Sci-Thurs AM: YIS-09: Detection of Lung Remodeling Following Radiation Therapy Using Hyperpolarized 3He Magnetic Resonance Imaging

Abstract

Purpose: To assess subjects with clinically diagnosed radiation-induced lung injury (RILI) at a single center using hyperpolarized magnetic resonance imaging (MRI). Measurements of the apparent diffusion coefficient (ADC) and percnt ventilated volume (PVV) were examined at an initial visit and 6-month follow-up. Furthermore, feasibility of , — CT and — radiation planning image registration was evaluated. Methods: Seven subjects diagnosed with RILI provided written informed consent and were enrolled. Four subjects returned 22.0 ± 0.8 weeks later for a second visit. All subjects underwent spirometry and plethysmography. Three MR image sets were acquired at 3.0T; a image, a static ventilation image, and a diffusion weighted image. PVV and ADC were evaluated for each lung independently and combined, and differences between lungs and between scans were evaluated. Feasibility of registration was assessed for the center slice using the overlap coefficient. Results: At baseline, PVV was significantly different (), and lower in the ipsilateral lung as compared to the contralateral lung. Significant increases were found in the contralateral lung for both ADC (mean increase , ), and PVV (mean increase 16% ± 6%, ). Good feasibility of — CT registration was observed, with a center slice overlap coefficient of 75.3 ± 10.5. Conclusions: Hyperpolarized MRI is well tolerated in subjects with moderate to severe RILI. Using this technique, structural and functional differences between lungs and over time can be detected in RILI subjects. Multi-modality image registration incorporating MRI is feasible.