W Xiong

SU‐E‐T‐545: Dose Comparison between Intravenous Contrast‐Enhanced CT and Non Contrast CT in Treatment Planning

PURPOSE With increasing concern for patient dose from CT scan, we are trying to reduce CT scan and use intravenous contrast-enhanced CT (contrast CT) in treatment planning. This study is to investigate dose calculation accuracy using contrast CT in treatment planning for lung, esophagus and pancreas cancer. METHODS We analyzed treatment plans for 8 patients for whom CT simulation was performed both with and without intravenous contrast agent (CA) (non-contrast CT). IMRT/3D plans were generated with inhomogeneity correction on the non-contrast CT scan. Contrast CTs were fused to the non-contrast studies and all contours and plans were copied to the contrast CT scans. For each patient, we analyzed dose-volume histograms (DVH) for planning volumes (PTV) and the organs-at-risk (OAR), comparing the doses generated on non-contrast CT scans with those generated on contrast CT scans. RESULTS Maximum doses ratio Dmax(contrast)/Dmax (non-contrast) in PTVs was 1.0009±0.0013. The ratio of D05 (contrast)/D05 (non-contrast) was 0.996±0.005. The ratio of mean PTV dose Dmean(contrast)/Dmean(non-contrast) was 0.990±0.005%. The ratio of minimum dose Dmin(contrast)/Dmin(non-contrast) and D95(contrast)/D95(non-contrast) were 0.970±0.030 and 0.984±0.009, respectively. Contrast CT raised cord dose slightly. The ratio of cord Dmax was 1.005±0.026. However there were two cases the ratio of cord Dmax were 1.035. CONCLUSIONS The PTV D95 is usually normalized to prescription dose and the D95 differences between contrast and regular CT were within 2%. In most cases, the contrast CT could be used to treatment planning clinically. However more attention should be paid to maximum cord dose if it is already close to criteria limit.

SU-F-T-226: QA Management for a Large Institution with Multiple Campuses for FMEA

PURPOSE To redesign our radiation therapy QA program with the goal to improve quality, efficiency, and consistency among a growing number of campuses at a large institution. METHODS A QA committee was established with at least one physicist representing each of our six campuses (22 linacs). Weekly meetings were scheduled to advise on and update current procedures, to review end-to-end and other test results, and to prepare composite reports for internal and external audits. QA procedures for treatment and imaging equipment were derived from TG Reports 142 and 66, practice guidelines, and feedback from ACR evaluations. The committee focused on reaching a consensus on a single QA program among all campuses using the same type of equipment and reference data. Since the recommendations for tolerances referenced to baseline data were subject to interpretation in some instances, the committee reviewed the characteristics of all machines and quantified any variations before choosing between treatment planning system (i.e. treatment planning system commissioning data that is representative for all machines) or machine-specific values (i.e. commissioning data of the individual machines) as baseline data. RESULTS The configured QA program will be followed strictly by all campuses. Inventory of available equipment has been compiled, and additional equipment acquisitions for the QA program are made as needed. Dosimetric characteristics are evaluated for all machines using the same methods to ensure consistency of beam data where possible. In most cases, baseline data refer to treatment planning system commissioning data but machine-specific values are used as reference where it is deemed appropriate. CONCLUSION With a uniform QA scheme, variations in QA procedures are kept to a minimum. With a centralized database, data collection and analysis are simplified. This program will facilitate uniformity in patient treatments and analysis of large amounts of QA data campus-wide, which will ultimately facilitate FMEA.

SU-E-T-52: Beam Data Comparison for 20 Linear Accelerators in One Network

PURPOSE To compare photon beam data for the 20 Varian linear accelerators (TrueBeam, iX, and EX models) in use at five centers in the same network with the intent to model with one set of beam data in Eclipsec. METHODS Varian linear accelerators, TrueBeam (3), 21 EX, iX, and Trilogy (14), and 6 EX (3), installed between 1999 and 2014 have their 6 MV and 15 MV x-ray beams reevaluated. Full commissioning, including output factors (St), percent depth doses (PDD), and off-axis profiles, was recently performed for a TrueBeam with a cc04 ion chamber in an IBA Blue phantom. Similarly, a subset of beam data for each of the other accelerators was measured recently as follows: for 3×3, 10×10, and 30×30 cm2 field sizes, flatness and penumbra (80-20%) were measured at dmax and 10 cm depths, PDD were measured at 10 and 20 cm depths, and St were measured at 5 cm depth. Measurement results for all machines were compared. RESULTS For 15 high-energy (6 and 15 MV) and 3 low-energy machines (6MV only): 1) PDD agreed within 1.4% at 10 and 20 cm depths; 2) penumbra agreed within 1.0 mm at dmax and 10 cm depths; 3) flatness was within 1.3% at dmax and 10 cm depths; and 4) with exception of the three low energy machines, output factors were within 1.1% and 0.5% for 3×3 and 30×30 cm2 , respectively. Measurement uncertainty, not quantified here, accounts for some of these differences. CONCLUSION Measured beam data from 15 high-energy Varian linacs are consistent enough that they can be classified using one beam data set in Eclipse. Two additional high-energy machines are removed from this group until their data are further confirmed. Three low-energy machines will be in a separate class based upon differences in output factors (St).

