gms | German Medical Science

Objective: Dysembryoplastic neuroepithelial tumors (DNET) were first described by Daumas-Duport et al. in 1988 and are considered rare, benign and associated with chronic epilepsy. We present the case of a 28-year-old man with a history of epilepsy since age 12. Surgery of an occipital cortical lesion with low signal in T1-weighted and high signal in T2 weighted images and no contrast enhancement in 2009 revealed a DNET WHO I. After 5 years, a MRI showed a recurrent tumor with new contrast enhancement at the edge of the resection cavity. The tumor was removed and the tissue was processed for intensive histopathological and molecular diagnostic workup.

Method: Histopathological analysis included histological staining (H&E and Tibor) and immunostaining including GFAP, Map-2, Ki67, NeuN, IDH1-R132 and Kluever-staining. Molecular diagnostics with 450k methylation array and copy number profile were performed on tissue from the primary and the recurrent tumor. Array raw data were normalized by SWAN and pre-processed using the RnBeads included pipeline. Cluster-analysis and Principal Components Analysis were done by R-Software. Reference GBM methylation data from TCGA were for comparison of the methylation profiles of standard GBM with the patient's tumor.

Results: While the first tumor was unequivocally characterized as dysembryoplastic neuroepithelial tumor WHO I, the recurrent tumor was diagnosed as a glioblastoma multiforme WHO grad IV by two independent neuropathology departments based on typical histopathological morphological features. The tumor has an IDH1-wild type and no MGMT promoter methylation was detected. Because of this unusual clinical course and because of the discrepancy between the histologies of the primary and recurrent tumors, a 450k analysis was performed to further characterize the recurrent tumor. The methylation profile of this tumor is significantly different from those of other GBM (PCA, Cluster analysis) and revealed a DNET-origin.

Conclusions: Molecular characterization of this tumor suggests malignant transformation of a previously benign DNET WHO I. Modern high throughput analysis might add essential molecular information in addition to standard histopathology for proper identification of rare brain tumors which present with an unusual clinical course.