Abstract
PURPOSE: This study aimed to assess prognostic values of the POST-Treatment Extent of tumor (POSTTEXT) system and clinical factors after neoadjuvant chemotherapy in hepatoblastoma patients and evaluate benefits of post-treatment imaging and clinical factors concomitant with Children's Hepatic Tumors International Collaboration-Hepatoblastoma Stratification (CHIC-HS) system. MATERIALS AND METHODS: This single-center retrospective study of hepatoblastoma cases (2006-2022) included pediatric patients receiving ≥ 4 cycles of neoadjuvant chemotherapy, with pre- and post-treatment imaging and complete medical records. Clinical data included age, sex, and serum α-fetoprotein (AFP) levels. Cox regression analyses identified predictors of event-free survival (EFS). Time-dependent receiver operating characteristic curves assessed the predictive power of combining the CHIC-HS risk stratification with post-treatment factors. Inter-reader agreement was analyzed using weighted kappa. RESULTS: Among the 109 hepatoblastoma patients, 73 (mean age, 2.2±2.7 years) met the inclusion criteria. Prognostic factors for EFS included AFP levels after the fourth cycle of neoadjuvant chemotherapy (hazard ratio [HR], 1.233; 95% confidence interval [CI], 1.086 to 1.400; p=0.001), tumor size change ratio (HR, 0.654; 95% CI, 0.448 to 0.955; p=0.030), and POSTTEXT annotation factor M (HR, 5.209; 95% CI, 1.639 to 16.553; p=0.005). Incorporating AFP levels after the fourth cycle of neoadjuvant chemotherapy into the CHIC-HS improved predictive power (p=0.043). POSTTEXT system showed better inter-reader agreement than PRE-Treatment Extent of tumor (PRETEXT). CONCLUSION: Predictors of EFS in hepatoblastoma include AFP levels after the fourth cycle of neoadjuvant chemotherapy, tumor size change ratio, and metastasis (POSTTEXT M). Combining AFP levels after the fourth cycle of neoadjuvant chemotherapy to the CHIC-HS improved the predictive ability.