Comparison of intensity-modulated radiation therapy (IMRT), 3D conformal proton therapy and intensity-modulated proton therapy (IMPT) for the treatment of metastatic brain cancer.

The purpose of this study has been to compare photon intensity modulated radiation therapy (IMRT) against both conformal and intensity modulated proton therapy (IMPT) plans for metastatic brain cancer. Ten IMRT patients with brain cancer were chosen retrospectively, with prescription doses in the range of 20 to 40 Gy, delivered in 3 to 5 fractions using Varian TrueBeam STx machine. Three proton plans with proton double scattering, single collimation static-IMPT, and energy layer by layer collimation dynamic-IMPT were then generated for the same patients using the Mevion S250 system for conformal plans and the S250i system for IMPT plans. Each plan had respective treatment planning systems that include Brainlab iPlan for IMRT, Varian Eclipse for proton double scattering, and RaySearch Raystation for IMPT. Dosimetric and radiobiologic comparisons were made through dose-volume histogram (DVH) analysis of the target and the organs at risk (OAR); and with parameters of equivalent uniform dose (EUD), tumor control probability (TCP), and normal tissue complication probability (NTCP), respectively. A set of variables α/ß ratio, survival fraction, and clonogenic cell density were selected and varied to observe their effect on the abovementioned parameters. Doses were observed to be more homogeneous for patients with brain malignancies with photon IMRT treatments, while dose conformity is improved with proton PBS treatments. Normal tissue is, on average, spared more through both proton treatment options. The minimum doses, closely approximated by dose to 98% of the target volume, are similar across treatment modalities with slight variations.

Dosimetric Effect of Biozorb Markers for Accelerated Partial Breast Irradiation in Proton Therapy.

PURPOSE: To investigate dosimetric implications of biodegradable Biozorb (BZ) markers for proton accelerated partial breast irradiation (APBI) plans. MATERIALS AND METHODS: Six different BZs were placed within in-house breast phantoms to acquire computed tomography (CT) images. A contour correction method with proper mass density overriding for BZ titanium clip and surrounding tissue was applied to minimize inaccuracies found in the CT images in the RayStation planning system. Each breast phantom was irradiated by a monoenergetic proton beam (103.23 MeV and 8×8 cm2) using a pencil-beam scanning proton therapy system. For a range perturbation study, doses were measured at 5 depths below the breast phantoms by using an ionization chamber and compared to the RayStation calculations with 3 scenarios for the clip density: the density correction method (S1: 1.6 g/cm3), raw CT (S2), and titanium density (S3: 4.54 g/cm3). For the local dose perturbation study, the radiographic EDR2 film was placed at 0 and 2 cm below the phantoms and compared to the RayStation calculations. Clinical effects of the perturbations were retrospectively examined with 10 APBI plans for the 3 scenarios (approved by our institutional review board). RESULTS: In the range perturbation study, the S1 simulation showed a good agreement with the chamber measurements, while excess pullbacks of 1∼2 mm were found in the S2 and S3 simulations. The film study showed local dose shadowing and perturbation by the clips that RayStation could not predict. In the plan study, no significant differences in the plan quality were found among the 3 scenarios. However, substantial range pullbacks were observed for S3. CONCLUSION: The density correction method could minimize the dose and range difference between measurement and RayStation prediction. It should be avoided to simply override the known physical density of the BZ clips for treatment planning owing to overestimation of the range pullback.

Characterization of penumbra sharpening and scattering by adaptive aperture for a compact pencil beam scanning proton therapy system.

PURPOSE: To quantitatively access penumbra sharpening and scattering by adaptive aperture (AA) under various beam conditions and clinical cases for a Mevion S250i compact pencil beam scanning proton therapy system. METHODS: First, in-air measurements were performed using a scintillation detector for single spot profile and lateral penumbra for five square field sizes (3 × 3 to 18 × 18 cm2 ), three energies (33.04, 147.36, and 227.16 MeV), and three snout positions (5, 15, and 33.6 cm) with Open and AA field. Second, treatment plans were generated in RayStation treatment planning system (TPS) for various combination of target size (3- and 10-cm cube), target depth (5, 10, and 15 cm) and air gap (5-20 cm) for both Open and AA field. These plans were delivered to EDR2 films in the solid water and penumbra reduction by AA was quantified. Third, the effect of the AA scattered protons on the surface dose was studied at 5 mm depth by EDR2 film and the RayStation TPS computation. Finally, dosimetric advantage of AA over Open field was studied for five brain and five prostate cases using the TPS simulation. RESULTS: The spot size changed dramatically from 3.8 mm at proton beam energy of 227.15 MeV to 29.4 mm at energy 33.04 MeV. In-air measurements showed that AA substantially reduced the lateral penumbra by 30% to 60%. The EDR2 film measurements in solid water presented the maximum penumbra reduction of 10 to 14 mm depending on the target size. The maximum increase of 25% in field edge dose at 5 mm depth as compared to central axis was observed. The substantial penumbra reduction by AA produced less dose to critical structures for all the prostate and brain cases. CONCLUSIONS: Adaptive aperture sharpens the penumbra by factor of two to three depending upon the beam condition. The absolute penumbra reduction with AA was more noticeable for shallower target, smaller target, and larger air gap. The AA-scattered protons contributed to increase in surface dose. Clinically, AA reduced the doses to critical structures.

