Program Information
Indications for Online Adaptive Radiotherapy Based On Dosimetric Consequences of Interfractional Pancreas-To-Duodenum Motion in MRI-Guided Pancreatic Radiotherapy
K Mittauer1*, S Rosenberg1 , M Geurts1 , M Bassetti1 , I Chen2 , L Henke2 , J Olsen3 , R Kashani2 , A Wojcieszynski1 , P Harari1 , Z Labby1 , P Hill1 , B Paliwal1 , P Parikh2 , J Bayouth1 , (1) University of Wisconsin, School of Medicine and Public Health, Madison, WI, (2) Washington University School of Medicine, St. Louis, MO, (3) University of Colorado- Denver, Aurora, CO
Presentations
TU-AB-BRA-11 (Tuesday, August 2, 2016) 7:30 AM - 9:30 AM Room: Ballroom A
Purpose: Dose limiting structures, such as the duodenum, render the treatment of pancreatic cancer challenging. In this multi-institutional study, we assess dosimetric differences caused by interfraction pancreas-to-duodenum motion using MR-IGRT to determine the potential impact of adaptive replanning.
Methods: Ten patients from two institutions undergoing MRI-guided radiotherapy with conventional fractionation (n=5) or SBRT (n=5) for pancreatic cancer were included. Initial plans were limited by duodenal dose constraints of 50 Gy (0.5 cc)/31 Gy (0.1 cc) for conventional/SBRT with prescriptions of 30 Gy/5 fractions (SBRT) and 40-50 Gy/25 fractions (conventional). Daily volumetric MR images were acquired under treatment conditions on a clinical MR-IGRT system. The correlation was assessed between interfractional GTV-to-duodenum positional variation and daily recalculations of duodenal dose metrics. Positional variation was quantified as the interfraction difference in Hausdorff distance from simulation baseline (ΔHD) between the GTV and proximal duodenal surface, or volume overlap between GTV and duodenum for cases with HD₀=0 (GTV abutting duodenum). Adaptation was considered indicated when daily positional variations enabled dose escalation to the target while maintaining duodenal constraints.
Results: For fractions with ΔHD>0 (n=14, SBRT only), the mean interfraction duodenum dose decrease from simulation to treatment was 44±53 cGy (maximum 136 cGy). A correlation was found between ΔHD and dosimetric difference (R²=0.82). No correlation was found between volume of overlap and dosimetric difference (R²=0.31). For 89% of fractions, the duodenum remained overlapped with the target and the duodenal dose difference was negligible. The maximum observed indication for adaptation was for interfraction ΔHD=11.6 mm with potential for adaptive dose escalation of 136 cGy.
Conclusion: This assessment showed that Hausdorff distance was a reasonable metric to use to determine the indication for adaptation. Adaptation was potentially indicated in 11% of the treatments (fractions where GTV-to-duodenum distance increased from simulation), with a feasible average dose escalation of 7.0%.
Funding Support, Disclosures, and Conflict of Interest: MB, LH, JO, RK, PP: research and/or travel funding from ViewRay Inc. PP: research grant from Varian Medical Systems and Philips Healthcare
Contact Email: