Maximum Dose Difference Research Articles

SU‐E‐T‐441: Sensitivity of a 3D‐Verification System to VMAT Delivery Errors

Purpose: VMAT offers continuous dose delivery in the form of arcs while dynamically changing position of collimators and dose rate. The aim of the present study is to investigate the sensitivity of a 3D‐verification system to different VMAT delivery errors. Methods: Three VMAT plans were generated with MONACO TPS (v2.0, Elekta‐CMS, Maryland Heights, MO) for brain tumor paradigms. With in‐house software, we modified the plans introducing the following errors: (i) One set of leaf bank was opened by 1mm, 2mm and individual leaves were randomly opened (0mm–2mm). (ii) Both sets of leaf banks were shifted by +/−1mm and +/− 2mm. (iii) Gantry angle was shifted by +/−0.5°, +/−1°, +/−2° and randomly shifted (−2° to +2°). All these plans (a total of 45) were delivered to a MatriXX‐based COMPASS system (v2.0, IBA‐Dosimetry,Germany). 3D‐dose distribution was reconstructed onto the patient anatomy based on the measured fluences. Dose—volume parameters for target (D95) and critical organs (Davg) were compared and analyzed. Results: For three reference plans, the maximum dose‐difference (MDD) between MONACO and COMPASS was about 10% (Davg‐lens). However, the absolute dose‐difference (ADD) for the same was 4cGy. For plans with errors in one set of leaf bank, the MDD and ADD were 10% and 13cGy for Davg‐optic nerve, respectively. With errors in both leaf banks, the MDD and ADD were 15% and 21cGy for Davg‐optic nerve, respectively. With gantry angle errors, the MDD and ADD were 50% and 20cGy, respectively. For PTV‐D95, the MDD noticed was 6% and the ADD was 16cGy for different kind of errors. Conclusions: Any increase in aperture size (error in one leaf bank) showed systematic deviations. Shifts in both leaf banks up to 2mm and gantry angle difference (2°) showed considerable deviations. Errors greater than 2mm/2° can be detected with COMPASS system in most of the cases.

SU-E-T-398: 3D Dosimetry Verification for Prostate Implant Treatment Plans

Purpose: A 3D dose calculation program (PIDOSE) for prostate implant was developed for comparison to treatment plans from a commercial treatment planning system (Variseed V7.1, Varian Medical System, Palo Alto, CA). Methods: Although a simple dose verification (one source) was done when Variseed was commissioned, we believe more complicated treatment plans should also be verified. These verifications include 3D dose distributions, dose histograms, isodose lines, maximum, median and minimal doses as a second hand check. A file with needle‐source information was exported from the Variseed planning system to PIDOSE. A 3D dose matrix was created using a tri‐exponential fitting function for the seed model (1). All single‐source doses were summed and selected isodose lines were superimposed on the ultrasoundimage slices together with those from Variseed. Dose‐volume histograms for prostate, urethra and partial rectum were calculated and compared. Results: : Although maximum dose, mean dose and standard deviation were slightly different, there were no large deviations from typical isodose lines (< 1 mm), minimum dose (<2%), median dose (<2%) and DVHs. Conclusions: The 3D dose verification as a second hand check confirms the accuracy of dose calculations provided by the Variseed Planning System. The small differences between maximum dose, mean dose and standard deviation are due to the differences in dose calculation resolution, dose volume interpolation, and radial dose fitting functions. The maximum doses from PIDOSE at some calculation points near the seeds were high due to the exponential fitting function.

SU‐E‐T‐149: 3D Proton Gel Dosimetry

Purpose: To evaluate the water equivalency of PRESAGE and genipin gel for 3D proton dosimetry. Methods: The GEANT4 Hadrontherapy example was modified to calculate depth doses in four different formulations of PRESAGE dosimeters and in water for a proton beam with a mean energy of 62 MeV. To compare the depth doses in each phantom, the water equivalent thickness (WET) was calculated for each PRESAGE dosimeter according to IAEA technical report 398. Results: Our GEANT4 results show that the maximum energy deposition occurs at 2.9 cm depth in water, at 2.7 cm in the two new chemical formulations of PRESAGE®, at 2.6 cm in the original formulation of PRESAGE®, at 2.55 cm in PRESAGETM and at 2.85 cm in genipin gel. The deposited energy in PRESAGETM is 15% higher than that in water, while it is approximately 5% higher than water in the new formulations of PRESAGE®. For the calculated relative dose versus WET, the new PRESAGE® dosimeters are in better agreement with the depth doses in water. At Bragg peak the maximum relative dose difference with that of water is up to 1% for the PRESAGE® formulation A and more than 2% for the PRESAGE® formulation B and C and the PRESAGETM dosimeter. For the genipin gel the maximum relative dose difference with water is less than 0.5%. Conclusions: The PRESAGE® dosimeters with the new chemical formulations are more water equivalent than the existing PRESAGE® and PRESAGETM dosimeters and are thus more suitable for accurate 3D dosimetry of proton beams. The depth dose in the genipin gel dosimeter shows better agreement with water than the PRESAGE dosimeters. Further experimental studies are planned to investigate possible saturation in the new PRESAGE® formulations and in genipin gel for high LET radiation fields.