SU‐E‐T‐540: Comparison of CT and MRI Based Monte Carlo Simulation for Gamma Knife Stereotactic Radiosurgery

PURPOSE MRI image is most frequently used for target contouring and treatment planning in Gamma Knife stereotactic radiosurgery (SRS). This study is to compare geometric and dosimetric accuracy of CT and MRI-based Monte Carlo (M.C.) simulation for Gamma Knife SRS. METHODS A cylindrical water phantom with scale for MRI QA was scanned and the MRI images were transferred to a planning system for geometric analysis. M.C. simulation was applied on patient geometries reconstructed from CT and MRI data for dosimetric comparison. In the M.C. simulation, Gamma Knife (Model C) unit geometry and material were reconstructed according to original unit. A heterogeneous patient MRI geometry was created by putting a 1.8 g/cc skull in the unity homogeneous MRI geometry based on MRI anatomy knowledge. The dose was calculated using M.C. simulation in both homogenous and inhomogeneous CT and MRI geometries with identical beam parameters. The dose distribution was compared by overlapping the isodose-lines for each calculation. The DVH was derived by collecting dose on a small volume around isocenter. RESULTS In MR image, the maximum errors along all directions are within 0.5 mm in the volume of interest (VOI) which is about 15cm high and 20cm diameter in x and y plane. There is no observable difference of relative isodose lines in CT and MRI geometries. However, the absolute dose in heterogeneous CT geometry was 3.2% lower than the dose in homogeneous CT geometry from the DVH comparison. The absolute dose in homogeneous MRI phantom was 3.3% higher that dose in heterogeneous CT geometry. After applying heterogeneity correction to the skull for MRI, the difference was reduced to less than 2%. CONCLUSION MRI image distortion is small with the maximum distortion within 0.5mm in VOI. MRI-based Monte Carlo planning for Gamma Knife is feasible after applying proper skull heterogeneity correction.

SU‐E‐T‐557: Dose Calculation Accuracy Using Cone‐Beam CT (CBCT) for Lung Stereotactic Body Radiotherapy (SBRT)

Purpose: To evaluate the dose calculation accuracy using Varians cone‐beam CT(CBCT) for lung stereotactic body radiotherapy(SBRT). Methods: Treatment plans of 2 lungSBRT patients (1 and 2) were retrospectively studied. Both plans used 3 or 4 6MV coplanar IMRT fields, generated on an in‐house Treatment Planning System (TPS) using Pencil‐Beam (PB) dose calculation with radiological path‐length correction. CBCTimages were taken prior to each treatment and a total 7 sets of CBCTimages were exported into TPS. Each CBCTimages set was registered with the planning CT based on matching of PTV PTV. Patient contours from planning CT were also imported into CBCT and only external contours were adjusted accordingly. CT numbers (HU+1000) were compared at chosen spots between CBCT and planning CT. Treatment fields were copied to CBCT and CBCT plans were generated. Results: CBCT CT‐numbers agree with planning CT‐number at low CT‐number (< 1100) region. For CT‐number 99, 887 and 1049, CBCT shows 83±24, 836±48 and 1054±26. For high CT‐ number 1571 and 1981, the CBCT CT‐numbers are 1730±55 and 2248±174. Mean dose for CBCT plans are slightly higher than those in treatment plans, which are 101.8% and 100.8% vs 102.0±1.1% and 102.1±0.1% of prescribed dose. Average D95 in CBCT plans are closed to those in treatment plan: 100.2±0.2%, 100.2±0.8% in CBCT 99.3% and 100.8% in treatment plan. However target inhomogeneity is higher in the CBCT plans, with average PTV D05 of 104.1±0.5% and 103.9±1.4% compared to 102.4% and 103.0% respectively. Critical normal tissuedoses in CBCT plans were approximately the same as the acceptable values predicted by treatment plan. Conclusions: The dosimetric differences between CBCT and planing‐CT were within 2% of each other. CBCT may be used to patient radiation dose verification besides patient setup and target verification for SBRT patients.

