Program Information
A Retrospective Analysis of the Impact of Range Uncertainty in Brain Patients
K Grantham*, L Santanam , S Goddu , B Sun , T Zhang , S Mutic , C Robinson , J Huang , S Perkins , C Tsien , J Bradley , T Zhao , Washington University School of Medicine, Saint Louis, MO
Presentations
SU-F-T-135 (Sunday, July 31, 2016) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: We retrospectively evaluate the dosimetric impact of a 3.5% range uncertainty on CTV coverage and normal organ toxicity for a cohort of brain patients.
Methods: Twenty treatment plans involving 20 brain cancer patients treated with Mevions S250 were reviewed. Forty uncertain plans were made by changing the ranges in original plans by ±3.5% while keeping all devices unchanged. Fidelity to the original plans was evaluated with gamma index. Changes in generalized equivalent uniform dose (gEUD) were reported for the following structures: CTV coverage, brainstem, optic chiasm, and optic nerves. Comparisons were made by plotting the relevant endpoints from the uncertain plans as a function of the same endpoints from the original clinical plan.
Results: Gamma-index analysis resulted in a 50% pass rate of the uncertain plans using a 90% passing rate and 3%/3mm criterion. A 9.5% decrease in the slope of gEUD plot for the CTV was observed for the 3.5% downward range shift. However, the change in slope did not result in a gEUD change greater than 1.1% for the CTV. The slopes of the gEUD plots for normal structures increased by 3.1% 3.9% 2.4% and 0.2% for the chiasm, brainstem, left optic nerve and right optic nerve respectively. The maximum deviation from the gEUD of the clinical plan for normal structures was: 64% in the chiasm, 31% for the brainstem, and 19% for both optic nerves.
Conclusion: A retrospective review shows moderate radiobiological impact of range uncertainty in passively scattered proton therapy with sporadic catastrophe. The linear regression analysis on the statistical data indicates a systematic deviation of gEUD from treatment planning in the light of range uncertainty.
Contact Email: