Feasibility of CEUS LI-RADS Nonradiation Treatment Response Assessment for Early Evaluation of Intrahepatic Metastases After TACE.

RATIONALE AND OBJECTIVES: To explore the feasibility and reproducibility of applying contrast-enhanced ultrasound liver imaging reporting and data system nonradiation treatment response (CEUS LI-RADS TRA v2024) criteria for assessing intrahepatic metastases within 2 months after transarterial chemoembolization (TACE). MATERIALS AND METHODS: In this prospective study, 38 patients (mean age, 64 years; range, 41-89 years) with intrahepatic metastases undergoing TACE were included. US and CEUS were performed before and after TACE treatment, with a follow-up duration of 1-2 months. Imaging features, including treated lesion (TL) size, shape, echogenicity, margin, and perilesional echogenicity, were evaluated based on interval changes before and after TACE on conventional B-mode US, along with intralesional and perilesional enhancement patterns on CEUS. CEUS findings were independently evaluated by two readers according to CEUS LI-RADS TRA v2024. Interobserver agreement was assessed using Cohen's kappa and intraclass correlation coefficient (ICC). Multiple assessment strategies were compared based on intralesional enhancement alone and in combination with perilesional enhancement and conventional US features. Diagnostic performance was assessed using sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV), as well as the proportion of indeterminate cases. Statistical analyses were performed using Fisher's exact test, independent samples t-test, and Mann-Whitney U test, and receiver operating characteristic (ROC) curve analysis to determine optimal cutoff values. The reference standard was based on contrast-enhanced magnetic resonance imaging (CEMRI). RESULTS: Significant post-treatment changes were observed in TL echogenicity, margin, intralesional enhancement, and perilesional enhancement (all P < 0.05). Interobserver agreement for LI-RADS TRA v2024 CEUS features was substantial to excellent, with κ values ranging from 0.78 to 0.87 and ICC values ranging from 0.69 to 0.96 (all p < 0.001). Intralesional enhancement demonstrated the highest reproducibility (κ = 0.87, ICC = 0.96) while maintaining the best overall diagnostic performance across both readers, with sensitivity of 93.3-96.7% [95% Confidence Interval (CI):77.9-99.9], specificity of 75.0-87.5% (95% CI:34.9-99.7), accuracy of 89.5-94.7% (95% CI:75.2-99.4), PPV of 100.0% (95% CI:87.7-100.0), NPV of 87.5-100.0% (95% CI: 47.3-100.0), and only 2.6%-10.5% indeterminate cases. The addition of perilesional enhancement increased sensitivity (93.3%-100.0%) and NPV (100.0%) but reduced specificity (12.5%-25.0%), accuracy (76.3-84.2%), and diagnostic clarity, resulting in 15.8-23.7% cases deemed equivocal. When intralesional CEUS features were further combined with changes in TL echogenicity, TL margin, and perilesional enhancement before and after TACE, NPV consistently reached 100.0%, but overall diagnostic performance did not improve and in some cases decreased, with sensitivity ranging from 96.7% to 100.0%, specificity from 37.5% to 75.0%, accuracy from 84.2% to 94.7%, PPV from 88.2% to 100.0%, and indeterminate cases ranging from 0.0% to 15.8%. Notably, the inclusion of perilesional enhancement substantially increased the number of equivocal cases, indicating that additional imaging features increased sensitivity at the expense of diagnostic specificity and clarity. CONCLUSION: CEUS LI-RADS TRA v2024 is feasible and reproducible for early assessment of intrahepatic metastases after TACE. Intralesional enhancement alone provides the most balanced diagnostic performance and reproducibility during the first 1-2 months after treatment, suggesting it may serve as a simple and reliable biomarker for early treatment response assessment. Additional perilesional and conventional US features do not improve diagnostic performance and may reduce diagnostic clarity shortly after TACE.