Water equivalent PRESAGE for synchrotron radiation therapy dosimetry

Transcription

1 Water equivalent PRESAGE for synchrotron radiation therapy dosimetry Frank M. Gagliardi a) Alfred Health Radiation Oncology, The Alfred, Melbourne, Vic 3004, Australia School of Health and Biomedical Sciences, RMIT University, Bundoora, Vic 3083, Australia Liam Day, Christopher M. Poole, and Rick D. Franich School of Science, RMIT University, Melbourne, Vic 3000, Australia Moshi Geso School of Health and Biomedical Sciences, RMIT University, Bundoora, Vic 3083, Australia (Received 23 June 2017; revised 17 November 2017; accepted for publication 16 December 2017; published 20 January 2018) Purpose: Synchrotron Radiation Therapy techniques are currently being trialed and commissioned at synchrotrons around the world. The patient treatment planning systems (TPS) developed for these treatments use simulated data of the synchrotron x-ray beam to produce the dosimetry in the treatment plan. The purpose of this study was to investigate a water equivalent PRESAGE â dosimeter capable of 3D dosimetry over an energy range suitable for synchrotron x-ray beams. Methods: Water equivalent PRESAGE â dosimeters were fabricated with a radiological effective atomic number similar to water over an energy range of 10 kev to 10 MeV. The dosimeters were irradiated at various energies, scanned using optical CT (OCT) scanning and compared to ion chamber measurements. Percentage depth dose and beam profiles of the synchrotron beam were compared to Monte Carlo (MC) model simulations. Results: The PDD profiles of the water equivalent PRESAGE â agreed with ion chamber measurements and MC calculations within 2% for all kev energies investigated. The PRESAGE â also showed good agreement to the MC model for depths below 5 mm of the synchrotron beam where ion chamber data do not exist. The spatial resolution of the OCT was not sufficient to accurately measure the penumbra of the synchrotron beams compared to MC calculations or EBT3 film; however, the water equivalent PRESAGE â was able to verify dose profile characteristics of the MC model. Conclusions: The radiological response of a water equivalent PRESAGE â dosimeter has been validated for synchrotron x-ray beam energies along with the ability to independently verify dose distributions of a MC model American Association of Physicists in Medicine [ /mp.12745] Key words: optical CT scanning, PRESAGE â dosimeter, synchrotron radiation therapy 1. INTRODUCTION Stereotactic Synchrotron Radiation Therapy (SSRT) and Microbeam Radiation Therapy (MRT) are experimental techniques that use synchrotron radiation to treat tumors as an alternative to conventional radiotherapy which uses megavoltage (MV) x-ray beams. 1 SSRT uses open field synchrotron x-ray beams to perform a stereotactic treatment to attain full dose coverage to a target volume. While MRT consists of a parallel array of highly collimated x-ray microbeams lm wide with lm peak-to-peak spacing to cover the target volume. The synchrotron x-ray beams used for SSRT and MRT are typically 2 mm high and 25 mm wide. The beams are collimated by custom-made tungsten or cerrobend masks and for MRT a multislit collimator is also used. To attain target coverage greater than mm, the target must be scanned through the beam. Dosimetric characteristics of the delivered fields can be measured with an ionization chamber with a small volume or in 2D using radiochromic films in suitable phantoms; however, the 3D dosimetric characteristics of the x-ray beams cannot be measured in a single delivery. Patient treatment planning systems (TPS) using Monte Carlo modeling have been developed for future clinical trials; 2 4 however, the progression of these experimental techniques requires rigorous practical validation of the calculation models with comparable accuracy to those employed in a conventional radiotherapy TPS. 5 The radiochromic PRESAGE â dosimeter offers a unique opportunity to validate dosimetry models in 3D. 6,7 Researchers have formulated PRESAGE â that is radiologically equivalent to water by altering the original formulation. 8 Experimentally and through Monte Carlo simulations it has been shown that slight alterations in the chemical composition of the PRESAGE â can change its dosimetric properties to be radiologically equivalent to water over an energy range suitable for kv and MV dosimetry. 9,10 The water equivalent PRESAGE â dosimeter has the capability of being accurate enough for 3D verification of IMRT, VMAT, and gated treatments for external beam radiotherapy 1255 Med. Phys. 45 (3), March /2018/45(3)/1255/ American Association of Physicists in Medicine 1255