Performance evaluation of adaptive aperture's static and dynamic collimation in a compact pencil beam scanning proton therapy system: A dosimetric comparison study for multiple disease sites.

A compact pencil beam scanning (PBS) proton therapy system, Mevion S250i with Hyperscan, is equipped with adaptive aperture (AA) to collimate the beam with 2 different techniques: Static aperture (SA) and dynamic aperture (DA). SA (single aperture) collimates the outermost contour of the target and DA (multi-layer aperture) collimates each energy layer of the proton beam. This study evaluates dosimetric performance of SA and DA for different disease sites. This study includes 5 disease sites (brain, head and neck (HN), partial breast, lung, and prostate), and 8 patients for each. A total of 80 patient treatment plans (5 sites × 8 patients per site × 2 collimation techniques) were created using 2 to 4 proton beams. Both SA and DA plans were made using the same plan and optimization parameters calculated by a Monte Carlo dose algorithm. Multi-field optimization (MFO) was used for HN treatment plans, whereas treatment plans for the other sites were made with single-field optimization (SFO). All plans were robustly optimized with 3 mm (brain and HN) or 5 mm (breast, lung, and prostate) position uncertainty along with 3.5% range uncertainty. Treatment plans were normalized such that 99% of the clinical target volume (CTV) received 100% of the prescribed dose. Dose volume histogram (DVH) parameters were evaluated for CTV and organs at risk (OARs). The CTV was also evaluated for dose homogeneity, dose conformity, and dose gradient. In general, the DA plan made CTV hotter, while it saved OARs better. DA produced better conformity with sharper dose falloff around CTV, while SA generated better homogenous target coverage. DA decreased Dmax to brainstem (1.2% = [(SA-DA)/DA × 100%]) for brain, Dmax to the spinal cord (137.3%) for HN, D1% of the ipsilateral lung (50.5%) for breast, and Dmax to the spinal cord (74.0%) for lung. The dose reduction in bladder and rectum for prostate plans with DA was less than 2.5%. The DA plans reduced the dose to OARs for all disease sites but escalated the target maximum dose for the same target coverage than the SA plans. The OAR saving and dose escalation depended on CTV size, proximity of the OARs to CTV, and the plan complexity.

Clinical commissioning of intensity-modulated proton therapy systems: Report of AAPM Task Group 185.

Proton therapy is an expanding radiotherapy modality in the United States and worldwide. With the number of proton therapy centers treating patients increasing, so does the need for consistent, high-quality clinical commissioning practices. Clinical commissioning encompasses the entire proton therapy system's multiple components, including the treatment delivery system, the patient positioning system, and the image-guided radiotherapy components. Also included in the commissioning process are the x-ray computed tomography scanner calibration for proton stopping power, the radiotherapy treatment planning system, and corresponding portions of the treatment management system. This commissioning report focuses exclusively on intensity-modulated scanning systems, presenting details of how to perform the commissioning of the proton therapy and ancillary systems, including the required proton beam measurements, treatment planning system dose modeling, and the equipment needed.

Prediction of the output factor using machine and deep learning approach in uniform scanning proton therapy.

PURPOSE: The purpose of this work is to develop machine and deep learning-based models to predict output and MU based on measured patient quality assurance (QA) data in uniform scanning proton therapy (USPT). METHODS: This study involves 4,231 patient QA measurements conducted over the last 6 years. In the current approach, output and MU are predicted by an empirical model (EM) based on patient treatment plan parameters. In this study, two MATLAB-based machine and deep learning algorithms - Gaussian process regression (GPR) and shallow neural network (SNN) - were developed. The four parameters from patient QA (range, modulation, field size, and measured output factor) were used to train these algorithms. The data were randomized with a training set containing 90% and a testing set containing remaining 10% of the data. The model performance during training was accessed using root mean square error (RMSE) and R-squared values. The trained model was used to predict output based on the three input parameters: range, modulation, and field size. The percent difference was calculated between the predicted and measured output factors. The number of data sets required to make prediction accuracy of GPR and SNN models' invariable was also evaluated. RESULTS: The prediction accuracy of machine and deep learning algorithms is higher than the EM. The output predictions with [GPR, SNN, and EM] within ± 2% and ± 3% difference were [97.16%, 97.64%, and 92.95%] and [99.76%, 99.29%, and 97.18%], respectively. The GPR model outperformed the SNN with a smaller number of training data sets. CONCLUSION: The GPR and SNN models outperformed the EM in terms of prediction accuracy. Machine and deep learning algorithms predicted the output factor and MU for USPT with higher predictive accuracy than EM. In our clinic, these models have been adopted as a secondary check of MU or output factors.

