Index of content:
Volume 35, Issue 6, June 2008
- Therapy Scientific Session: Room 351
- IMRT — Optimization
MO‐D‐351‐01: (Jack Fowler Junior Investigator Competition Winner) Optimization Tools for Modulated Electron Radiotherapy35(2008); http://dx.doi.org/10.1118/1.2962352View Description Hide Description
Purpose: To devise optimization tools to plan Modulated Electron Radiotherapy (MERT) and to compare MERT plans to conventional or IMRT plans. Method and Materials: There are no commercially available treatment systems to plan or deliver MERT. Tertiary electron MLCs for MERT has been met with practical limitations. We've investigated use of the inherent photonMLCs. We optimized MERT plans for a left sided post‐mastectomy breast cancer patient. From CTimages, target organs such as chest wall, axillas, inter‐mammary nodes were contoured. The MERT plan was performed in a three‐step optimization process. Distances from the external contour‐to‐distal location of the PTV were calculated in transverse CTimages. Energies were selected for efficient target coverage and energy bins were generated. Optimization of the energy bins was accomplished using a custom MERT Planning graphic user interface (MERTgui). Monte Carlo simulations were then performed using different MLC instructions for each segment generated by the MERTgui. Dose distributions of each segment were imported to CERR. Finally we used a custom built dose superposition GUI to combine doses for each segment using different weights which yields their relative MUs. The resulting dose distribution was used to calculate DVHs using CERR and compared to the conventional plan. Results: The MERT plan resulted in good target organ coverage and less dose to the organs at risk than the conventional plan. The dose that 20% of the ipsilateral lung received was reduced from 45Gy (conventional) to 26Gy (MERT). Using MERT, 80% volume of both axilla and IM received the prescription dose without consequence of excess dosage as with the conventional plan. Contralateral breast dose decreased significantly to 1/10 the dose of the conventional plan. Conclusion: MERT is ideal for treatment of shallow targets such as post‐mastectomy chest wall. We have devised optimization tools to facilitate the MC based planning.
35(2008); http://dx.doi.org/10.1118/1.2962353View Description Hide Description
Purpose: To develop and evaluate an efficient algorithm for inverse treatment planning of volumetric arc therapy (VMAT). Method and Materials: VMAT plans were generated by first applying a direct machine parameter optimization algorithm from a research version of a commercial treatment planning system (Pinnacle 8.1v) to 18 beams spaced equidistantly at 20 degrees. At each beam angle, the number of segments was variable (typically 3 to 8) to match the intensity modulation fluence map. Low weight or small area segments were eliminated during optimization. The resulting segments were redistributed around the arc such that each individual segment was at a unique arc angle. Segments were sorted according to shape similarity in order to reduce leaf travel in between arc angle positions. To ensure accuracy of the final dose calculation, additional interpolated segments were generated, followed by a segment weight optimization to identify beam‐off segments that would otherwise conflict with plan objectives. The approach was evaluated for six cases (three head‐and‐necks, two prostate, and one brain) with respect to treatment delivery time, and overall plan quality in comparison to step‐and‐shoot IMRT using a conventional number of static beams and identical objectives and constraints as in the VMAT case. Results: The dynamic arc plans consistently demonstrated target coverage and critical structure sparing comparable to step‐and‐shoot plans. For the head‐and‐neck cases, sparing of the brainstem and the spinal cord could be improved substantially. Treatment parameter optimization and dose calculation required 15 to 20 minutes on standard hardware. Estimated delivery times were between 1.5 and 4 minutes, which is mainly determined by the leaf trajectories, gantry speed and dose rate. Conclusion: A method for inverse planning for VMAT delivery was developed. The method produces treatment plans with high dosimetric quality and efficient delivery time with clinically feasible user time and effort.
MO‐D‐351‐03: A Unified Approach to Beam Angle Selection and Dose Optimization with High‐Throughput Computing for IMRT35(2008); http://dx.doi.org/10.1118/1.2962354View Description Hide Description
Purpose: To present a unified approach to solving the Beam Angle Selection (BAS) and DoseOptimization (DO) problems in radiation treatment planning using a Nested Partitions (NP) framework. Method and Materials: The NP framework is a powerful new optimization paradigm that combines adaptive global sampling with local heuristic search. It uses a flexible partitioning method to divide the search space into regions that can be analyzed individually and then coordinates the results to determine how to continue the search, that is, where to concentrate additional computational effort. This partitioning/sampling approach makes the NP framework uniquely well‐suited for high‐throughput computing. Beam angle space is partitioned and sampled. DO algorithms are incorporated during the evaluation of the quality of a selected angle set. After the execution of our proposed method, we not only obtain a set of beam, but also the optimized intensity for each beam in 3DCRT or “intensity maps” for each angle in IMRT.Results: Using a 3DCRT data set of a pancreas case, we demonstrated the following improvements in OAR dose relative to a equispaced beam set: cord, 66%; kidney, 78%; liver 36%. We also considered an IMRT head‐and‐neck case, and obtained a 28% reduction in dose to normal tissue as well as improvements in the right parotid dose with no significant changes in dose to other tissues.Conclusion: We have demonstrated that our framework provides an effective and automated approach to obtaining high‐quality solutions to the unified BAS and DO problems in both 3DCRT and IMRT. Relative to beam‐angle sets constructed via expert clinical judgment and other approaches, the beams and doses via NP with HTC showed significant reduction in the radiation delivered to non‐cancerous tissue near the tumors.
35(2008); http://dx.doi.org/10.1118/1.2962355View Description Hide Description
Purpose: Although intensity modulated proton therapy (IMPT) is a flexible and powerful treatment technique, its inverse planning can be computationally challenging in terms of memory and speed. Here, we propose a novel optimization algorithm with the hybrid strategy to overcome those challenges. The algorithm worked very well in our tested phantoms and patients. Method and Materials: A common strategy to speed up the dose calculation during optimization is to use a Kij matrix which stores precalculated dose contributed by the scanning spot j to voxel i. However, this often requires large amount of memory due to the large dose voxel size in patients and multiple energy layers as the extra dimension of data for proton scanning beams. To speed up the dose calculation while keeping the memory usage low, our algorithm separately caches only major dose contributions from scanning spots with full intensities to each dose voxel; the sum of contributions from the remaining scanning spots is cached but unchanged except when they are accurately updated at some specified moment. A lower cut‐off threshold can be adjusted to establish the set of spots considered to make major dose contributions. We developed a variant of Newton's method, which automatically determines the patient‐customized damping parameter to attain convergence efficiency. Results: In addition to the greatly improved memory usage by the partial Kij , our tests showed that each iteration is also much faster than that using the full Kij matrix. The algorithm worked well for patients with tumors at different sites, including one with a large tumor in the lung. It showed that significant improvements of critical structure sparing were achieved. Conclusion: The proposed algorithm was computationally efficient in terms of computing time and memory usage. It worked well in complicated situations involving different treatment sites.
35(2008); http://dx.doi.org/10.1118/1.2962356View Description Hide Description
Purpose: To present a two‐stage optimization approach that explicitly considers aperture‐specific dose distributions prior to dose optimization.Method and Materials: Approximate and fast non‐Monte Carlo dose calculation models have typically been utilized for pencil beam weight optimization. Such approximate dose calculations do not take into account aperture specific collimator effects. This leads to potentially significant differences between the dose distribution corresponding to the optimization and a final more accurate dose calculation is typically used to evaluate the treatment plan (conventional planning). We propose a 2‐stage approach in which the aperture specific dose distributions were calculated using Monte Carlo and then optimized using sequential linear programming techniques. Our approach was compared with the “conventional” IMRT planning approach and a single‐stage process known as “direct aperture optimization”. Results: We have tested our approach on a head and neck case in the which, the target involved the primary tumor plus the nodal volume. The dose prescription was a simultaneous integrated boost. The conventional plan used 159 apertures, the 2‐stage plan used 123 apertures (36 of the original apertures yielded weights of zero) and the DMPO plan used 89 apertures. The 2‐stage plan outperformed both the conventional IMRT plan and the DMPO plans. The 2‐stage optimization approach resulted in an improvement of almost 10–14% for the RP compared with the other optimization approaches and in the lowest maximum cord and brainstem doses. The 2‐stage approach also resulted in the least volume of the CTV (both primary and nodal volumes) receiving “hot spots”. These improvements were achieved with greater delivery efficiency relative to the conventional plan. Conclusion: Our results demonstrate that the 2‐stage optimization process in which an initial set of apertures is obtained and then re‐optimized using linear programming yields superior results relative to the conventional approach and to the DMPO direct aperture‐based approach.
35(2008); http://dx.doi.org/10.1118/1.2962357View Description Hide Description
Purpose: The recent development of new linaccontrol systems that are capable of delivering Volumetric Modulated Arc Therapy (VMAT) has attracted significant attention. There remains, however, a lack of robust inverse planning tools for VMAT. In this study, we will present a generalized inverse planning tool that can provide highly conformal VMAT solutions using either single‐arc or multiple‐arc deliveries for both Varian and Elekta MLCs.Method and Materials: To generate VMAT plans, we first created optimized multi‐field IMRT plans with equal‐spaced beam angles in Pinnacle3 using direct machine parameter optimization (DMPO). A “deliverable” fluence map was reconstructed using the resulting apertures for each beam. Next, we applied our home‐grown arc sequencer to translate these fluence maps into VMAT plans. Based on the user‐defined requirements, the sequencer can provide either single‐arc or multiple‐arc plans that meet the predefined VMAT leaf‐motion constraints. The obtained VMAT plans were then loaded into Pinnacle3 for a final dose calculation. In this study, 10 cases were tested in this study covering a variety of treatment sites including head‐&‐neck(5), prostate(3), lung(1) and brain(1). Results: A total of 24 VMAT plans were created using these 10 cases. Results demonstrated that highly conformal VMAT dose distributions can be achieved with an average sequencing time of under 8 minutes. On average, the VMAT plans required 513 MUs to deliver between 1 to 3 arcs. The average standard deviation in the target dose was 5.79 cGy/fraction; while the average target volume covered by 95% prescribed dose was 98.1%. Our results show that comparable VMAT plans can be achieved using either the Elekta 80‐leaf MLC or the Varian 120‐leaf Millennium MLC.Conclusion: Our generalized arc‐sequencing algorithm serves as a robust inverse planning solution for VMAT. Highly conformal single‐arc or multiple‐arc VMAT plans can be created for Elekta and Varian MLCs.
MO‐D‐351‐07: A System to Dynamically Balance Dose Sparing Between Critical Structures in Intensity Modulated Radiation Therapy (IMRT)35(2008); http://dx.doi.org/10.1118/1.2962358View Description Hide Description
Purpose: The IMRTtreatment planning optimization process typically follows one, somewhat arbitrary, path, based on the manner in which the planner changes the optimization constraints/weights during planning. Thus, the planner usually never fully understands the trade‐off in doses between structures and, thereby, can not optimally distribute critical structuredose sparing. We present a system that enables the planner to generate an optimally balanced plan in real‐time by fully exploiting dose exchange between critical structures.Method and Materials: The system uses pre‐optimized fluences that are generated automatically. An initial baseline optimization is generated by “restricting” doses out of all critical structures equally. From this baseline, optimizations are generated for each critical structure further restricted, with all other critical structures completely relaxed to absorb excess dose (the optimizations explicitly respect target constraints) The fluences during the restricting process are stored, providing a history of the pathway taken by the optimization. A fast strategy is used to generate these optimizations prior to planning. At the time of planning, the system enables the planner to dynamically combine the pre‐computed dose distributions corresponding to the extent that they would like each critical structure to be spared. The results are displayed in histogram‐like form, allowing the planner to evaluate and dynamically balance critical structuredose sparing. Results: The system is illustrated for prostate and paranasal sinus cancer cases. In the prostate case, below a certain extent of rectal and bladder dose sparing, decreasing dose to one structure leads to increasing dose to the other structure. In the paranasal sinus cancer case, lowering the dose below the baseline optimization results in dose trade‐off between the the left eye/optic nerve, chiasm, right optic nerve, and left/right temporal lobes. Conclusion: The system presented here enables maximized dose sparing by optimally balancing IMRTdoses between critical structures in real‐time.