SU‐E‐T‐663: Extended TrueBeam Patient‐Independent Phase Space Library for the 6X, 6XFFF, 10X and 10XFFF Radiotherapy Beams

Purpose: To develop a patient‐independent IAEA‐compliant phase space database using accurate geometry representation of the treatment head components imported from CAD (Computer‐Aided‐Design) drawings and a physics list exploration for the specific materials of the linac treatment head. Methods: Geant4.9.4 was employed to simulate the TrueBeam 6X, 6XFFF, 10X and 10XFFF photon beams. Electron beam transport in the linac was simulated with Parmela and used as input into dose calculations. Geometry components were tessellated solids, i.e. Generalized‐Dynamic‐Markup‐Language files, extracted from mechanical CAD drawings. A phase space scorer was implemented above the upper jaws on a cylindrical surface. The simulations were run on the Amazon Elastic Compute Cloud. Dose calculations in a water phantom (with 6×6×4 mm3 voxels) were performed using two physics lists, Standard Option‐0 and Low‐Energy Livermore/Penelope, with a 10um range‐cut. The Urban93 and Goudsmit‐Saunderson models were considered for electron multiple‐scattering, while Tsai and 2BN models for electron Bremsstrahlung angular generators. Results: The maximum dose difference between simulation and experiment was &lt;1% for the percent depth dose curves and &lt;3% for in‐field dose profiles. For large fields, the simulated dose profile horns were smaller by 2% for the 6X beam and larger by 3% for the 10X beam compared to the experiment. Simulations using the Goudsmit‐Saunderson model reproduced the experiment with higher accuracy than Urban93, causing a 3x CPU increase. No CPU penalty or accuracy differences were observed in simulations using the Tsai versus the 2BN angular generators. The Standard Option‐0 list has the advantage of providing 2x CPU speedup and can be used with a range cut of 0.1mm for medical physics applications above 1MeV, ensuring the same accuracy as the Low‐Energy Livermore/Penelope list. Conclusions: We developed an extended TrueBeam phase space database using high precision geometry implementation from CAD that can be used for future medical physics user applications.Work supported by Varian Medical Systems

TU‐E‐BRC‐11: Clinical Impact of Dose Mapping Sensitivity to Deformable Image Registration Parameter Selection

Purpose: 4D dose mapping accuracy is limited by the accuracy of the displacement vector fields (DVF) generated by deformable image registration (DIR) algorithms. DIRs depend on user adjustable parameters whose election may result in different resulting DVFs. This work evaluates the clinical impact of DIR parameter selection on 4D accumulated doses. Methods: A 180 cGy/fraction PTV dose clinical treatment plan is generated, with dose calculated on each phase of a ten‐phase 4D lung CT image. Per‐ phase doses are mapped to a reference phase using DVFs from the demons DIR implemented in a research version of Pinnacle 9.1 for dose accumulation using a variety of settings for DIR user‐adjustable parameters. Dose‐volume histograms, TCP, and NTCP values are evaluated for each accumulated dose as are differences in accumulated dose values. Results: Cumulative dose values vary with the user adjustable parameters. The maximum dose difference is 62.5 cGy/fraction when the number of image filter histogram bins changes from 32 to 16 and 14.9 cGy/fraction when the number of level 4 iterations (where the image is downsampled to 1/8 of the original resolution) changes from 50 to 10. Dose variance tends to be in specific locations that accumulate most differences. These regions are suspect for dose accumulation errors. Despite the dose differences found, DVHs, TCP, and NTCP values varied little, indicating minimal clinical impact of DIR parameters. Conclusions: Accumulated dose variations due to DIR parameter selection occur in limited regions of an image. Such variation can be used to estimate the precision of 4D dose. Although minimal clinical impact is found for the case presented, additional study with more patients and different planning features are required prior to establishing global conclusions.Supported by NIH‐P01CA116602 and Philips Medical Systems

SU-E-T-483: Investigations of Dose Delivery Accuracy of High Dose Rate Flattening Filter Free (FFF) Beams in RapidArc™ and IMRT Deliveries on TrueBeam™

