Background: Isocitrate dehydrogenase (IDH) mutation is the initiating event
that defines major clinical and prognostic classes of gliomas, but the potential
mechanisms have not been well interpreted yet. The main objective of the current study
was to better understand the underlying biology of IDH mutant gliomas as captured by
gene expression profiles.
Methods: RNA sequencing data of WHO grade II-IV gliomas from the Chinese Glioma
Genome Atlas (CGGA, N=325) were used to assess differentially expressed genes
between IDH mutant and wild type gliomas and to construct a gene expression-based
classifier to detect IDH mutant samples with high sensitivity and specificity. The classifier
was validated in independent RNA sequencing data from the Cancer Genome Atlas
(TCGA, N=699), and the prognostic value of the classifier was also assessed in the two
Results: A 58-gene-pair IDH mutation signature was developed by using the top scoring
pairs algorithm. In CGGA dataset, 98.5% and 100% IDH mutant samples were also
predicted to be mutant by gene expression based IDH status in grade II-III and grade IV
gliomas, respectively. In TCGA dataset, the proportions were 99.8% and 100%,
respectively. The signature remained to be a prognostic marker in multivariate cox
analysis both in CGGA and TCGA datasets.
Conclusion: A characteristic gene expression signature is associated with and
accurately predicts IDH mutation status. This suggests a common biology between
these tumors and adds prognostic and biologic information that is not captured by the
mutation status alone. These results may help in population stratification for clinical
trials. As RNA-seq is more and more prevalent and cost-effective in glioma molecular
diagnosis, this gene signature would provide a precise method to predict IDH mutation
status with RNA-seq data.