Evaluating Maximum MU per Spot Thresholds on CNS Pencil Beam Proton Therapy Plan Quality

Abstract

There are concerning reports of CNS toxicity in patients treated with proton therapy at doses lower than would be expected based on photon data estimates. Given the desire to mitigate risk of serious unexpected events, we propose limiting the maximum monitor unit weighting per spot (maxMU/spot) to the entire plan with the hypothesis that heavily weighted spots may pose an unforeseen risk of toxicity despite meeting objective dose constraints. This dosimetric study assesses the impact of maxMU/spot restrictions on pencil beam proton therapy (PBPT) plan quality. We hypothesize that plan quality may be preserved while still restricting the maxMU/spot, but there may be a threshold below which restrictions negatively impact plan quality. PBPT plans of 11 patients (9 consecutive, 2 additional with notably challenging plan objectives) were included, of which 3 plans included boosts, for a total of 14 nominal PBPT cases. Per nominal plan, a single dosimetrist created 4 test plans as outlined in Table 1, iteratively reducing the maxMU/spot in the plan. Test Plan 1, unrestricted in maxMU/spot, was the reference for all analyses (comparison sets 2 vs. 1; 3 vs. 1; and 4 vs. 1). The differences in each plan quality metric in each comparison set were analyzed using the Wilcoxon signed rank test, with p<0.05 significance testing. Beam-specific metrics in each comparison set were analyzed, defined by the root mean square deviation in beam metrics from the unrestricted plan. Linear regression with log transformation was performed to analyze the impact of restriction conditions on changes in the beam-specific max dose, heterogeneity, and % of target receiving the Rx/#beams. A total of 14 PBPT plans, 7 (50%) SFO, 7 (50%) MFO, 9 (64%) delivering >3500 cGy, 9 (64%) with 3 beams, 5 (36%) with 4-5 beams, and 7 (50%) without a range shifter were included. With increasing MU restriction, there were no differences in plan quality metrics of target coverage (V95% and V100%Rx), conformality and gradient indices, hot spot volume (V105%Rx), and dose to unaffected brain (V10%/30%/50%/70%/90%/100%Rx) across all comparison sets (p>0.05). However, there was a mean increase of 854 (2 vs. 1, p = 0.02), 869 (3 vs. 1, p = 0.025), and 892 (4 vs. 1, p = 0.007) total spots in the more restricted MU plans. MaxMU restriction did not significantly impact the beam-specific max dose (p = 0.259), heterogeneity (p = 0.305), and % of target receiving the Rx/#beams (p = 0.409). MaxMU/spot restriction did not negatively impact overall plan quality and beam-specific metrics within the thresholds evaluated in this study. Future studies should evaluate spot weighting with LET/RBE to determine if spot weighting manipulation may truly impact clinical outcomes and mitigate toxicity.Tabled 1Abstract 2601; Table; Iterative reduction maximum allowable MU per spot*PlanNo Range shifter (RS)2cm RS3cm RS5cm RS1No limitNo limitNo limitNo limit230405070320304060410203050∗1 MU = 1 cGY/mm2 at a 2 cm for monoenergetic proton beam Open table in a new tab