F Cao

Whole brain radiotherapy with hippocampal avoidance and simultaneous integrated boost for 1-3 brain metastases: a feasibility study using volumetric modulated arc therapy.

PURPOSE To evaluate the feasibility of using volumetric modulated arc therapy (VMAT) to deliver whole brain radiotherapy (WBRT) with hippocampal avoidance and a simultaneous integrated boost (SIB) for one to three brain metastases. METHODS AND MATERIALS Ten patients previously treated with stereotactic radiosurgery for one to three brain metastases underwent repeat planning using VMAT. The whole brain prescription dose was 32.25 Gy in 15 fractions, and SIB doses to brain metastases were 63 Gy to lesions >or=2.0 cm and 70.8 Gy to lesions <2.0 cm in diameter. The mean dose to the hippocampus was kept at <6 Gy(2). Plans were optimized for conformity and target coverage while minimizing hippocampal and ocular doses. Plans were evaluated on target coverage, prescription isodose to target volume ratio, conformity number, homogeneity index, and maximum dose to prescription dose ratio. RESULTS Ten patients had 18 metastases. Mean values for the brain metastases were as follows: conformity number = 0.73 +/- 0.10, target coverage = 0.98 +/- 0.01, prescription isodose to target volume = 1.34 +/- 0.19, maximum dose to prescription dose ratio = 1.09 +/- 0.02, and homogeneity index = 0.07 +/- 0.02. For the whole brain, the mean target coverage and homogeneity index were 0.960 +/- 0.002 and 0.39 +/- 0.06, respectively. The mean hippocampal dose was 5.23 +/- 0.39 Gy(2). The mean treatment delivery time was 3.6 min (range, 3.3-4.1 min). CONCLUSIONS VMAT was able to achieve adequate whole brain coverage with conformal hippocampal avoidance and radiosurgical quality dose distributions for one to three brain metastases. The mean delivery time was under 4 min.

Dosimetric evaluation of Plastic Water Diagnostic Therapy

High‐precision radiotherapy planning and quality assurance require accurate dosimetric and geometric phantom measurements. Phantom design requires materials with mechanical strength and resilience, and dosimetric properties close to those of water over diagnostic and therapeutic ranges. Plastic Water Diagnostic Therapy (PWDT: CIRS, Norfolk, VA) is a phantom material designed for water equivalence in photon beams from 0.04 MeV to 100 MeV; the material has also good mechanical properties. The present article reports the results of computed tomography (CT) imaging and dosimetric studies of PWDT to evaluate the suitability of the material in CT and therapy energy ranges. We characterized the water equivalence of PWDT in a series of experiments in which the basic dosimetric properties of the material were determined for photon energies of 80 kVp, 100 kVp, 250 kVp, 4 MV, 6 MV, 10 MV, and 18 MV. Measured properties included the buildup and percentage depth dose curves for several field sizes, and relative dose factors as a function of field size. In addition, the PWDT phantom underwent CT imaging at beam qualities ranging from 80 kVp to 140 kVp to determine the water equivalence of the phantom in the diagnostic energy range. The dosimetric quantities measured with PWDT agreed within 1.5% of those determined in water and Solid Water (Gammex rmi, Middleton, WI). Computed tomography imaging of the phantom was found to generate Hounsfield numbers within 0.8% of those generated using water. The results suggest that PWDT material is suitable both for regular radiotherapy quality assurance measurements and for intensity‐modulated radiation therapy (IMRT) verification work. Sample IMRT verification results are presented. PACS number: 87.53Dq

On using the dosimetric leaf gap to model the rounded leaf ends in VMAT/RapidArc plans

