Room: AAPM ePoster Library
Purpose: To evaluate if simple MR-based online adaptive radiotherapy planning strategies can be effectively implemented to account for prostate cancer treatment interfraction motion variation.
Methods: Five prostate cancer patients with MR scans acquired prior to treatment, 2-weeks, and 3-months post were analyzed in this retrospective study. The MR prior to treatment was used as the reference treatment planning dataset. All plans were computed in Pinnacle v16.2 using appropriate anatomical density overrides. To emulate the Elekta Unity system, plans were generated using 6XFFF and seven step-and-shoot IMRT beams with 90° collimator angle. Two different fractionation schemes were used: hypofractionated (70Gy/28 fractions) and stereotactic (36.25Gy/5 fractions). Rigid image registration was performed between the reference MR and the subsequent MRs with alignment focused on the prostate/seminal vesicles region and the evaluation of three simple adaptive strategies was performed. The strategies evaluated for each MR were: (1) recalculation of original segments, (2) adapt segments by isocenter shift for new PTV, and (3) adapt segments by isocenter shift and segment weight re-optimization using reference contours. Target and OARs were contoured on each MR and adaptive strategies were evaluated using NRG-GU005 objectives.
Results: In the hypofractionated plans, CTV coverage was >99% for strategies (2) and (3); and >95% in strategy (1). OAR dose constraints were maintained using adaptive strategy (1), however, optimizing target coverage with strategies (2) and (3) led to an increase in dose to OARs. In the stereotactic plans, CTV coverage was >90% for strategies (1), and >95% for strategy (2) and (3). OAR dose constraints were compromised for all strategies and were worst for strategy (3).
Conclusion: Simple adaptive strategies are sufficient with reasonable CTV coverage and similar OAR sparing for the hypofractionated regimen, however for the stereotactic scheme full replanning is warranted.
Not Applicable / None Entered.