Click here to


Are you sure ?

Yes, do it No, cancel

Examining the Influence of Variable RBE Models On Passive Scattering Proton Therapy Esophageal Cancer Patients by Means of the Fast Dose Calculator at The University of Texas MD Anderson Cancer Center

A Adair1*, Q Wang1 , P Yepes1 , S Lin2 , M Aldehaim2 , U Titt2 , D Mirkovic2 , R Mohan2 , (1) Rice University & University of Texas MD Anderson Cancer Center, Houston, TX, (2) University of Texas MD Anderson Cancer Center, Houston, TX,


(Sunday, 7/14/2019)  

Room: ePoster Forums

Purpose: Evaluating how proton models incorporating dosimetric indices are affected by variable relative biological effective dose (RBE)(variable) as opposed to the standard fixed RBE model of 1.1 (fixed), for a large cohort (300+ patients) of esophageal cancer patients treated with passive scattering proton therapy (PSPT).

Methods: Treatment plans for a cohort of PSPT esophageal cancer patients were analyzed. The Fast Dose Calculator (FDC) was utilized for dose calculations. The FDC computed the fixed as well as three variable models: Wedenberg (WDG), McNamara (MGH), and Repair-misrepair-fixation (RMF), for each treatment plan. Various Dosimetric indices were assessed for the target volumes (TVs) and the organs at risk (OARs). The dosimetric indices and their differences were compared between the variable and fixed models.

Results: For the TVs, the variable models mean dose dosimetric index (mean) is within 4% of the fixed, the cohort average is within 2%. The variable is often less then the fixed. For the OARs, MGH and WDG models systematically predict the variable greater than the fixed. The variable is often greater than 10% of the fixed for heart and lungs. The RMF model systematically predicts lower increases as opposed to the other variable models. However, it too predicts analogously high values. The RMF models variable mean for heart is often greater than 5% of the fixed.

Conclusion: The variable models are within a few percent of the fixed model in the TV analysis, indicating under prediction is not an issue for the TVs. The variable models generally give greater dose than the fixed to the OARs. This suggests that it might be pertinent to develop strategies to reduce high-dose to such structures, with the goals of possibly minimizing toxicity. Further work, developing normal tissue complication probability (NTCP) models is being investigated.

Funding Support, Disclosures, and Conflict of Interest: This work is supported by the Cancer Research and Prevention Institute of Texas (CPRIT) under contract RP160232.




TH- External Beam- Particle therapy: Proton therapy - computational dosimetry-Monte Carlo

Contact Email