Partial transmission through rounded leaf ends of Varian multileaf collimators (MLC) is accounted for with a parameter called the dosimetric leaf gap (DLG). Verification of the value of the DLG is needed when the dose delivery is accompanied by gantry rotation in VMAT plans. We compared the doses measured with GAFCHROMIC film and an ionization chamber to treatment planning system (TPS) calculations to identify the optimum values of the DLG in clinical plans of the whole brain with metastases transferred to a phantom. We noticed the absence of a single value of the DLG that properly models all VMAT plans in our cohort (the optimum DLG varied between 0.93±0.15 mm and 2.2±0.2 mm). The former value is considerably different from the optimum DLG in sliding window plans (about 2.0 mm) that approximate IMRT plans. We further found that a single‐value DLG model cannot accurately reproduce the measured dose profile even of a uniform static slit at a fixed gantry, which is the simplest MLC‐delimited field. The calculation overestimates the measurement in the proximal penumbra, while it underestimates in the distal penumbra. This prompted us to expand the DLG parameter from a plan‐specific number to a mathematical concept of the DLG being a function of the distance in the beams eye view (BEV) between the dose point and the leaf ends. Such function compensates for the difference between the penumbras in a beam delimited with a rounded leaf MLC and delimited with solid jaws. Utilization of this concept allowed us generating a pair of step‐and‐shoot MLC plans for which we could qualitatively predict the value of the DLG providing best match to ionization chamber measurements. The plan for which the leafs stayed predominantly at positions requiring low values of the DLG (as seen in the profiles of 1D slits) yielded the combined DLG of 1.1±0.2 mm, while the plan with leafs staying at positions requiring larger values of the DLG yielded the DLG 2.4±0.2 mm. Considering the DLG to be a function of the distance (in BEV) between the dose point and the leaf ends allowed us to provide an explanation as to why conventional single‐number DLG is plan‐specific in VMAT plans. PACS numbers: 87.56.jf, 87.56.nk

Template-based breast IMRT planning for increased workload efficiency

BackgroundTo be less resource intensive, we developed a template-based breast IMRT technique (TB-IMRT). This study aims to compare resources and dose distribution between TB-IMRT and conventional breast radiation (CBR).MethodsTwenty patients with early stage breast cancer were planned using CBR and TB-IMRT. Time to plan, coverage of volumes, dose to critical structures and treatment times were evaluated for CBR and TB-IMRT. Two sided-paired t tests were used.ResultsTB- IMRT planning time was less than CBR (14.0 vs 39.0 min, p < 0.001). Fifteen patients with CBR needed 18 MV, and 11 of these were planned successfully with TB-IMRT using 6 MV. TB-IMRT provided better homogeneity index (0.096 vs 0.124, p < 0.001) and conformity index (0.68 vs 0.59, p = 0.003). Dose to critical structures were comparable between TB-IMRT and CBR, and treatment times were also similar (6.0 vs 7.8 min, p = 0.13).ConclusionsTB- IMRT provides reduction of planning time and minimizes the use of high energy beams, while providing similar treatment times and equal plans compared to CBR. This technique permits efficient use of resources with a low learning curve, and can be done with existing equipment and personnel.

SU‐E‐T‐455: Evaluation of the Influence of the Dosimetric Leaf Gap on RapidArc Plans Using Gafchromic Film and Monte Carlo Simulations

Purpose: Correct setting the dosimetric leaf gap (DLG) in the planning system is essential for proper calculation of the dose distribution for RapidArc plans. We evaluated the influence of the DLG by using Gafchromic EBT2 film, as well as running Monte Carlo(MC) simulations. Methods: Gafchromic EBT2 film was sandwiched between acrylic slabs, and irradiated during delivery of RapidArc plans. The films were scanned with Epson V700 scanner, corrected for scanner nonuniformity, and converted to dose. The agreement between the dose distribution measured with film (for DLG values ranging from 1.3mm to 5.0mm) was compared to the corresponding dose calculated in Eclipse using the AAA8615 algorithm. After registering the film images to the calculated dose, agreement was quantified by comparing line profiles and examining gamma maps. No normalization or spatial shifting was employed. MC simulations were carried out on a phantom; gamma maps were calculated between the simulated and planned doses. Results: Comparing line profiles from film to the corresponding profiles from AAA revealed the best agreement at DLG of 1.6mm. For larger values of the DLG the AAA calculations overestimated the film profiles, while smaller values underestimated the film data. The gamma maps in the region conformal to the PTV had nearly 100% passing rate for gamma (3%, 3mm). MC simulations yielded an optimum DLG value of 2.2mm Conclusions: EBT2 films proved to be sensitive to detect variations in dose distributions calculated for various values of the DLG. Such sensitivity was only possible in the absence of normalization of the data sets, as the DLG influences the average calculated dose in the PTV. There is a small disagreement of the optimum value of DLG established using these two methods, which may be attributed to the film data being a subset of the MC calculated 3D data set.