2 1256 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1256 with 6MV photons, 11,12 and for high-resolution 3D imaging of synchrotron generated microbeams. 13 In this study we have investigated the response of a water equivalent PRESAGE â over an energy range suitable for synchrotron radiation therapy and compared the dosimetric properties of SSRT beams with ion chamber measurements, film measurements, and Monte Carlo simulations. The radiological response of the water equivalent PRESAGE â was found to be within acceptable limits (2%) and has the potential to validate a future synchrotron TPS. 2. METHODS 2.A. PRESAGE dosimeter fabrication PRESAGE â dosimeters were fabricated using polyurethane precursors Crystal Clear â 206 Part A and Part B (Smooth-On Inc., Easton, PA, USA). Before mixing Part A and Part B together to produce the polyurethane compound, luecomalachite green (LMG) (Sigma-Aldrich â ) (2 wt%) was dissolved into Part A (48.95 wt%) and chloroform (Chem-Supply) (5 wt%) was dissolved into Part B (44 wt%). The two parts were then mixed thoroughly before the catalyst dibutyltin dilaurate (DBTDL) (Merck KGaA, Darmstadt, Germany) was added (0.05 wt%) and the final mixture poured into polypropylene cylinders and cured for 48 hr at 60 psi pressure. The addition of DBTDL in the PRESAGE â fabrication reduces curing time and is known to increase sensitivity and post response stability of the dosimeters. 8 The main advantage of adding DBTDL to the PRESAGE â formulation however is its ability to alter the effective atomic number (Z eff ) of the PRESAGE â to be radiologically equivalent to water over a wide range of x-ray energies. The radiological effective atomic number of a material can be calculated with the Auto Z eff software 14 which considers all photon energy absorption processes. The calculation of Z eff for this PRESAGE â formulation using the Auto Z eff software matches the Z eff of water within 1.7% over the effective energy range of 10 kev to 10 MeV (Fig. 1). The Z eff of the original PRESAGE â formulation 15 differs from water by 10% 30% for energies below 100 kev making it unsuitable for SSRT dosimetry. The PRESAGE â dosimeters were cast as cylindrical rods with length 70 mm and width 43 mm (Fig. 2). Typically this formulation of water equivalent PRESAGE â has small intrabatch (<0.5%) and interbatch (<3.5%) variability B. Optical CT scanning and image processing The PRESAGE â dosimeters were scanned pre and post irradiation on a VISTA TM Optical CT scanner (Modus Medical Devices Inc, CA, USA). The LED light source of wavelength of 633 nm was given sufficient time to warm-up and stabilize before the samples were imaged. A dark field projection was acquired at the end of each pre and post scan to minimize any stray light during image reconstruction. The dosimeter was immersed in a tank of ethyl benzoate as the refractive index matching medium. Ethyl benzoate has a refractive index of which closely matches the refractive index of PRESAGE â which is approximately Image reconstruction was performed with the VISTA 3D Reconstruction software using a 3D image voxel size of 0.25 mm. Percentage depth dose curves (PDD) and beam profiles were generated using ImageJ (RRID: SCR_003070) software from the 3D reconstruction images. 2.C. Irradiations Irradiation of the dosimeters was performed on the Imaging and Medical Beamline (IMBL) at the Australian Synchrotron. 