High-grade gliomas remain one of the most fatal tumors with median survivals ranging from 3–5 years for WHO (World Health Organizaiton) grade III anaplastic astrocytoma and 12–14 months for WHO grade IV glioblastoma.1 Until recently, few treatment options were available for patients once standard chemoradiation therapy with temozolomide had failed. With the identification of multiple signaling pathways and growth factors essential for tumor angiogenesis, several new anti-angiogenic drugs have been developed. One of these, bevacizumab, is a recombinant humanized monoclonal IgG1 antibody that binds to human vascular endothelial growth factor (VEGF) and inhibits angiogenesis.2 Bevacizumab received accelerated FDA approval for the treatment of recurrent glioblastoma on May 5, 2009, and has since become the standard of care for treatment of high-grade glioma. Patients with recurrent high-grade glioma treated with bevacizumab have demonstrated an excellent radiographic response rate and improved clinical outcome when compared with historical data.3 In 2 of the early prospective phase II clinical trials,4,5 the rate of progression-free survival (PFS) at 6 months was 29%–42.6% compared with the 15% historical control rates. However, a statistically significant increase in overall survival (OS) was not demonstrated. In these trials, the traditional approach for assessing response, the MacDonald Criteria,6 were used. The MacDonald criteria are based on the 2-dimensional measurement of enhancing tumor on MRI or CT. In later studies using a modified MacDonald criteria that included fluid-attenuated inversion recovery (FLAIR) imaging, increases in both PFS and OS were reported.3 Subsequently, the RANO (Response Assessment in Neuro-Oncology) criteria were published in 2010 as an update to the MacDonald criteria.7 The RANO criteria newly incorporates FLAIR imaging to assess the degree of peritumoral edema and is therefore the standard approach used by many institutions for assessing disease progression and treatment response in glioblastoma. Though these assessment criteria are considered standard, it is becoming increasingly clear that these anatomic measures of tumor response to bevacizumab are often unreliable. Contrast-agent enhancement and FLAIR hyperintensities on imaging primarily reflect the breakdown of the blood-brain-barrier. Bevacizumab acts as a powerful corticosteroid decreasing the permeability of the blood-brain-barrier.8 Thus, rapid decreases in the degree of contrast enhancement and FLAIR hyperintensity may not necessarily reflect true changes in tumor biology or cellular burden. As such, measures derived from standard imaging may neither be able to predict OS reliably nor be an accurate representation of PFS. The antivascular permeability effect of VEGF inhibition likely accounts for the high initial radiographic response rate, with profound reductions in enhancement on MRI being observed as soon as 24 hours after the first dose of bevacizumab.5,9 Yet, as demonstrated in a recent study,5 this is clearly too early for an equally profound cellular antitumor effect since rapid subsequent progression developed in nearly half of the initial responders. Still, patients do derive clinical benefit from treatment as manifested by decreased cerebral edema, improved neurological symptoms, and decreased requirement for corticosteroids. In addition, some bevacizumab-mediated antumor effect likely occurs in at least a subpopulation of patients because nearly half of the initial radiographic responders were progression free for more than 6 months. Consequently, biomarkers that more directly measure the biologic effects of these agents are vital for better stratification into various treatment arms, as well as the improvement of research pursuing new therapies and strategies for high-grade tumors. Relative cerebral blood volume (rCBV) imaging, which is derived from dynamic susceptibility contrast (DSC) MRI, has the potential to serve as a predictive biomarker of response. With DSC-MRI, T2 or T2*-weighted images are acquired with high temporal resolution during the bolus adminstration of a gadolinium (Gd) contrast agent.10 The resulting rCBV image maps have demonstrated the ability to predict tumor grade11–16 and survival17 and to distinguish posttreatment radiation effects from recurrent tumor.18–20 In preliminary studies the, potential of rCBV to predict response to anti-angiogenic therapy more reliably than standard MRI has been demonstrated.21 Though DSC-MRI has shown great promise for the evaluation of brain tumors, its role as a prognostic measure for response to bevacizumab has not been fully evaluated. Consequently, the goal of this study is to investigate the utility of rCBV, or more specifically standardized rCBV (stdRCBV), to predict the overall survival (OS) and progression free survival (PFS) in response to bevacizumab in comparison with FLAIR-hyperintense and contrast-agent enhancing tumor volumes.