Evaluation of Various Spot MU Correction Methods for Range Adaptive Scanning-Beam Proton Planning
J Cheung1*, L Court1, P Park1, X Zhu1, S Frank1, R Kudchadker1, L Dong2, (1) UT MD Anderson Cancer Center, Houston, TX, (2) Scripps Proton Therapy Center, San Diego, CASU-E-T-666 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: For CT-based image-guided proton therapy, it is possible to adapt to new patient anatomy by correcting the range of protons. In this work we investigated the use of several correction methods to adjust proton spot monitor units (MU) for single field uniform dose for real-time range adaptive proton therapy.
Methods: Spot MU correction factors for range-adapted plans were calculated based on two different methods: differences in central-axis peak dose per MU, and differences in number of protons per MU. These factors were applied either only to the highest energy spot for each ray, or to the entire spot matrix. For each of these methods, the difference between the dose distributions from the verification (non-adaptive) and adaptive plans to the original treatment plan was assessed. We examined target coverage (V95%), target hot spots (V110%), and dose to the organs-at-risk including the spinal cord (Dmax), esophagus (Dmean), contralateral lung (Dmean), and heart (Dmean). A scanning-beam plan with single-field optimization was created for a lung cancer patient, simulating a single-field delivering a uniform target dose pointing directly at organs-at-risk. Average change in water equivalent thickness (WET) was used to quantify anatomic change.
Results: Dose to normal tissues distal to the target was reduced for the adaptive plans compared to the verification plans for cases where WET decreased, but were similar for cases where WET increased. The adaptive plans almost always improved target coverage when WET increased, but the verification plans tended to have better target coverage when WET decreased. The hot spots in the target were always greater in the adaptive plans compared to the verification plans.
Conclusion: The simple proton range correction adaptive scheme shows promise to reduce normal tissue doses without a full reoptimization. However, more sophisticated spot MU correction is warranted to increase dose homogeneity throughout the target volume.