18 The synchrotron was operating in top-up mode with a 200 ma ring current and 3 GeV electron energy in the storage ring. The polychromatic x-ray beam produced by the superconducting multipole wiggler on the IMBL was filtered both in vacuo and ex vacuo (details below) producing an x-ray beam mm (width 9 height) of 95.3 kev mean energy with a 262 Gy/s dose rate. The dose rate was calculated at the surface of a water phantom using the computer program spec.exe which is based on a comprehensive model of the IMBL beamline, and has been validated against a number of experimental measurements. 18 FIG. 1. Radiological effective atomic number (Z eff ) based on all photon energy absorption processes 14 for water, PRESAGE â original formulation, and PRESAGE â water equivalent formulation. [Color figure can be viewed at wileyonlinelibrary.com] FIG. 2. Cylindrical PRESAGE â dosimeters used in this study. [Color figure can be viewed at wileyonlinelibrary.com]

3 1257 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1257 The filter combination used to produce the spectrum was: (a) in vacuo filter set 4, F4 18 (0.45 mm graphene mm high density graphite mm Cu), combined with (ii) ex vacuo filtration (6.8 m He m air mm Be mm diamond mm Kapton mm Al). The photon spectrum for this combination, calculated using spec.exe, is shown in Fig. 3. The intensity of the IMBL beam is at a maximum at the center of the beam and decreases monotonically in both the horizontal and vertical directions. The central part of the beam is uniform in the horizontal direction to within 5% across 90% of the FWHM for a field size of mm. 19 The decrease in intensity near the edge of the x-ray field in the horizontal direction is referred to as the roll off effect. PRESAGE â samples were irradiated dynamically through the x-ray beam (to 50 Gy at the surface) with field sizes of mm and mm collimated by custom-made 4 mm thick pure tungsten (>99%) masks. Samples used for PDD curves were irradiated end-on in a custom-made rectilinear Perspex phantom (to provide full scatter conditions) attached to the sample stage goniometer for the irradiations [Fig. 4(a)]. A second set of samples used for extracting field profiles, was irradiated in a vertical orientation using a custom-made Perspex sleeve to provide a flat surface for the beam entrance into the dosimeter [Fig. 4(b)]. Further irradiations were performed at Alfred Health Radiation Oncology (The Alfred, Australia) at 120 kvp on a Pantak Therapax Series3 150T SXRT machine and at 6 and 18 MV on a Varian 21EX (Varian Medical Systems, Palo Alto, USA) linear accelerator. The dosimeters were placed in a NE type mini-water phantom ( cm) [Figs. 4(c) and 4(d)] and irradiated to 30 Gy at the surface with a 10 mm circular field at FSD 15 cm on the SXRT machine and 15 Gy to D max with a cm field at 100 cm SSD on the linear accelerator. The dosimeters were immediately stored after irradiation in a light-free environment at 18 o C to prevent any fading or exposure to UV light. The post response of PRESAGE â dosimeters over time is stable when DBTDL is used as the catalyst. 8,20 2.D. Ion chamber measurements Percentage depth dose (PDD) profiles were measured with detector phantom combinations appropriate for the respective x-ray energy: a plane-parallel chamber (PTW 23342) in a full scattering kv equivalent solid water phantom at different depths for the SXRT (120 kvp) beam; a pinpoint chamber (PTW 31014) in a mini water tank for the synchrotron (95.3 kv average energy) beam; and a plane-parallel Roos chamber (PTW 34001) in a water tank for the 6 and 18 MV beams. Recombination corrections were not applied to the PDDs as the recombination factor is less than 0.2% for planeparallel chambers at low dose rates typically used in SXRT and linac treatments, 21 and only 0.8% for the pinpoint chamber in the IMBL synchrotron beam E. GEANT4 Monte Carlo simulations Synchrotron beams produced on the IMBL were simulated using Monte Carlo (MC) GEANT4 (Version 9.4.6). 22 The low-energy electromagnetic Livermore polarized physics model was used as the synchrotron beam produced by the superconducting multipole wiggler is highly polarized. Initial energies were sampled from measured beamline spectrum data with a mean energy of 95.3 kev. The particle source was defined via Gaussian position and momentum distributions beginning with a 30 mm wide beam 1 mm in height. The simulation geometry then included the tungsten field-defining mask of mm or mm into a 20 cm 3 water phantom. The cutoff ranges were set to 10 lm for photons and electrons/positrons in water and histories were simulated. A superposition technique summing doses over voxel dimensions ( mm) encapsulating the total field height was used to attain the mm and mm field sizes. PDD profiles along with lateral and vertical dose profiles of the and mm fields at the surface, 10 and 20 mm depths were generated. 