SU‐GG‐T‐101: Dosimetric Analysis of Matching 6 MV Photon and Electron Fields in Chestwall Treatment

Purpose: To investigate dose distribution for matching 6MV photon and electron fields in chestwall treatment. Method and Materials: A rectangular water equivalent phantom was used to investigate dose distribution at junction of adjacent photon and electron fields. Both 3D and simple IMRT plans have been created for photon fields. Electron plans with different electron energy were created and normalized to 90% iso‐dose line. Photon and electron fields were matched based on skin markings. Monte Carlo simulation has been applied for photon plans. We compared plans with a single photon‐electron match line to “mixed treatment” plans using 2 photon‐electron match lines (ie day1/day2) Results: Hot spots are observed at junction of photon field for all energy electron fields. The highest dose are 129%, 134%, 137%, 139% and 139% of dose for 6, 9, 12, 16 and 20 MeV electrons, respectively, with 6MV photon fields. The depths of these hot spots from skin surface are 1.5, 2.5, 3.4, 4.6 and 5.5cm, as compared with dmax for the electron fields individually (1.4, 2.2, 3.0, 3.0 and 2.0 cm, respectively). The hot spots for “mixed treatment” arrangement (day 1 and day 2) are 120%, 123%, 126%, 129% and 130% of dose, much lower than for the single match‐line fields. The depth for hot spot for day 1 and day 2 arrangement stays similar with day1 treatment. Results from Monte Carlo simulation agree with pencil beam dose calculation algorithm in skin region. Conclusion: Hot spots are reduced to clinically acceptable level by mixed treatment arrangement with two or even more matching lines between photon and electron fields. Further study is warranted to compare these results with intensity and/or energy modulated electron treatment, which may provide a more homogeneous dose distribution for chestwall treatment.

SU‐FF‐T‐527: Patient's Variation On Tumor Control Probability for Lung Cancer Treatment

Purpose: Many protocols have been applied on Stereotactic Body Radio Therapy (SBRT) and Stereotactic Radio Surgery (SRS) for lungcancers. This work is to investigate the effect of patient variation on tumor control probabilities (TCP) for SBRT,SRS as well as standard fractionated radio therapy for lungcancer treatment. Method and Materials: Linear‐quadratic (LQ) model was used in our TCP analysis. Three different protocols were investigated with the same parameters for LQ model: 1 fraction with 22 Gy total dose, 4 fractions with 48 Gy total dose and 33 fractions with 6600 Gy total dose. The dose inhomogeneity was assumed with a Gaussian distribution with a deviation of σdose. The patients variation of radio‐sensitivity for a population was added assuming Gaussian distributions for LQ parameters αand β with σα and σβ, respectively. Results: Although dose inhomogeneitys existence requires high total dose to achieve the same TCP, for most clinic cases, the dose increase is very limited and TCP could be evaluated base on single prescribed dose for these plans. Patients variation σα has more impact on 33 fractions IMRT treatment, the total dose needs to increase by 15%, 19% and 26% to maintain 95% TCP in SRS,SBRT and IMRT when σα increases from 0.0 to 0.2. At the same time, Patients variation σβ has more impact on SRS, the total dose needs to increase by 17%, 12% and 5% to keep 95% TCP in SRS,SBRT and IMRT while patient variation σβ increases from 0.0 to 0.2. Conclusion: LungSRS and SBRT provide better tumor control for lungcancer than standard IMRT. However, patient variation should be considered when SRS and SBRT are designed for lungcancer treatment. SRS and SBRT also may lead to more lung complications.

SU‐FF‐T‐190: Monte Carlo‐Guided Improvement of Therapeutic Ratio in Pencil Beam Dose‐Based IMRT Plans of Small Tumors in Lung