SU‐F‐T‐284: The Effect of Linear Accelerator Output Variation On the Quality of Patient Specific Rapid Arc Verification Plans

PURPOSE The aim of the current study is to investigate the effect of machine output variation on the delivery of the RapidArc verification plans. METHODS Three verification plans were generated using Eclipse™ treatment planning system (V11.031) with plan normalization value 100.0%. These plans were delivered on the linear accelerators using ArcCHECK- device, with machine output 1.000 cGy/MU at calibration point. These planned and delivered dose distributions were used as reference plans. Additional plans were created in Eclipse- with normalization values ranging 92.80%-102% to mimic the machine output ranging 1.072cGy/MU-0.980cGy/MU, at the calibration point. These plans were compared against the reference plans using gamma indices (3%, 3mm) and (2%, 2mm). Calculated gammas were studied for its dependence on machine output. Plans were considered passed if 90% of the points satisfy the defined gamma criteria. RESULTS The gamma index (3%, 3mm) was insensitive to output fluctuation within the output tolerance level (2% of calibration), and showed failures, when the machine output exceeds ≥3%. Gamma (2%, 2mm) was found to be more sensitive to the output variation compared to the gamma (3%, 3mm), and showed failures, when output exceeds ≥1.7%. The variation of the gamma indices with output variability also showed dependence upon the plan parameters (e.g. MLC movement and gantry rotation). The variation of the percentage points passing gamma criteria with output variation followed a non-linear decrease beyond the output tolerance level. CONCLUSION Data from the limited plans and output conditions showed that gamma (2%, 2mm) is more sensitive to the output fluctuations compared to Gamma (3%,3mm). Work under progress, including detail data from a large number of plans and a wide range of output conditions, may be able to conclude the quantitative dependence of gammas on machine output, and hence the effect on the quality of delivered rapid arc plans.

Sci-Thur AM: Planning - 03: Extensive patient specific QA for field junction regions for craniospinal irradiation with Jagged-Junction IMRT approach without beam edge matching for field junctions

PURPOSE Jagged-Junction IMRT was developed for craniospinal irradiation. An extensive QA was performed for the field junction regions. METHODS AND MATERIALS The Jagged-Junction IMRT plan employed three field sets, each with unique isocentres (Iso1,2,3). Fields from adjacent sets were overlapped and the dose was smoothly integrated inside the overlapped junction. The delivered dose in the junction regions were verified with film and ion chamber measurements on phantoms. An anthropomorphic-wax phantom was created for verifying the cranio-spinal junction. For measuring at the spinal-spinal junction, a solid water phantom was used. The influence of beam mismatching due to setup and mechanical inaccuracy was investigated by shifting all the fields from Iso1 and Iso3 superiorly and inferiorly by 3 mm and at the same time keeping all the fields from Iso2 without any shift. RESULTS The patient-averaged difference between the measured dose with ion chamber and planned dose in the cranio-spinal junction is 0.34 % ± 0.40% and in the spinal-spinal junction this difference is 0.03% ± 0.71%. The dose profile comparison shows that measured and planned dose profiles match well to each other over a junction region. The patient-averaged dose difference discrepancy between the film measurement and the planned is 1.1% ± 1.3% at the cranio-spinal junction and -0.14% ± 1.8% at the spinal junction. CONCLUSIONS Jagged-Junction IMRT planning provided smooth dose coverage to the target in the field junction region. The junction dose for the Jagged-Junction IMRT plan is not sensitive to the setup error during the treatment.

Poster — Thur Eve — 57: Craniospinal irradiation with jagged‐junction IMRT approach without beam edge matching for field junctions

PURPOSE Craniospinal irradiation were traditionally treated the central nervous system using two or three adjacent field sets. A intensity-modulated radiotherapy (IMRT) plan (Jagged-Junction IMRT) which overcomes problems associated with field junctions and beam edge matching, improves planning and treatment setup efficiencies with homogenous target dose distribution was developed. METHODS AND MATERIALS Jagged-Junction IMRT was retrospectively planned on three patients with prescription of 36 Gy in 20 fractions and compared to conventional treatment plans. Planning target volume (PTV) included the whole brain and spinal canal to the S3 vertebral level. The plan employed three field sets, each with a unique isocentre. One field set with seven fields treated the cranium. Two field sets treated the spine, each set using three fields. Fields from adjacent sets were overlapped and the optimization process smoothly integrated the dose inside the overlapped junction. RESULTS For the Jagged-Junction IMRT plans vs conventional technique, average homogeneity index equaled 0.08±0.01 vs 0.12±0.02, and conformity number equaled 0.79±0.01 vs 0.47±0.12. The 95% isodose surface covered (99.5±0.3)% of the PTV vs (98.1±2.0)%. Both Jagged-Junction IMRT plans and the conventional plans had good sparing of the organs at risk. CONCLUSIONS Jagged-Junction IMRT planning provided good dose homogeneity and conformity to the target while maintaining a low dose to the organs at risk. Jagged-Junction IMRT optimization smoothly distributed dose in the junction between field sets. Since there was no beam matching, this treatment technique is less likely to produce hot or cold spots at the junction in contrast to conventional techniques.

