NIMG-28. DIFFUSION DIFFERENCES IN LOW TO HIGH GRADE GLIOMA TRANSFORMATION

Abstract

Abstract PURPOSE Apparent diffusion coefficient (ADC) maps offer tremendous insight into cellularity and local tissue environment. Proliferation of tumor cells results in increased local cellularity and cell density, thus impairing the free diffusion of water molecules. Therefore, hypointense ADC signals are thought to reflect regions of increased cellularity, and are useful imaging markers for distinguishing neoplastic cellular processes. In this study, we tested the hypothesis that changes in ADC intensity over the course of disease progression differ in progressing gliomas within the tumor-delineated region. METHODS Sixty-three glioma patients from our brain bank were separated into two groups. Group 1 included patients with Grade 1-3 gliomas at surgical diagnosis and progressed to a Grade 4 glioma at autopsy. The second group of patients were those with an initial and final Grade 4 glioma diagnosis. Contrast-enhancing tumor regions on MRI sequences were annotated with an established machine-learning algorithm for each patient on their initial and final MRIs. Average ADC intensity within the contrast enhancing regions was extracted and change in signal intensity between the first and last MRI was calculated for each patient. Finally, the mean intensity difference between Group 2 and Group 1 were evaluated using a linear mixed model. RESULTS We report a negative mean difference between the Group 2 and Group 1s change in ADC intensity over time (p=0.03). CONCLUSIONS These findings suggest changes in cellularity, identified by ADC, are greater in patients who progress from low-to-high grade gliomas. With these results, it may be concluded that change in ADC hypointensity, as a proxy for cellularity, may be a defining imaging characteristic to identify progressing gliomas.