The role of amide proton transfer imaging in detecting active malignant glioma.

person J. O. Blakeley The Johns Hopkins University School of Medicine, Baltimore, MD info_outline J. O. Blakeley, X. Ye, M. Lim, H. Zhu, L. Blair, A. Quinones-Hinojosa, C. Eberhart, P. Barker, J. Laterra, P. C. M. van Zijl, J. Zhou

Authors person J. O. Blakeley The Johns Hopkins University School of Medicine, Baltimore, MD info_outline J. O. Blakeley, X. Ye, M. Lim, H. Zhu, L. Blair, A. Quinones-Hinojosa, C. Eberhart, P. Barker, J. Laterra, P. C. M. van Zijl, J. Zhou Organizations The Johns Hopkins University School of Medicine, Baltimore, MD, Johns Hopkins University School of Medicine, Baltimore, MD, F. M. Kirby Research Center for Functional Brain Imaging; The Johns Hopkins University, Baltimore, MD Abstract Disclosures Research Funding Other Foundation Background: The inability to distinguish active tumor from treatment-related injury is a major challenge in neuro-oncology. Amide Proton Transfer (APT) is a new magnetic resonance imaging (MRI) technique thought to image protein content at the cellular level. We assessed if APT-weighted MRI can improve the diagnostic value of MRI for malignant gliomas (MG). Methods: Two groups of patients were accrued. Group 1: patients with MG were imaged 1mo after radiation (RT) and temozolomide (TMZ) to assess APT, FLAIR, and T1w Gadolinium (T1wGd) enhancing volumes. Group 2: pre-operative glioma patients were imaged to correlate APT hyperintensity with pathology. Whole brain APT-weighted imaging was performed (FOV = 212x186 mm2; resolution = 2.2x2.2 mm2; 15 slices at 4.4 mm thickness) with FLAIR and T1wGd enhanced MRI. Regions of interest (ROI) were: (a) high APT-weighted signal intensity (≥ 2.5%), (b) T1wGd and (c) FLAIR enhancement. APT-based ROIs were digitally overlaid on T1wGd and FLAIR images and uploaded to the surgical navigation system for tissue sampling. A neuro-pathologist blinded to ROI reviewed all samples. The primary endpoints were volume of abnormal signal on MRI (group 1) and correlation between APT-weighted % hyperintensity and pathology (group 2). Results: Nine patients were studied in group 1. FLAIR volume was largest (median 34 cm3, range 2-117), APT-weighted signal intensity was intermediate (9 cm3, range 0-31), and T1wGd volume was smallest (median: 5 cm3, range 0-31). Ten patients were studied in group 2. Regions of high pre-operative APT-weighted signal intensity (≥2.5%) were associated with pathologically confirmed active MG in 7/7 patients; 3/3 patients with low APT-weighted signal intensity (<2.5%) had pathologically confirmed low grade glioma. Conclusions: APT imaging provides spatially distinct visual data from either FLAIR or T1wGd in patients with glioma. We observed 100% correspondence of high or low APT-weighted signal intensity with pathologically confirmed active MG (≥2.5%) or not (<2.5%) in patients with both newly diagnosed and recurrent/progressive gliomas. Additional longitudinal and pathologic correlation studies are ongoing to confirm the sensitivity and specificity of APT for active MG.