Experimental Assessment of Proton Dose Calculation Accuracy in Small-Field Delivery Using a Mevion S250 Proton Therapy System.

PURPOSE: Dose calculation accuracy of the Varian Eclipse treatment planning system (TPS) is empirically assessed for small-aperture fields using a Mevion S250 double scattering proton therapy system. MATERIALS AND METHODS: Five spherical pseudotumors were modeled in a RANDO head phantom. Plans were generated for the targets with apertures of 1, 2, 3, 4, or 5 cm diameter using one, two, and three beams. Depth-dose curves and lateral profiles of the beams were taken with the planned blocks and compared to Eclipse calculations. Dose distributions measured with EBT3 films in the phantom were also compared to Eclipse calculations. Film quenching effect was simulated and considered. RESULTS: Depth-dose scans in water showed a range pullback (up to 2.0 mm), modulation widening (up to 9.5 mm), and dose escalation in proximal end and sub-peak region (up to 15.5%) when compared to the Eclipse calculations for small fields. Measured full width at half maximum and penumbrae for lateral profiles differed <1.0 mm from calculations for most comparisons. In the phantom study, Eclipse TPS underestimated sub-peak dose. Gamma passing rates improved with each beam added to the plans. Greater range pullback and modulation degradation versus water scans were observed due to film quenching, which became more noticeable as target size increased. CONCLUSIONS: Eclipse TPS generates acceptable target coverage for small targets with carefully arranged multiple beams despite relatively large dose discrepancy for each beam. Surface doses higher than Eclipse calculations can be mitigated with multiple beams. When using EBT3 film, the quenching effect should be considered.

Comparability of three output prediction models for a compact passively double-scattered proton therapy system.

The purpose of this study was to investigate comparability of three output prediction models for a compact double-scattered proton therapy system. Two published output prediction models are commissioned for our Mevion S250 proton therapy system. Model A is a correction-based model (Sahoo et al., Med Phys, 2008;35(11):5088-5097) and model B is an analytical model which employs a function of r = (R'-M')/M' (Kooy et al., Phys Med Biol, 2005;50:5487-5456) where R' is defined as depth of distal 100% dose with straggling and M' is the width between distal 100% dose and proximal 100% dose with straggling instead of the theoretical definition due to more accurate output prediction. The r is converted to ((R-0.31)-0.81 × M)/(0.81 × M) with the vendor definition of R (distal 90% dose) and M (distal 90% dose-to-proximal 95% dose), where R' = R-0.31 (g cm-2 ) and M' = 0.81 × M (g cm-2 ). In addition, a quartic polynomial fit model (model C) mathematically converted from model B is studied. The outputs of 272 sets of R and M covering the 24 double scattering options are measured. Each model's predicted output is compared to the measured output. For the total dataset, the percent difference between predicted (P) and measured (M) outputs ((P-M)/M × 100%) were within ±3% using the three different models. The average differences (±standard deviation) were -0.13 ± 0.94%, -0.13 ± 1.20%, and -0.22 ± 1.11% for models A, B, and C, respectively. The p-values of the t-test were 0.912 (model A vs. B), 0.061 (model A vs. C), and 0.136 (model B vs. C). For all the options, all three models have clinically acceptable predictions. The differences between models A, B, and C are statistically insignificant; however, model A generally has the potential to more accurately predict the output if a larger dataset for commissioning is used. It is concluded that the models can be comparably used for the compact proton therapy system.

Implementation of output prediction models for a passively double-scattered proton therapy system.

