1323 - Development and Validation of a RNA-Seq Based Prognostic Signature in Neuroblastoma

Background

Credible prognostic stratification remains a challenge for neuroblastoma (NBL) with variable clinical manifestations. RNA expression signatures might predict the outcomes; notwithstanding, independent cross-platform validation is still rare.

Methods

RNA expression data were obtained from NBL patients and then analyzed. In TARGET-NBL data, an RNA-based prognostic signature was developed and validated. Survival prediction was assessed using a time-dependent receiver operating characteristic (ROC) curve. Functional enrichment analysis of the RNAs was conducted using bioinformatics methods.

Results

A total of 1,119 differentially expressed RNAs and 149 prognosis-related RNAs were identified sequentially. Then, in the training cohort, 12 RNAs were identified as significantly associated with overall survival (OS) and were combined to develop a model that stratified NBL patients into low- and high-risk groups. Twelve RNA signature high-risk patients had poorer OS in the training cohort (n = 105, Hazard Ratios (HR)= 0.10 (0.05-0.20), P < 0.001) and in the validation cohort (n = 44, HR = 0.25 (0.09-0.69), P = 0.008). ROC curve analysis also showed that both the training and validation cohorts performed well in predicting OS (12-month AUC values of 0.852 and 0.438, 36-month AUC values of 0.824 and 0.737, and 60-month AUC values of 0.802 and 0.702, respectively). Moreover, these 12 RNAs may be involved in certain events that are known to be associated with NBL through functional enrichment analysis.

Conclusions

This study identified and validated a novel 12-RNA prognostic signature to reliably distinguish NBL patients at low and high risk of death. Further larger, multicenter prospective studies are desired to validate this model.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

National Natural Science Foundation of China (Grant No. 81660512).

Disclosure

All authors have declared no conflicts of interest.