2.F. Film irradiations Radiochromic film (Gafchromic TM EBT3, Ashland Inc.) was irradiated on the surface and at 20 mm depth inside a solid water phantom consisting of Gammex 457 slabs (Gammex Inc., MI, USA) to a dose of 20 Gy for the mm and mm field sizes. The irradiated films were scanned on an EPSOM Perfection V700 flatbed scanner in transmission mode at 5.9 pixel/mm and dose profiles of the fields plotted using FilmQA (V2.0) software. FIG. 3. Calculated photon spectrum for filter set F4. [Color figure can be viewed at wileyonlinelibrary.com] 3. RESULTS 3.A. Radiological properties Depth dose profiles were taken from the central region of the reconstructed optical CT images of the irradiated water equivalent PRESAGE â dosimeters using a line profile of 20 pixel width in ImageJ, as shown in Fig. 5(a). A total of 512

4 1258 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1258 (a) (b) (c) (d) FIG. 4. Setup for PRESAGE â dosimeter irradiations; (a) enclosed by a Perspex phantom and mounted to the sample stage goniometer on the IMBL, (b) fitted with a Perspex sleeve to form a flat entrance face, (c) submersed in a water phantom for linear accelerator irradiation, and (d) submersed in a water phantom for SXRT irradiation. [Color figure can be viewed at wileyonlinelibrary.com] projections with step size were taken with the maximum reconstructing resolution of 4 pixels/mm. The PDD profiles show similar dosimetrical properties to the ion chamber (IC) measurements for 120 kvp SXRT, 95.3 kev (average) synchrotron, 6 and 18 MV linac x-ray beams, Fig. 6. The agreement between the ion chamber and PRESAGE â PDDs are within 2% except at the entrance of the PRE- SAGE â dosimeter for the 6 and 18 MV beams. The under response of the dosimeter in the first few millimeters of the 6 and 18 MV profiles is a known characteristic 9,23 due to the flat end of the cylindrical dosimeter and high dose gradient of the MV beams at the surface. Surface roughness of the dosimeter, imperfect alignment of the dosimeter central axis in the OCT scanner and orthogonality on the beam entrance typically affect the first 1 2 slices before the full field is seen in the OCT reconstruction. Therefore, the PDD curves have some uncertainty in the surface position in the order of 0.5 mm. To account for this, the PDD curves are registered at the depth of dose maximum to achieve the best fit. The effect is less pronounced for the kev beams where the maximum (a) FIG. 5. Slices taken from the 3D OCT reconstruction of the irradiated PRE- SAGE â dosimeters (a) along the depth axis for a mm field, and across (b) a mm, and (c) a mm field at 20 mm depth. [Color figure can be viewed at wileyonlinelibrary.com] (b) (c)

5 1259 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1259 FIG. 6. Measured percentage depth dose curves for water equivalent PRESAGE â and ion chamber at 18 MV, 6 MV, 95.3 kev (average), and 120 kvp x-ray beams. [Color figure can be viewed at wileyonlinelibrary.com] dose is less than 2 mm depth. The 95.3 kev (average) synchrotron beam was normalized to 96.5% at 5 mm depth (since the measured ion chamber data do not reach the water surface for the horizontal synchrotron beam due to the wall of the water tank) to compare the shape of the profile with the PRESAGE â profile which was normalized to 100% at the surface. 