TMET-16. 3D spectroscopic imaging detects hidden peritumoral activity in gliomas

Abstract Gliomas extend beyond visible tumor borders, yet standard imaging fails to detect this infiltration. We applied volumetric proton MR spectroscopic imaging (3D-MRSI) with the MIDAS pipeline to identify metabolic changes beyond MRI-defined margins in IDH-mutant gliomas. Seven patients (2 glioblastomas wild-type, 3 astrocytomas IDH-mutants, 2 oligodendrogliomas IDH-mutants) underwent 3T echo-planar spectroscopic imaging. MIDAS processed MRSI data through a standardized pipeline: spatial and spectral Fourier transforms with B₀ correction, lipid extrapolation via LITE, spectral fitting (FITT), signal normalization, and quality mapping (QMAPS), yielding high-resolution metabolite maps co-registered with structural MRI. Metabolite concentrations (NAA, Cho, Cr, mIno, Lac) and ratios (Cho/Cr, NAA/Cr) were quantified in tumoral, peritumoral, and contralateral regions. Significant differences were observed for Cho/Cr and NAA/Cr across regions (ANOVA, p = 0.0162 and 0.0272), with Cho/Cr notably elevated in tumor vs. peritumoral tissue (p = 0.0226). NAA/Cr was reduced peritumorally (p = 0.005), reinforcing the idea of a metabolic gradient. Cr and Lac levels remained unchanged. Importantly, when stratified by IDH status, CHO/Cr and mIno levels exhibited moderate-to-large group differences within peritumoral and contralateral regions (Cohen’s d = 0.70–1.21), despite normal FLAIR signal. This study provides a spectroscopic evidence of metabolic alterations beyond visible tumor borders in gliomas using a clinically deployable MRSI platform. MIDAS-enhanced 3D-MRSI reliably captured subtle peritumoral changes undetectable by standard imaging. These findings suggest that metabolic mapping could refine surgical and radiotherapy planning by identifying biologically active tumor margins. Our results validate MIDAS as a robust tool for high-throughput, quality-controlled MRSI analysis and position 3D-MRSI as a frontier imaging modality in glioma management. Future larger-scale studies are warranted to validate these peritumoral biomarkers and integrate spectroscopic imaging into clinical workflows.