MO‐D‐351‐08: Analysis of the Effects of Multiple GEUD‐Type Constraints On Dose Distribution for IMRT Optimization35(2008); http://dx.doi.org/10.1118/1.2962359View Description Hide Description
Purpose: To analyze the effects of multiple gEUD‐type (convex) constraints on the resulting dose‐volume distribution/histogram (DVH). A key motivation for this work is to find a convex optimization alternative to (non‐convex) partial volume constraints in IMRT optimization. Method and Materials: A formal mathematical framework for analysis of the effects of multiple gEUD‐type constraints on the resulting DVH is proposed. The framework relies on interpreting DVH as a cumulative probability distribution of the underlying “random”, i.e., unknown variable, representing dose to a voxel. Consequently, the analysis of the effects of gEUD‐based constraints on DVH is rephrased in terms of the effects of moments of the random variable on its distribution, which corresponds to a well‐recognized moment problem in mathematics. Results: Given a set of gEUD‐type constraints, we demonstrate how to compute the worst —in a sense of generating the largest volume ratio receiving a fixed dose— dose‐volume distributions that satisfy these constraints. A generalization of gEUD‐based constraints, the Generalized Moment Constraints (GMC's), is proposed, with the rationale behind GMC's to provide more modeling flexibility for IMRT optimization. A potential applicability of the approach is discussed. Applicability analysis is based on proximity of a family of dose distributions, satisfying a fixed set of GMC's, to the desired “ideal” dose distribution. We use a hypothetical prostate cancer patient's rectum as an example. Conclusion: The newly proposed convex GMC‐based dose‐distribution modeling framework has a potential to serve as a viable alternative to partial volume constraints in IMRT optimization, at least for some critical structures. To name a few potential benefits of our approach, we mention global solution to IMRT optimization problem that minimizes proximity of a physically deliverable plan to the desired “ideal” physician‐prescribed plan, and better control over the resulting dose distributions. Further investigation of the approach is required and is ongoing.
35(2008); http://dx.doi.org/10.1118/1.2962745View Description Hide Description
Purpose: Functional disorders of the brain, such as dystonia and neuropathic pain, may exhibit poor response to medical therapy. The Globus Pallidus pars interna (GPi) and the medial‐thalamus are potential targets for surgical or stimulation interventions in these two pathologies, respectively. Radiosurgery is an attractive non‐invasive technique to treat dystonia or pain through a radio‐induced lesion to create the internal pallidotomy (IP) and medial thalamotomy (MT), respectively, but does not provide the possibility of confirming the target location through microelectrode recording. Computer‐aided atlas‐based target identification may provide a tool to maximize the anatomic detection of the target. The aim of this work is to show the feasibility of atlas‐based functional radiosurgery and evaluate the early results. Method and Materials: The coordinates of GPi, centro‐median (CMN) and medio‐dorsal (MD) nuclei were identified in the Talairach‐Tournoux atlas and transformed to the corresponding regions of the Montreal Neurological Institute (MNI) electronic atlas. Binary masks describing the target nuclei were created. The MNI electronic atlas was deformed onto the patient T1‐MRI by applying an affine transformation followed by a local nonrigid registration. The obtained deformation field was then applied to the target masks, which were superimposed on patient T2‐MRIs. The radiosurgical targets were thus identified on the patient T2‐MRI registered on the treatment CT. Results: Two patients affected by dystonia and neuropathic pain were treated through IP and MT (120 Gy prescribed at 65% and at 77%, respectively) delivered by the CyberKnife. Six month after the treatment, T2‐ and contrast‐enhanced‐T1‐MRIs showed edematous regions around the targets, whose correct lesion placements were confirmed through DW‐MRIs. Complete resolution of dystonia and pain was obtained within 6 months with consequent discontinuation of medical therapy so far required. Conclusion: This work shows that atlas‐based target reconstruction is a valid tool to maximize target detection during functional radiosurgery.
