Incidence trends and a nomogram for predicting overall survival in children with hepatoblastoma: A population-based analysis

Background: In recent years, the incidence of pediatric hepatoblastoma has increased significantly. The goals of our study were to analyze incidence trends, identify independent risk factors influencing prognosis, and develop a nomogram based on these risk factors to guide clinical treatment. Methods Clinicopathological data from children diagnosed with hepatoblastoma between 2000 and 2018 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database to analyze incidence trends. The annual percentage change (APC) was assessed using joinpoint regression analysis. After multiple interpolations, the data selected from 810 children who met the study criteria were randomized into data from the training cohort (n = 568) and validation cohort (n = 242). COX, LASSO, and BSR univariate analyses were used to screen potential risk factors affecting prognosis. Independent risk factors were identified by a multivariate COX regression analysis, and a line-graph model was developed to predict overall survival. The risk score for each child was obtained according to the model. The optimal cutoff value was determined using X-tile software, and the high-risk and low-risk groups were determined. Survival curves were plotted for subgroups using the Kaplan–Meier method. The nomogram was evaluated and verified by the consistency index (C-index) and other analyses: area under the curve (AUC) for receiver operating characteristic (ROC) curve, calibration diagram, and decision curve analysis (DCA). Results A total of 810 children with hepatoblastoma were included in the study. The APC was 1.63% (95% confidence interval [CI] -0.6–3.9%, P < 0.05). In particular, boys had a higher incidence than girls; however, the incidence of girls has risen in recent years. The incidence in African-American children has risen. The incidence in children under 2 years old is higher and is on the rise compared to other age groups. Race, age, tumor size, surgical type, and chemotherapy were independent risk factors and thus were selected to develop the nomogram. The time-dependent AUC (> 0.7) and the time-dependent C-index (> 0.7) indicated the satisfactory discriminative ability of the nomogram. The calibration plots showed favorable consistency between the prediction of the nomogram and actual observations in both the training and validation cohorts. Furthermore, DCA showed that the nomogram was clinically useful and had a discriminative ability. We developed a web-based calculator that implements a dynamic nomogram that automatically generates a survival curve (with a 95% confidence interval [CI]) after inputting clinical data for the patient. Using a smartphone, clinicians can access the calculator to predict the prognosis of an individual patient and guide treatment decisions. Conclusions The incidence of pediatric hepatoblastoma has increased in recent years, according to data from the SEER database. We developed a nomogram to predict prognosis and guide treatment. A combined treatment approach of surgery and chemotherapy is highly likely to prolong survival and improve patient outcomes.