Prediction of Sensitivity of IMPT Plans to Anatomical Change Due to Plan Complexity
J Cheung1*, L Dong2, P Park1, X Zhu1, R Kudchadker1, S Frank1, M Palmer1, W Liu1, L Court1, (1) UT MD Anderson Cancer Center, Houston, TX, (2) Scripps Proton Therapy Center, San Diego, CASU-E-T-683 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To assess the significance of the treatment plan complexity on the sensitivity of multifield optimized intensity modulated proton therapy (IMPT) plans to inter-fraction anatomical changes.
Methods: Eight lung cancer patients were planned with IMPT to meet clinical criteria. These plans were retrospectively analyzed for this study. A complexity score based on the in-field dose heterogeneity within the target weighted by the number of regions of dose homogeneity was devised to evaluate the complexity of IMPT plans. The complexity score was calculated for each of the treatment plans and the change in target coverage over the course of treatment was assessed for each patient based on weekly CT imaging. The change in proton range to the distal edge of the clinical target volume was used as a quantitative metric for anatomical change. Greater than 5% drop to the target D99% was chosen as an action level to signal a clinically significant drop in target coverage that would require adaptive replanning.
Results: For plans with a low complexity score, target coverage was well preserved over the course of treatment for even a large change in proton range to the target distal edge. Plans with high complexity scores, however, were sensitive to changes in anatomy, and even a small WET change resulted in a clinically significant drop in target coverage. Plans with complexity score below 1 preserved target coverage in 16/18 cases, while plans with complexity score above 1 preserved target coverage in only 3/25 cases.
Conclusion: IMPT plan with highly modulated field-to-field dose distribution can suffer clinically significant target coverage changes due to anatomical change. The complexity score function devised in this work can be a good predictor of robustness to anatomical change and could potentially be a part of routine plan check.