Purpose: Medical linacs with dose rates as high as 2400 cGy/min at 10MV and 1400 cGy/min at 6MV have been clinically installed in many institutions. In reality, little is known about the accuracy of delivery at these high dose rates, particularly for complex treatment plans with simultaneous motion of gantry and multileaf‐collimator (MLC) aperture. We present measurements that examine the dosimetric accuracy of dynamically‐shaped RapidArc™ and IMRT beams at the highest available dose rates. Methods: An Eclipse™ ver. 8.9 planning system commissioned with the analytic anisotropy algorithm (AAA™) for the flattening‐filter‐free X‐ray beams of the Varian TrueBeam™ linear accelerator with a high‐definition 120‐leaf MLC (HDMLC) accessory was used to create RapidArc™ and IMRT lung‐ and abdominal‐target patient treatment plans deliverable in the FFF mode of TrueBeam™. Verification plans applied to a 22‐cm diameter cylindrical polymethylmethacrylate phantom configured as water‐ equivalent material with an electron density of 1.147 were delivered at the highest available dose rates. Dose was measured with TG‐51 cross‐ calibrated ScandiDos® Delta4™dual orthogonal‐plane 1069‐diode system for conventional 2 Gy fractions and for 15Gy–16.5 Gy fractions characteristic of stereotactic body radiotherapy(SBRT). Delivery was repeated with conventional dose rate (400 cGy/min). Measured results for the two dose rates were compared with each other, and with Eclipse treatment plans. Results: Excellent agreement was found between 2400 and 400 MU/min delivery. A sample 2‐arc RapidArc™ plan gave absolute relative dose difference ratio of 0.42%±0.21% for the 439 points measuring above 10% of prescription dose, with maximum dose difference over all 1069 measurement points being 0.76% of prescription. For a small‐field 1‐ arc plan, quantities were 0.96%±0.14% for 103 points, and maximum difference over all 1069 points was 1.61%. Gated versus ungated delivery for 200 cGy/fx showed negligible differences at both 400 and 2400 dose rates. Eclipse calculations demonstrated reasonable agreement with absolute dose measurement.

MO-D-BRB-09: Evaluation of Dose to Normal/critical Structures from Lung Hypo-Fractionated Stereotactic Body Radiation Therapy

Purpose: The purpose of this work was to quantify the dose received by the normal/critical structures during lung stereotactic body radiation therapy(SBRT) when registered with respect to the tumor. Methods: For this retrospective work, 16 lungSBRT patients were treated with total dose of 50 Gy in 4 fractions. Cone‐beam CT was done for all fractions and registered with planning CT for target localization. Isocenter shifts were determined from the differences between the boney and tumor alignments. Once the isocenter shifts were determined for each fraction, doses were recalculated based on the new isocenter and summed over all four fractions to compare against the planned normal/critical tissuedose. The normal/critical structures evaluated were total and ipsilateral lung, spinal cord, and esophagus. Two data sets were collected from this work. The first data set was the collection of isocenter coordinate shifts in all three Cartesian coordinates. The second data set was the doses to the normal/critical structures from the planned and recalculated doses. Results: The study showed that while the maximum isocenter coordinate shifts in any direction could be as much as 1.60 cm, the normal/critical structure dose variations between the original plans and the simulated plans showed almost no change. The average V20 difference for total lung, ipsilateral lung, and esophagus were 0.01%, −0.03%, and −0.35%, respectively. For spinal cord, the maximum and mean dose differences were −0.08 Gy and −0.02 Gy, respectively. Conclusions: Target localization using daily CBCT with soft tissue registration was appropriate for minimizing the dose to the normal/critical structures without the need to re‐plan due to the changes in the tumor position. For tumors located close to a critical structure, it is advised that a daily CBCT is used to determine appropriate isocenter shifts.

