Search

Purpose: The intention of this study is to determine whether previously observed large external surrogate residual motion at end‐or‐inhale (EOI) translates into large tumor residual motion, and if improving the reproducibility at this phase can lessen the internal residual motion. Method and Materials: We simulate gated treatment at the EOI phase, using a set of recently measured internal/external correlated patient data. The 3D locations of internal fiducial markers placed near the target are tracked in real‐time with stereoscopic x‐ray fluoroscopy. An external surrogate respiratory gating system is synchronized with the fluoroscopic unit so that the log files contain the three‐dimensional marker position and the abdominal surface position at every time point. The internal and external measurements are taken even when the MV beam is gated off, throughout each treatment, so large amounts of internal/external‐correlated data were collected. Results: We found that under free‐breathing conditions the residual motion of the tumors is much larger for EOI phase than for end‐of‐exhale (EOE) phase. The mean value of residual motion at EOI was found to be 2.2 mm and 2.7 mm for amplitude and phase‐based gating, respectively; and, at EOE, 1.0 mm and 1.2 mm for the same quantities. However, the residual motion in the EOI gating window is correlated well with the reproducibility of the external surface position in the EOI phase. Using the results of a published breath‐coaching study, we deduce that the tumor residual motion at EOI would approach that at EOE under breath‐coaching conditions. Conclusion: We conclude that the same reproducibility of tumor location can be achieved at EOI as at EOE if breath coaching is implemented. Based on these results, we believe that inhale gating is preferable to exhale gating as long as proper margins are employed and breath coaching is performed.

Purpose: Respiratory gating based on external surrogates may have large uncertainty due to the unstable correlation between external and internal signals. The purpose of this work is to use the image guidance techniques to improve the precision of external gating. Method and Materials: The image‐guided gating procedure developed in this work including the following key components: 1) Varian RPM system for monitoring patient's respiration, 2) breath coaching technique to produce a reproducible breathing pattern intra‐ and inter‐fractionally, 3) 4D CT simulation to acquire the accurate tumor geometric information in the gating window, 4) gated on‐board kV x‐ray imaging to align the target position in the gating window, 5) gated treatment delivery, and 6) kV or EPID fluoros for treatment verification. Results: This procedure has been applied to liver and lung patients. Here we use a liver patient with implanted fiducial seeds as an example. The patient had a peak‐to‐peak motion of 2 cm under free breathing, which was reduced to less than 5 mm when gated at 30% duty cycle. The patient's breathing pattern was greatly improved with the coaching technique. The fiducial seeds detected in the gated on‐board x‐ray images were used for patient alignment, which indicated a 5 mm systematical error and ±5 mm random error in all three directions. The treatment was verified using the EPID in cine mode and it was found that the intra‐fraction residual seed motion was about 5 mm and the inter‐fraction seed motion was within +8.3/−4.5 mm in the SI direction. Conclusion: The utilization of image guidance techniques in simulation, patient setup, and treatment verification, along with breath coaching technique, can greatly increase the precision of the gating treatment based on external surrogates. Conflict of Interest: This work was partially supported by a research grant from Varian Medical Systems, Inc..

In cancer therapy, effective delivery of the therapeutic plays a critical role in determining the success of the treatment planning system. In last decade, there is a major thrust in developing novel nanoparticles based delivery systems to improve the therapeutic benefits in cancer. An efficient delivery system should be able to target the diseased site with minimal systemic toxicity and a slow sustained release of the drug. Here, we have fabricated a nanoparticles based smart ‘INCeRT’ (Implantable Nanoplatform for Chemo-Radiation Therapy) implant for localized delivery of radiosensitizing nanoparticles/ drugs in prostate cancer in conjunction with brachytherapy. These implants are physically similar to the clinically used brachytherapy spacers but have the added capability of imaging and local drug delivery. We have fabricated INCeRT spacers with biocompatible and biodegradable polymer, PLGA (poly-(lactide-co-glycolide)) impregnated with nanoparticles encapsulating imaging probe (Cyanine 7.5) and chemotherapeutic drug, docetaxel (DTX). Using a similar approach, we have also fabricated PLGA spacers impregnated with high Z (atomic number) gold nanoparticles (Hi-Z-CuRE: High Z-Customizable Radiotherapy Enhancement) for effectively boosting the radiation dose locally. The morphology, composition and nanoparticle's distribution inside the spacers was studied by SEM (scanning electron microscopy) and EDS (Energy-dispersive X-ray spectroscopy). Further, preliminary in vivo imaging experiments with subcutaneous prostate cancer tumored mice implanted with INCeRT spacers showed a size dependent diffusion of nanoparticles from the spacers in the tumor matrix. Also, from in vivo therapeutic studies with these spacers showed a sustained and slow release of the DTX from the spacers and showed a better response in suppressing the tumor as opposed control mice with saline injections. Further experiments for studying the combined chemo-radiation therapy are underway. This new treatment approach would be of crucial benefit for patients with local relapse who require salvage radiotherapy but have reached their radiotherapy normal tissue dose limits. This work was supported partially by NSF-DGE-0965843, HHS/1U54CA151881 CORE1, 1R03 CA164645-01 and a seed grant from the BWH Biomedical Research Institute. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A82. Citation Format: Rajiv Kumar, Jodi E. Belz, Stacey Markovic, Houari Korideck, Wilfred F. Ngwa, Mark Niedre, Ross I. Berbeco, Robert Cormack, Mike G. Makrigiorgos, Srinivas Sridhar. Localized tumor delivery of radiosensitizers and chemotherapeutics using ‘INCeRT’ implants. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A82.

DVHs were generated and compared. Monte Carlo calculations were performed to simulate ISA and to verify plan robustness. Results: On average the total number of seeds compared to the standard plan is 32% and 70% less for hemi and ultra focal plans respectively, whereas the average needles were 38% and 58% less for hemi and ultra focal respectively. Target volume coverage was adequately achieved for all plans with V100 above 97.8%. D90 was increased by 19% and 69% for hemi and ultra focal plans respectively. Meanwhile, D10 of urethral dose was reduced by 7% and 55% for hemi and ultra focal plans respectively. D2cc of rectum was reduced by 28% and 60% for hemi and ultra focal respectively. Plan robustness shows that a random shift in seed position of 4 mm would result in V100 and D90 dropping by 7.2% and 14%, respectively, for standard plans, with a corresponding reduction of 7.3% and 20% for hemi plans and 5% and 32% for ultra focal plans. Despite the fact that seed density is higher for the hemi and ultra focal approaches, ISA has similar effect on DHV parameters. For example, V100 and D90 are reduced by less than 1% and 3%, respectively, for all types of plans. Conclusions: Treatment planning for hemi and ultra focal options is feasible. Dose constraints are easily met with a notable reduction to organs at risk such as the bladder, urethra and rectum. Treating smaller targets, such as for ultra focal, makes seed positioning more critical. This should be taken into consideration during treatment planning. ISA has the same effect on DVH calculations for standard, hemi and ultra focal plans. Author Disclosure: B. Al-Qaisieh: G. Consultant; GE Oncura Ltd. J. Mason: None. A. Henry: None. P. Bownes: None. L. Dickinson: None. A. Hashim: None. M. Emberton: None. S. Langley: None.

Purpose: To provide real-time lung tumor motion estimation during radiotherapy treatment delivery without the need for implanted fiducial markers or additional imaging dose to the patient. Methods: 2D radiographs from the therapy beam's-eye-view (BEV) perspective are captured at a frame rate of 12.8 Hz with a frame grabber allowing direct RAM access to the image buffer. An in-house developed real-time soft tissue localization algorithm is utilized to calculate soft tissue displacement from these images in real-time. The system is tested with a Varian TX linear accelerator and an AS-1000 amorphous silicon electronic portal imaging device operating at a resolution of 512 × 384 pixels. The accuracy of the motion estimation is verified with a dynamic motion phantom. Clinical accuracy was tested on lung SBRT images acquired at 2 fps. Results: Real-time lung tumor motion estimation from BEV images without fiducial markers is successfully demonstrated. For the phantom study, a mean tracking error

Purpose: The XCAT phantom is a realistic 4D digital torso phantom that is widely used in imaging and therapy research. However, lung mass is not conserved between respiratory phases of the phantom, making detailed dosimetric simulations and dose accumulation unphysical. A framework is developed to correct this issue by enforcing local mass conservation in the XCAT lung. Dose calculations are performed to assess the implications of neglecting mass conservation, and to demonstrate an application of the phantom to calculate the accumulated delivered dose in an irregularly breathing patient. Methods: A displacement vector field (DVF) between each respiratory state and a reference image is generated from the XCAT motion model and its divergence is calculated and used to correct the lung density. A series of phantoms with regular and irregular breathing (based on patient data) are generated and modified to conserve mass. Monte Carlo methods are used to simulate conventional and SBRT treatment delivery. The calculated dose is deformed and accumulated using the DVF. Results from the mass-conserving and original XCAT are compared. A 4DCT is simulated for the irregularly breathing patient, and a 4DCT-based dose estimate is compared with the accumulated delivered dose. Results: The presented framework successfully conserves mass in the XCAT lung. The spatial distribution of the lung dose was qualitatively changed by the use of a mass conservation in the XCAT; however, the corresponding DVH did not change significantly. The comparison of the delivered dose with the 4DCT-based prediction shows similar lung metric results, however dose differences of 10% can be seen in some spatial regions. Conclusions: The XCAT phantom has been successfully modified so that it conserves lung mass during respiration, enabling it to be used as a tool to perform dose accumulation studies in the lung without relying on deformable image registration. Neglecting mass conservation can result in erroneous spatial distributions of the dose in the lung. Using this tool to simulate patient treatments reveals differences between the planned dose and the calculated delivered dose for the full treatment. The software is freely available from the authors.

