Skip to content

Artificial intelligence-based clinical decision support tool for personalized HCC risk stratification

Artificial intelligence-based clinical decision support tool for personalized HCC risk stratification - ARCTIC

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
DRKS
Registry ID
DRKS00035975
Enrollment
340
Registered
2025-01-22
Start date
2025-02-01
Completion date
Unknown
Last updated
2025-04-07

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

C22.0

Interventions

Group 1: Volunteers in the TRAINING cohort: Patients =18 years old, which have alcohol-related or NAFLD-induced cirrhosis (plus/minus HCC), and have an indication for liver transplantation or resectio

Sponsors

Deutsche Krebshilfe
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: TRAINING cohort: Patients can be included if they: - have given written informed consent for participation in the study - have liver cirrhosis as proven by radiologic criteria and FS =12 kPa, stratified 1:1 by NAFLD or alcohol consumption as most probable causes - have an indication for liver transplantation or resection. - have been screened for the TRAINING cohort and receive a screening MRI as part of the standard preoperative diagnostic algorithm. - present non-classifiable hepatic nodules (LI-RADS -2/-3) - have biopsies successfully be taken intraoperatively from the imaging-predefined nodules as well as surrounding nodules not prominent on imaging (up to 5 LI-RADS -2/-3 nodules and up to 5 surrounding nodules) SCREENING cohort: Patients can be included if they: - have given written informed consent for participation in the study - have compensated liver cirrhosis due to either alcohol-related liver disease or MASLD - meets risk criteria for the presence of unclassified nodules including at least one of: o liver stiffness =12 kPa by elastography o inhomogeneous multinodular liver structure on standard HCC surveillance ultrasonography or lack of assessability by ultrasonography due to obesity o diabetes mellitus o presence of cirrhosis over =3 years, o AFP levels between 6 and 14 ng/mL - present in gadoxetic acid-enhanced multiparametric liver MRI with at least one liver nodule = 1 cm fulfilling LI-RADS -2/-3 criteria - have an estimated life expectancy of = 5 years. - have successfully undergone MRI-guided biopsy of the nodule(s) as well as a biopsy from the liver tissue surrounding the nodule

Exclusion criteria

Exclusion criteria: TRAINING cohort: Patients will be excluded if they: - have a contraindication for undergoing MRI have liver disease other than alcohol- or NAFLD-related liver cirrhosis - have no histologic signs of cirrhosis in the explanted or resected liver tissue - have HCC or other liver malignancies in the liver - have not given written informed consent for participation in the study - pregnancy or lactation, or intention of becoming pregnant during study treatment. The ß-HCG value must be determined at each visit, unless the woman is menopausal SCREENING cohort: Patients will be excluded if they: - have a contraindication for undergoing MRI - have liver disease other than alcohol- or NAFLD-related liver cirrhosis - fail to receive biopsy of at least one liver nodule and surrounding tissue - have HCC or other malignancies in the liver proven by liver biopsy - receive treatment with anticoagulants that cannot be paused during the biopsy - have increased risk of bleeding or with congenital coagulation disorders - have not given written informed consent for participation in the study - pregnancy or lactation, or intention of becoming pregnant during study treatment. The ß-HCG value must be determined at each visit, unless the woman is menopausal

Design outcomes

Primary

MeasureTime frame
The primary endpoint of the TRAINING cohort is to create a multiscale atlas of cirrhotic nodules with low or intermediate risk of transition to HCC and an AI based decision support tool. Based on these multiscale data, the hypothesis for the SCREENING cohort will be generated The primary endpoint of the SCREENING cohort is the rate of HCC development, and its predictability by the decision support tool.

Secondary

MeasureTime frame
The feasibility of the AI and multiscale-based screening approach, the refinement of the multiscale model and identification of potential biomarkers The assessment of safety and applicability (patient acceptance) of the protocol.

Countries

Germany

Contacts

Public ContactThomas Berg

Universitätsklinikum Leipzig, Medizinsiche Fakultät

thomas.berg@meidzin.uni-leipzig.de+49 341 9712330

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: Feb 4, 2026