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

Jian-Guo Zhou; Bo Liang; Su-Han Jin; Hui-Ling Liao; Guo-Bo Du; Long Cheng; Hu Ma; Udo S Gaipl

doi:10.3389/fonc.2019.01361

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

Front Oncol. 2019 Dec 4:9:1361. doi: 10.3389/fonc.2019.01361. eCollection 2019.

Authors

Jian-Guo Zhou^{1

2}, Bo Liang³, Su-Han Jin⁴, Hui-Ling Liao⁵, Guo-Bo Du⁶, Long Cheng⁶, Hu Ma¹, Udo S Gaipl²

Affiliations

¹ Department of Oncology, Affiliated Hospital of Zunyi Medical University, Zunyi, China.
² Department of Radiation Oncology, Universitätsklinikum Erlangen, Erlangen, Germany.
³ Affiliated Nanjing Hospital of Chinese Medicine, Nanjing University of Chinese Medicine, Nanjing, China.
⁴ Department of Orthodontics, Affiliated Stemmatological Hospital of Zunyi Medical University, Zunyi, China.
⁵ College of Integrated Traditional Chinese and Western Medicine, Southwest Medical University, Luzhou, China.
⁶ Department of Oncology, Affiliated Hospital of North Sichuan Medical College, Nanchong, China.

Abstract

Objective: The stratification of neuroblastoma (NBL) prognosis remains difficult. RNA-based signatures might be able to predict prognosis, but independent cross-platform validation is still rare. Methods: RNA-Seq-based profiles from NBL patients were acquired and then analyzed. The RNA-Seq prognostic index (RPI) and the clinically adjusted RPI (RCPI) were successively established in the training cohort (TARGET-NBL) and then verified in the validation cohort (GSE62564). Survival prediction was assessed using a time-dependent receiver operating characteristic (ROC) curve and area under the ROC curve (AUC). Functional enrichment analysis of the genes was conducted using bioinformatics methods. Results: In the training cohort, 10 gene pairs were eventually integrated into the RPI. In both cohorts, the high-risk group had poor overall survival (OS) (P < 0.001 and P < 0.001, respectively) and favorable event-free survival (EFS) (P = 0.00032 and P = 0.06, respectively). ROC curve analysis also showed that the RPI predicted OS (60 month AUC values of 0.718 and 0.593, respectively) and EFS (60 month AUC values of 0.627 and 0.852, respectively) well in both the training and validation cohorts. Clinicopathological indicators associated with prognosis in the univariate and multivariate regression analyses were identified and added to the RPI to form the RCPI. The RCPI was also used to divide populations into different risk groups, and the high-risk group had poor OS (P < 0.001 and P < 0.001, respectively) and EFS (P < 0.05 and P < 0.05, respectively). Finally, the RCPI had higher accuracy than the RPI for the prediction of OS (60 month AUC values of 0.730 and 0.852, respectively) and EFS (60 month AUC values of 0.663 and 0.763, respectively) in both the training and validation cohorts. Moreover, these differentially expressed genes may be involved in certain NBL-related events. Conclusions: The RCPI could reliably categorize NBL patients based on different risks of death.

Keywords: event-free survival; gene pairs; neuroblastoma; overall survival; prognostic model; validation.