Purpose: This project sought to create a patient specific dynamic phantom setup utilizing 3D printing to fabricate realistic target volumes derived from planning CTs.Materials/Methods: The anthropomorphic phantom is a modified Alderson Phantom (The Dynamic Breathing Phantom by Radiology Support Devices). The original lung system was removed and replaced with a new foam rubber structure having lung‐equivalent density. Using patient CT data, tumor models were manufactured using a 3D printer. With the planning image set as a guide, the tumor was positioned within the anthropomorphic phantom to match the relative tumor positioning within the patient. The patient‐specific phantom setup was then imaged under both CT and MV fluoroscopy and compared to the previously acquired patient imaging. Results: These patient‐specific phantom configurations accurately reflect their respective patient's treatment planning CTs. MV fluoroscopy of the phantom demonstrated both accurate positioning when compared to patient DRRs and accurate recreation of respiratory motion as compared to 4DCT data. Conclusion: This patient‐specific anthropomorphic breathing phantom will allow for further realistic motion management studies with applications in both imaging and dose delivery.

Purpose: To develop a method of verifying delivered dose based on time‐varying 3D treatment images estimated from single kV projection images and a 4DCT‐based patient‐specific motion model. Methods: Delivered dose calculation is performed in two steps. In the first step 3D fluoroscopic (time‐varying) images are generated from single 2D kV projection images. In the second step a set of 3D fluoroscopic images are used to calculate the delivered dose. 3D fluoroscopic images are estimate by creating a patient‐specific lung motion model through deformable registration of 4DCT, and then iteratively optimization the parameters of the motion model. Using a modified digital XCAT phantom, the accuracy of the delivered dose calculated with our method was compared to the dose that would be calculated during routine 4DCT‐based treatment planning and to the actual delivered dose. The planning dose was calculated on an average intensity projection (AIP) image. The actual delivered dose is calculated based on the defined motion of the digital XCAT phantom. The same treatment plan was delivered to all three phantoms. The treatment plan was developed from the AIP and designed as a 3×18 Gy SBRT plan. Results: The dose calculated on 3D fluoroscopic images closely matches to the actual delivered dose. Both the calculated and delivered doses vary from AIP‐based planned dose. D95 tumor dose values of planned, delivered and calculated dose are 54.0, 59.0 and 58.4 Gy. Corresponding V20 lung values are 4.5%, 4.8% and 4.8% respectively. Conclusions: We introduced a novel method for delivered dose calculation. Our algorithm successfully generates a 3D fluoroscopic image from a single kV images and then uses these time‐varying 3D images to accurately calculate delivered dose. The project described was supported, in part, by an RSNA Research Scholar Grant and Award Number R21CA156068 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the RSNA, National Cancer Institute or the National Institutes of Health.

Purpose The aim of this study is to quantify and to compare the dose enhancement factor from gold nanoparticles (AuNP) to tumor endothelial cells for different concentrations of AuNP, and clinical MV beam configurations. Methods Tumor endothelial cells are modeled as slabs measuring 10 × 10 × 2 lm. A spherical AuNP is simulated on the surface of the endothelial cell, within the blood vessel. 6 MV photon beams with and without the flattening filter are investigated for different field sizes, depths in material and beam modulation. The incident photon energy spectra for each configuration is generated using EGSnrc. The dose enhancement in the tumor endothelial cell is found using an analytical calculation. The endothelial dose enhancement factor is defined to be the ratio of the dose deposited with and without AuNPs. Results It is found that clinical beam parameters may be chosen to maximize the effect of gold nanoparticles during radiotherapy. This effect is further amplified ∼20% by the removal of the flattening filter. Modulation of the clinical beam with the multileaf collimator tends to decrease the proportion of low energy photons, therefore providing less enhancement than the corresponding open field. Conclusions The results of this work predict a dose enhancement to tumor blood vessel endothelial cells using conventional therapeutic (MV) X-rays and quantify the relative change in enhancement with treatment depth and field size.

