Abstract

Abstract Brain tumors are the most common cause of cancer-related death in childhood. Ependymomas, are the third most common pediatric brain tumor. The disease remains incurable for about 45% of patients even after gross total resection and radiotherapy. Despite showing a very homogeneous histological picture, ependymomas display distinct molecular behavior, which supports the existence of several independent entities of the disease. We examined two non-overlapping cohorts of 102 and 75 ependymomas by mRNA expression profiling, on two different array platforms (Affymetrix, Agilent). When performing multiple statistical clustering methods (unsupervised consensus NMF and consensus HCL), we could consistently identify three major clusters, including two subgroups of posterior fossa (PF) ependymoma, a variant common in children and associated with heterogeneous clinical outcome. Subgroup-specific chromosome aberrations of PF tumors were detected by aCGH, and biological signaling pathways distinguishing PF subgroups were identified by gene set enrichment analysis and visualized in Cytoscape. We validated the most significantly classifying markers of each subgroup by immunohistochemistry on a tissue microarray containing an independent set of 265 PF ependymomas. Our findings delineate two subgroups of PF ependymoma (groups A and B) which are demographically, transcriptionally, genetically, and clinically distinct. Group A patients are younger, have laterally located tumors with a balanced genome, more frequently develop secondary metastases and are much more likely to have an extremely poor outcome as compared with group B patients. Based on a multi-variate Cox proportional-hazards model, our identified markers have the strongest independent prognostic value among demographic and molecular variables with Hazard ratios of 8.45 (PFS) and 10.55 (OS). Prognostic significance and predictive impact is being validated in the GPOH HIT2000 Ependymoma study. The identification of two distinct subgroups of PF ependymoma, and markers applicable for their clinical distinction, will allow for better prognostication of individual cases, independent of age, level of resection and WHO grade, and also for stratification in future ependymoma clinical trials. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 1432. doi:1538-7445.AM2012-1432

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call