Multicentre prospective trial of abbreviated MRI using gadoxetic acid versus CT for detection of late recurrent HCC (AMRICT): study protocol.

Abstract

Hepatocellular carcinoma (HCC) recurs in most patients after curative treatment. Late recurrence (>2 years after curative treatment) typically indicates de novo tumours in the residual liver. Although contrast-enhanced computed tomography (CECT) and MRI are widely used for post-treatment follow-up, they each have limitations including radiation exposure, high cost and limited access. The abbreviated MRI using gadoxetic acid versus multiphasic CECT for detection of late recurrent HCC after curative treatment (AMRICT) trial aims to compare gadoxetic acid-enhanced abbreviated MRI using hepatobiliary phase imaging (HBP-AMRI) and multiphasic CECT for detecting late recurrent HCC after curative treatment. This prospective multicentre intra-individual comparison trial will enrol 455 participants who have undergone surgical resection or local ablation for HCC and remained recurrence-free for over 2 years. Each participant will undergo two imaging sessions at 6±2 month intervals, using both HBP-AMRI and multiphasic CECT. The primary endpoint is the detection rate of all-stage HCC. The secondary endpoints include the false referral rate of all-stage HCC and detection and false referral rates of Barcelona Clinic Liver Cancer stage 0-A HCC and of stage 0 HCC. Structured imaging protocols and quality assessments will be implemented for both modalities. This study was approved by the Institutional Review Boards of the three participating institutions (approval number: 2023-1630 (Asan Medical Center), H-2407-146-1556 (Seoul National University Hospital) and B-2410-929-401 (Seoul National University Bundang Hospital)) and registered at ClinicalTrials.gov (NCT06537193). Findings will be disseminated through peer-reviewed journals, scientific meetings, public forums and guideline updates. ClinicalTrials.gov: NCT06537193. Participant enrolment began on 12 December 2024, and is ongoing.