WE‐D‐351‐02: Feasibility Study of Frameless Angiographic Localization of Arteriovenous Malformations (AVMs) for Image‐Guided Radiosurgical Interventions35(2008); http://dx.doi.org/10.1118/1.2962746View Description Hide Description
Purpose: We investigated the feasibility of novel approach for frameless 3D AVM localization within the coordinate space of an angiographic cone‐beam CT(CBCT) dataset. The localization is based on characterization and calibration of 2D digital subtraction angiography(DSA) and CBCT acquisition modes available on a commercial flat panel detector C‐arm neuroangiography system. This localization method in combination with image‐guided Cyberknife delivery could provide consistent approach to frameless AVM radiosurgery.Method and Materials: We introduced preset AVM localization protocols comprising anterior‐posterior (AP) and lateral (LAT) DSA series combined with CBCT without couch displacement between acquisitions. The AP and LAT C‐arm positions were selected by matching AP/LAT phantom images to corresponding projections from the CBCT series. We subsequently evaluated the accuracy and the reproducibility of the 2D–3D correspondence for the various protocols by imaging a CBCTcalibration phantom with embedded markers. Paired marker centers were automatically extracted in the AP/LAT images and the corresponding CBCT acquisition frames. The maximum and the mean distance between the marker centers were calculated as a metrics for 2D‐to‐3D correspondence accuracy. The reproducibility was investigated by repeating the imaging protocols in different sessions and displacing the C‐arm between the sessions. Results: The accuracy of the DSA‐CBCT correspondence depended on the protocol and the C‐arm position for the DSA acquisitions within the protocol. For further clinical investigations we retained a protocol with an AP and left LAT DSA modes that reproducibly resulted in maximum/mean paired marker distance of 0.68 mm and 0.4 mm (AP) and 0.29 mm and 0.15 mm (left LAT) in the C‐arm isocenter plane. Conclusion: With proper patient immobilization, frameless AVM localization within the angiographic CBCT dataset based on calibrated AP and left LAT views is feasible with an uncertainty of 0.7–0.8 mm. Conflict of Interest: This work is supported by Siemens Medical Solutions.
35(2008); http://dx.doi.org/10.1118/1.2962747View Description Hide Description
Purpose: Age‐related macular degeneration (AMD) is a leading cause for vision loss for people over the age of 65 in the United States. There are a number of treatments used to help slow and in some cases stabilize the process of AMD, but these require frequent invasive injections into the eye. This paper presents a potential radiotherapytreatment used to destroy the leaky vasculature while minimizing dose to surrounding tissues, allowing for combination therapy to treat AMD. Method and Materials: An adult male head phantom was modeled with Rhinoceros 4.0 using ICRP Publication 89 reference values. The head phantom was voxelized and modeled using the MCNPX radiation transport code with a photon beam treatment operated at 100 kVp. Doses to the macula target, lens, optic nerve, brain, thyroid, and salivary glands were tabulated using MCNPX at 8 beam angles. These angles can be characterized using a spherical 3D polar coordinate system with the macula at the origin, the z‐axis as the axis of vision, a radius of 13 cm, a polar angle of 30 degrees, and azimuth angles in 8 increments of 45 degrees. Results: For each beam angle, the doses to the organs of interest were several orders of magnitude less than the 8 Gy dose to the macula target and significantly lower than the thresholds for serious implications. Conclusion: The preliminary results bode well for the proposed radiotherapytreatment.Dose Volume Histograms (DVH) will be tabulated using MCNPX so that the maximum dose to localized portions of each organ of interest can be determined. The safest beam angle for treatment will be determined from the results of this study.
