PREOPERATIVE DYNAMIC CONTRAST-ENHANCED MRI CORRELATES WITH MOLECULAR MARKERS OF HYPOXIA AND VASCULARITY IN SPECIFIC AREAS OF INTRATUMORAL MICROENVIRONMENT AND IS PREDICTIVE OF PATIENT OUTCOME

摘要

BACKGROUND: Measures of tumor vascularity and hypoxia have been correlated with glioma grade and outcome. Dynamic contrast-enhanced MRI (DCE-MRI) can noninvasively map multiple parameters of tumor hypoxia, blood flow, and vascularity. In this prospective observational cohort pilot study, our goal was to use preoperative imaging to correlate specific glioma samples with molecular markers of hypoxia, tumor vascularity, and tumor proliferation along with progression-free and overall patient survival. METHODS: Pharmacokinetic modeling methods were used to generate maps of tumor blood flow (F), extraction fraction (E), permeability-surface area product (PS), transfer constant (Ktrans), washout rate (kep), interstitial volume (v_e), blood volume (vb), capillary transit time (tc), and capillary heterogeneity (α(-1)) from preoperative DCE-MRI data in human glioma patients. Tissue samples were obtained using an MRI-guided intraoperative navigation system from areas of peritumoral edema (PE), active tumor (AT), hypoxic penumbra (HP), and necrotic core (NC). Each sample was evaluated for tumor vascularity, proliferation and expression of hypoxia-regulated molecules including hypoxia inducible factor-1 (HIF-1) and vascular endothelial growth factor (VEGF). Mean DCE-MRI parameter values corresponding to tissue sample sites and marker expression were compared with imaging results. RESULTS: Patient survival correlated with DCE parameters in two cases: α(-1) in AT and ve in areas of PE. Statistically significant correlations between DCE parameters and tissue markers were also observed. Specifically, 1) Vb correlated with VEGF (r = 0.604, p = 0.032) and HIF-1 (r = 0.747, p = 0.043) expression in AT; 2) tc correlated with HIF-1 in AT (r = 0.659, p = 0.017), HP (r = 0.761, p = 0.023), and PE (r = 0.697, p = 0.031) and with VEGF in AT (r = 0.588, p = 0.034) and HP (r = 1.00, p = 0.003); and 3) α(-1) correlated with VEGF in AT (r = 0.599, p = 0.034) and PE (r = 0.973, p < 0.0001). kep correlated with VEGF in PE (r = 0.761, p = 0.0065) and NC (r = 0.931, p = 0.007) and E correlated with HIF-1 expression in NC (r = 0.943, p = 0.017). In addition, MIB-1 index was found to be predictive of OS (p = 0.044) and is correlated with VEGF expression in HP (r = 0.7933, p = 0.0071) and PE (r = 0.4546). Increased microvascular density (MVD) found to correlate with worse patient outcome (p = 0.026). With these results we describe a new clinical trial that we will implement to use these biomarkers in guiding therapeutic interventions for patients with malignant gliomas. CONCLUSIONS: Our findings suggest that it may be possible to use DCE-MRI to make noninvasive preoperative predictions of areas of tumor with increased hypoxia and proliferation. This has the potential to both make unprecedented prognostic decisions and to guide therapies to specific tumor areas SECONDARY CATEGORY: Imaging.