Abstract LB-248: A single-institution prospective study evaluating the functional diffusion map (fDM) as an early imaging biomarker for overall survival in high-grade glioma

Introduction: Standard assessment of radiologic response in patients with malignant glioma is traditionally carried out using T1 gadolinium-enhanced MRI at approximately 10 weeks following treatment initiation. Diffusion-weighted MRI of gliomas, analyzed by the functional diffusion map (fDM), may provide an early measure of response during treatment to predict patient survival (OS). Tumor cells restrict the random diffusion of water molecules within a tumor; therefore, diffusion will increase during successful therapy. The fDM imaging biomarker is a voxel-wise method for analysis of DW-MRI data providing for quantification of treatment-associated spatial alterations in tumor diffusion values earlier than radiologic response. Methods: A prospective single-institution trial of early assessment of brain tumor response by MRI was performed early during radiation therapy. Sixty patients with high grade glioma undergoing radiation therapy were enrolled in an IRB-approved study of intra-treatment MRI 1, 3, and 10 weeks after the initiation of treatment. Receiver operating characteristic curve analysis was used to evaluate imaging parameters as a function of patient survival at one-year. Log Rank and Cox Proportional Hazards models were utilized to assess OS. Results: Measurable changes in diffusion MRI were observed as early as one week into treatment. The volume of tumor with increased diffusion as assessed by fDM 3 weeks after the start of therapy was the strongest predictor of patient survival at one year, with larger fDM predicting longer median survival [52.6 vs. 10.9 months, log-rank, P Conclusions: Compared to conventional neuro-imaging, fDM provided an earlier assessment of equal predictive value during the course of treatment, providing the opportunity to individualize therapy. The combination of fDM and RR provided a more accurate prediction of patient survival than either metric alone.