SU‐E‐T‐431: Dosimetric Uncertainties in RapidArc Treatment

Purpose: The purpose of this study is to investigate the dosimetric effects of the beam delivery system on the delivered dose distributions of RapidArc treatments and their impact on RapidArc patient‐specific quality assurance. Methods: Monte Carlo dose calculations based on intensity maps that were built from the actual deliverable leaf sequences with and without considering the characteristics of the beam delivery system were performed in this study using our in‐house Monte Carlo software. The resolution of the intensity map is 0.2×0.2mm2 at the isocenter plane. The characteristics studied in this work include the source distribution, MLC leaf thickness, leakage, tongue‐and‐groove structure, effective leaf offset and dynamic delivery. The mean dose of the treatment target volume and the voxel doses in the high dose region of 5 RapidArc plans were used for comparison and evaluation. Results: The extra‐focal source, MLC leaf thickness, leakage, tongue‐and‐groove structure, effective leaf offset and dynamic delivery can affect the mean dose of a RapidArc treatment by up to 1.0%, 4.8%, 3.2%, 6.7%, 9.4% and 0.4%, respectively, when these factors are considered separately. They also cause significant uncertainties to the voxel dose with standard deviations up to 1.0%, 0.6%, 0.2%, 1.0%, 1.6% and 1.1%, respectively. The overall effect of all these characteristics on the mean dose is up to 2.2% and the standard deviation of the voxel dose uncertainty is up to 1.8%. The maximum voxel dose difference caused by these factors is up to 7.6%, 6.5%, 4.8%, 11.5%, 16.8% and 21.5% in the high dose region. Conclusions: The characteristics of the beam delivery system have great contribution to point dose uncertainty of a RapidArc treatment. And the effects are case‐dependent and location‐dependent. Treatment planning system should consider all these effects more precisely in the final dose calculation to reduce the measured dose uncertainty.

MLC mediated beam hardening effects in IMRT

Abstract Purpose Beam hardening effects due to photons passing through the multi-leaf collimator (MLC) frequently exist in Intensity Modulated Radiotherapy (IMRT) fields. A fast online dose transmission verification system, MBMC (Measurement Based Monte Carlo), can verify IMRT delivery by using a fluence efficiency map to replace MLC geometry and movement simulation. This system, however, ignores beam hardening effects, and assumes that dose disturbances through the MLC are not significant for IMRT fields. This assumption has to be justified before it can be applied in the clinic. Methods In this study, we simulated several field sizes (0.5 × 0.5, 1 × 1, 3 × 3, 5 × 5, and 10 × 10 cm2) to evaluate the dose influence of beam hardening effects under clinical conditions. In addition, a LATCH technique was used during simulation processes, which can record each particle interaction with specific gantry components, to distinguish between dose contribution from the total beam and MLC mediated beam. Results The MLC indeed caused significant beam hardening effects, but the dose contribution fraction from the MLC was noticeable only for field sizes less than 1 × 1 cm2. Furthermore, in mixed fields containing both the total beam and MLC mediated beam, the maximum dose deviation due to the presence of the MLC is small even for the 0.5 × 0.5 cm2 field size (∼2%). Conclusions The MLC causes noticeable beam hardening effects, but this effect results in only slight dose differences that are only noticeable for small field sizes in IMRT delivery. The use of a fluence efficiency map was feasible in our MBMC system.

Stereotactic radiosurgery-radiotherapy: Should Monte Carlo treatment planning be used for all sites?

PurposeTo evaluate a Monte Carlo (MC) treatment planning system for CyberKnife treatments of cranial and extracranial lesions and determine whether it is necessary for all treatment sites. Dose distributions are compared to those calculated with a ray-tracing algorithm. Maximum doses and dose-volume histograms for the target and selected critical structures are analyzed. Methods and MaterialsThe CyberKnife is used for stereotactic radiosurgery-radiotherapy of intracranial lesions (91) as well as stereotactic body radiotherapy for lesions in the spine (24), lung (58), and pelvis (36). The Multiplan system is an inverse treatment planning system which uses an effective path length (EPL) algorithm (sometimes referred to as ray-trace) for dose calculations. In addition, an MC algorithm became clinically available in late 2007. ResultsThe maximum doses calculated by the EPL to targets in the lung were uniformly larger than the doses calculated by MC by up to a factor of 1.32. In addition, large differences in target and critical organs' dose coverage were observed. In general, more beams traversing larger distances through low density lung are associated with larger differences. For other sites such as brain and pelvis targets the differences in maximum doses and tumor coverage were generally less than 5% between the 2 calculation methods. ConclusionsThe MC algorithm should be consistently used for treatment plans of lung lesions and lesions near large air cavities, but the faster EPL algorithm is adequate for treatment sites with less tissue heterogeneity.

