LAUSR

Purpose: Glioblastoma (GBM) is the most common and malignant form of primary human brain tumor in adults, with an average survival at diagnosis of 18 months. Metabolism is a new attractive therapeutic target in cancer; however, little is known about metabolic heterogeneity and plasticity within GBM tumors. We therefore aimed to investigate metabolic phenotyping of primary cultures in the context of molecular tumor heterogeneity to provide a proof of concept for personalized metabolic targeting of GBM. Experimental Design: We have analyzed extensively several primary GBM cultures using transcriptomics, metabolic phenotyping assays, and mitochondrial respirometry. Results: We found that metabolic phenotyping clearly identifies 2 clusters, GLNHigh and GLNLow, mainly based on metabolic plasticity and glutamine (GLN) utilization. Inhibition of glutamine metabolism slows the in vitro and in vivo growth of GLNHigh GBM cultures despite metabolic adaptation to nutrient availability, in particular by increasing pyruvate shuttling into mitochondria. Furthermore, phenotypic and molecular analyses show that highly proliferative GLNHigh cultures are CD133neg and display a mesenchymal signature in contrast to CD133pos GLNLow GBM cells. Conclusions: Our results show that metabolic phenotyping identified an essential metabolic pathway in a GBM cell subtype, and provide a proof of concept for theranostic metabolic targeting. Clin Cancer Res; 23(20); 6292–304. ©2017 AACR.