Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Share the Article
Krishnan A. Imaging tumor heterogeneity to predict response to transarterial chemoembolization in hepatocellular carcinoma. World J Gastroenterol 2026; 32(39): 120606 [DOI: 10.3748/wjg.120606]
World J Gastroenterol. Oct 21, 2026; 32(39): 120606 Published online Oct 21, 2026. doi: 10.3748/wjg.120606
Imaging tumor heterogeneity to predict response to transarterial chemoembolization in hepatocellular carcinoma
Arunkumar Krishnan
Arunkumar Krishnan, Department of Supportive Oncology, Atrium Health Levine Cancer, Atrium Health Wake Forest Baptist Comprehensive Cancer Center, Charlotte, NC 28204, United States
Author contributions: Krishnan A conceptualized the manuscript and conducted the assessment, prepared the manuscript draft, which was subsequently reviewed and approved for final publication.
AI contribution statement: AI-based tool (Microsoft Copilot) was used in this manuscript for language refinement and editorial polishing - improving grammar and readability - consistent with most current publisher policies on AI-assisted language editing. No AI tool was used to generate the substantive analysis, conclusions, or scientific arguments or writing.
Conflict-of-interest statement: The author reports no relevant conflicts of interest for this article.
Corresponding author: Arunkumar Krishnan, MD, Department of Supportive Oncology, Atrium Health Levine Cancer, Atrium Health Wake Forest Baptist Comprehensive Cancer Center, 1021 Morehead Medical Dr, Charlotte, NC 28204, United States. dr.arunkumar.krishnan@gmail.com
Received: March 3, 2026 Revised: April 20, 2026 Accepted: May 25, 2026 Published online: October 21, 2026 Processing time: 191 Days and 3.5 Hours
Core Tip
Core Tip: The study by Lv et al integrated contrast-enhanced magnetic resonance imaging-derived habitat features, conventional radiomics, and clinical variables to predict early response to transarterial chemoembolization in hepatocellular carcinoma, achieving area under the curve values of 0.97 and 0.93 in the training and validation cohorts, respectively. While the ecological diversity index and multicenter design are meaningful advancements, several key methodological issues remain unresolved: Feature selection outside nested cross-validation, which risks data leakage and overly optimistic performance estimates; confounding from mixed protocols; outcome misclassification due to the use of both computed tomography and magnetic resonance imaging follow-up; and insufficient inter-site harmonization. Prospective validation, integration of multi-sequence imaging, and studies analyzing the impact on decision-making are necessary before implementing this in clinical practice.