Purpose: To quantitatively assess the impact of real time tumor tracking on the delivered dose during thoracic radiotherapy in the presence of realistic breathing. Methods: The 4D XCAT digital torso phantom was used in this study. The phantom was created with irregular breathing motion based on observed patient tumor motion. A locally mass‐conserving framework was applied to modify the XCAT to enable dose accumulation. The first 6 seconds of phantom data were used to simulate 4DCT acquisition. Two SBRT treatment plans were developed to assess the effects of tracking. The first plan, simulating a standard SBRT treatment, used an ITV+5mm aperture from the maximum intensity projection of the 4DCT. This plan was normalized to deliver 54 Gy to 95% of the PTV based on the 4DCT dose. The second plan simulated couch tracking by defining apertures from a single 4DCT phase (GTV+2mm), and translating the phantom during the delivery to keep the tumor centered in the aperture. Delivered dose was calculated using Monte Carlo techniques for the subsequent 60 seconds of phantom data at 1‐second time resolution, and was deformed and accumulated on to a reference phase. Results: The tumor D95 was 55.5 Gy for the ITV‐based plan and 54.6 Gy for the tracking plan. For the ITV‐based plan, the lung V5 was 22.9%, V20 was 3.6% and the mean dose was 3.9 Gy. For the tracking plan, the V5 was 14.5%, the V20 was 1.8% and the mean dose was 2.5 Gy. Conclusion: A method for assessing the dosimetric impact of tumor tracking during respiration has been demonstrated using the digital XCAT phantom. For the phantom simulated in this study, the tracking and conventional SBRT plans both delivered the prescribed dose to the tumor, and the lung dose is substantially reduced with tracking. The project described was supported, in part, by an RSNA Research Scholar Grant and Award Number R21CA156068 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the RSNA, National Cancer Institute or the National Institutes of Health.

Purpose: Recent studies have cogently demonstrated major dose enhancement to tumor endothelial cells (EC) when targeted with gold nanoparticles (GNPs) during external beam radiotherapy (EBR). This study investigates the endothelial dose enhancement for FDA approved concentrations of platinum‐based nanoparticles: cisplatin nanoparticles (CNPs), Oxaliplatin nanoparticles (ONPs) and carboplatin nanoparticles (CBNPs) as potential alternative to GNPs. Methods: As in previous work for GNPs, analytic calculations were carried out to estimate the dose to the EC caused by radiation‐induced photoelectrons from platinum nanoparticles during EBR. Here, Monte Carlo generated 6 MVp energy spectra at depth of 20 cm (10 × 10 cm2 field) was employed. The endothelial dose enhancement factor (EDEF), representing the ratio of the dose to the EC with and without the presence of nanoparticles was calculated. The Result was then appropriately scaled to reflect the percentage of platinum in CNPs, ONPs, and CBNPs. For the first time, the investigated concentration range took into account a potential clinical scenario where the nanoparticles are released in situ from routinely used fiducials loaded with the nanoparticles. Results: As expected, the results showed increase in EDEF with increased concentration of the nanoparticles for both GNPs and platinum NPs, with the highest contribution to the EDEF coming from the lower kV‐energy part of the EBR beam spectrum. At FDA concentration limits of in situ released nanoparticles results revealed EDEF values of up to ca. 1.8 for 60 mg/g CNPs, 1.6 for 50 mg/g ONPs and 2.2 for 120 mg/g CBNPs. Conclusions: The results predict that substantial photon‐induced dose boost to tumor endothelial cells can be achieved by applying tumor vasculature‐targeted CNPs, ONPs or CBNPs during external beam radiotherapy. Such radiation boosting or vasculature dose painting could be customized to work complementarily with the chemotherapy effect of these nanoparticles for more effective treatment outcomes.

Purpose: To develop the 2D/3D registration of cine mode electronic portal imaging device (EPID) images acquired during radiotherapy treatment to the planning computed tomography (CT) images and combine it with relative, markerless EPID tumor tracking. Together the methods will provide an automatic absolute tracking between physician defined volumes such as the gross tumor volume (GTV) the treatment field. Methods: During treatment of lung SBRT cases, EPID images were continuously acquired. The relative motion was tracked using a markerless multitemplate algorithm whose accuracy was previously confirmed and assessed with manual tracking. Each image then underwent an intensity based 2D/3D registration to the planning CT. In order to minimize the effect of motion blur, the end‐of‐exhale phase of the four dimensional computed tomography (4DCT) was used. The volume was converted from Hounsfield units into electron density by a calibration curve and DRRs were generated for the beam geometry. Using normalized cross correlation (NCC) between the DRR and EPID image, the best in plane rigid transformation was found. It was then applied to contours in the planning CT, mapping them into the EPID image domain. The best breathing phase for the registration was found with a large set of patient data. Results: The success of 2D/3D registration proved accurate only over certain phases of the breathing cycle. By registering at this time and using relative tracking, we successfully track target volumes in the EPID images throughout the entire treatment delivery. Conclusions: Through the combination of relative tracking and phase dependent EPID/4DCT registration, it is possible to track clinical volumes on EPID images. This knowledge of tumor volumes relative to the treatment field provides powerful information for future applications like motion management, adaptive radiotherapy and delivered dose calculations.

