Tumor Blood Volume Measurement in Brain Gliomas Using Velocity-Selective Arterial Spin Labeling

Abstract Background To evaluate the feasibility and performance of velocity selective (VS) ASL based cerebral blood volume (CBV) mapping among glioma patients in clinical practice, comparing with the VSASL based cerebral blood flow (CBF) mapping and dynamic susceptibility contrast-enhanced perfusion-weighted imaging (DSC-PWI). Methods This study included patients with histologically proven brain glioma who underwent preoperative MRI including VSASL based CBV, CBF and DSC-PWI between 2017 and 2021. Visual inspection was performed to evaluate the lesion conspicuity on VSASL derived CBV maps based on 1–3 criteria by comparing to the surrounding parenchyma. The relative values of maximum tumor blood volume (rTBV) and tumor blood flow (rTBF), derived from VSASL or DSC-PWI were compared between low-grade and high-grade glioma. Linear regression and Bland–Altman analyses were constructed to evaluate the correlation and agreement of rTBV measurements between VSASL method and DSC-PWI. The diagnosis ability for discrimination between low- and high-grade glioma were evaluated using ROC curves. Results Forty-eight participants (mean age, 45 ± 13 years; 25 men; 23 high-grade gliomas) were evaluated. The lesion conspicuity of VSASL based CBV maps was good on visual inspection (averaged scores: 2.26 ± 0.76, weighted kappa of 0.8 between readers). Moreover, VSASL provided highly correlated quantifications of rTBV (R2 = 0.83, p < 0.001) compared to DSC-PWI, and further improved diagnostic performance than VSASL derived rTBF measurements (ROC AUC = 0.94 vs. 0.89). Conclusions VSASL served as a promising and accurate means in quantification of TBV for glioma stratification in clinical patients, indicating its potential as a viable non-contrast alternative to DSC-PWI for brain tumor applications.