PURPOSE: Two output (cGy/MU) prediction models (one existing and one newly developed) for a passively double-scattered proton therapy system are implemented and investigated for clinical use. Variations of each model are tested for accuracy in order to determine the most viable prediction model. METHODS: The first output prediction model [model (1)] is a semianalytical model proposed by Kooy et al. [Phys. Med. Biol. 50, 5847-5856 (2005)], which employs three main factors. The first factor (basic output prediction) uses a unique combined parameter [r = (R - M)/M] of range (R) and modulation [M; spread-out Bragg peak (SOBP) width] along with option specific fitting parameters. The second factor takes into account minor source shifts using a linear fit due to varying beamline configurations for different options. The final factor accounts for a condition where the point of measurement is not at the isocenter or away from the middle of the SOBP based on an inverse-square correction. The second model [model (2)] is a novel quartic polynomial fit of the basic output prediction whose idea was inspired by the first model. Different variations in the definition of R and M at distal (D) and proximal (P) ends resulted in the exploration of three variations of r for both models: r1 = (RD90 - MD90-P95)/MD90-P95, r2 = [(RD90 + ΔR1) - m × (MD90-P95 + ΔR1)]/[m × (MD90-P95 + ΔR1)], where ΔR1 is an offset between RD80 and RD90 and m is a ratio between MD90-P95 and theoretical MD100-P100', and r3 = [(RD90 - 0.305) - 0.801 × MD90-P95]/(0.801 × MD90-P95), where 0.305 (ΔR2) is an offset between RD90 and RD100 and 0.801 is a ratio between MD90-P95 and measured MD100-P100. Output measurements for 177 sets of R and M from all 24 options are compared to outputs predicted by both the models of three variations of r. RESULTS: The mean differences between measurements and predictions ([predicted - measured]/measured × 100%) were -0.41% ± 1.78% (r1), 0.03% ± 1.53% (r2), and 0.05% ± 1.20% (r3) for model (1), and 0.27% ± 1.36% (r1), 0.71% ± 1.51% (r2), and -0.05% ± 1.20% (r3) for model (2). For a passing prediction rate with a difference threshold of ±3%, model (1) showed slightly worse results than model (2) using r1 (91.5% vs 94.4%). In general, small (M < 4 g/cm2) and close-to-full modulations produced larger discrepancies. However, 100% output predictions using r3 were confined within ±3% of measurements for both models and the difference between the models was not substantial (mean difference: 0.05% vs -0.05%). CONCLUSIONS: The first existing model has proven to be a successful predictor of output for our compact double-scattering proton therapy system. The new model performed comparably to the first model and showed better performance in some options due to a great degree of flexibility of a polynomial fit. Both models performed well using r3. Either model with r3 thus can serve well as an output prediction calculator.

Quantitative evaluation of patient setup uncertainty of stereotactic radiotherapy with the frameless 6D ExacTrac system using statistical modeling.

The purpose of this study is to evaluate patient setup accuracy and quantify indi-vidual and cumulative positioning uncertainties associated with different hardware and software components of the stereotactic radiotherapy (SRS/SRT) with the frameless 6D ExacTrac system. A statistical model is used to evaluate positioning uncertainties of the different components of SRS/SRT treatment with the Brainlab 6D ExacTrac system using the positioning shifts of 35 patients having cranial lesions. All these patients are immobilized with rigid head-and-neck masks, simu-lated with Brainlab localizer and planned with iPlan treatment planning system. Stereoscopic X-ray images (XC) are acquired and registered to corresponding digitally reconstructed radiographs using bony-anatomy matching to calculate 6D translational and rotational shifts. When the shifts are within tolerance (0.7 mm and 1°), treatment is initiated. Otherwise corrections are applied and additional X-rays (XV) are acquired to verify that patient position is within tolerance. The uncertain-ties from the mask, localizer, IR -frame, X-ray imaging, MV, and kV isocentricity are quantified individually. Mask uncertainty (translational: lateral, longitudinal, vertical; rotational: pitch, roll, yaw) is the largest and varies with patients in the range (-2.07-3.71 mm, -5.82-5.62 mm, -5.84-3.61 mm; -2.10-2.40°, -2.23-2.60°, and -2.7-3.00°) obtained from mean of XC shifts for each patient. Setup uncer-tainty in IR positioning (0.88, 2.12, 1.40 mm, and 0.64°, 0.83°, 0.96°) is extracted from standard deviation of XC. Systematic uncertainties of the frame (0.18, 0.25, -1.27mm, -0.32°, 0.18°, and 0.47°) and localizer (-0.03, -0.01, 0.03mm, and -0.03°, 0.00°, -0.01°) are extracted from means of all XV setups and mean of all XC distributions, respectively. Uncertainties in isocentricity of the MV radiotherapy machine are (0.27, 0.24, 0.34 mm) and kV imager (0.15, -0.4, 0.21 mm). A statisti-cal model is developed to evaluate the individual and cumulative systematic and random positioning uncertainties induced by the different hardware and software components of the 6D ExacTrac system. The uncertainties from the mask, local-izer, IR frame, X-ray imaging, couch, MV linac, and kV imager isocentricity are quantified using statistical modeling.

Dosimetric effects of positioning shifts using 6D-frameless stereotactic Brainlab system in hypofractionated intracranial radiotherapy.