Proton Treatment of Peripheral Lymph Nodes in Breast Cancer

The treatment of peripheral lymph nodes in locally advanced breast cancer patients often entails increased dose to organs at risk (OARs). The goal of this study is to evaluate the potential advantages of using proton therapy in treating peripheral lymph nodes vs. 3-dimensional conformal radiotherapy (3D-CRT) for left sided breast cancer patients. Treatment plan comparison was performed on planning computed tomography of 7 consecutive left-sided breast cancer patients. Patients were immobilized in the supine position using a Vac Loc device and Wing Board. All photon plans used a 4-field, mono-isocenter technique. Wide tangents were used to cover the internal mammary nodes. In the matched proton-photon technique, a proton field covering the internal mammary nodes, supraclavicular nodes and level II and III axillary nodes was matched to the opposing tangent photon fields. The proton field was planned using forward-planned, double-scattered 3-dimensional proton therapy techniques. The breast and/or chest wall were adequately treated in both modalities. There was no significant difference in the coverage of supraclavicular nodes, level I or III axillary nodes. Protons improved the coverage of level II lymph nodes, with a mean D95 of 50 Gy compared to D95 of 37 Gy with 3D-CRT (p = 0.01). Protons also improved the mean D95 of internal mammary nodes to 51 Gy vs. 42 Gy with 3D-CRT (p = 0.01). When protons were used to cover the peripheral lymph nodes, there was a significant decrease in the dose to the heart and ipsilateral lung for all isodose levels. In comparing protons vs. 3D-CRT, the mean left lung dose was 2 Gy vs. 13 Gy (p = 0.01); left lung V20 was 24% vs. 38% (p = 0.01), left lung V5 was 42% vs. 84% (p = 0.01), respectively. The mean heart V20 was 0% using proton vs. 8% with 3D-CRT (p = 0.01). The use of protons also lowered the cardiac V5 from 44% to 2% (p = 0.01). There was no significant difference in the mean maximum skin dose, which were 53 Gy in 3D-CRT and 56 Gy with the use of a matched proton field. Proton treatment of peripheral lymph nodes improved dose coverage of the peripheral lymph nodes and decreased the dose to the heart and ipsilateral lung. The dosimetric advantages may improve local control of breast cancer and reduce late effects from radiation treatment.

Do we need fractionation-corrected doses in sequential two-phase treatments? A quantification of dose differences between non-corrected and corrected combined non-uniform dose distributions in normal tissue

Background. For many tumour sites, external beam radiation therapy (EBRT) is delivered with a sequential two-phase treatment regime. Yet, there is a lack of consensus of how to add two different non-uniform dose distributions in order to evaluate the late radiation effect for normal tissue. The purpose of this novel investigation is to quantify the dose differences between non-corrected and fractionation-corrected combined non-uniform dose distributions. Material and methods. We used a model of an organ at risk (OAR) located in six different positions relative the treated volume giving 16 clinically representative two-phase treatment situations (46 Gy + 22 Gy). The linear-quadratic model was applied to correct for fractionation effects in each voxel before the doses were added. Dose differences were quantified using mean and maximum doses with corresponding fractionation-corrected doses as reference. Results. Non-corrected doses were higher than fractionation-corrected doses in all treatment situations (mean dose: p<0.001; maximum dose: p=0.003). With the OAR outside the treated volume, non-corrected doses were 3–6 Gy higher representing 10–50% of the reference dose (10–25 Gy); with the OAR included in the treated volume, 1–6 Gy higher representing 1–15% (30–60 Gy). Mean dose differences were generally larger than maximum dose differences. Conclusion. Substantial dose differences were present in all of the simulated treatment situations but more apparent when the OAR was located outside the treated volume in both phases. Our findings require verification in clinical cases but nevertheless indicate a need for fractionation-corrected doses in two-phase treatments both in the daily clinical routine as well as in the modelling of late radiation effects. Our data suggest that adjusting for fractionation effects would lead to lower tolerance doses than currently suggested, in particular for OARs with parallel tissue architecture.

Dosimetric Impact of Online Correction via Cone-Beam CT-Based Image Guidance for Stereotactic Lung Radiotherapy

To evaluate the dosimetric impact of online cone-beam computed tomography (CBCT) guided correction in lung stereotactic body radiation therapy (SBRT). Twenty planning and 162 CBCT images from 20 patients undergoing lung SBRT were analyzed. The precorrection CBCT (CBCT after patient setup, no couch correction) was registered to planning CT using soft tissue; couch shift was applied, with a second CBCT for verification (postcorrection CBCT). Targets and normal structures were delineated on CBCTs: gross tumor volume (GTV), clinical target volume (CTV), cord, esophagus, lung, proximal bronchial tree, and aorta. Dose distributions on all organs manifested on each CBCT were compared with those planned on the CT. Without CBCT guided target position correction, target dose reduced with respect to treatment plan. Mean and standard deviation of treatment dose discrepancy from the plan were -3.2% (4.9%), -2.1% (4.4%), -6.1% (10.7%), and -3.5% (7%) for GTV D(99%), GTV D(95%), CTV D(99%), and CTV D(95%), respectively. With CBCT correction, the results were -0.4% (2.6%), 0.1% (1.7%), -0.3% (4.2%), and 0.5% (3%). Mean and standard deviation of the difference in normal organ maximum dose were 2.2% (6.5%) before correction and 2.4% (5.9%) after correction for esophagus; 6.1% (14.1%) and 3.8% (8.1%) for cord; 3.1% (17.5%) and 6.2% (9.8%) for proximal bronchial tree; and 17.7% (19.5%) and 14.1% (17%) for aorta. Online CBCT guidance improves the accuracy of target dose delivery for lung SBRT. However, treatment dose to normal tissue can vary regardless of the correction. Normal tissues should be considered during target registration, according to target proximity.

