The degree of angiogenesis and ability to recruit vessels are important factors in the evaluation of brain tumor malignancy. Cerebral blood volume (CBV) and flow (CBF) are important indicators of the homeostasis of the microvasculature. CBV alterations may be more sensitive and may occur earlier than CBF changes in tumors(1). Dynamic susceptibility contrast (DSC) MRI is currently the standard perfusion MRI approach in brain tumors(2,3). Nevertheless, the gadolinium-based contrast agents used in DSC are associated with nephrogenic systemic fibrosis in subjects with renal insufficiency(4,5), which often occurs in tumor patients undergoing chemotherapy. This makes follow-up evaluation of tissue perfusion during chemotherapy difficult with DSC(6,7). The recently developed inflow-based vascular-space-occupancy (iVASO) MRI can quantify arteriolar CBV (CBVa) without need for an exogenous contrast agent(8-10). Because the generation of arterioles may occur before capillary growth and CBF changes in angiogenesis(11), the investigation of CBVa may furnish information that is not obtainable from total CBV/CBF measures. Here, we applied iVASO MRI to measure CBVa and arterial transit time (ATT) in primary cerebral glioma patients, and compared the results with DSC MRI to evaluate its potential additional value for tumor perfusion imaging.Twelve brain tumor patients were scanned (Table 1). Each subject underwent a clinical and a research MRI session with DSC and iVASO MRI, respectively. The clinical session was conducted on a 3T Siemens scanner with a 12-channel head coil: MPRAGE, FLAIR; DSC MRI (Gd-DTPA) with gradient-echo echo-planar-imaging (GRE-EPI), TR/TE=1640/35ms, voxel=2x2x4mm³, 6min. The research session was performed on a 7T Philips scanner with a 32-channel head coil. Dielectric pads(12) were used to mitigate B1 inhomogeneity(13). Single-slice iVASO MRI was performed with GRE-EPI: TR/TI=10000/1626, 5000/1382, 3100/1081, 2000/797, 1700/691, 1300/563ms; voxel=2.5x2.5x3mm³; crusher gradients V_enc=10cm/s, z-direction; 7min. The iVASO slice was prescribed on MPRAGE images and centered on the tumoral region. Analysis: All data were pre-processed using SPM. DSC and iVASO images were analyzed using DSCoMAN/ImageJ and in-house Matlab routines(8), respectively. ROIs of tumoral and similar-sized contralateral gray matter (GM) regions were selected on anatomical images. The same ROIs were used for DSC and iVASO. Efforts were made to avoid necrotic/cystic regions and major macrovessels. GM was chosen as the internal reference mainly because iVASO is less sensitive in regions with very long ATT such as white matter(14). This choice is justified by the fact that most of the lesions in our study involved GM, and should not affect the correlation analysis (main finding, see below). To reduce influence from individual baseline differences, the ratio of tumor/contralateral was calculated. Group difference was examined by t-tests. Correlations were assessed using multiple regression with age included as an independent variable. Multiple comparisons were corrected with false discovery rate.All DSC and iVASO measures are listed in Table 1 and statistical results are summarized in Table 2. DSC CBV in tumors were all higher than contralateral regions in grade IV patients, but were comparable in grade II/III patients. DSC mean transit time (MTT) was comparable between tumoral and contralateral regions in all patients. The iVASO CBVa was significantly lower in grade II tumors than in contralateral GM. CBVa in all grade IV patients was greater in tumor than contralateral (but not significant). The iVASO ATTs were all shorter in tumor than contralateral except for case 1. Correlations: The iVASO CBVa ratio showed the strongest correlation with tumor grade (Figure 1b). The DSC CBV ratio showed a trend for correlation with tumor grade (Figure 1a). Subject age showed negligible contribution in all correlations analyzed. A trend of correlation between DSC CBV ratio and iVASO CBVa ratio was observed (Figure 2).The DSC results are consistent with the literature(15,16). Our main finding is that iVASO CBVa in tumors demonstrated a stronger correlation with tumor grade than DSC CBV in the same patients. Also, the average difference between grades II and III oligodendrogliomas was greater in iVASO CBVa. This implies that changes in CBVa might be a better classifier than changes in the total microvasculature for the stratification of brain tumors, especially between grades II and III. This may be of potential diagnostic value since differences between grade II and IV tumors are usually evident on contrast-enhanced MRI images, while differentiation between grade II and III tumors is often more difficult. Further investigation is warranted to validate our findings in a larger cohort. We will use the recently improved 3D iVASO MRI(17) with whole-brain coverage and same scan time in follow-up studies to evaluate CBVa changes in the entire tumor and potential heterogeneities among different sub-regions.