NIMG-64. IN VIVO METABOLIC PROFILING FOR RECURRENT GLIOMA

Abstract

Abstract INTRODUCTION The 2021 WHO classification has identified the importance of molecular features in diagnosing adult glioma. The goal of this study was to use short echo MR spectroscopic imaging (MRSI) to evaluate the metabolite profiles at recurrence and characterize in vivo metabolism in relation to pathology by combining it with image-guided tissue samples. METHODS Fifty-one patients with an initial diagnosis of glioma (18 astrocytoma, 14 oligodendroglioma, and 19 glioblastoma) received MR examinations before surgical resection of progressive disease confirmed histologically. The 3D MRSI data were combined, processed, and quantified by LCModel with a simulated basis set. Metabolite levels were compared between contrast-enhancing lesions (CEL), non-enhancing lesions (NEL), and patient groups. Spectra data centered at the tissue target location were correlated to pathological parameters from the samples (0-4 samples/subject). RESULTS 74.5% (38/51) patients had enhancing tumors, with median CEL volume of 4.2 cm3 (min-max: 0.2-28.5 cm3) and median NEL volumes of 50.6 cm3 (4.8-167.2 cm3). The patients with GBM had significantly larger CEL volumes than the others (p=0.009).18/29 (62%) patients with lower-grade glioma had malignant progression to higher grades. Those who progressed from grade 2 to 3 or 4 had higher myo-inositol/choline (mean±SD, 2.62±0.64, N=18) compared to those who recurred as grade 2 (1.88±0.52, N=5, p = 0.015). Of the 51 patients, 40/65 in vivo spectra voxels centered where the tissues were sampled displayed good quality, with 27 having active tumors, 7 having treatment effects, and 6 having mixed tumors/treatment effects. The levels of choline/creatine and myo-inositol+glycline/creatine were significantly correlated with MIB-1 (p< 0.0001, =0.034; r=0.563, 0.349), but no difference was found between the locations with active tumors and treatment effects. DISCUSSION: This study demonstrated metabolite variations at the recurrence. Ongoing studies will further characterize the metabolite profiles for malignant progression and differences between active tumors and treatment effects.