SU‐GG‐T‐521: Analysis of Average Path Length and Inhomogeneity Correction Effect for Cranial SRS Cases

Purpose: This study is designed to analyze the properties of beam path length at different brain locations and the effect of Inhomogeneity orrection (IC) on isocenter dose of cranial Stereotactic Radiosurgery (SRS) treatment planning. Method: Fifty five clinical cranial SRS plans, with total 63 lesions were analyzed in this study. Lesions were grouped as frontal (16 lesions), parietal (10 lesions), temporal (10 lesions), occipital (10 lesions), cerebellum (10 lesions), and acoustic neuroma (7 lesions). All plans were planned on Brainlab iPlan RT Dose 3.0.2. Pencil beam convolution dose calculation algorithm was used both with and without IC with identical beam parameters. Nine to 11 beams were used for each lesion. Average beam path length, and differences in isocenter dose for each lesion with and without IC were analyzed. Result: The mean and standard deviation (SD) of average beam path length for frontal, parietal, temporal, occipital, cerebellum and acoustic neuroma lesions without IC were 47.54±16.03mm, 45.33±8.69mm, 64.15±12.33mm, 51.27±9.73mm, 78.70±10.89mm and 79.92±7.46mm, respectively. The corresponding results with IC were 56.08±14.86mm, 52.80±8.79mm, 71.85±12.27mm, 59.97±10.20mm, 86.57±11.26mm and 90.07±7.57mm, respectively. The differences in average beam path length between with and without IC were 8.54±1.62mm, 7.47±1.04mm, 7.70±1.66mm, 8.71±1.56mm, 7.87±1.39mm and 10.15±3.52mm respectively. The mean and SD of isocenter dose for frontal, parietal, temporal, occipital, cerebellum, acoustic neuroma and all lesions combined without IC were 3.83±0.72%, 3.35±0.61%, 3.47±0.88%, 3.99±0.76%, 3.69±0.69%, 4.83 ±1.77% and 3.81±0.96% higher then the isocenter dose with IC, respectively. The maximum dose difference was 7.05% for an acoustic neuroma case and minimum dose difference was 1.59% for a frontal lobe lesion. Conclusion: In conclusion, an isodose difference between dose calculation with and without IC of about 4% was found for cranial SRS case. The effect of planning cranial SRS with or without IC increases with the average path length differences.

MO‐FF‐A3‐05: Dosimetric Interplay Effects for Proton Uniform Scanning Lung Treatments

Purpose: To investigate the dosimetric impact of respiratory motion on proton lung treatments delivered by uniform scanning. Method: Uniform scanning is an aperture and compensator based technique in which the pristine energy layers are delivered sequentially (‘energy stacking’). Because of high scanning frequency (3Hz) compared to breathing motion, the study ignores lateral effects and focuses on interplay between energy stacking and tissue motion/deformation. This exercise is based on actual lung patient plans that, however, were delivered by simultaneously irradiating the whole SOBP. To simulate the breathing motion the full SOBP dose is recalculated on all 10 4DCT phases. To simulate the energy stacking, the per phase dose distributions are deconvolved into energy layers with appropriate weighting. Based on delivery parameters such as total irradiation time (doserate), time offset, and switching times and patient specific parameters like average breath rate, a series of dose distributions belonging to various breathing phases and energy layers are identified to constitute a treatment. For total dose accumulation deformable registration methods are used to deform all dose maps to a mid‐exhale phase. The interplay effect of breathing motion and energy stacking is deduced by comparison of the total dose distributions over multiple breathing cycles for both all‐energies and by energy layer deliveries. Results: No dramatic discrepancies on beam dose distributions produced by the two methods are observed. For a doserate of 2Gy/min and a respiratory rate of 20bpm the PTV coverage, maximum dose, and lung mean dose differences are within 2, 2.5, and 1.5% of prescribed dose (per beam). Doserate is the sole parameter that significantly influences the interplay effect which becomes insignificant for values below 0.5Gy/min. Conclusions: Based on the cases that we applied our uniform scanning evaluation method we conclude that lung treatment can safely be delivered by uniform scanning for low doserates.

