18F-FDG PET/CT as a Potential Predictor of Lung Inflammation After Arc-Based Radiotherapy for Esophageal Cancer: A Pilot Study

Abstract PurposeFluorine-18-fuorodeoxyglucose (18F-FDG) positron emission tomography/computerized tomography (PET/CT) scan serves as a useful tool not only for tumor detection, radiotherapy (RT) target volume delineation but also for the assessment of the inflammatory changes in normal organs. Previously we proposed the volume-based algorithm (VBA) method to reduce low dose-volume of lung by improving the arc angle in dynamic arc-based RT. The aim of this study was to assess lung inflammation by integrating 18F-FDG PET/CT with VBA before and after arc-based RT for esophageal cancer (EC).MethodsThirty EC patients underwent 18F-FDG PET/CT imaging before RT (pre-RT) and after RT (post-RT) on a retrospective pilot study. The VBA was used to define the high dose (HD) (≧ 5 Gy) region and the low dose (LD) (< 5 Gy) region in the lungs. The maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean), and global lung glycolysis (GLG) of the HD and LD regions in the lungs were quantified. The pre-RT and post-RT SUVmax, SUVmean, and GLG of HD and LD regions in lungs were analyzed in radiation pneumonitis (RP) ≧ grade 1 and non-radiation pneumonitis (nRP) lungs. The mean lung dose (MLD), V5, V10, V15, V20, V25, V30 in lungs were analyzed. Vx indicates the organ volume percentage exceeding a radiation dose of x (Gy). Receiver-operating characteristic curves were used to identify optimal cut-off values for RP after RT.ResultsIn RP lungs (n = 30), the SUVmax, SUVmean and GLG of the HD region between pre-RT and post-RT showed significant increases (all p < 0.05). Whereas there were no significant differences in those of the HD and LD regions in nRP lungs (n = 22). The post-RT SUVmax (2.78 vs. 2.07, p = 0.000) and post-RT SUVmean (0.64 vs. 0.52, p = 0.015) of the HD region in RP lungs were significantly higher than those in nRP lungs. The MLD (10.15 Gy vs. 8.11 Gy, p = 0.041), lung V5 (49.78% vs. 38.07%, p = 0.010) and lung V10 (32.25% vs. 24.71%, p = 0.017) in the RP lungs were significantly higher than those in nRP lungs. The area under the curve (AUC) of post-RT SUVmax of the HD regions was 0.852, and the AUC of the lung V5 was 0.727. For predicting RP, the optimal cut-off values of post-RT SUVmax and lung V5 were > 2.28 and > 47.14%. respectively. ConclusionThis study successfully integrated 18F-FDG PET/CT with VBA to assess RP in EC patients undergoing dynamic arc-based RT. The post-RT SUVmax and post-RT SUVmean of HD (≧5 Gy) regions can be used to evaluate RP. The post-RT SUVmax > 2.28 of HD regions and lung V5 > 47.14% may be a potential predictor of RP. 18F-FDG PET/CT is a promising tool to detect RP for EC patients treated with arc-based RT.