Purpose: Lung toxicities encountered during the early lung SBRT clinical trials have been largely abated by adhering to strict planning limits on lung dose and PTV size. This restricts the maximum safely treatable GTV size in the presence of respiratory motion. We hypothesize that real‐time motion management (e.g. tracking) will enable the maximum GTV size to increase, thereby making more patients eligible for SBRT. Methods: Consensus maximum PTV size (50 cc) was determined from recent clinical trials and our own clinical experience. PTV as a function of GTV was found for four treatment scenarios: 1 cm motion, no tracking; 0.5 cm motion, no tracking; no motion, no tracking; and, arbitrary motion, with tracking. We included a 0.5 cm/1.0 cm asymmetric CTV to PTV margin for no tracking (RTOG recommendation) and a 0.3 cm margin for tracking, as suggested by the results of Rottmann (2010). No GTV to CTV margin is included. A large population study is used to determine the increase in numbers of patients eligible for SBRT. Results: PTV 250%, going from a limit of 13 cc, for tumors moving more than 5 mm, to 33 cc for all motions. This translates to a ∼50% increase in the numbers of patients eligible for lung SBRT. Conclusions: Restrictions on PTV size limit patient eligibility for SBRT, excluding patients with smaller GTVs and respiratory motion. Real‐time motion management will enable a substantial increase in the numbers of patients eligible for SBRT. The project described was supported, in part, by Award Number R21CA156068 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.

Purpose: The XCAT phantom is a realistic 4D digital torso phantom that is widely used in imaging and therapy research. However, lung mass is not conserved between respiratory phases of the phantom, making detailed dosimetric simulations unphysical. A framework is presented for enforcing local mass conservation in the XCAT lung and is combined with a dose calculation algorithm to create a tool for dose accumulation studies. As an application of this technique, the impact of irregular breathing on 4DCT-based predictions of accumulated lung dose is assessed. Methods: A displacement vector field (DVF) between a respiratory state and a reference image is directly determined from the XCAT motion model and its divergence is used to correct the lung density. A series of phantoms with regular and irregular breathing (based on patient data) are generated and corrected. Monte Carlo simulations of conventional and SBRT treatments are performed, and the dose is deformed and accumulated. A 4DCT is simulated for the irregular breathing patient, and the 4DCT dose calculation is compared with the full accumulated delivered dose. Results: The presented framework successfully conserves mass in the XCAT lung. The spatial distribution of the lung dose was qualitatively changed by the use of mass conservation; however the DVH does not change significantly. The comparison of the delivered dose with the 4DCT-based prediction shows similar lung metric results, however dose differences of 10% can be seen in different regions of the lung. Conclusion: The XCAT phantom has been successfully modified so that it conserves lung mass during respiration, enabling it to be used as a tool to perform dose accumulation studies in the lung without relying on deformable image registration. These studies can reveal potentially significant differences between the predicted dose and the actual delivered dose. The software is freely available from the authors. The project described was supported, in part, by an RSNA Research Scholar Grant and Award Number R21CA156068 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the RSNA, National Cancer Institute or the National Institutes of Health.

Purpose: The aim of this study is to quantify and to compare the dose enhancement factor from gold nanoparticles (AuNP) to tumor endothelial cells for different concentrations of AuNP, and clinical MV beam configurations. Methods: Tumor endothelial cells are modeled as slabs measuring 10 × 10 × 2μm. A spherical AuNP is simulated on the surface of the endothelial cell, within the blood vessel. 6 MV photon beams with and without the flattening filter are investigated for different field sizes, depths in material and beam modulation. The incident photon energy spectra for each configuration is generated using EGSnrc. The dose enhancement in the tumor endothelial cell is found using an analytical calculation. The endothelial dose enhancement factor is defined to be the ratio of the dose deposited with and without AuNPs. Results: It is found that clinical beam parameters may be chosen to maximize the effect of gold nanoparticles during radiotherapy. This effect is further amplified ∼20% by the removal of the flattening filter. Modulation of the clinical beam with the multileaf collimator tends to decrease the proportion of low energy photons, therefore providing less enhancement than the corresponding open field. Conclusions: The results of this work predict a dose enhancement to tumor blood vessel endothelial cells using conventional therapeutic (MV) x-rays and quantify the relative change in enhancement with treatment depth and field size.

