Program Information
Evaluation of Proton Dose Distribution Between Analytical Pencil Beam and Monte Carlo Algorithms for Head Neck Treatment
L Gentry Swan*, L Song , S Tang , C Chang , Texas Center for Proton Therapy, Irving, Texas
Presentations
SU-I-GPD-T-187 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: To compare pencil beam algorithm (PB) and Monte Carlo algorithm (MC) dose distributions for pencil beam scanning head neck (HN) intensity modulated proton therapy (IMPT) and assess effect of air gap.
Methods: Ten patients receiving HN IMPT from 2015-2017 at a single institution were selected for comparison. A single treatment planning system was used for both PB and MC dose calculations. All clinical plans were created using PB inversely planned with robust optimization. They were subsequently forward-calculated with MC employing 1% dose uncertainty. All plans were also computed with an air gap reduction to 1cm for all fields. Differences in target coverage were assessed using percentage of prescription dose covering 98% of the CTV. Dmean and Dmax values were compared for organs at risk (OARs).
Results: In these 10 cases, PB overestimated the CTV_D98 by 11% (range 6-17%) in comparison to MC. In contrast, PB underestimated OAR dose. On average, PB overestimated spinal cord Dmax by 1 Gy and larynx Dmean by 3 Gy. With an air gap reduction to 1cm, PB still overestimated CTV_D98, but by less (7% overestimation, range 5-12%). Similarly, decreased air gap reduced the OAR dose difference.Air gap change resulted in opposing effects on target coverage when comparing plans forward-calculated with PB and MC. PB plans with an air gap reduction to 1cm had lower CTV coverage (average 2% decrease). Using MC, however, the CTV_D98% increases by 1% when minimal air gap is used.
Conclusion: For HN IMPT, PB and MC algorithms calculate 11% differences in CTV coverage, which can be reduced to 7% with a reduction in air gap to 1cm. The PB algorithm should be employed with caution, and air gap reduction is recommended.
Contact Email: