Microarray gene expression profiling in meningiomas: Differential expression according to grade or histopathological subtype

Meningiomas, one of the largest subgroup of intracranial tumours are generally benign, but can progress to malignancy. They are classified into the three World Health Organization grades: benign, atypical and anaplastic meningiomas. Various histopathological features have been associated with aggressiveness or recurrence. Several genes have been suggested as prognostic factors, but molecular signatures have not permitted the classification of the tumours into the three grades. We have performed a microarray transcriptomic study on 17 meningiomas of different malignancy using CodeLink Uniset Human Whole Genome Bioarrays to try to distinguish the different grades and histopathological subtypes. Unsupervised hierarchical clustering classified the meningiomas into groups A, B and C, which corresponded to the three grades except for 3 benign meningiomas with higher proliferation indexes and/or recurrence, included in the atypical group. Several genes involved in cell adhesion (CD44, LOX), cell division (CKS2, BIRC5 and UBE2C), cell differentiation (NotchI) or signal transduction (ARHGAP28) were upregulated, whereas tumour suppressor genes (LRIB, DRRI, PLZF, GPX3, SYNPO, TIMP3 and HOPS) and genes involved in cell adhesion (PROS1), proliferation (SERPINF1 and PDGFD) and differentiation (AOX1) were downregulated in groups B and C compared to group A. In the benign tumours, we identified genes with signatures specific for fibroblastic meningiomas (FBLN1, Tenascin C and MMP2 encoding extracellular matrix proteins) and for meningothelial meningiomas (MLPH, DEFB1 and FAT3), suggesting different mechanisms involved in the tumorigenesis of these subtypes. This microarray-based expression profiling study revealed candidate genes and pathways that may contribute to a better understanding of the recurrence of a benign meningioma. Our results might make it possible to determine which benign meningiomas might recur despite complete resection, and will provide helpful information for neurosurgeons in the follow-up of the patients.