Purpose: The treatment efficacy of radiation therapy for lung tumors can be increased by compensating for breath-induced tumor motion. In this study, we quantitatively examine a mathematical model of pseudomechanical linkages between an external surrogate signal and lung tumor motion. Methods: A spring-dashpot system based on the Voigt model was developed to model the correlation between abdominal respiratory motion and tumor motion during lung radiotherapy. The model was applied to clinical data obtained from 52 treatments (“beams”) from 10 patients, treated on the Mitsubishi Real-Time Radiation Therapy system, Sapporo, Japan. In Stage 1, model parameters were optimized for individual patients and beams to determine reference values and to investigate how well the model can describe the data. In Stage 2, for each patient the optimal parameters determined for a single beam were applied to data from other beams to investigate whether a beam-specific set of model parameters is sufficient to model tumor motion over a course of treatment. Results: In Stage 1, the baseline root mean square (RMS) residual error for all individually optimized beam data was 0.90 ± 0.40 mm (mean ± 1 standard deviation). In Stage 2, patient-specific model parameters based on a single beam were found to model the tumor position closely, even for irregular beam data, with a mean increase with respect to Stage 1 values in RMS error of 0.37 mm. On average, the obtained model output for the tumor position was 95% of the time within an absolute bound of 2.0 and 2.6 mm in Stages 1 and 2, respectively. The model was capable of dealing with baseline, amplitude and frequency variations of the input data, as well as phase shifts between the input abdominal and output tumor signals. Conclusions: These results indicate that it may be feasible to collect patient-specific model parameters during or prior to the first treatment, and then retain these for the rest of the treatment period. The model has potential for clinical application during radiotherapy treatment of lung tumors.

Purpose/Objective(s): Accurate treatment verification is of particular importance when delivering high doses per fraction, as in lung SBRT. In this study, we characterize and validate the performance of a real-time markerless lung tumor-tracking algorithm, using continuous MV portal imaging. MV portal imaging is advantageous, as beam’s-eye-view, images allow motion capture in both directions of steep dose fall-off, and the patient is not exposed to additional imaging dose. Furthermore, a markerless tracking system avoids the risk associated with implanting fiducial markers. Materials/Methods: We retrospectively analyzed the delivery of 9 lung SBRT treatments carried out at our institution. Radiation therapy was delivered with a 6 MV beam in 3 fractions to 54 Gy or in 5 fractions to 60 Gy, using 9-11 beams, including non-coplanar geometries. Treatments were monitored by continuous acquisition of portal images with a frame rate of 2 Hz. Soft tissue motion was tracked on the portal images with a multi-region algorithm developed for this purpose. For each field and each fraction a set of landmark candidates was automatically defined on the first image. The tracker then identified this set on each subsequent image by means of a similarity measure and calculated a centroid trajectory. In this way, the images were analyzed in a prospective manner to simulate in-treatment real-time analysis. The first 40 images of 54 fractions from 34 beam angles were used for comparison with expert human examiners. To minimize bias, the images were presented to the examiner in random sequence and each image was shown 3 times. Tracking error in a realistic phantom was determined as a baseline. Computation time was measured for real-time processing of the images. Results: In-treatment EPID images were successfully acquired for 9 lung SBRT patients with 3 or 5 fractions each (19,942 portal images), including non-coplanar geometries. By comparing with the manual tracking, the RMS tracking accuracy of the automatic marker-less algorithm was found to be (2.1 1.7) mm. Excluding the worst 9 series (out of 54) reduced the RMSE to (1.5 1.1) mm. If we define tracking success as having a RMSE

Purpose: The implementation of a real‐time adaptive therapy system. Automatic soft tissue localization (STiL) is used to drive DMLC adaptation maintaining the position of the dynamic treatment aperture relative to the tumor location during the entire breathing cycle. The STiL component utilizes electronic portal images and operates without the need for fiducial markers. The proposed system has the potential to improve treatment accuracy, dose conformity and sparing of healthy tissue.Methods: The system is implemented and tested on a clinical linear accelerator featuring an electronic portal imaging device(EPID) and a DMLC control system.EPIDimages are continuously acquired at a frame rate of 12.86 Hz. The STiL component processes the images in real‐time, sending its output ‐ the current tumor position ‐ to the DMLC component, which moves the aperture to that position. Image transfer, tumor position calculation and DMLC motion introduce a time lag between tumor position at acquisition time and at the time the treatment aperture reaches this position. We analyze this latency with a dynamic chest phantom driving a 1D sinusoidal motion (20mm superior‐inferior motion range and 4.5s period). We estimate the resulting average geometric systematic error in a clinical setting by driving the phantom with several patient traces (recorded from fiducial tracking during lungSBRT).Results: The individual geometric errors of the StiL and the DMLC component are each smaller than 1 mm. The overall system latency was found to be 210 ms. The average rms‐error of 11 patients traces (172 beams) was found to be (1.8 ± 0.8) mm with this latency. Conclusions: We have implemented a real‐time adaptive therapy system integrating automatic soft tissuetumor localization with DMLC adaptation of the treatment aperture. The functionality of the combined system was tested successfully and the systemic latency and resulting rms error measured. Varian Medical Systems Inc., NCI grant CA93626

