Meeting Banner
Abstract #3476

Correlations with Gene Expression Profile May Generate New MR Spectroscopy Markers for Glioblastoma Prognosis and Treatment Planning

Alan James Wright1, Greg Fellows2, Ken Laing3, John Griffiths4, Brian Anthony Bell2, Franklyn Arron Howe5

1Basic Medical Sciences, St. George's University of London, London, UK; 2Academic Neurosurgery Unit, St. George's University of London, London, UK; 3Cellular and Molecular Medicine, St. George's University of London, London, UK; 4CRUK Cambridge Research Institute, Cambridge, UK; 5Cardiac and Vascular Sciences, St. George's University of London, London, UK


Surgically resected gliomas were analysed for gene expression data using microarrays and for metabolite concentrations with HR-MAS NMR. Hierarchical clustering of expression data produced two distinct groups. The differential expression of genes between the two groups suggested that the glioblastomas in cluster 1 had expressed genes similarly to astrocytic tumours (also in cluster 1) and were likely to be transformed secondary glioblastomas as opposed to the more common primary glioblastomas (cluster 2). Concentrations measured by NMR showed significant differences in some metabolites between the two groups that should be investigated as novel markers for tumour prognosis and treatment planning.