3.B. Synchrotron beams PDD (PRESAGE â, IC and MC) Using the depth recommended by international protocols for medium energy x-ray beams (TRS-398) 24 of 20 mm, we chose the mm field size to be a suitable reference point for comparison with the ion chamber measured data, MC calculated data, and PRESAGE â dosimeter data. The mm field size was chosen as there is less variability in the dose uniformity across the field and less change in the energy spectrum across the field compared to the mm field. The PDDs for mm and mm fields were normalized to 20 mm depth of the mm field and plotted in Fig. 7. The MC calculations match the ion chamber measurements within 1% over the entire PDD from depth 5 to 100 mm. The PRESAGE â PDD profiles also match the ion chamber measured data; however, the curves are not as smooth as the ion chamber fitted PDD profile. FIG. 7. Percentage depth dose profiles of water equivalent PRESAGE â, ion chamber, and MC for mm and mm synchrotron beams normalized to 20 mm depth of the mm beam. Insert is a zoomed in region of the curve over the first 10 mm in depth. [Color figure can be viewed at wileyonlinelibrary.com]

6 1260 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1260 At depths less than 5 mm where there is no ion chamber data, the MC calculations and PRESAGE â profiles agree within 2% and display the characteristic maximum dose inside the phantom at a depth of <2 mm. The insert in Fig. 7 shows the PDD curves for the region 0 to 10 mm in depth; however, the exact position of the maximum dose is obscured by the resolution of the OCT reconstruction (0.25 mm) and the MC voxel size (0.25 mm). 3.C. Synchrotron beam profiles (PRESAGE â and MC) Lateral dose profiles of the mm and mm beams were taken from slices of the PRE- SAGE â OCT near the surface and at depths of 10 and 20 mm [Figs. 5(b) and 5(c)] and compared to the corresponding depths in the MC calculation. The near-surface FIG. 8. MC and PRESAGE â lateral dose profiles of the x-ray beam for a mm field at the surface, 10 and 20 mm depths normalized to 20 mm depth. [Color figure can be viewed at wileyonlinelibrary.com] FIG. 9. MC and PRESAGE â lateral dose profiles of the x-ray beam for a mm field at the surface, 10 and 20 mm depths normalized to 20 mm depth. [Color figure can be viewed at wileyonlinelibrary.com]

7 1261 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1261 profiles in the case of the PRESAGE â samples are taken within the first 2 mm in the first slice that features a full image of the field due to the curvature at the surface of the rod. For these scans the number of projections taken by the OCT was increased to 1024 with step size and reconstructed at 4 pixels/mm to obtain the sharpest possible field penumbra. The profiles normalized to 20 mm depth are shown in Figs. 8 and 9. Both Figs. 8 and 9 display agreement in terms of shape of the profile across the fields; however, there is a notable FIG. 10. Calibration curve of the EBT3 film on a 6 MV linear accelerator. [Color figure can be viewed at wileyonlinelibrary.com] difference in the characteristics of the penumbra and out of field dose for the PRESAGE â dosimeters. Both the MC and water equivalent PRESAGE â profiles display dose out of the field which decreases with distance from the edge of the field and increases with depth and field size. The PRESAGE â out of field dose is times higher than the MC model predicts at 2 mm from the field edge and decreases to comparable out of field doses of the MC model at 3 10 mm from the field edges. 3.D. Synchrotron beam profiles (PRESAGE â, MC, and EBT3) Investigation into the discrepancy of the PRESAGEâ profiles and the MC model was carried out with Gafchromic TM EBT3 films. Since it is difficult to accurately achieve low doses on the synchrotron due to speed constraints of the sample stage goniometer, the film was calibrated on a 6MV linear accelerator from 0.25 to 30 Gy (Fig. 10). Lateral and vertical dose profiles of the PRESAGE â,mc model, and EBT3 films were compared in Figs. 11 and 12 near the surface and at 20 mm depth for both the mm and mm fields. The film and MC model both exhibit a steep gradient at the edge of the x-ray FIG. 11. MC, water equivalent PRESAGE â, and EBT3 lateral dose profiles of the synchrotron x-ray beams (a) mm near surface, (b) mm at 20 mm depth, (c) mm near surface, and (d) mm at 20 mm depth. [Color figure can be viewed at wileyonlinelibrary.com]

