Quantitative Hepatobiliary Phase Signal Analysis of HCC Nodules on Gd-EOB-DTPA-Enhanced MRI: A Potential Imaging Biomarker Associated with Disease Control After Conventional Transarterial Chemoembolization (cTACE)

Background: To evaluate whether quantitative signal intensity (SI) measured on the hepatobiliary phase (HBP) of gadoxetic-acid-enhanced MRI is associated with disease control rate (DCR) after conventional transarterial chemoembolization (cTACE) in patients with hepatocellular carcinoma (HCC). Methods: In this retrospective single-center study, 112 patients with HCC treated with cTACE between April 2018 and September 2020 and with available pre-treatment gadoxetic acid-enhanced MRI were included, with one target HCC lesion analyzed per patient. Mean HBP SI was quantitatively measured using manually drawn regions of interest (ROIs) within the target tumor, adjacent liver parenchyma, and reference tissues. Treatment response was assessed using mRECIST criteria at 1, 3, and 6 months. Logistic regression and receiver operating characteristic analyses were performed to evaluate the association between HBP SI and DCR. Results: Lower tumor HBP SI was associated with DCR at 6 months after cTACE (unadjusted p = 0.010; Bonferroni-adjusted p = 0.030), whereas no significant association was observed at 1 or 3 months. A tumor HBP SI cut-off value of 580 demonstrated moderate discriminatory performance for DCR (AUC = 0.714, 95% CI: 0.585–0.843; p = 0.008), with 69.1% sensitivity and 80.0% specificity. Conclusions: Lower baseline tumor HBP SI was associated with 6-month radiological disease control after cTACE in this exploratory single-center cohort. However, its moderate discriminatory performance, the absence of significant associations at earlier response assessments, and the methodological limitations of the study preclude its use as a stand-alone clinical decision-making tool. HBP SI should therefore be regarded as a potential imaging biomarker, and these findings should be considered hypothesis-generating and require validation in larger, independent cohorts.