Purpose: The internal target volume (ITV) accounts for uncertainties in tumor position and shape and is defined from images acquired with 4DCT. In this work, the locations of gold fiducial markers implanted in lungtumors of seven patients are used to represent tumor motion and investigate the role of the CTscanner geometry and scan duration on ITV definition. Methods: All of the simulations are geometric simulations performed in MATLAB (The Mathworks, Nattick,MA) and variations in image reconstruction are not considered. The CTscanner geometries simulated include a 4‐slice CTscanner (with an axial field of view of 10 mm) and an ‘ideal’ CTscanner geometry that covers all potential tumor positions in the same bed position (axial field of view of 100 mm). The scan durations were varied from 6 seconds per bed position to 30 seconds per bed position. For all of the simulations, the tumor was modeled as a sphere of diameter 25 mm. The ITVs were compared by calculating the mean ITV for each patient and scan parameters. Results: For all of the patients, the mean ITV increased as the scan duration increased. The maximum increase in mean ITV was 25%. Conclusions: This work highlights the potential benefit of modified CT protocols to determine the ITV. One example of a modified protocol would be to increase the scan duration (to decrease ITV uncertainties) while decreasing the mA of the scanner (to avoid giving extra radiation dose). The largest factor influencing the ITV is the scan duration (and not the scanner geometry).

Purpose: To quantify PET treatment response using a novel image registration approach, and to investigate the impact of 4D PET/CT pre‐ radiotherapy scan for predicting residual high‐uptake areas toward dose‐ boosting in adaptive radiotherapy.Methods: 3D and 4D FDG‐PET/CT scans were obtained from ten patients with non‐small cell lungcancer(NSCLC) for approximately two weeks before radiotherapy (pre‐RT) and approximately one month after therapy (post‐RT). The 4D scan was taken immediately following a 3D PET scan. The areas of high uptake in the pre‐ RT scans were aligned with those in the post‐RT scans using a novel rigid‐ penalized deformable registration. The global lung bronchi and chest wall were non‐rigidly transformed to minimize the respiration effects while the target lesion is rigidly transformed to preserve the volume and SUV values. A region of interest (ROI) was defined by a threshold of 40% of the maximum SUV for both 3D and 4D PET scans. The maximal SUV within the ROI were measured. Volumetric overlap fraction (VOF) was calculated as the overlap volume of pre‐RT and post‐RT target volumes divided by the residual volume in the post‐RT scans. Results: The maximal SUVs for 4D pre and post PET have a 16.3% increase compared to those in 3D PET. The pre‐to‐post maximal SUV variations showed 57% lesions has nontrivial residual uptake (SUVmax>2.5) for 3D PET, and 64% for 4D PET. The results of the VOFs showed that the residual metabolic‐active areas are mainly located within the primary tumor volume, 92% for 3D and 89% for 4D. Conclusions: Residual metabolic‐active areas have strong physiological and geometric correspondence to the primary tumor. With better signal recovery than standard 3D PET, high uptake areas of 4D PET pre‐RT can predict areas where residual metabolic activity will be found after radiotherapy.

Purpose: While real time imaging of treatment through an electronic portal imaging device(EPID) is a powerful tool to monitor treatment, limited field of view and lower contrast from an MV beam can make assessment difficult for physicians. This work will develops a method to register and project contour outlines for the internal target volume (ITV) and planning target volume (PTV) of lungtumor cases onto cine mode EPIDimages to help physicians in interpretation during treatment. Methods: A sequence of EPIDimages, acquired during treatment, was registered to treatment planning computed tomography(CT) by machine geometry and patient setup with cone‐beam computed tomography(CBCT). The planning CT was converted from Hounsfield scale to electron density by calibration curves of our CT simulator and digitally reconstructed radiographs(DRRs) were produced to match the EPID geometry, pixel for pixel. ITV and PTV structures as defined on the planning CT were then projected onto the DRRs. The DRRs were registered to the EPIDimages using cross correlation of a single template defined within the treatment aperture of each image. Once registered, the contours from the DRR were transferred to the EPID.Results: We were able to successfully register MV DRRs to EPIDimages and display the projected target volumes. Without introduced motion, geometric registration and CBCT guided patient setup up were sufficient to register the contours within a single pixel, as normalized cross correlations produced no additional shift. We expect the DRR/EPID registration to be an important step when looking at cases with substantial tumor movement. Conclusions: The visualization of target volumes provides a tool for physicians to interpret EPIDimages and assess treatment, especially in cases with tumor movement. The methods developed will serve as the basis for a clinical tool providing real time contours.

