A Novel Diagnostic Algorithm for Subcentimeter Hepatocellular Carcinoma Utilizing Gd-EOB-DTPA-Enhanced MRI:...

Area of Science:

• Radiology and Medical Imaging
• Oncology
• Hepatology

Background:

• Subcentimeter hepatocellular carcinoma (scHCC) diagnosis is challenging due to limitations in current imaging guidelines, particularly for lesions lacking arterial phase hyperenhancement (APHE).
• Accurate early detection of scHCC is crucial for timely treatment and improved patient outcomes.
• Gadolinium-ethoxybenzyl-diethylenetriaminepentaacetic acid (Gd-EOB-DTPA)-enhanced MRI offers valuable information for liver lesion characterization.

Purpose of the Study:

• To develop and validate a novel decision-tree algorithm for diagnosing scHCC (≤1.0 cm) using Gd-EOB-DTPA-enhanced MRI.
• To overcome the limitations of existing guidelines in detecting scHCC that lacks APHE.
• To integrate key MRI features into a robust diagnostic model.

Main Methods:

• A multicenter retrospective study analyzed 419 patients with focal liver nodules ≤ 1.0 cm.
• A decision-tree algorithm was developed using classification and regression tree (CART) modeling, incorporating restricted diffusion, non-rim APHE, transitional phase hypointensity, and mild-moderate T2WI hyperintensity.
• Performance was evaluated on training (n=225), internal test (n=96), and external validation (n=98) sets and compared against LI-RADS and JSH criteria.

Main Results:

• The algorithm demonstrated high diagnostic performance: Training set (93.8% accuracy), internal test set (90.6% accuracy), and external validation set (90.8% accuracy).
• The algorithm significantly outperformed LI-RADS LR-4, modified LR-4, and JSH criteria (P < 0.001 to P = 0.005).
• Crucially, the algorithm achieved 80.0%-87.5% accuracy for scHCC cases that lacked APHE, highlighting its utility in challenging cases.

Conclusions:

• The developed decision-tree algorithm effectively diagnoses scHCC by integrating key MRI features and reducing reliance on APHE.
• This novel algorithm shows high accuracy across multicenter cohorts, supporting its clinical translation for improved early hepatocellular carcinoma detection.
• The findings suggest a streamlined and more accurate approach to diagnosing small liver lesions.