P07.10 Integrative Network-based Analysis links Vessel-Size Imaging (VSI) and Cerebral Blood Volume (CBV) to specific Molecular Patterns in Malignant Gliomas

Abstract

Objective: Perfusion weighted imaging (estimating the cerebral blood volume, CBV) and its specialisation ‘vessel size imaging’ (VSI) are well-established MRI techniques to assess tumor vascularization. Currently, no studies exist which link these MRI-based parameters with genetic profiles in malignant gliomas. The molecular background of both perfusion parameters is largely unknown and needs to be explored.Method: We retrospectively analyzed 25 patients diagnosed with a primary glioblastoma between 2010 and 2014. In a presurgical MRI CBV and VSI were calculated in addition to standard imaging. During the following surgery tumour samples were removed from a predefined contrast-enhancing tumour region controlled by neuronavigation. RNA was extracted and used for array based transcriptional analysis (Affymetrix). Integrative analysis was performed by individual designed pipelines based on bioconductor (www.bioconductor.org) available R-packages and Weighted Gene Co-Expression Network Analysis (WGCNA).Results: Transcriptional data were analyzed by WGCNA and split into modules by its network-based affiliation. 10 modules were highly correlated either to VSI and/or CBV. One module was exclusively associated to VSI. This module is highly related to hypoxia (Gene Set Enrichment Analysis (GSEA) of intramodule connectivity (FDR<0.001)). Another module was exclusively correlated to CBV and not to VSI and showed strong enrichments in the EGF pathway and Epithelial-to-Mesenchymal-Transition (EMT). In addition, patients with increased CBV or VSI mostly showed a mesenchymal gene-expression phenotype.Conclusion: We show that CBV/VSI-based data may present surrogate imaging markers for specific genetic tumor subtypes. Further analysis with an increased number of patients should be done in the future.