Purpose: IMRT plan optimization commonly uses pencil beam (PB)‐based dose calculation with limited inhomogeneity correction, often predicting better target coverage than is delivered for small tumors in lung. We examine using Monte‐Carlo (MC)dose calculation to improve the therapeutic ratio for stereotactic body radiotherapy(SBRT) of lungcancer.Method and Materials:Treatment plans of ten patients (13 tumors) were retrospectively studied, with average GTV and PTV of 13.2cc and 72.3cc. All plans used IMRT with 3–5 6MV coplanar fields per target, generated on an in‐house planning system using PB dose calculation with radiological pathlength correction. Initial plans (TPS1) were recalculated with an in‐house MC algorithm (MC1), yielding lower PTV and GTV dose indices D05 and D95 relative to TPS1. A second IMRT plan (TPS2) used an objective function that increased the GTV and PTV doses to compensate for the deficit observed in MC1 but using the same normal tissue constraints as TPS1. The new plan was recalculated with MC (MC2). Results: Although the target dose observed in TPS1 plans is the clinical goal, it is overestimated by the PB algorithm. Average PTV D95 and GTV D95 in MC1 plans are (85±6)% and (92±4)% of those in TPS1 for 13 sites. MC calculation following the second optimization showed target dose indices to be closer to clinical goals: average PTV D95 and GTV D95 in MC2 plans are (98 ±4)% and (102±2)% of those in TPS1. However, target inhomogeneity is higher, with average PTV D05 and GTV D05 of (109±4)% and (104±3)% respectively. Critical normal tissuedoses in MC2 were approximately the same as the acceptable values predicted by TPS1. Conclusion: MC‐guided second optimization of PB‐based IMRT plans substantially improves target coverage for small lungtumors while maintaining normal tissue constraints. However increased hot spots in tumor require careful clinical consideration.

SU-GG-T-443: Effect of Inhomogeneity On Dose Distribution for Gamma Knife(R) Perfexion(TM) Treatment

Purpose: To investigate the dose perturbation in Gamma Knife® Perfexion™ Stereotactic Radiosurgery due to inhomogeneity. Method and Materials: We applied Monte Carlo(MC) simulation for Gamma Knife® Perfexion™. Approximation parameters for geometries and materials were used to reconstruct 4mm, 8mm and 16mm collimators in Gamma Knife® Perfexion™ unit. A total of 192 Cobalt‐60 sources with the same activity were simulated. Phase space files stored all particles passing through collimator and reaching a sphere surface of surrounding the isocenter. Patients geometries were rebuilt from CT data. The doses were calculated using MC simulation with and without inhomogeneity correction. Dose perturbation effect was derived from the comparison of DVH between homogeneous and inhomogeneous geometries. Results: Three patients were investigated. For each patient, at least 2 sites were selected: one in brain and another one close to air cavity. A total of 14 Monte Carlo simulations were performed for 16mm collimator with and without inhomogeneity correction. The statistical error for isocenter dose is around 0.2%. Monte Carlo results show that inhomogeneity effects lowered the dose, on average, by 4.6% in brain site. For sites close to the air cavity, the inhomogeneity effects are mixed. The inhomogeneity effect ranged from 2.7% to −1.3%. Less attenuation from air cavity increases the dose to the site. However, the high density area such as spine results more attenuation and it eventually reduces the dose. Conclusion:Monte Carlo shows that the inhomogeneity effect could cause lower delivered doses from Gamma Knife® Perfexion™ in brain sites and higher or lower doses in those sites close to air cavity. More investigations are necessary to determine accurate dose for treatment sites close to air cavity for Gamma Knife® Perfexion™.

SU‐FF‐T‐247: Implementation of MRI‐Based Monte Carlo Simulation for Gamma Knife SRS

Purpose: To verify the dosimetry accuracy for MRI‐based planning of Gamma Knife Stereotactic Radiosurgery.Method and Materials: In Monte Carlo simulation, Gamma Knife unit geometry was reconstructed exactly as the original unit. Materials were selected as close to the actual one as possible. Patients were scanned on both a CT unit and a 1.5 T MRI scanner. Simulations were performed for these patients using homogeneous geometry based on CT and MRI, and heterogeneous geometry built based on CT numbers or different densities for MR contoured skull structures. The homogenous density was chosen as 1.0g/cc and the density of skull was chosen in the range 1.5 – 2.0 g/cc. Isodose distributions and DVHs were used in comparison. Results: The dose in the homogeneous geometries based on CT was about 3.2% higher than the dose in heterogeneous geometry based on CT. The difference in the DVH between homogeneous MRI and heterogeneous CT geometry was also around 3.3%. After applying heterogeneity correction to the skull for MRI, the difference was reduced to less than 2%. The 90%, 50% and 10% isodose lines matched each other very well between homogeneous CT and MRI geometries and heterogeneous CT and MRI geometries. Conclusion: A useful CT and MRI based Monte Carlo simulation has been developed for Gamma Knife dose verification. For many Gamma Knife centers, MRI is the only choice for Gamma Knife planning. Our results show that MRI‐based Monte Carlo planning for Gamma Knife is feasible after applying proper skull heterogeneity correction.