Dosimetric consequences of positional shifts were studied using frameless Brainlab ExacTrac X-ray system for hypofractionated (3 or 5 fractions) intracranial stereo-tactic radiotherapy (SRT). SRT treatments of 17 patients with metastatic intracranial tumors using the stereotactic system were retrospectively investigated. The treatments were simulated in a treatment planning system by modifying planning parameters with a matrix conversion technique based on positional shifts for initial infrared (IR)-based setup (XC: X-ray correction) and post-correction (XV: X-ray verification). The simulation was implemented with (a) 3D translational shifts only and (b) 6D translational and rotational shifts for dosimetric effects of angular correction. Mean translations and rotations (± 1 SD) of 77 fractions based on the initial IR setup (XC) were 0.51 ± 0.86 mm (lateral), 0.30 ± 1.55 mm (longitudinal), and -1.63 ± 1.00 mm (vertical); -0.53° ± 0.56° (pitch), 0.42° ± 0.60° (roll), and 0.44°± 0.90° (yaw), respectively. These were -0.07 ± 0.24 mm, -0.07 ± 0.25 mm, 0.06± 0.21 mm, 0.04° ± 0.23°, 0.00° ± 0.30°, and -0.02° ± 0.22°, respectively, for the postcorrection (XV). Substantial degradation of the treatment plans was observed in D95 of PTV (2.6% ± 3.3%; simulated treatment versus treatment planning), Dmin of PTV (13.4% ± 11.6%), and Dmin of CTV (2.8% ± 3.8%, with the maximum error of 10.0%) from XC, while dosimetrically negligible changes (< 0.1%) were detected for both CTV and PTV from XV simulation. 3D angular correction significantly improved CTV dose coverage when the total angular shifts (|pitch| + |roll| + |yaw|) were greater than 2°. With the 6D stereoscopic X-ray verification imaging and frameless immobilization, submillimeter and subdegree accuracy is achieved with negligible dosimetric deviations. 3D angular correction is required when the angular deviation is substantial. A CTV-to-PTV safety margin of 2 mm is large enough to prevent deterioration of CTV coverage.

Interplay effect of angular dependence and calibration field size of MapCHECK 2 on RapidArc quality assurance.

The purpose of this study is to investigate an effect of angular dependence and calibration field size of MapCHECK 2 on RapidArc QA for 6, 8, 10, and 15 MV. The angular dependence was investigated by comparing MapCHECK 2 measurements in MapPHAN-MC2 to the corresponding Eclipse calculations every 10° using 10× 10 cm2 and 3 × 3 cm2 fields. Fourteen patients were selected to make RapidArc plans using the four energies, and verification plans were delivered to two phantom setups: MapCHECK 2/MapPHAN phantom (MapPHAN QA) and MapCHECK 2 on an isocentric mounting fixture (IMF QA). Migration of MapCHECK 2 on IMF was simulated by splitting arcs every 10° and displacing an isocenter of each partial arc in the Eclipse system (IMFACTUAL QA). To investigate the effect of calibration field size, MapCHECK 2 was calibrated by two field sizes (10 × 10 cm2 and 3 × 3 cm2) and applied to all QA measurements. The γ test was implemented using criteria of 1%/1 mm, 2%/2 mm, and 3%/3 mm. A mean dose of all compared points for each plan was compared with respect to a mean effective field size of the RapidArc plan. The angular dependence was considerably high at gantry angles of 90° ± 10° and 270° ± 10° (for 10 × 10/3 × 3 cm2 at 90°, 30.6% ± 6.6%/33.4%± 5.8% (6 MV), 17.3% ± 5.3%/15.0% ± 6.8% (8 MV), 8.9%± 2.9%/7.8% ± 3.2% (10 MV), and 2.2% ± 2.3%/-1.3% ± 2.6% (15 MV)). For 6 MV, the angular dependence significantly deteriorated the γ passing rate for plans of large field size in MapPHAN QA (< 90% using 3%/3 mm); however, these plans passed the γ test in IMFACTUAL QA (> 95%). The different calibration field sizes did not make any significant dose difference for both MapPHAN QA and IMFACTUAL QA. For 8, 10, and 15 MV, the angular dependence does not make any clinically meaningful impact on MapPHAN QA. Both MapPHAN QA and IMFACTUAL QA presented clinically acceptable γ passing rates using 3%/3 mm. MapPHAN QA showed better passing rates than IMFACTUAL QA for the tighter criteria. The 10 × 10 cm2 calibration showed better agreement for plans of small effective field size (< 5 × 5 cm2) in MapPHAN QA. There was no statistical difference between IMF QA and IMFACTUAL QA. In conclusion, MapPHAN QA is not recommended for plans of large field size, especially for 6 MV, and MapCHECK2 should be calibrated using a field size similar to a mean effective field size of a RapidArc plan for better agreement for IMF QA.

Dependency of planned dose perturbation (PDP) on the spatial resolution of MapCHECK 2 detectors.

