Encrypted login | home

Program Information

Evaluation of Analytical Proton Dose Predictions with a Lung-Like Plastic Phantom

no image available
M Sell

M Sell1,2*, U Titt1, L Perles1, D Mirkovic1, R Mohan1, M Bangert2, U Oelfke2, (1) The University of Texas MD Anderson Cancer Center, Houston, Texas (2) German Cancer Reseach Center DKFZ, Heidelberg, Germany

WE-E-BRB-2 Wednesday 2:00:00 PM - 3:50:00 PM Room: Ballroom B

Purpose: Former studies have shown that inhomogeneities in the path of therapeutic proton
beams can lead to a degradation of the distal edge of the Bragg peak. These studies mostly
investigated bone-air interfaces. This study focuses on distal edge degradation caused by finely
structured soft tissue - air interfaces, which can be found in lung tissue.

Methods: A randomly filled voxelized lung-like phantom was designed and produced using
rapid prototyping methods. The results of transmission measurements on this phantom were
used to validate Monte Carlo (MC) calculations, which were then used as gold standard to
calculate doses in several lung equivalent geometries (phantoms). The results were compared
to the results of analytical dose calculation engines.

Results: Transmission measurements showed that the distal falloff width (from 90 % of the
peak dose to 10 %) in water increased from 3.32 mm by 117 % to 7.19 mm for an initial proton
energy of 140 MeV, and from 5.95 mm to 9.03 mm (52 %) for 200 MeV. The peak dose in the
degraded beam was only 70 % (for 140 MeV) and 84 % (for 200 MeV) of the value observed in
non-degraded beams. These findings were in contrast to the results obtained with analytical
dose computation engines, but are in agreement with MC calculations.

Conclusions: If not predicted correctly, Distal Edge Degradation in lung cancer therapy can lead
to severe under-dosage of the target region and unwanted dose in organs at risk distal to the
Bragg peak. Therefore clinically used dose calculation algorithms have to be extended to take
lateral inhomogeneities into account.

Contact Email