SU‐GG‐J‐83: Interplay Study between Respiration Motion and Dynamic MLC Delivery for 4D EMRT Plan

Purpose Interplay between MLC movement and organ motion can affect treatment dose. The study investigates interplay effect of 4D inverse planning and delivery, and discusses possible ways to reduce it. Method and Materials: Three 4D phantom images were used in this study. Respiratory movement was in S‐I direction with amplitude of 1cm, 2cm and 3cm. 4D inverse planning was accomplished with the prescription dose to the isocenter. The expected (planed) dose is weighted sum over all phases of motion. The delivery dose, however, depends on the starting phase and lasting time of each segment. Starting phase was selected randomly. With the target motion extracted from 4DCT images, we calculated position, phase and dose rate at point of interest for any given time. With an appropriate sampling rate, the dose was accumulated to mimic the treatment delivery in one fraction. Other fractions was simulated the same way. Here we did 100 simulations for each case and compared difference on point dose and equivalent uniform dose (EUD). We also studied time evolution of the delivery dose and its sensitivity on dose rate. Results: The target point dose difference could be 6.3% (the maximum in one fraction). Tumor volume with point dose difference greater than +/−3% was less than 8%. EUD differs from the expected dose by less than 1% for 1cm and 2cm phantom in all simulations. For phantom with 3cm motion amplitude, the mean EUD reduced to 1% difference in 6 fractions. Reducing dose rate makes EUD variance and maximum point dose difference smaller. Conclusion: The result suggests a small interplay effect in our 4D IMRT plans. Limiting number of segments, increasing minimum MU per segment and decreasing dose rate could reduce interplay effect. Conflict of interest: The work is supported by Elekta.

SU-GG-T-617: Contura Balloon Rotations and Its Impact on Patient Dosimetry for APBI

Purpose: To determine rotations of Contura™ multilumen HDR balloons from axial CT scanimages and coronal or sagittal reconstructions and their impact in the dosimetry of Accelerated Partial Breast Irradiation (APBI) patients. Methods and Materials: A methodology for calculating device rotation angles based on matrix rotations was utilized in order to evaluate setup variations. This is especially important for patients with small balloon‐skin distance. Catheter 1 and 5 positions in the axial and coronal images were employed to determine the rotations. A phantom was built in order to assess the accuracy of the calculation and dosimetry. The calculated rotation angles were compared with measured ones from phantom scans and from patient data as determined using the treatment planning system (BrachyVision). Dwell position and time data from patient plans were used to evaluate dosimetric changes as a function of balloon rotation. Relative dose differences and distance‐to‐agreement between the original and rotated dose distributions were performed using DoseLab as well as using the treatment planning system. Results: The comparison between calculated and measured rotation angles show good agreement, with an average deviation of 1.2 degrees and 3.2 degrees of absolute deviation on average for the cases studied. Dosimetric differences depend on the degree of asymmetry of the dose distribution which is related to the distance between the balloon and dose limiting structures such as the skin and the chest wall. The maximum skin dose is sensitive to the balloon rotation with maximum point dose differences of up to 64% for 180 degrees rotations. PTV coverage is less sensitive to rotations. Conclusions: A methodology to evaluate and track Contura balloon rotations was devised and its accuracy evaluated. Patient dosimetry can be seriously affected if large rotations are allowed.

SU-GG-T-39: Grid-Based Boltzmann Solver (GBBS) vs TG-43 for Ir-192 HDR Intracavitary Brachytherapy: A Retrospective Dosimetric Study

Purpose: To determine dosimetric differences between a grid‐based Boltzmann solver (GBBS) and TG‐43 for a cohort of HDR cervical cancer cases treated using the CT/MR Fletcher‐Suit‐Delclos (FSD) type applicator. Method and Materials:Dose distributions from a cohort (n=10) of cervical cancer patients treated with the VS2000 192Ir HDR source were analyzed retrospectively using the BrachyVision‐Acuros v8.8 TPS. A single physician contoured the rectum and bladder. 6 Gy was prescribed to point A. Doses were recorded for ICRU rectal and bladder points as well as vaginal surfaces. Rectum, bladder, and entire dose grid DVHs were visually assessed and compared at relevant points. Dose grids were exported to IDL to calculate mean voxel percent differences. Average, minimum, and maximum percent dose differences are reported for 1) D2cc and D90 for rectum and bladder and for 2) dose grid mean voxel percent difference. Results: Visual inspection of DVH showed lateral shifting of GBBS curves to lower doses. The shifting was noticeable for dose levels within 4 Gy. Percent difference results from the DVH output: D2cc‐Rectum: −2.1%(mean) [−3.0%(min), −0.5%(max)]; D90‐Rectum: −2.3% [−7.5%, 4.6%]; D2cc‐Bladder: −0.7% [−1.8%, 0.7%]; D90‐Bladder: −4.4% [−7.0%, −2.2%]. All ICRU points studied had dose differences within −3.3% relative to TG‐43. The average of the mean voxel percent difference for the entire dose grid was −1.3% and ranged from −3.7% to 1.2%. Visualization of individual percent difference maps overlaid on CT revealed lower GBBS doses behind contrast filled balloon in the bladder. Conclusions: For this cohort of patients that underwent intracavitary HDR brachytherapy with the CT/MR FSD applicator, the GBBS calculated dose to clinical reference points and hot spots (D2cc) was within 3.3% of TG‐43. Over the entire dose grid, however, lateral shifting of DVHs to lower doses was observed. This may be partially due to contrast filled balloon in bladder.