The purpose of this study is to determine the dependency of the planned dose perturbation (PDP) algorithm (used in Sun Nuclear 3DVH software) on spatial resolution of the MapCHECK 2 detectors. In this study, ten brain (small target), ten brain (large target), ten prostate, and ten head-and-neck (H&N) cases were retrospectively selected for QA measurement. IMRT validation plans were delivered using the field-by-field technique with the MapCHECK 2 device. The measurements were performed using standard detector density (standard resolution; SR) and a doubled detector density (high resolution; HR) by merging regular with shifted measurements. SR and HR measurements were fed into the 3DVH software and ROI (region of interest), planning target volume (PTV), and organ at risk (OAR)) dose statistics (D95, Dmean, and Dmax) were determined for each. Differences of the dose statistics normalized to prescription dose for ROIs between original planning and PDP-perturbed planning were calculated for SR (ΔDSR) and HR (ΔDHR), and difference between ΔDSR and ΔDHR (ΔDSR-HR = ΔDSR - ΔDHR) was also calculated. In addition, 2D and 3D γ passing rates (GPRs) were determined for both resolutions, and a correlation between GPRs and ΔDSR or ΔDHR for PTV dose metrics was determined. No considerably high mean differences between ΔDSR and ΔDHR were found for almost all ROIs and plans (< 2%); however, |ΔDSR|, |ΔDHR|, and |ΔDSR-HR| for PTV were found to significantly increase as the PTV size decreased (e.g., PTV size < 5 cc). And statistically significant differences between SR and HR were observed for OARs proximal to targets in large brain target and H&N cases. As plan modulation represented by fractional MU/prescription dose (MU/cGy) became more complex, the 2D/3D GPRs tended to decrease; however, the modulation complexity did not make any noticeable distinctions in the DVH statistics of PTV between SR and HR, excluding the small brain cases whose PTVs were extremely small (PTV = 11.0 ± 10.1 cc). Moderate to strong negative correlations (-1 < r < -0.3) between GPRs and PTV dose metrics indicated that small clinical errors for PTV occur at the higher GPRs. In conclusion, doubling the detector density of the MapCHECK 2 device is recommended for small targets (i.e., PTV < 5 cc) and multiple targets with complex geometry with minimum setup error in the DVH-based plan evaluation.

A comprehensive comparison study of three different planar IMRT QA techniques using MapCHECK 2.

The purpose of this study is to determine comparability of three different planar IMRT QA techniques: patient gantry angle composite (PGAC), single gantry angle composite (SGAC), and field by field (FBF), using MapCHECK 2 device and the γ test as performance metrics; and to assess the dependency of these techniques on intensity modulation, couch attenuation, and detector position (angular dependency). Ten highly modulated head and neck (H&N) and ten moderately modulated prostate IMRT validation plans were delivered using different techniques and were intercompared using the Student's t-test. The IMRT QA measurements were evaluated by percentage of points passing the γ test for three different criteria: 1% (dose difference)/1 mm (distance to agreement (DTA)) (C1), 2%/2 mm (C2), and 3%/3 mm (C3). To investigate dependency of the IMRT validation on treatment couch, ionization chamber measurements, as well as the conventional MapCHECK 2 QAs, were performed with PGAC and PGAC-WOC (without couch; using an extended tennis racket-type insert with negligible attenuation assumed). To determine angular dependency of the MapCHECK 2, patient gantry field-by-field (PG-FBF) technique was delivered and evaluated separately for each field. The differences of γ passing rates between SGAC and FBF were statistically insignificant, while these were statistically significant when compared to PGAC. SGAC and FBF techniques showed statistically insignificant differences between different levels of intensity modulation (H&N vs. Prostate) at C2 and C3 criteria, while PGAC could not for any criteria. The treatment couch has a significant impact on γ passing rates (PGAC vs. PGAC-WOC), but an ionization chamber-based IMRT validations showed clinically insignificant dose errors (< 2%) in all cases. This study showed that the MapCHECK 2 device has large angular dependency, especially at gantry angles of 90° and 270°, which dramatically affected the γ passing rates of PGAC. With proper consideration of couch attenuation and beam arrangement, the MapCHECK 2 will produce clinically comparable QA results using the three different planar IMRT QA techniques.

Dosimetric characterization of whole brain radiotherapy of pediatric patients using modulated proton beams.

This study was designed to investigate dosimetric variations between proton plans with (PPW) and without (PPWO), a compensator for whole brain radiotherapy (WBRT). The retrospective study on PPW and PPWO in Eclipse and XiO systems and photon plans (XP) using controlled segments in Pinnacle system was performed on nine pediatric patients for craniospinal irradiations. DVHs and derived metrics, such as the homogeneity index (HI), the doses to 2% (D(2%)) and 5% (D(5%)) volumes, and mean dose (D(mean)) of the whole brain (i.e., PTV), and the organs at risk (OARs) such as lens and skull, were obtained. The PPW plans from both Eclipse and XiO systems uncovered the following advantages: (1) encompassing a cribriform plate area with the 100% isodose line was better than either PPWO or XP, according to calculated two-dimensional distributions of one patient; (2) the mean value of D(5%) for lens was reduced to 23.6% of D(P) from 54.1% for PPWO or 41.6% for XP; and (3) the mean value of D(mean) for skull was reduced to 94.8% of D(P) from either 98.4% for PPWO or 98.3% for XP. However, the PPW plans also exposed several disadvantages including: (1) the HI of PTV increased to 7.7 from 4.7 for PPWO or 3.7 for XP; (2) D(2%) to PTV increased to 108.8% of D(P) from 104.8% for PPWO or 105.1% for XP; and (3) D(5%) to the skull increased to 104.9% of D(P) from 101.6% for PPWO or 103.4% of for XP. One-half of the observed variations were caused by different penumbra on lateral profiles and distal fall-off depth doses of protons in Eclipse and XiO. Because the utilization on the sharp proton distal fall-off was limited for WBRT, the difference between PPW and PPWO or XP indicated no distinguishable improvement by using a compensator in proton plans.

