MRI: short protocols (<10 minutes), based on the sequences with the better detection sensitivities (Se> 83%).

The hypothesis is that Fast-MRI used as a screening examination in patients at high risk of HCC (> 3% per year) could increase the rates of patients detected at an early stage accessible to curative treatment and demonstrate its cost-effectiveness in this population.

Hypothesis/Objective: The main objective is to assess the cost / QALY and / patient detected with an early HCC BCLC 0 (single tumor <2cm) by semi-annual monitoring by liver US and Fast-MRI, compared to conventional semi-annual monitoring by liver US alone in patients with cirrhosis and an anticipated HCC incidence>3%.

Conclusion: If positive, this trial could modify international practice guidelines and set MRI as the optimal tool for early HCC detection in high-risk patients.

Method: This is a randomized controlled, multicenter, 2 parallel arm, superiority trial carried out in patients at high risk of HCC>3%. Patients with cirrhosis of non-viral cause or controlled/eradicated for HBV/HCV infection will be included if their estimated yearly HCC incidence is above 3% according to clinical risk stratification scoring systems previously developed (and published) in French population. Randomization will be individual according to a 1: 1 allocation ratio, centralized and stratified on the center. After inclusion in the trial, each patient will be randomized to be assigned to the experimental group (six-month liver US and fast-MRI) or control (six-month liver US only). At each semi-annual visit, a patient will be considered free from nodules if neither ultrasound nor Fast-MRI detects a nodule. If a nodule is detected by either of the two exams, the patient will undergo a characterization process according to international recommendations, using a combination of injected sectional imaging and/or liver biopsy. The diagnosis of HCC will be definitively assessed in each center during a multidisciplinary consultation meeting. The primary analysis will be carried out by intention to treat.

The rates of BCLC 0 stage HCC will be compared between the two arms. Medico-economic efficiency criterion will be based on an analysis of the different costs from the point of view of the healthcare system and on an analysis of clinical effectiveness in real life and will be supplemented by a budget impact analysis from the point of view of Health Insurance. The time horizon extends from inclusion up to 3 years with an annual update of costs and benefits at 2.5%.Quality of life will be assessed using the EQ-5D5L scale, their variations to the total costs evaluated for each arm will be compared. QALYs will be calculated in each group. The costs and QALYs will be compared for the 2 strategies.

Inclusion Criteria:

• Age ≥ 18 years
• Patient enrolled in a screening program for at least 6 months in a tertiary hepatology center
• Cirrhosis histologically proven or unequivocally suggested by non-invasive tests
• Absence of HCC on imaging less than 3 months o
• Liver parenchyma explorable by ultrasound
• Child-Pugh A or B
• Cirrhosis of non-viral or viral B/C cause controlled/healed
• With an estimated annual risk of HCC>3%
• Written informed consent
• Affiliation to a social security system

Exclusion Criteria:

• Child-Pugh C score
• Active hepatitis B or C
• Estimated annual risk of HCC<3%
• No prior enrollment in a screening program
• Contraindication to Fast-MRI
• Non-echogenic patient
• Patient deprived of liberty
• Patient under legal protection
• Pregnant or breastfeeding woman
• Patient on AME (state medical aid)