Utilization of metabolomics as a tool to identify a novel diagnostic and prognostic plasma biomarker for glioblastoma.

Glioblastoma is the most common malignant primary brain tumor that universally carries a poor prognosis. Despite aggressive treatment, this tumor recurs because surviving tumor cells acquire new mutations, which can alter cellular metabolism. These mutations cause metabolic changes in the plasma that can be detected using metabolomic approaches. Little is known about these metabolomic changes after chemoradiation. Such an understanding holds promise for identification of biomarkers of treatment response and even serve prognostic value. 

In this study, we prospectively enrolled a cohort of patients with isocitrate dehydrogenase (IDH) wild type glioblastoma and performed untargeted metabolomics of patient plasma before and after surgery, as well as before and after concurrent chemoradiation. Here we utilize untargeted metabolomics to examine changes in the levels of 157 metabolites in the serum of glioblastoma patients at each stage of treatment. We then draw correlations between metabolite levels and each treatment stage. Finally, we draw associations between metabolite levels and disease progression or overall survival to identify clinically relevant biomarkers. 

Preliminary results have demonstrated positive associations between the levels of multiple metabolites at the time of diagnosis with eventual overall survival: trans-4-hydroxyproline (p=0.041), ribose (p=0.017), pipecolinic acid (p=0.008), phenol (p=0.017), kynurenine (p=0.021), inositol-4-monophosphate (p=0.007), indole-3-propionic acid (p=0.038), glucose (p=0.003), arachidonic acid (p=0.007), 5-methoxytryptamine (0.001), and 3-aminopiperidine-2,6-dione (p=0.046). No metabolites were negatively associated with overall survival. 

Ongoing analysis will further explore survival associations with metabolite levels at different stages in treatment. Further, we will utilize machine learning approaches to gain deeper insight into our results. These results demonstrate that metabolomics may be an important tool in the development of clinically relevant biomarkers for glioblastoma.