LI-RADS Treatment Response Algorithm: Performance and Diagnostic Accuracy

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• The 2017 Liver Imaging Reporting and Data System (LI-RADS) introduced a treatment response algorithm for hepatocellular carcinoma (HCC).
• This algorithm aims to standardize evaluation of treatment response to guide subsequent therapies.
• Validation of the LI-RADS 2017 Treatment Response algorithm in literature is limited.

Purpose of the Study:

• To evaluate the performance of the LI-RADS 2017 Treatment Response algorithm.
• To assess its accuracy in predicting histopathologic viability of HCC after bland arterial embolization.

Main Methods:

• Retrospective study of patients who underwent bland arterial embolization for HCC and subsequent liver transplantation.
• Three radiologists independently applied the LI-RADS 2017 Treatment Response algorithm to CT/MRI scans.
• Comparison of radiologic assessment with posttransplant histopathology findings.

Main Results:

• The LR-TR Viable category showed 60%-65% accuracy and 86%-96% positive predictive value for predicting incomplete tumor necrosis.
• The LR-TR Nonviable category demonstrated 67%-71% accuracy and 81%-87% negative predictive value for predicting complete tumor necrosis.
• Moderate interreader agreement (κ = 0.55) was observed for the LR-TR category, with 27% of lesions deemed Equivocal.

Conclusions:

• The LI-RADS 2017 Treatment Response algorithm exhibits high predictive value for assessing HCC viability post-bland arterial embolization.
• The algorithm demonstrates moderate interreader association when lesions are categorized as Viable or Nonviable.
• Further validation and refinement may be needed, especially for Equivocal cases.