Application of a novel dose-uncertainty model for dose-uncertainty analysis in prostate intensity-modulated radiotherapy.

PURPOSE: To analyze dose uncertainty using a previously published dose-uncertainty model, and to assess potential dosimetric risks existing in prostate intensity-modulated radiotherapy (IMRT). METHODS AND MATERIALS: The dose-uncertainty model provides a three-dimensional (3D) dose-uncertainty distribution in a given confidence level. For 8 retrospectively selected patients, dose-uncertainty maps were constructed using the dose-uncertainty model at the 95% CL. In addition to uncertainties inherent to the radiation treatment planning system, four scenarios of spatial errors were considered: machine only (S1), S1 + intrafraction, S1 + interfraction, and S1 + both intrafraction and interfraction errors. To evaluate the potential risks of the IMRT plans, three dose-uncertainty-based plan evaluation tools were introduced: confidence-weighted dose-volume histogram, confidence-weighted dose distribution, and dose-uncertainty-volume histogram. RESULTS: Dose uncertainty caused by interfraction setup error was more significant than that of intrafraction motion error. The maximum dose uncertainty (95% confidence) of the clinical target volume (CTV) was smaller than 5% of the prescribed dose in all but two cases (13.9% and 10.2%). The dose uncertainty for 95% of the CTV volume ranged from 1.3% to 2.9% of the prescribed dose. CONCLUSIONS: The dose uncertainty in prostate IMRT could be evaluated using the dose-uncertainty model. Prostate IMRT plans satisfying the same plan objectives could generate a significantly different dose uncertainty because a complex interplay of many uncertainty sources. The uncertainty-based plan evaluation contributes to generating reliable and error-resistant treatment plans.

A study on target positioning error and its impact on dose variation in image-guided stereotactic body radiotherapy for the spine.

PURPOSE: To investigate the amount of target positioning error and evaluate its dosimetric impact during image-guided stereotactic body radiotherapy for single-fraction spine treatment. METHODS AND MATERIALS: A prescription dose of 15 Gy and five to nine coplanar intensity-modulated beams were used. The patient was immobilized with a custom-fit vacuum mold, and the target was localized with a volumetric cone-beam CT image. A robotic couch with six degrees of freedom was used for target adjustment. For evaluation a cone-beam CT image was obtained at the end of treatment. Both target positioning error and its dosimetric impact were investigated for the first 9 cases. RESULTS: For cases studied, translational errors were 0.9 +/- 0.5 mm (lateral), 1.2 +/- 0.9 mm (longitudinal), 0.7 +/- 0.6 mm (vertical), and 1.8 +/- 1.0 mm (vector), and rotational errors were 1.6 degrees +/- 1.3 degrees (pitch), 0.8 degrees +/- 0.9 degrees (roll), and 0.8 degrees +/- 0.4 degrees (yaw). For the clinical target volume, D(95) (dose to 95% of target volume), D(90), D(max), and D(mean) were evaluated. Only 1 case showed significant dose variations, reaching up to 18% in D(95). The spinal cord dose was evaluated by observing D(0.1) (dose to 0.1 cm(3)), D(0.5), D(1.0), and D(max). Although 1 case showed a dose change reaching up to 30% in D(max), cord dose was within the planning tolerance limit in all but 2 cases (3% higher in one and 0.4% higher in the other). CONCLUSION: The implemented image-guided stereotactic body radiotherapy provides precise target localization. However, despite reasonably precise spatial precision, dosimetric perturbation can be significant because of both extremely steep dose gradients and close distances between the target and the spinal cord.

A generalized a priori dose uncertainty model of IMRT delivery.

