Pilot study shows DR-CSI differentiates treatment effects from viable tumors in brain metastasis, suggesting improved diagnosis.
INTRODUCTION Differentiating treatment effects (TxE), including cases with radiation necrosis and/or with gliosis and hyalinized vessels on histopathology, from tumor progression in irradiated brain tumors remains a clinical challenge due to overlapping imaging characteristics on conventional MRI. The presence of mixed pathology further complicates diagnosis. Diffusion-Relaxation Correlation Spectroscopic Imaging (DR-CSI) is an emerging technique that characterizes tissue microstructure by measuring the joint variations in T2 relaxation and diffusivity. This study evaluates the feasibility of DR-CSI in distinguishing TxE from viable tumor and benchmarks it against conventional diffusion/perfusion imaging and sodium MRI. METHODS Sixteen patients with suspected brain tumors (glioma or brain metastasis) and/or TxE underwent preoperative MRI, including DR-CSI, sodium, and a standard clinical protocol with diffusion and perfusion sequences, on 3T scanners. DR-CSI was acquired using a 30-point TE-b-value spin-echo EPI sequence. Histopathological diagnosis was: 8 recurrent tumors, 4 pure TxE, and 4 mixed pathology cases. Voxel-wise diffusion-relaxation (D-T2) spectra were calculated from DR-CSI datasets using dictionary-based fitting. Lesion segmentations, including the contrast-enhancing and necrotic tumor tissue, were obtained using an automated AI-based tool. Lesion-averaged D-T2 spectra were compared between tumor and pure TxE groups using a point-by-point Mann-Whitney U-test to identify discriminative spectral regions representing TxE. A “TxE index” for each lesion was calculated by integrating spectral power over these regions. TxE indexes were subsequently compared across groups. RESULTS In TxE, mixed, and tumor groups, the TxE indexes were 0.30±0.04, 0.27±0.05, and 0.14±0.08, respectively. TxE indexes showed significant differences across groups: tumor vs. non-tumor (p=0.02, AUC=0.81), TxE vs. non-TxE (p=0.01, AUC=0.81), and pure TxE vs. pure tumor (p=0.02, AUC=0.90, sensitivity=1.0, specificity=0.83). In contrast, conventional imaging biomarkers (ADC and rCBV) and sodium showed no significant group differences (all p>0.4). CONCLUSION These preliminary results support the potential of DR-CSI as a non-invasive imaging biomarker to resolve microstructural differences and distinguish TxE from tumor, especially in complex or ambiguous cases where conventional imaging is limited.
No takes yet. Share an insight, caveat, or question.
Shao et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: