Establishment and validation of a nomogram to predict cancer-specific survival in pediatric neuroblastoma patients

The term "neuroblastoma (NB)" refers to a type of solid pediatric tumor that develops from undivided neuronal cells. According to the American Cancer Society report, between 700 and 800 children under the age of 14 are diagnosed with NB every year in the United States (U.S.). About 6% of all cases of pediatric cancer in the U.S. are caused by NB. NB is the most frequent malignancy in children younger than 1 year; however, it is rarely found in those over the age of 10 and above.

To accurately predict cancer-specific survival (CSS) in children with NB, this research developed and validated an all-encompassing prediction model.

The present retrospective study used the Surveillance, Epidemiology, and End Results (SEER) database to collect information on 1,448 individuals diagnosed with NB between 1998 and 2019. The pool of potentially eligible patients was randomly split into two groups, a training cohort (N = 1,013) and a validation cohort (N = 435). Using multivariate Cox stepwise regression, we were able to identify the components that independently predicted outcomes. The accuracy of this nomogram was measured employing the consistency index (C-index), area under the time-dependent receiver operating characteristic curve (AUC), calibration curve, and decision-curve analysis (DCA).

In this study, we found that age, primary location, tumor size, summary stage, chemotherapy, and surgery were all significant predictors of CSS outcomes and integrated them into our model accordingly. The C-index for the validation cohort was 0.812 (95% CI: 0.773-0.851), while for the training cohort it was 0.795 (95% CI: 0.767-0.823). The C-indexes and AUC values show that the nomogram is able to discriminate well enough. The calibration curves suggest that the nomogram is quite accurate. Also, the DCA curves demonstrated the prediction model's value.

A novel nomogram was developed and validated in this work to assess personalized CSS in NB patients, and it has been indicated that this model could be a useful tool for calculating NB patients' survival on an individual basis and enhancing therapeutic decision-making.