Purpose: We have previously shown that adaptive lung SBRT planning can mitigate dose to adjacent organs at risk (OARs). The aim of this study is to investigate factors that affect dosimetric effectiveness of adaptive lung SBRT, so that we may predict which patients benefit most from an adaptive strategy. Methods: 20 patients who underwent SBRT for lung cancer at our institution were included. Cone‐beam CT(CBCT) images acquired before each fraction were retrospectively analyzed to determine tumor volume changes throughout the course of treatment. Adaptive plans were created for each CBCT dataset using the same planning parameters as the original CT‐based plan. Two accumulative dose plans, non‐adaptive(PNON) and adaptive(PADP), were generated for each patient using deformable image registration to sum dose from each fraction. Dosimetric comparisons between PNON and PADP were performed for cord, lung, chest wall, and esophagus. Statistical significances were tested using paired t‐test(significance level=0.05). Correlations were evaluated between dosimetric changes from adaptive planning(Delta_D) and potential impacting factors, including tumor‐to‐OAR distances(dT‐OAR), initial ITV(ITV1), ITV change(Delta_ITV), and effective ITV diameter change(Delta_dITV). Results: Compared to PNON, PADP significantly reduced dosimetric metrics of all OARs (p=0∼0.03). Dealta_D was found to positively correlate with Delta_dITV/dT‐OAR for dose‐to‐5cc (D5cc) of esophagus (r=0.77), Dmax of esophagus (r=0.76), and D30cc of chest wall (r=0.72). Stronger correlations between Delta_dITV/dT‐OAR and Delta_D30cc of chest wall were discovered for peripheral (r=0.82) and central tumors (r=0.94), respectively. No correlation was found between Delta_dITV/dT‐OAR and Delta_D1.2cc or Delta_D0.35cc of spinal cord, largely due to two IMRT patients planned with tight cord constraints. After removing these two cases, good correlations were recovered between Delta_dITV/dT‐OAR and Delta_D1.2cc, Delta_D0.35cc of spinal cord (r=0.75 and 0.71, respectively). Conclusion: Adaptive SBRT planning for selected patients resulted in significant dose reduction in adjacent OARs. Tumor diameter change and tumor‐to‐OAR distance are potential indicators for dosimetric effectiveness of adaptive lung SBRT.
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging