Vasile Stupar1, 2, Coquery Nicolas2, 3, Farion Rgine1, 2, Emmanuel Luc Barbier2, 3, Chantal Rmy2, 3, Florence Fauvelle4
1Precilinical MRI Facility, Grenoble, France; 2U836, INSERM, Grenoble, France; 3Universit Joseph Fourier, Grenoble, France; 4IRBA antenne CRSSA, La Tronche, France
In Vivo 1H MRS can provide information regarding glioma growth and response to treatment. A wider range of metabolites can be obtained ex vivo in biopsies using HRMAS 1H MRS. The metabolic data can be interpreted and classified using multivariate pattern recognition methods, such as Projection to Latent Structure-Discriminant Analysis (PLS-DA). Comparison of metabolic profiles between 1H MRS and HRMAS 1H MRS is essential and the ability of both approaches to discriminate tumoral from normal tissue with statistical tools such as PLS-DA might help for diagnosis. We have used this approach in the rat RG2 model of glioma.