8 1262 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1262 FIG. 12. MC, water equivalent PRESAGE â, and EBT3 vertical dose profiles of the synchrotron x-ray beams (a) mm near surface, (b) mm at 20 mm depth, (c) mm near surface, and (d) mm at 20 mm depth. [Color figure can be viewed at wileyonlinelibrary.com] beam and similar out of field doses. The out of field dose decreases with distance from the field edge and increases with depth and field size closely matching the PRESAGE â profiles at distances greater than 2 mm from the field edge for the mm and mm fields. There is a subtle difference between the lateral and vertical dose profiles in the out of field dose measured by the PRE- SAGE â and film compared to the MC calculations. The measured out of field dose is slightly higher in the vertical dose profiles for the PRESAGE â and film. There is also a sharper penumbra in the vertical dose profiles of the PRESAGE â although still not as sharp as the MC calculation of the film measurements. To examine the effect of polarization and the roll off effect, we compared the lateral and vertical dose profiles of the MC calculation, PRESAGE â and film for the mm x-ray field (Fig. 13). The vertical dose profiles have steeper penumbra than the lateral dose profiles for MC calculation, PRESAGE â, and film. The out-of-field dose is the same for the MC calculation but slightly higher in the vertical profile for the PRESAGE â and film. 4. DISCUSSION The radiological response of the water equivalent PRE- SAGE â dosimeters compared to water over an energy range consisting of SXRT, synchrotron, and linac x-ray beams was consistent with the calculated radiological response (Fig. 1). The PDD profiles determined by the water equivalent PRESAGE â show agreement with those of the ion chamber within 2% except in the first few millimeters of the buildup region for megavoltage beams (Fig. 5) which is an inherent feature of optical CT scanning of PRESAGE â dosimeters. 9,23 The PDD profiles of the water equivalent PRESAGE â show agreement with ion chamber measurements and MC calculated PDD profiles for synchrotron BB field sizes of mm and mm within 1% for depths greater than 5 mm when normalized to 20 mm depth of a mm field (Fig. 7). The agreement between the water equivalent PRESAGE â and MC model for depths below 5 mm is within 2% and display similar characteristics of the buildup region. For depths below 5 mm, the PRESAGE â dosimeters are able to validate the MC model where ion chamber measurements are not possible; however, it must be noted that the OCT resolution (0.25 mm) and the voxel size of the MC calculation (0.25 mm) is too large to precisely define the maximum dose point. The ability to accurately measure the change in PDD near the surface is a fundamental challenge facing medical physicists implementing treatment planning systems in general and in particular for patients potentially undergoing synchrotron radiotherapy using BB or MRT in the near future.

9 1263 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1263 FIG. 13. Lateral versus vertical dose profiles of the mm synchrotron x-ray beam near the surface and at 20 mm depth for (a) MC, (b) PRESAGE â, and (c) EBT3 film. [Color figure can be viewed at wileyonlinelibrary.com] An OCT system with a higher resolving capability or a highresolution microscopy technique such as laser scanning confocal microscopy 13 must be used to accurately verify the change in PDD near the surface. The field width profiles for the mm MC calculations and water equivalent PRESAGE â measurements have similar relative intensities at the surface, 10 and 20 mm depth; however, there are differences in the shape of the penumbra regions. The MC data have a sharper field edge than the water equivalent PRESAGE â. The water equivalent PRESAGE â profiles display rounded shoulders of the delivered field. The MC and water equivalent PRESAGE â measurements however show minimal divergence of the field widths over the first 20 mm depth. As the penumbral shoulder discrepancy does not appear in the EBT3 film profiles, it apparently results from the OCT scanning of the water equivalent PRESAGE â dosimeters irradiated with a highly collimated minimally divergent synchrotron x-ray beam of small field size. Although the angular step size of the OCT scanner was 0.35 and the highest 3D reconstruction resolution available of 4 pixels/mm was used, the penumbra was significantly exaggerated when compared to the EBT3 film profiles. The EBT3 film was scanned with a resolution of 5.9 pixel/mm and displays a sharp penumbra for the and mm fields at the surface and at 20 mm depth, indistinguishable from the MC model (Fig. 9). A literature search was conducted to explore previous experience with profiling sharp field edge penumbra using PRE- SAGE. Although there appears to be little evidence, Clift et al. 23 reported that 80% 20% penumbral widths measured using PRESAGE were typically times broader than