35(2008); http://dx.doi.org/10.1118/1.2962748View Description Hide Description
Purpose: To quantify the background absorbed dose to the uninvolved brain as a function of target volume for varying isocenter spacing and collimator sizes using Perfexion™. Method and Materials: Multiple treatment plans were generated for a series of spherical targets, using a sphere‐packing algorithm to automatically determine the isocenter coordinates. Target volumes investigated ranged from 2.1 cm3 (diameter=16 mm) to 33.5 cm3 (diameter=40 mm), with isocenter spacing for each volume ranging between 3.2 mm and 8 mm. The dose was determined for each scenario at fixed distances from 3 cm to 7 cm from the edge of the target volume and expressed as a percentage of the dose to the target edge. Dose was calculated using Gamma Plan (v8.0) and measured using EBT Gafchromic film inserted in a plastic spherical phantom (8 cm radius). Results: No statistically significant difference in background dose was found between isocenter spacing levels. Relative dose at each point from the edge of the target increased with increasing target volume. Using only 4‐mm collimated isocenters, the background dose at 7 cm from the target edge was 0.5% (23 isocenters) and 4.6% (226 isocenters) for 2.1 cm3 and 33.5 cm3 targets, respectively. For a given volume size, background dose was higher when using 16‐mm compared to 4‐mm collimated isocenters. Using only 16‐mm collimated isocenters, the average dose rate at 7 cm from the target edge was 2.3% (1 isocenter) and 6.8% (9 isocenters) for 2.1 cm3 and 33.5 cm3 targets, respectively. Conclusion:Treating larger target volumes with Perfexion™ results in larger background dose to uninvolved brain, irrespective of what spacing is used between isocenters. The use of 4‐mm collimated isocenters yields lower background dose levels than 16‐mm collimated isocenters for equivalent target volumes. Conflict of Interest: Research sponsored by Elekta Instruments AB.
35(2008); http://dx.doi.org/10.1118/1.2962749View Description Hide Description
Purpose: Linac‐based radiosurgery is performed using arcs delivered at different couch angles. The goal of this study is to present the idea of track‐based, instead of arc‐based planning. Various beam tracks can be delivered to a given isocenter if the couch and gantry are rotating simultaneously. Such a technique would be a modification of the McGill technique introduced in 1987. Method and Materials: In theory, track‐based planning should have advantages over the arc‐based planning because every set of acceptable arc‐based plans is a subset of all acceptable track‐based plans. The total‐arc‐degree parameter in arc‐based planning would be replaced by the track length over a sphere with a unity radius. In order to illustrate track‐based planning, several different plans using large number of 10‐degree arcs and appropriate couch rotations between adjacent arcs were generated. The dose distributions were calculated using the XKnife TPS (Radionics, Boston, MA). Several treatment plans were created for several targets shapes and sizes. A track‐based plan for a lesion with AP elongation (the most difficult ellipsoid target) was compared with two conventional single isocenter plans (15‐mm cone). Results: The three plans for the AP‐elongated lesion achieved the same conformality (PITV 1.4). However, the track‐based plan delivered significantly lower doses to the surrounding normal brain defined by the volume receiving 20% and 50% of the prescription dose (20.8 cc & 4.7vs. 32.3 cc & 5.6). The size of the isodose lines was also smaller for the track‐based plan (5.9 cm & 3 vs. 19.7 cm & 4.2). Similarly, the track‐based planning achieved similar conformalities with all targets. Conclusion: There is no doubt that the track‐based planning will be more flexible and offer a greater variety of solutions compared to arc‐based planning. Track‐based planning could be the future of automated forward or inverse planning.