SU‐E‐T‐130: Evaluation of Optically Stimulated Luminescent Dosimeters as An In‐Vivo Dosimeter for Breast Surface Dose Measurement

Purpose: Evaluation of the optically stimulated luminescence(OSL) system for surface dosemeasurement for breast radiation therapy techniques. Methods: Landauer InLight OSL were calibrated on a Varian 21EX linac using an ion chamber with 6 MV beam. The response of OSL,TLD and a parallel plate chamber were compared at the surface of a solid water phantom and also under reference conditions (10×10cm2 at depth 5cm). OSL angular dependence, reproducibility, linearity, and inter‐OSL variability were investigated. Radiation therapytreatment plans were generated for an anthropomorphic breast phantom (RANDO). Eleven locations were identified and 2 OSL and 3 TLD were placed at each location. Treatment was delivered under clinical conditions. The OSL and TLD were evaluated for dose response. Results: The average response of the OSL to dose as compared to the ion chamber was found to be within 6% under reference conditions and the TLD within 2.5%. The factor of surface response to response at depth was found to be 0.284 for the Markus chamber, 0.425 for the OSLs and 0.434 for the TLD. No angular dependence was found for the OSL. The reproducibility of the batch response given 100 cGy was within 1%. The linearity of the OSL displayed an R2 value of 0.9997. The OSL given 100 cGy under reference conditions had a maximum deviation from average of 2.6%. Discrepancies on the IMRT plan between TLD and OSL ranged from 0.1% to 6.5% for the dose>50cGy and between 3.5% and 38.9% for the dose≤50cGy Conclusions: OSL and TLD are over‐responding to surface dose with respect to the parallel plate ion chamber.OSL have been found to respond similarly to TLD at the surface of an anthropomorphic phantom.

SU‐FF‐T‐212: Quality Assurance of IMRT Plans Using IMSure QATM Software as a Substitute for Measurement

Purpose: To study appropriateness of replacing patient‐specific IMRT QA with independent calculations using IMSure QA software. Also, our aim was to develop acceptability criteria of IMRT plan QA and to determine site‐specific recommendations. Method and Materials: Point dose data and fluence maps from Eclipse treatment planning system, IMSure, and measurements were compared for head and neck and prostate IMRT plans. In all 25 prostate and 20 head and neck IMRT plans were included in this study. A three‐dimensional portal image‐based dose reconstruction program in a virtual phantom (Epidose) was utilized for the comparison of the isocentre dose with Eclipse. Results: For 5‐field prostate IMRT plans, the average percentage discrepancy between IMSure and Eclipse was 0.3% with two third of the data agreed between +/− 0.5%. While the average discrepancy between Epidose and Eclipse was −0.3%+/−0.6%. Similar results were obtained for the 7‐field IMRT head and neck; the average discrepancy between IMSure and Eclipse was −0.7% +/−0.8%, while the average discrepancy between Epidose and Eclipse was 0.1% +/− 1.4%. Fluence map comparison of pixels was performed and gamma function values were calculated. The results for the 2% and 3% level discrepancies were derived for prostate and head and neck cases, respectively. For prostate, the average percentage of cells failing the 2% limit was 4.2%+/−1.5%, while for head and neck with 3% limit it was 7.5% +/−2.6%. Conclusions: Our investigation of the IMSure QA software for IMRT showed good agreement with both Eclipse treatment planning system and EPIDOSE measurements to within 1%. The results showed that IMSure software can be a reliable tool for IMRT QA, and measurements need to be invoked only for few select patients (less than one third) when point dose discrepancies exceed 1% for prostate and 2% for head and neck cases.