Prognostic Factors for Event-Free Survival in Pediatric Patients with Hepatoblastoma Based on the 2017 PRETEXT and CHIC-HS Systems

• Published in: Cancers Vol. 11; no. 9; p. 1387
• Main Authors: Yoon, Hee Mang, Hwang, Jisun, Kim, Kyung Won, Namgoong, Jung-Man, Kim, Dae Yeon, Koh, Kyung-Nam, Kim, Hyery, Cho, Young Ah
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 18.09.2019; MDPI
• Subjects: Abdomen; Age; Annotations; Blood clots; Clinical medicine; Collaboration; Disease; hepatoblastoma; imaging; Magnetic resonance imaging; Medical imaging; Medical prognosis; Medical records; Metastases; Metastasis; Patients; pediatric; Pediatrics; Portal vein; Prediction models; PRETEXT; prognostic factor; Statistical analysis; Statistics; Survival; Tumors; Variables; Veins & arteries
• ISSN: 2072-6694; 2072-6694
• DOI: 10.3390/cancers11091387

This study aimed to evaluate the prognostic value of variables used in the 2017 PRE-Treatment EXTent of tumor (PRETEXT) system and the Children’s Hepatic tumors International Collaboration-Hepatoblastoma Stratification (CHIC-HS) system in pediatric patients with hepatoblastoma. A retrospective analysis of data from the pediatric hepatoblastoma registry of a tertiary referral center was conducted to evaluate the clinical and imaging variables (annotation factors) of the PRETEXT staging system. The primary outcome was event-free survival (EFS). Data from 84 patients (mean age: 2.9 ± 3.5 years) identified between 1998 and 2017 were included. Univariable Cox proportional hazards analysis revealed that PRETEXT annotation factors P (portal vein involvement), F (multifocality of tumor), and M (distant metastasis) showed a significant negative association with EFS. Multivariable Cox proportional hazard analysis showed that factor F was the strongest predictor (HR (hazard ratio), 2.908; 95% CI (confidence interval), 1.061–7.972; p = 0.038), whereas factor M showed borderline significance (HR, 2.416; 95% CI, 0.918–6.354; p = 0.074). The prediction model based on F and M (F + M) showed good performance to predict EFS (C-statistic, 0.734; 95% CI, 0.612–0.854). In conclusion, the PRETEXT annotation factor F was the strongest predictor of EFS, and the F + M model showed good performance to predict EFS in pediatric patients with hepatoblastoma.