Multileaf collimator-based intensity modulated radiation therapy (IMRT) is complex because each intensity modulated field consists of hundreds of subfields, each of which is associated with an intricate interplay of uncertainties. In this study, the authors have revised the previously introduced uncertainty model to provide an a priori accurate prediction of dose uncertainty during treatment planning in IMRT. In the previous model, the dose uncertainties were categorized into space-oriented dose uncertainty (SOU) and nonspace-oriented dose uncertainty (NOU). The revised model further divided the uncertainty sources into planning and delivery. SOU and NOU associated with a planning system were defined as inherent dose uncertainty. A convolution method with seven degrees of freedom was also newly applied to generalize the model for practical clinical cases. The model parameters were quantified through a set of measurements, accumulated routine quality assurance (QA) data, and peer-reviewed publications. The predicted uncertainty maps were compared with dose difference distributions between computations and 108 simple open-field measurements using a two-dimensional diode array detector to verify the validity of the model parameters and robustness of the generalized model. To examine the applicability of the model to overall dose uncertainty prediction in IMRT, a retrospective analysis of QA measurements using the diode array detector for 32 clinical IM fields was also performed. A scatter diagram and a correlation coefficient were employed to investigate a correlation of the predicted dose uncertainty distribution with the dose discrepancy distribution between calculation and delivery. In addition, a gamma test was performed to correlate failed regions in dose verification with the dose uncertainty map. The quantified model parameters well correlated the predicted dose uncertainty with the probable dose difference between calculations and measurements. It was visually validated with the scatter diagrams. The average correlation coefficient between uncertainty and dose difference of 108 verification measurements was 0.80 +/- 0.04, indicating a strong linear correlation. In the clinical IM field studies, the dose uncertainty map mimicked the probable dose difference distribution. The average correlation coefficient between the overall dose uncertainty and the dose difference of 32 QA measurements (total 13 184 comparison points) was 0.75 +/- 0.07, which also indicated a strong linear correlation between them. The failed regions of the gamma test remarkably corresponded to relatively high dose uncertainty. In conclusion, the dose uncertainty map was able to highlight high dose uncertainty regions, where more care should be taken during the treatment plan. The a priori accurate prediction of dose uncertainty in IMRT will significantly improve the treatment plan evaluation process, thus improving the quality of radiation treatments.

Dose variations with varying calculation grid size in head and neck IMRT.

Ever since the advent and development of treatment planning systems, the uncertainty associated with calculation grid size has been an issue. Even to this day, with highly sophisticated 3D conformal and intensity-modulated radiation therapy (IMRT) treatment planning systems (TPS), dose uncertainty due to grid size is still a concern. A phantom simulating head and neck treatment was prepared from two semi-cylindrical solid water slabs and a radiochromic film was inserted between the two slabs for measurement. Plans were generated for a 5,400 cGy prescribed dose using Philips Pinnacle(3) TPS for two targets, one shallow ( approximately 0.5 cm depth) and one deep ( approximately 6 cm depth). Calculation grid sizes of 1.5, 2, 3 and 4 mm were considered. Three clinical cases were also evaluated. The dose differences for the varying grid sizes (2 mm, 3 mm and 4 mm from 1.5 mm) in the phantom study were 126 cGy (2.3% of the 5,400 cGy dose prescription), 248.2 cGy (4.6% of the 5,400 cGy dose prescription) and 301.8 cGy (5.6% of the 5,400 cGy dose prescription), respectively for the shallow target case. It was found that the dose could be varied to about 100 cGy (1.9% of the 5,400 cGy dose prescription), 148.9 cGy (2.8% of the 5,400 cGy dose prescription) and 202.9 cGy (3.8% of the 5,400 cGy dose prescription) for 2 mm, 3 mm and 4 mm grid sizes, respectively, simply by shifting the calculation grid origin. Dose difference with a different range of the relative dose gradient was evaluated and we found that the relative dose difference increased with an increase in the range of the relative dose gradient. When comparing varying calculation grid sizes and measurements, the variation of the dose difference histogram was insignificant, but a local effect was observed in the dose difference map. Similar results were observed in the case of the deep target and the three clinical cases also showed results comparable to those from the phantom study.

Evaluation of surface and build-up region dose for intensity-modulated radiation therapy in head and neck cancer.

Despite much development, there remains dosimetric uncertainty in the surface and build-up regions in intensity-modulated radiation therapy treatment plans for head and neck cancers. Experiments were performed to determine the dosimetric discrepancies in the surface and build-up region between the treatment planning system (TPS) prediction and experimental measurement using radiochromic film. A head and neck compression film phantom was constructed from two semicylindrical solid water slabs. Treatment plans were generated using two commercial TPSs (PINNACLE3 and CORVUS) for two cases, one with a shallow (approximately 0.5 cm depth) target and another with a deep (approximately 6 cm depth) target. The plans were evaluated for a 54 Gy prescribed dose. For each case, two pieces of radiochromic film were used for dose measurement. A small piece of film strip was placed on the surface and another was inserted within the phantom. Overall, both TPSs showed good agreement with the measurement. For the shallow target case, the dose differences were within +/- 300 cGy (5.6% with respect to the prescribed dose) for PINNACLE3 and +/- 240 cGy (4.4%) for CORVUS in 90% of the region of interest. For the deep target case, the dose differences were +/- 350 (6.5%) for PINNACLE3 and +/- 260 cGy (4.8%) for CORVUS in 90% of the region of interest. However, it was found that there were significant discrepancies from the surface to about 0.2 cm in depth for both the shallow and deep target cases. It was concluded that both TPSs overestimated the surface dose for both shallow and deep target cases. The amount of overestimation ranges from 400 to 1000 cGy (approximately 7.4% to 18.5% with respect to the prescribed dose, 5400 cGy).