Semi-solid MT and APTw CEST-MRI predict clinical outcome of patients with glioma early after radiotherapy
Florian Kroh, Nikolaus von Knebel Doeberitz, Johannes Breitling, Srdjan Maksimovic, Laila König, Sebastian Adeberg, Moritz Scherer, Andreas Unterberg, Martin Bendszus, Wolfgang Wick, Peter Bachert, Jürgen Debus, Mark E. Ladd, Heinz-Peter Schlemmer, Andreas Korzowski, Steffen Goerke, Daniel Paech
Abstract
Purpose
The purpose of this study was to compare the potential of asymmetry-based (APTwasym), Lorentzian-fit-based (PeakAreaAPT and MTconst), and relaxation-compensated (MTRRexAPT and MTRRexMT) CEST contrasts of the amide proton transfer (APT) and semi-solid magnetization transfer (ssMT) for early response assessment and prediction of progression-free survival (PFS) in patients with glioma.
Methods
Seventy-two study participants underwent CEST-MRI at 3T from July 2018 to December 2021 in a prospective clinical trial four to 6 wk after the completion of radiotherapy for diffuse glioma. Tumor segmentations were performed on T2w-FLAIR and contrast-enhanced T1w images. Therapy response assessment and determination of PFS were performed according to response assessment in neuro oncology (RANO) criteria using clinical follow-up data with a median observation time of 9.2 mo (range, 1.6–40.8) and compared to CEST MRI metrics. Statistical testing included receiver operating characteristic analyses, Mann–Whitney-U-test, Kaplan–Meier analyses, and logrank-test.
Results
MTconst (AUC = 0.79, p < 0.01) showed a stronger association with RANO response assessment compared to PeakAreaAPT (AUC = 0.71, p = 0.02) and MTRRexMT (AUC = 0.71, p = 0.02), and enabled differentiation of participants with pseudoprogression (n = 8) from those with true progression (AUC = 0.79, p = 0.02). Furthermore, MTconst (HR = 3.04, p = 0.01), PeakAreaAPT (HR = 0.39, p = 0.03), and APTwasym (HR = 2.63, p = 0.02) were associated with PFS. MTRRexAPT was not associated with any outcome.
Conclusion
MTconst, PeakAreaAPT, and APTwasym imaging predict clinical outcome by means of progression-free survival. Furthermore, MTconst enables differentiation of radiation-induced pseudoprogression from disease progression. Therefore, the assessed metrics may have synergistic potential for supporting clinical decision making during follow-up of patients with glioma.