Performance of initial LI-RADS 2018 treatment response in predicting survival of patients with hepatocellular carcinoma following TACE: a retrospective, single-center cohort study

PURPOSE Treatment response following transarterial chemoembolization (TACE) is frequently evaluated with Liver Imaging Reporting and Data System Treatment Response (LR-TR) algorithm, but its association with patients' outcomes is not supported in the literature. The purpose of this study was to provide such data. METHODS A retrospective analysis of 99 TACE patients with stage A/B hepatocellular carcinoma according to Barcelona-Clinic Liver Cancer staging system was performed. Two radiologists assessed LR-TR, while a third radiologist re-assessed divergent results. Overall survival (OS) and time to disease progression (TTP) were the primary endpoints of the study, while the Cox proportional hazard model was used for outcome analyses. RESULTS Interobserver agreement was substantial between the two readers with κ = 0.69 (95% CI 0.58-0.81). The median OS in viable, equivocal, and non-viable groups were 27, 27, and 73 months, respectively (p < 0.001). However, after adjustment for confounding factors, there was no significant association between initial viable response and OS (HR 0.98 [95% CI 0.37-2.63], p = 0.97), while equivocal response remained statistically significant (HR 3.52. [95% CI 1.27-9.71], p = 0.015). No significant association was noted when viable and equivocal groups were analyzed in aggregate (HR 1.03 [95% CI 0.4-2.4], p = 0.96). The median TTP did not differ between non-viable and viable groups (23 vs 18 months, respectively; p = 0.98). None of the analyzed predictors was associated with TTP. CONCLUSION Initial LR-TR response was not an independent predictor for OS nor TTP. The preliminary results suggest the necessity for more aggressive management of equivocal patients.