35(2008); http://dx.doi.org/10.1118/1.2962750View Description Hide Description
Purpose: We present an automated method of Linac mechanical quality assurance (QA) for MLC‐based Stereotactic Radiosurgery. The test is designed to determine 3D deviation of the Linac beam central axis defined by the center of the collimator aperture defined by the MLC leaves from the machine isocenter defined by the small tungsten sphere target. Method and Materials: The automation is achieved using Portal Imager as a 2D detector. Based on the approach of Winston‐Lutz Alignment Test, we monitor the alignment of the isocentrically‐placed tungsten sphere target, mounted on a metal stem, with the center of the MLC‐defined small square field. Multiple projection images are acquired for a specified set of gantry angles. After acquisition, the EPIDimages are saved in DICOM format by the imager's software for subsequent automated analysis on a Windows‐based PC running MATLAB®. Results: The center of the sphere deviated by 0.75 ±0.17 mm (1SD) from the MLC defined field center. The lateral, longitudinal and vertical sphere displacements were [0.66±0.17, 0.20±0.18, 0.217±0.13] mm, respectively. The average extent of the MLC motion in the direction parallel to the leaf travel was within 0.2 mm maximum span, while the carriage motion in the perpendicular direction ranged from 0.2 mm for 5×5cm2 field to 1 mm motion for 2×2cm2 field. All results were automatically processed using GUI. Conclusion: Our test is capable of sub‐millimeter sensitivity to the MLC leaf position and carriage position, as well as gantry sag and carriage shift due to the effect gravity at multiple gantry angles. The system employs a graphic interface, which is intended to facilitate the frequent clinical use of the presented QA technique. Conflict of Interest: This work is partially supported by Varian Medical Systems (PaloAlto, CA).
WE‐D‐351‐07: Daily CT Guidance in Liver Stereotactic Body Radiation Therapy; Impact of a Non‐Rigid Patient Anatomy On OAR Dose35(2008); http://dx.doi.org/10.1118/1.2962751View Description Hide Description
Purpose: To investigate the influence of daily tumor‐based setup corrections, derived from pre‐treatment CT‐scans, on the organs‐at‐risk (OARs) dose distributions in stereotactic body radiation therapy(SBRT) of livertumors.Method and Materials: Fifteen patients diagnosed with liver metastases and treated with SBRT were included in this study. Patients were positioned in a stereotactic body frame and abdominal compression was applied to decrease the respiratory tumor displacement. A total dose of 37.5 Gy at the 65% isodose was delivered in three fractions. Sixty CT data sets, corresponding to the planning and consecutive treatment days, were reviewed. Relevant OARs were delineated on all CT sets. Daily 3D dose distributions were calculated using the daily CT sets, both without and with taking the clinically applied setup correction into account. Dose‐volume histograms and relevant dosimetric parameters for the PTV and all OARs were calculated. Results: Large shape and volume variations were seen for OARs (especially the oesophagus, duodenum and stomach), causing large variations in the dose to these organs. In 27% of the treatment fractions, it was seen that after setup correction the dose to an OAR exceeded the constraints, while the planning was within constraints; in one case, the constraint was exceeded by a factor of two. Setup correction yielded a significant increase in PTV coverage, but for the OARs no significant dosimetric effect was seen. Conclusion: Daily tumor‐based CT‐guidance is effective to increase PTV coverage, but does not significantly affect dose to the OARs. This is due to their non‐rigid motion and volume variations. Both may cause substantial OAR constraint violations, even after set‐up correction. However, for most fractions the dose to the OARs is within constraints. To obtain optimal sparing of OARs, adaptive treatment explicitly accounting for non‐rigid anatomy may be required, especially when the tumordose is escalated.
