Program Information
Analysis of Differences in Tumor Control Probability Calculated From Monte Carlo and Treatment Planning System Dose Distributions for Proton Therapy
CR Peeler*, U Titt , DR Grosshans , Z Liao , R Mohan , D Mirkovic , The University of Texas M.D. Anderson Cancer Center, Houston, TX
Presentations
SU-F-FS1-1 (Sunday, July 30, 2017) 2:05 PM - 3:00 PM Room: Four Seasons 1
Purpose: We assessed differences in calculated tumor control probability (TCP) based on Monte Carlo and treatment planning system (TPS) dose distributions for patients treated with passive scattering proton therapy (PSPT) for brain and lung cancer.
Methods: Thirty patients treated for ependymoma and nine treated for non-small cell lung cancer (NSCLC) with PSPT at our institution were selected. All NSCLC patients had local recurrence and were being analyzed for another study. For each patient, the original treatment plan was used to generate input for Monte Carlo (MC) simulations, which were then completed using our in-house system based on MCNPX. For both MC and TPS dose distributions, dose statistics, including mean dose and equivalent uniform dose (EUD) for the target volume, were collected in addition to dose-volume histogram data. TCP was calculated with an EUD-based model.
Results: Average relative differences (MC – TPS) in mean dose were found to be -0.04% (±2%) and -2% (±1%) for brain and lung plans, respectively. Relative differences in EUD were found to be greater at -0.6% (±2%) and -6% (±8%). Relative differences in predicted TCP were subsequently found to be much greater for lung plans where the mean difference was -32% (±27%) (plausibly attributable to the recurrent population only). Brain treatment plan TCP differences were smaller at -2% (±4%), which is in keeping with other previous studies.
Conclusion: MC calculated dose was on average lower than the dose calculated by the TPS. Differences in the dose from MC and the TPS analytical algorithm lead to substantial differences in predicted TCP for the heterogeneous anatomy of the lung plans. Predicted TCP differences were much smaller for the more homogeneous brain anatomy. Further study is warranted to evaluate the exact cause of MC and TPS dose differences and whether the differences in TCP translate to poorer outcomes.
Funding Support, Disclosures, and Conflict of Interest: Funding provided by NIH grant 5U19CA021239-35 and Cancer Prevention and Research Institute of Texas grant RP160232.
Contact Email: