L Zhao

Dosimetric comparison between proton and photon beams in the moving gap region in cranio-spinal irradiation (CSI)

Abstract Purpose. To investigate the moving gap region dosimetry in proton beam cranio-spinal irradiation (CSI) to provide optimal dose uniformity across the treatment volume. Material and methods. Proton beams of ranges 11.6 cm and 16 cm are used for the spine and the brain fields, respectively. Beam profiles for a 30 cm snout are first matched at the 50% level (hot match) on the computer. Feathering is simulated by shifting the dose profiles by a known distance two successive times to simulate a 2 × feathering scheme. The process is repeated for 2 mm and 4 mm gaps. Similar procedures are used to determine the dose profiles in the moving gap for a series of gap widths, 0–10 mm, and feathering step sizes, 4–10 mm, for a Varian iX 6MV beam. The proton and photon dose profiles in the moving gap region are compared. Results. The dose profiles in the moving gap exhibit valleys and peaks in both proton and photon beam CSI. The dose in the moving gap for protons is around 100% or higher for 0 mm gap, for both 5 and 10 mm feathering step sizes. When the field gap is comparable or larger than the penumbra, dose minima as low as 66% is obtained. The dosimetric characteristics for 6 MV photon beams can be made similar to those of the protons by appropriately combining gap width and feathering step size. Conclusion. The dose in the moving gap region is determined by the lateral penumbras, the width of the gap and the feathering step size. The dose decreases with increasing gap width or decreasing feathering step size. The dosimetric characteristics are similar for photon and proton beams. However, proton CSI has virtually no exit dose and is beneficial for pediatric patients, whereas with photon beams the whole lung and abdomen receive non-negligible exit dose.

Factors Affecting Accuracy in Proton Therapy

Introduction: To examine the various processes involved and to assess their effects on the accuracy of proton therapy. Materials and methods: Proton therapy involves several processes: (1) Beam commissioning. (2) CT of patient to obtain the anatomical information of the patient. (3) Contouring of the tumors and organs at risks from the CT images. (4) Treatment planning to determine the beam configurations, the aperture and compensators, and the beam weightings, satisfying a specific set of dose-volume constraints. (5) Output factor measurements for each field. (6) Patient setup verification with image guidance. (7) Dose delivery. (8) Neutron dose and proton RBE at the distal edge. Within each step, there are several sub- processes that may contribute to the uncertainty in the treatment. By analyzing each of the sub-processes within each process, we estimated an uncertainty to each sub-process and/or an uncertainty on the proton range (in millimeter). A total uncertainty in dose delivery and the location of the distal edge can be estimated. Results: Examples of the uncertainties assessed for the various processes are : (1) ±1.5%; (4) ±3.0%, and 1-3mm; (5) ±2.0%; (6) ±2 mm; (7) ±2.0%, ±2mm. The uncertainties in (2), (3)and (8) neutron dose strongly depend on the location and type of the tumor. The proton RBE value at the distal edge is still debatable. The overall uncertainty in proton therapy is at least ±4.5% and ±4 mm (by adding the various uncertainties in quadrature), without consideration of processes (2), (3) and (8).Discussion: Due to the complexity in proton therapy, it is far more complicated to assess the accuracy in proton therapy than that for photon therapy. We showed that the accuracy in proton therapy is at least ± 4.5% in dose delivered to a tumor with an uncertainty of ±4mm to the distal edge of the SOBP.

Temporal response of gafchromic EBT2 radiochromic film in proton beam irradiation

Introduction: Gafchromic EBT2 is being used for radiation dosimetry including photon, electrons and recently in proton beams, based on its color changes from transparent to blue due to polymerization. It is critical to understand the temporal response of the polymerization after irradiation to proton beam.

SU‐C‐BRCD‐05: A Failure Mode and Effects Analysis (FMEA) Approach for Craniospinal Irradiation (CSI) with Proton Therapy (PT)

Purpose: In this paper, we report an FMEA approach on CSI at the IU Health Proton Therapy Center. Methods: A process map in proton CSI is developed. Each process consists of a number of sub‐processes. For each sub‐process, possible failures that may affect the process are identified and their respective Risk Priority Number (RPN) are calculated based on their severities (S), likelihood of occurrence (O) and probability of being detected (D), following the TG100 guidelines. Failures that most adversely impact the treatment are identified and quality assurance procedures to safeguard these most serious failures are developed. Attention is also given to certain failures which have low RPN but which may have dire consequence if it occurs in a treatment. Results: Ten intermediate processes involved in the CSI are identified. The number of sub‐processes within each intermediate process varies, from as few as one to as many as 11. For the ten processes in PTCSI, there are total of 66 sub‐processes, 139 failure modes and 561 causes of failures. 6/10 processes have failures with RPN=300. Examples of failures with such large RPN values are errors in administering anesthesia, in‐correct patient setup for image guidance, problems in the handling of apertures and compensators, etc. Conclusions: The implementation of FMEA in CSI (or any treatments) is a team effort. Significant efforts are involved in setting up process trees, failure modes, estimating the RPN values for each cause of failures, etc. However, once a FMEA is properly aligned, it is relatively easy to identify the most critical factors that require special attentions or QA to ensure safe execution of the processes. A learning curve to implement FMEA in any radiation oncology department should be expected given the different analysis practice from traditional QA approaches.

SU‐E‐T‐295: Factors Affecting Accuracy in Proton Therapy

PURPOSEnTo examine the various processes involved and to assess their effects on the accuracy in proton therapy.nnnMETHODSnProton therapy involved several processes: (1) Beam commissioning. (2) CT scan of patient. (3) Contouring. (4) Treatment planning. (5) Output factor measurements for each field. (6) Patient setup verification with image guidance. (7) Dose delivery. (8) Neutron dose and proton RBE at the distal edge. Within each step, there are several sub-processes that each may contribute to the uncertainty in the treatment. By analyzing each of the subprocesseswithin each process, based on measurements or published data, we estimated a % uncertainty to each sub-process and/or a distance uncertainty (in millimeter) on the proton range. A total uncertainty in proton therapy is estimated.nnnRESULTSnThe uncertainties assessed for the various processes are : (1) ±1.5%; (4) ±3.0%, and 1-3mm; (5) ±2.0%; (6) ±2 mm; (7) ±2.0%, ±2mm. The uncertainties in (2) CT, (3) contouring and neutron dose in (8) strongly depend on the location and type of the tumor. On the other hand, the proton RBE at the distal edge in (9) is still debatable and may affect the dose uncertainty from 0-20% depending on which value we want to accept. Thus the overall uncertainty in proton therapy is at least ±4.5% and ±4 mm (by adding the various uncertainties in quadrature), without consideration of processes (2), (3) and (8), and the RBE effect.nnnCONCLUSIONSnDue to the complexity in proton therapy and the various factors that may affect the accuracy in proton therapy, it is far more complicated to assess the accuracy in proton therapy. Our preliminary study showed that the accuracy in proton therapy is at least ± 4.5% in dose delivered to a tumor with an uncertainty of ±4mm to the distal edge of the SOBP.

SU‐GG‐T‐61: Dose Correction in Lung for HDR Breast Brachytherapy

Purpose: To evaluate and compare the dose in water equivalent medium to a lung equivalent medium from HDR Ir‐192. Methods and Materials: In a 40×40×40 cm3 water tank, a Contura brachytherapy device, inflated to 4 cm diameter was situated directly below the center of a 30 cm×30 cm×1 cm solid water slab. In the first phase, 9 cm of solid water was stacked above the 1 cm base. A Scanditronix parallel plate ion chamber (Type NACP‐02) was centered above the 1 cm of solid water base. Ionization current measurements established the central HDR source dwell position for channels 1, 2, 3, and 5 of the Contura. Additional data was acquired in the 9 cm stack at 1 cm increments. The second measurement phase was performed after replacing the 9cm solid water stack with cork slabs. The ratios of measurements in the two phantoms were calculated. The Oncentra Masterplan treatment planning system was used to compute comparable ratios based on CT studies of the two experimental setups. Results: Lower dose was measured in the cork within 1 cm of the cork/solid water interface due to backscatter effects. Higher dose was measured beyond 1cm from the cork/solid water interface, increasing with path length up to 15% at 9 cm depth in cork. The treatment planning computations did not predict either of these dose changes. Conclusions:Dose from HDR Ir‐192 in a cork media has been shown to be significantly different than in unit density media. This model should represent the dose effects in the lung when the breast is treated with Ir‐192 brachytherapy. Commercial brachytherapy planning systems do not consider heterogeneous media and therefore do not compute accurate lungdose. Empirical corrections based on measurements from this study may be used to correct lungdose computations associated with HDR breast brachytherapy.

SU‐GG‐T‐336: Effect of Treatment and Beam Parameters on Surface Dose in Proton Beam Therapy

Purpose: The advantage of lower skin dose in protonbeam therapy may result in less radiation‐related side effects which are typically seen in hypo‐fractionated conventional external beam therapies. In this study, we evaluate the surface dose (SD) in proton therapy as a function of various beam parameters. Materials & Methods: SD is defined as the ratio of absorbed dose on CAX at surface to that at the middle of the Spread‐Out Bragg Peak (SOBP). SD in proton therapy is affected by several parameters: energy ( E ), SOBP , source‐to‐surface distance ( SSD ), air gap ( g ), field radius ( r ), material thickness upstream of surface besides air ( t ), atomic number of medium ( Z ), beam angle relative to surface (q), and nozzle type ( N ). The parameters, t, Z and q are not included in this study.. Results: Giving a proton range in water of 27 cm, SD rises from 30 to 90% with SOBP increasing from 0 to17cm. At high energy, SD is 5% higher in the uniform scanning (US) nozzle than in double scattering (DS) nozzle. This can be explained by the large difference in source‐to‐axis distances in US and DS nozzles (250 cm vs 320 cm). SSD and g have minimal impact on SD for r > 2.5cm. For small fields ( r < 2.5cm), SD increases significantly with field size decreases. Within 15mm of the surface there is a small but pronounced 2% buildup in protonbeam at high energies, which may be partially due to secondary electrons, secondary protons and heavy charged particles. Conclusions: In a clinical setting, SD is most significantly affected by SOBP extent, followed by field size, SAD, and beamenergy. In general, SD ranges from 50 to 95%. It is important to clearly understand and minimize SD in Proton therapy, especially in hypo‐fractioned treatments.

SU‐GG‐T‐473: Dose Uncertainty Due to High‐Z Materials in Clinical Proton Beam Therapy

Purpose: In proton therapy, high‐Z materials, such as dental alloys, sternal reconstruction plates, prosthesis, and metallic ports, can introduce significant dose perturbations. Our objective is to quantify the high‐Z induced dosimetry uncertainty in clinical proton beams. Method and Materials: Dose perturbations from one titanium vascular port and a steel injection port of a breast expander were studied. An extended CT‐electron density (ED) curve for MVCT was obtained with an RMI CT phantom and metal plates (Al, Sn, Ti, Pb). Measurements taken with a 2D ion chamber array placed at different depths downstream from the high‐Z‐solid water interface were compared with dose calculations on the XiO treatment planning system based on both the MVCT and kVCT images. The Monte Carlo code FLUKA was used to verify accuracy of inhomogeneity corrections in the pencil beam algorithm. Dose perturbation factor (DPF) was defined as the ratio of the doses with and without the high‐Z material. Results: For MVCT, the CT‐ED relationship is linear from lung to lead. There are considerable dose enhancement (>10%) near the high‐Z interface due to secondary electrons from the metallic port. DPF as large as 20% was observed within the spread‐out Bragg peak. MVCT images provided more accurate delineation of the metallic object compared to kVCT, which tends to overestimate the water equivalent thickness of the metal object, resulting in shallower proton depth than its actual value. The DPF calculated from MVCT planning agrees with the measured results within 10%. Results from Monte Carlo calculations are comparable to results from XiO although there are small differences. Conclusions: Understanding dose uncertainty induced by high‐Z material is very important in proton therapy. MVCT based treatment planning may be preferred with an extended CT‐ED curve. Difference between measured and calculated dose distribution shall be quantified.

SU‐GG‐T‐474: Feasibility Study of MVCT Imaging Guided Adaptive Proton Therapy for Head and Neck Cancers

Purpose: Extending MVCTs potential for daily patient position and anatomy verification, we investigated the feasibility of using MVCT to assess the daily delivered dose for head and neck patients treated with proton beams. Method and Materials: A Rando head phantom was scanned on a Philips Brilliance Big Bore CTscanner to obtain the planning kVCT images. The corresponding MVCT images were acquired on a helical Tomotherapy unit. To establish CT‐electron density curves, a Gammex RMI 465 electron density phantom was scanned on both kVCT and MVCT scanners. A typical 4‐field plan was created for the phantom and dose distribution was calculated using pencil beam algorithm from the XIO treatment planning system. After MVCT and kVCT images manually registered, kVCT‐based treatment plan with structure contours and beam configuration including beam shaping devices was mapped to MVCT images for dose re‐computation. Dose map, isodose curves, dose profiles, gamma maps, and dose‐volume histograms were used for dose comparison. Results: Although the noise level in MVCT images is higher than that with the comparabl kVCT the dose distribution computed on MVCT image is comparable to the dose distribution computed on the kVCT images. The dose comparison showed an agreement of higher than 97% between the two plans given the criteria of 3% dose difference and 3 mm distance to agreement. The biggest dose difference between two plans occurred around the periphery of the target. There was negligible difference in the DVH comparison. Conclusions: MVCT images acquired for daily setup verification in image‐guidedproton therapy are feasible for assessment of daily deliveredproton dose distributions with accuracy comparable to that of kVCT‐based dose calculation. The original treatment plan based on kVCT can be transferred to the daily MVCT image set to evaluate the actual dose distribution for adaptive proton therapy.

SU‐GG‐I‐03: Implications for Proton Therapy Treatment Planning of Tissue Characterization Curves from Different CT Scanners

Purpose: Accuracy in proton beam treatment planning is intimately related to the quality of the treatment planningCTimage set characterized by Hounsfield unit (HU). Relative electron density (ED) or relative stopping power (RSP) is associated with HU for inhomogeneity corrections. Variability of these parameters from several scanners is evaluated. Methods: HU‐ED curves were obtained from 15 CTscanners and a TomoTherapy unit operated in a scanner mode from nine different institutions. In all, five different RMI467 CT phantoms were used to obtain the different curves. The ED was converted to RSP using the Bethe‐Bloch formula and ionization energies in ICRU49. The various HU‐RSP curves were then compared. Results: The differences in material composition between the five RMI phantoms were well within the acceptable variation specified by the manufacturer (ranging from 8% for Lung300 to 0.5% for inner bone). The HU‐RSP curve for the TomoTherapy scanner was almost linear. The calibration curves for the 15 CTscanners exhibit similar shapes and can be described by three linear segments. In the HU range −700 to zero, the HU‐RSP curves are almost linear for all scanners, and are within 6% of each other. For the region immediately above water to RSP∼1.1, the RSP changed slowly with HU. Above RSP=1.1, the curves increases approximately linearly with HU. Except for one CTscanner, all other curves have 1.8, present HU‐RSP variation only allows linear extrapolation. Conclusion: While most CTscanners provide very similar tissue characterization, some exhibit larger variations than others probably because of the x‐ray spectrum. The RMI CT phantom could be modified to accommodate higher density materials covering the range RSP=2.0–7.0 (eg. Al‐Ta) for more accurate inhomogeneity correction for high Z materials. A point at RSP∼0.1–0.2 should also be added for the low density region.