WE‐D‐351‐08: Integrated Subvolume Boosting Technique to Reduce Normal Tissue Irradiation During SBRT Treatment Planning Using Helical Tomotherapy35(2008); http://dx.doi.org/10.1118/1.2962752View Description Hide Description
Purpose: To characterize the reduction in normal tissue volume irradiated in a SBRT treatment plan by delivering an inhomogeneous dose to the PTV—via integrated subvolumes prescribed to increasing doses above prescription dose—instead of a homogeneous dose using helical tomotherapy. Method and Materials: Three patients were retrospectively planned using helical tomotherapy with two treatment plans: (1) homogenous plan delivering a uniform prescription dose to the PTV and (2) inhomogeneous plan delivering a non‐uniform dose to the PTV. Subvolumes were created by volumetrically contracting the PTV by 4mm resulting in three subvolumes within the PTV (PTVi, PTVii, and PTViii). Dose prescriptions to the various targets were as follows: Homogenous: PTV: 95% of volume to receive 60.0Gy; Inhomogeneous: PTViii: 20% of volume to receive 80.0Gy, PTVii: 50% of volume to receive 75.0Gy, PTVi: 50% of volume to receive 70.0Gy, PTV: 95% of volume to receive 60.0Gy. The homogenous and inhomogeneous plans were compared on the basis of normal tissue volume reductions to various dose levels assuming identical target coverage of the PTV. Results: A significant reduction in the volume of normal tissue irradiated to high doses was achieved for all patients. Average reduction of V90%, V80%, V50%, and V20% were 18, 15, 9, and 7%, respectively. It was noted that as the dose level decreases, the percent volume reduction between the homogeneous and inhomogeneous plans decreases. Conclusion: The study served to quantify the volumetric reduction of normal tissue irradiated during lungSBRT treatments with helical tomotherapy when delivering non‐uniform doses to the PTV as compared to uniform doses. Our results indicate that an inhomogeneous dose distribution generated by the creation of integrated subvolumes prescribed to higher doses within the PTV significantly reduces the volume of normal tissue irradiated, especially at higher doses (>50% of prescription dose).
WE‐D‐351‐09: Target Delineation, Reposition, and Dose Delivering Accuracies in CBCT‐Guided Stereotactic Radiotherapy of Small Lung Tumors35(2008); http://dx.doi.org/10.1118/1.2962753View Description Hide Description
Purpose: To use daily cone‐beam computed tomography(CBCT) for assessment of image‐guidedstereotactic radiotherapy (IGSR) of early‐stage non‐small cell lungcancer(NSCLC).Method and Materials: Patients were CT simulated in a body‐frame with or without an abdominal compression device. PET‐CT scans, if available, were fused with the simulation CT for target delineation. Daily pre‐treatment CBCTs were acquired with the Elekta XVI system, and the lungtumors were manually aligned with the simulation CTimages at the treatment machine. The shifts required to align lungtumors were recorded. Later, the CBCTs were imported into our treatment planning system and the treatment isocenter was registered with the planning CT isocenter, accounting for the daily pretreatment table shifts. GTVs and PTVs for every treatment session were redefined on the daily CBCTimage sets. Individual and cumulative DVHs for the daily treatments were calculated and compared to the original planning DVHs. Results: Using CBCT‐guidance, daily setup errors ranging from 1 to 3 cm had been corrected, that would have altered the mean dose to GTV by >30%. We found that even with daily CBCT‐guided table shifts to correct setup error, the dose to GTV could vary by 20% due to daily changes in shape of the lungtumor, when no margin was added from GTV to PTV. This daily dose variability was reduced to < 3% if PTV included a 5 mm margin around GTV. Conclusion: The interfractional setup errors caused the greatest amount of potential geographic miss of GTV, which was almost entirely eliminated by daily CBCT guidance. The underdosing of GTV due to deformation of the lungtumor could be overcome by expanding GTV by 5mm in all directions to define the PTV.