Exploring Differences in Drug Doses Between Japan and Western Countries

Japan is unique among Asian countries in that it requires inclusion of substantial domestic clinical trial data in new drug application data packages. Some question the need for this and call for globalization of approved doses. The current study examines international differences in maximum daily dose of drugs approved in Japan between 2001 and 2007, and for all cardiovascular system (CVS) and central nervous system (CNS) drugs marketed in Japan. For 32% of the drugs approved in Japan between 2001 and 2007, the maximum recommended dose in the United States was > or =2 times higher than the maximum dose approved in Japan. Dose differences were rare for antitumor and antiviral drugs and also for priority-review and orphan drugs. Of all the price-listed CVS drugs currently available in Japan, 65% had maximum doses that were > or =2 times higher in the Netherlands than in Japan; similarly, 57% of the drugs had maximum doses that were > or =2 times higher in the United States than in Japan. For CNS drugs, these figures were 32% (the Netherlands) and 29% (United States). These results underscore the necessity to carry out quantitative analyses to determine the causes of these differences.

A new metric for assessing IMRT modulation complexity and plan deliverability

To evaluate the utility of a new complexity metric, the modulation complexity score (MCS), in the treatment planning and quality assurance processes and to evaluate the relationship of the metric with deliverability. A multisite (breast, rectum, prostate, prostate bed, lung, and head and neck) and site-specific (lung) dosimetric evaluation has been completed. The MCS was calculated for each beam and the overall treatment plan. A 2D diode array (MapCHECK, Sun Nuclear, Melbourne, FL) was used to acquire measurements for each beam. The measured and planned dose (PINNACLE3, Phillips, Madison, WI) was evaluated using different percent differences and distance to agreement (DTA) criteria (3%/ 3 mm and 2%/ 1 mm) and the relationship between the dosimetric results and complexity (as measured by the MCS or simple beam parameters) assessed. For the multisite analysis (243 plans total), the mean MCS scores for each treatment site were breast (0.92), rectum (0.858), prostate (0.837), prostate bed (0.652), lung (0.631), and head and neck (0.356). The MCS allowed for compilation of treatment site-specific statistics, which is useful for comparing different techniques, as well as for comparison of individual treatment plans with the typical complexity levels. For the six plans selected for dosimetry, the average diode percent pass rate was 98.7% (minimum of 96%) for 3%/3 mm evaluation criteria. The average difference in absolute dose measurement between the planned and measured dose was 1.7 cGy. The detailed lung analysis also showed excellent agreement between the measured and planned dose, as all beams had a diode percentage pass rate for 3%/3 mm criteria of greater than 95.9%, with an average pass rate of 99.0%. The average absolute maximum dose difference for the lung plans was 0.7 cGy. There was no direct correlation between the MCS and simple beam parameters which could be used as a surrogate for complexity level (i.e., number of segments or MU). An evaluation criterion of 2%/ 1 mm reliably allowed for the identification of beams that are dosimetrically robust. In this study we defined a robust beam or plan as one that maintained a diode percentage pass rate greater than 90% at 2%/ 1 mm, indicating delivery that was deemed accurate when compared to the planned dose, even under stricter evaluation criterion. MCS and MU threshold criteria were determined by defining a required specificity of 1.0. A MCS threshold of 0.8 allowed for identification of robust deliverability with a sensitivity of 0.36. In contrast, MU had a lower sensitivity of 0.23 for a threshold of 50 MU. The MCS allows for a quantitative assessment of plan complexity, on a fixed scale, that can be applied to all treatment sites and can provide more information related to dose delivery than simple beam parameters. This could prove useful throughout the entire treatment planning and QA process.