10 1264 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry 1264 those measured using IC or EBT film, respectively. The effect was observed to increase with decreasing field size. The EBT3 film does display dose outside the field slightly higher than the MC calculation which decreases with distance from the field edge and increases with depth. However, the magnitude of the dose outside the field is minimally higher than the MC model predicts at the field edge and closely matches the dose outside the field beyond the penumbra region. Thus, it is apparent that more dose is delivered outside the field due to lateral scatter in the phantom and possibly scatter from the edge of the field-defining masks than the MC model predicts. The comparison of lateral and vertical dose profiles all display the similar characteristic that the penumbra of the vertical profile is steeper than the lateral profile. This is primarily due to the roll off effect of the beam which is modeled by the MC code and seen in the PRE- SAGE â and film measurements. Polarization of the synchrotron beam is in the horizontal direction and will contribute to the out of field dose; 25 however, the effect is small compared to maximum dose of the x-ray field it was not observable as a difference in the out-of-field dose profiles in the lateral and vertical direction comparison for MC, water equivalent PRESAGE â and film (Fig. 13). Further work refining the MC model must be undertaken to accurately calculate the out of field dose at depth due to phantom scatter. If synchrotron treatments are to take advantage of the high dose gradients to spare healthy tissue adjacent to the target volume, the out-of-field dose is an important quantity that must be considered when attempting to limit the dose to organs at risk. 5. CONCLUSIONS Alterations to the original PRESAGE â formula produced water equivalent PRESAGE â dosimeters with similar radiological responses to water over a wide energy range that included synchrotron energies for SSRT and MRT treatments. PDD profiles of synchrotron beams for and mm field sizes show good agreement with ion chamber measurements and the MC model. The limitation of the water equivalent PRESAGE â dosimeters is the inability to accurately measure the field penumbra with the OCT scanning technique. However, the water equivalent PRESAGE â dosimeters were able to verify dose profile characteristics of the MC model which were also validated with EBT3 film. The water equivalent PRESAGE â dosimeters possess the ability to independently verify dose distributions calculated by a MC model for SSRT and potentially MRT treatments with sufficiently increased resolution of the PRESAGE â readout technique. ACKNOWLEDGMENTS The authors were grateful for financial assistance from J. Millar and T. Ackerly (Alfred Health Radiation Oncology, Alfred Health) and also acknowledge N. Brouwer (Alfred Health Radiation Oncology, Alfred Health), J. Crosbie, D. Pelliccia, S. Keehan (RMIT University), A. Blencowe (University of South Australia), A. Stevenson, C. Hall, and J. Livingstone (Australian Synchrotron) for their technical assistance and useful discussion. CONFLICT OF INTEREST The authors have no conflict of interest to report. a) Author to whom correspondence should be addressed. Electronic mail: frank.gagliardi@wbrc.org.au REFERENCES 1. Br auer-krisch E, Adam J-F, Alagoz E, et al. Medical physics aspects of the synchrotron radiation therapies: Microbeam radiation therapy (MRT) and synchrotron stereotactic radiotherapy (SSRT). Physica Med. 2015;31: Martinez-Rovira I, Sempau J, Prezado Y. Development and commissioning of a Monte Carlo photon beam model for the forthcoming clinical trials in microbeam radiation therapy. Med Phys. 2012;39: Cornelius I, Guatelli S, Fournier P, et al. Benchmarking and validation of a Geant4-SHADOW Monte Carlo simulation for dose calculations in microbeam radiation therapy. J Synchrotron Radiat. 