Transcriptional profiling of paediatric ependymomas identifies prognostically significant groups

J Pathol Clin Res. 2021 Nov;7(6):565-576. doi: 10.1002/cjp2.236. Epub 2021 Jul 27.

Abstract

The majority of supratentorial ependymomas in children contain oncogenic fusions, such as ZFTA-RELA or YAP1-MAMLD1. In contrast, posterior fossa (PF) ependymomas lack recurrent somatic mutations and are classified based on gene expression or methylation profiling into group A (PFA) and group B (PFB). We have applied a novel method, NanoString nCounter Technology, to identify four molecular groups among 16 supratentorial and 50 PF paediatric ependymomas, using 4-5 group-specific signature genes. Clustering analysis of 16 supratentorial ependymomas revealed 9 tumours with a RELA fusion-positive signature (RELA+), 1 tumour with a YAP1 fusion-positive signature (YAP1+), and 6 not-classified tumours. Additionally, we identified one RELA+ tumour among historically diagnosed CNS primitive neuroectodermal tumour samples. Overall, 9 of 10 tumours with the RELA+ signature possessed the ZFTA-RELA fusion as detected by next-generation sequencing (p = 0.005). Similarly, the only tumour with a YAP1+ signature exhibited the YAP1-MAMLD1 fusion. Among the remaining unclassified ependymomas, which did not exhibit the ZFTA-RELA fusion, the ZFTA-MAML2 fusion was detected in one case. Notably, among nine ependymoma patients with the RELA+ signature, eight survived at least 5 years after diagnosis. Clustering analysis of PF tumours revealed 42 samples with PFA signatures and 7 samples with PFB signatures. Clinical characteristics of patients with PFA and PFB ependymomas corroborated the previous findings. In conclusion, we confirm here that the NanoString method is a useful single tool for the diagnosis of all four main molecular groups of ependymoma. The differences in reported survival rates warrant further clinical investigation of patients with the ZFTA-RELA fusion.

Keywords: NanoString; ependymoma; molecular groups; prognosis; transcriptional profiling.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Age Factors
  • Biomarkers, Tumor / genetics*
  • Cluster Analysis
  • Ependymoma / genetics*
  • Ependymoma / mortality
  • Ependymoma / pathology
  • Ependymoma / therapy
  • Gene Expression Profiling*
  • Humans
  • Infratentorial Neoplasms / genetics*
  • Infratentorial Neoplasms / mortality
  • Infratentorial Neoplasms / pathology
  • Infratentorial Neoplasms / therapy
  • Predictive Value of Tests
  • Prognosis
  • Retrospective Studies
  • Supratentorial Neoplasms / genetics*
  • Supratentorial Neoplasms / mortality
  • Supratentorial Neoplasms / pathology
  • Supratentorial Neoplasms / therapy
  • Transcriptome*

Substances

  • Biomarkers, Tumor