Abstract
PURPOSE: Delayed ferumoxytol enhancement on T(1) -weighted images appears visually similar to gadoteridol enhancement. The purpose of this study was to quantitatively compare ferumoxytol T(1) enhancement to gadoteridol enhancement with an objective, semi-automated method. METHODS: 206 sets of post-gadoteridol and 24 h post-ferumoxytol T(1) -weighted scans from 58 high grade glioma patients were analyzed (9 pre-chemoradiation, 111 < 90 days post-chemoradiation, 21 > 90 days post-chemoradiation, 65 post-bevacizumab scans). Enhancement volumes and signal intensities normalized to normal appearing tissue proximal to enhancement were calculated with a semi-automated method. Enhancement cube root volumes (D) and signal intensities (SI) were compared between the 2 contrast agents, and relative difference of D and SI were compared in different treatment groups with multivariate analysis. Within patient differences in D and SI before and after treatment with bevacizumab or steroid were assessed in 26 patients in each treatment group. RESULTS: When compared to gadoteridol, ferumoxytol D was 13.83% smaller and SI was 7.24% lower (P < 0.0001). The relative differences in D and SI between the 2 contrast agents were not significantly different between treatment groups (P > 0.05). Relative difference in D and SI did not change significantly in response to bevacizumab (P = 0.5234 and P = 0.2442, respectively) or to steroid (P = 0.3774, P = 0.0741) in the within patient comparison. CONCLUSION: The correlation between the 2 contrast agents' enhancement size and signal intensity and their similar behavior in response to therapy suggest that ferumoxytol can be used for revealing enhancement in high grade glioma patients. Magn Reson Med 80:224-230, 2018. © 2017 International Society for Magnetic Resonance in Medicine.