2014;21: Poole CM, Day LRJ, Rogers PAW, Crosbie JC. Synchrotron microbeam radiotherapy in a commercially available treatment planning system. Biomed Phys Eng Express. 2017;3: International Atomic Energy Agency. Specification and Acceptance Testing of Radiotherapy Treatment Planning Systems. IAEA-TECDOC Vienna: IAEA Adamovics J, Maryanski MJ. Characterisation of PRESAGE: a new 3-D radiochromic solid polymer dosemeter for ionising radiation. Radiat Prot Dosimetry. 2006;120: Guo PY, Adamovics JA, Oldham M. Characterization of a new radiochromic three-dimensional dosimeter. Med Phys. 2006;33: Alqathami M, Blencowe A, Qiao G, Adamovics J, Geso M. Optimizing the sensitivity and radiological properties of the PRESAGE â dosimeter using metal compounds. Radiat Phys Chem. 2012;81: Alqathami M, Blencowe A, Geso M, Ibbott G. Characterization of Novel PRESAGE â dosimeters for megavoltage and kilovoltage X-ray beam dosimetry. Radiat Meas. 2015;74: Gorjiara T, Hill R, Kuncic Z, et al. Investigation of radiological properties and water equivalency of PRESAGE â dosimeters. Med Phys. 2011;38: Jackson J, Juang T, Adamovics J, Oldham M. An investigation of PRE- SAGE â 3D dosimetry for IMRT and VMAT radiation therapy treatment verification. Phys Med Biol. 2015;60: Brady SL, Brown WE, Clift CG, Yoo S, Oldham M. Investigation into the feasibility of using PRESAGE( TM )/optical-ct dosimetry for the verification of gating treatments. Phys Med Biol. 2010;55: Gagliardi FM, Cornelius I, Blencowe A, Franich RD, Geso M. High resolution 3D imaging of synchrotron generated microbeams. Med Phys. 2015; 42: Taylor ML, Smith RL, Dossing F, Franich RD. Robust calculation of effective atomic numbers: the Auto-Zeff software. Med Phys. 2012;39: Gorjiara T, Hill R, Kim J-H, Kuncic Z, Adamovics J, Baldock C. Study of dosimetric water equivalency of PRESAGE â for megavoltage and kilovoltage x-ray beams. J Phys: Conf Ser. 2010;250: Alqathami M, Blencowe A, Yeo UJ, Doran SJ, Qiao G, Geso M. Novel multicompartment 3-dimensional radiochromic radiation dosimeters for Nanoparticle-enhanced radiation therapy dosimetry. Int J Radiat Oncol Biol Phys. 2012;84:e549 e555.

11 1265 Gagliardi et al.: Water equivalent PRESAGE for SRT dosimetry Rankine L, Oldham M. How effective can optical-ct 3D dosimetry be without refractive fluid matching? J Phys: Conf Ser. 2013;444: Stevenson AW, Crosbie JC, Hall CJ, Hausermann D, Livingstone J, Lye JE. Quantitative characterization of the X-ray beam at the Australian Synchrotron Imaging and Medical Beamline (IMBL). J Synchrotron Radiat. 2017;24: Lye JE, Harty PD, Butler DJ, et al. Absolute dosimetry on a dynamically scanned sample for synchrotron radiotherapy using graphite calorimetry and ionization chambers. Phys Med Biol. 2016;61: Alqathami M, Blencowe A, Qiao G, Butler D, Geso M. Optimization of the sensitivity and stability of the PRESAGE â dosimeter using trihalomethane radical initiators. Radiat Phys Chem. 2012;81: Bruggmoser G, Saum R, Schmachtenberg A, Schmid F, Sch ule E. Determination of the recombination correction factor k S for some specific plane-parallel and cylindrical ionization chambers in pulsed photon and electron beams. Phys Med Biol. 2007;52: N Agostinelli S, Allison J, Amako K, et al. Geant4 - a simulation toolkit. Nucl Instrum Methods Phys Res, Sect A. 2003;506: Clift C, Thomas A, Adamovics J, Chang Z, Das I, Oldham M. Toward acquiring comprehensive radiosurgery field commissioning data using the PRESAGE â /optical-ct 3D dosimetry system. Phys Med Biol. 2010;55: International Atomic Energy Agency. Absorbed Dose Determination in External Beam Radiotherapy: An International Code of Practice for Dosimetry Based on Standards of Absorbed Dose to Water. TRS- 398 V.12. Vienna: IAEA Bartzsch S, Lerch M, Petasecca M, Br auer-krisch E, Oelfke U. Influence of polarization and a source model for dose calculation in MRT. Med Phys. 2014;41: