Premium
WE‐FG‐202‐08: Assessment of Treatment Response Via Longitudinal Diffusion MRI On A MRI‐Guided System: Initial Experience of Quantitative Analysis
Author(s) -
Qi X,
Yang Y,
Yang L,
Low D,
Sheng K
Publication year - 2016
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4957920
Subject(s) - nuclear medicine , medicine , effective diffusion coefficient , diffusion mri , magnetic resonance imaging , radiation therapy , diffusion imaging , radiology , radiation treatment planning
Purpose: To report our initial experience of systematic monitoring treatment response using longitudinal diffusion MR images on a Co‐60 MRI‐guided radiotherapy system. Methods: Four patients, including 2 head‐and‐necks, 1 sarcoma and 1 GBM treated on a 0.35 Tesla MRI‐guided treatment system, were analyzed. For each patient, 3D TrueFISP MRIs were acquired during CT simulation and before each treatment for treatment planning and patient setup purposes respectively. Additionally, 2D diffusion‐weighted MR images (DWI) were acquired weekly throughout the treatment course. The gross target volume (GTV) and brainstem (as a reference structure) were delineated on weekly 3D TrueFISP MRIs to monitor anatomy changes, the contours were then transferred onto the corresponding DWI images after fusing with the weekly TrueFISP images. The patient‐specific temporal and spatial variations during the entire treatment course, such as anatomic changes, target apparent diffusion coefficient (ADC) distribution were evaluated in a longitudinal pattern. Results: Routine MRI revealed progressive soft‐tissue GTV volume changes (up to 53%) for the H&N cases during the treatment course of 5–7 weeks. Within the GTV, the mean ADC values varied from −44% (ADC decrease) to +26% (ADC increase) in a week. The gradual increase of ADC value was inversely associated with target volume variation for one H&N case. The maximal changes of mean ADC values within the brainstem were 5.3% for the H&N cases. For the large size sarcoma and GBM tumors, spatial heterogeneity and temporal variations were observed through longitudinal ADC analysis. Conclusion: In addition to the superior soft‐tissue visualization, the 0.35T MR system on ViewRay showed the potential to quantitatively measure the ADC values for both tumor and normal tissues. For normal tissue that is minimally affected by radiation, its ADC values are reproducible. Tumor ADC values show temporal and spatial fluctuation that can be exploited for personalized adaptive therapy.