A Modified Model for Assessment for Retreatment with Transarterial Chemoembolization in Chinese Hepatocellular...

To develop a modified assessment for retreatment with transarterial chemoembolization (mART) score that may be more suitable for Chinese patients with hepatocellular carcinoma (HCC).

Chinese patients with HCC who were treated with transarterial chemoembolization in four hospitals were included. A univariate analysis and a multivariate forward Cox regression analysis were used to identify significant prognostic factors of overall survival (OS). A point scoring model was subsequently developed from the training cohort, and the validation process was performed in the validation cohort.

The study included 259 patients (124 patients in the training cohort and 135 patients in the validation cohort). Increase in Child-Pugh scores relative to the baseline (P < .001), Barcelona Clinic Liver Cancer (BCLC) stage B before first transarterial chemoembolization (P = .001), and absence of radiologic tumor response (P < .001) were identified as negative prognostic factors for OS and were used to create the mART scores. BCLC staging was substituted for aspartate aminotransferase increase in the mART scores. The mART scores differentiated two groups with distinct prognosis by a cutoff score of 2.5 points (22.9 mo [95% confidence interval (CI), 17.4-28.4] vs 8.9 mo [95% CI, 7.5-10.3] in median survival; P < .001). In the validation cohort, the C index in assessment for retreatment with transarterial chemoembolization (ART) criteria was 0.64, whereas it was 0.82 in mART criteria.

In Chinese patients with HCC, mART score of > 2.5 before second transarterial chemoembolization was associated with poor prognosis. The mART score was probably better validated compared with the ART score.