A linear relationship for the LET-dependence of Gafchromic EBT3 film in spot-scanning proton therapy
Sarah E Anderson,Michael P Grams,H Wan Chan Tseung,Keith M Furutani,Chris J Beltran
DOI: https://doi.org/10.1088/1361-6560/ab0114
IF: 3.5
2019-03-07
Physics in Medicine and Biology
Abstract:Radiochromic film (RCF) is a valuable dosimetric tool, primarily due to its sub-millimeter spatial resolution. For accurate proton dosimetry, the dependence of film response on linear energy transfer (LET) must be characterized and calibrated. In this work, we characterized film under-response, or 'quenching', as a function of dose-weighted linear energy transfer (LET<sub>d</sub>) in several proton fields and established a simple, linear relationship with LET<sub>d</sub>. We performed measurements as a function of depth in a PMMA phantom irradiated by a spot-scanning proton beam. The fields had energies of 71.3 MeV, 71.3 MeV with filter, and 159.9 MeV. At each depth (measurements taken in depth step sizes of 0.5-1 mm in the Bragg peak), we measured dose with a PTW Markus ion chamber and EBT3 RCF. EBT3 under-response was characterized by the ratio of dose measured with film to that with ion chamber. LET<sub>d</sub> values for our experimental setup were calculated using in-house clinical Monte Carlo code. Measured film under-response increased with LET<sub>d</sub>, from approximately 10% under-response for LET<sub>d</sub> = 5 keV µm<sup>-1</sup> to approximately 20% for LET<sub>d</sub> = 8 keV µm<sup>-1</sup>. The under-response for all measurements was plotted versus LET<sub>d</sub>. A linear fit to the data was performed, yielding a function for under-response, [Formula: see text], with respect to LET<sub>d</sub>: [Formula: see text]. Finally, the linear under-response relationship was applied to a film measurement within a spread-out Bragg peak (SOBP). Without correcting for LET<sub>d</sub>-dependence in the SOBP measurement, the discrepancy between film and Monte Carlo profiles was greater than 15% at the distal edge. With correction, the corrected film profile was within 2% and 1 mm of the Monte Carlo profile. RCF response depends on LET<sub>d</sub>, potentially under-responding by >15% in clinically-relevant scenarios. Therefore, it is insufficient to perform only a dose calibration; LET calibration is also necessary for accurate proton film dosimetry. The LET<sub>d</sub>-dependence of EBT3 can be described by a single, linear function over a range of clinically-relevant proton therapy LET<sub>d</sub> values.
engineering, biomedical,radiology, nuclear medicine & medical imaging