Purpose: Age‐related macular degeneration (AMD) is the leading cause of irreversible vision loss for people over the age of 60 in the United States. In this study the dosimetric feasibility of using goldnanoparticles (AuNP) as radiosensitizers to enhance stereotactic radiosurgery for neovascular AMD is investigated. Methods: Analytic calculations were carried out to estimate the nucleus dose enhancement factor (nDEF) due to photon‐induced photo‐ /Auger electrons from AuNP targeting neovascular AMD endothelial cells (EC). The nDEF represents the ratio of the dose to the nucleus with and without the presence of AuNP. As in previous studies, the EC is modeled as a slab of 2 μm (thickness) × 10 μm (length) × 10 μm (width) containing a nucleus of 5 μm diameter and thickness of 0.5 – 1 μm. The targeted AuNP are attached to the exterior of the EC. The nDEF was calculated for a range of feasible AuNP local concentrations (1–7 mg/g) using the clinically applicable 100 kVp x‐rays employed by the IRayTM system (Oraya Therapeutics Inc. Newark, CA), with total filtration of 0.75 mm Al and 0.8 mm Be. For comparison the nDEF for other energies: 80 kVp, 90 kVp, 110 kVp, and 120 kVp was also investigated. Results: For 100 kVp x‐rays, the results revealed nDEF values of 1.30 – 3.26 for the investigated concentration range of 1 – 7 mg/g, respectively. In comparison, for the same concentration range, nDEF values of 1.32 – 3.40, 1.31–3.33, 1.29 – 3.19, 1.28 – 3.12 were calculated for 80 kVp, 90 kVp, 110 kVp, and 120 kVp x‐rays, respectively. Conclusions: The results predict substantial dose enhancement to the sensitive nucleus of neovascular endothelial cells, targeted by AuNP during kilovoltage stereotactic radiosurgery. This suggests that AuNP may be employed as radiosensitizers to enhance therapeutic efficacy during radiosurgery for neovascular AMD.

Purpose: The suitability and performance of a mouse‐size MOSFET (‘Mousefet’) phantom is investigated for routine quality assurance (QA) of the Small Animal Radiation Research Platform (SARRP). Methods: The Mousefet phantom was constructed by custom integration of 5 micro MOSFETS, arranged in a quincunx pattern, within a miniature tissue equivalent phantom. The phantom was designed to facilitate SARRP QA tasks which may warrant daily evaluation: output constancy, isocenter congruency test, and cone beam computed tomography(CBCT)image geometric accuracy. For output constancy the phantom was irradiated with an open field and all 5 micro‐MOSFET readings taken. For the isocenter congruency test, an appropriate size collimator was used to irradiate one of the MOSFETs,imaged and positioned at CBCT isocenter. The acquired CBCTimage was then used to verify image geometric accuracy and other image quality parameters. Sample data from 10 days in a period of over one month was compared to reference measurements and evaluated for daily variation. Results: Output constancy measurements showed maximum daily variation of less than 3% for all MOSFETS, in consonance with observations from concurrent ion chamber measurements. The design of the Mousefet phantom allows the output check data to be used for prompt verification of beam energy and cone profile constancy. For the isocenter congruency test, it is demonstrated that the Mousefet Phantom can detect 0.3 mm deviations of the CBCT isocenter from the radiation isocenter. Meanwhile, results for CBCTimage geometric accuracy were consistently found to be within 2% of the expected value. Other CBCTimage‐quality parameters could also be assessed in terms of image intensity constancy, noise and image uniformity. Conclusions: Overall, the results establish the Mousefet phantom as a simple and time‐efficient multipurpose tool that could be employed effectively for routine quality assurance of the SARRP.

Nanoparticle formulations of gold have shown a tremendous potential in various biomedical applications. The use of high atomic number (Z) materials presents an attractive approach in enhancing the therapeutic efficacy of the radiation therapy. The high Z number for gold (Z=79) makes them an ideal candidate for radiosensitization enhancement. The ability of the anisotropic gold nanorods to sustain the resonating surface plasmon with minimal damping results in highly efficient two-photon induced photoluminescence imaging. Here we present the synthesis and in vitro characterization of PEGylated gold nanorods as efficient radiosensitizing agents which can be imaged using inherent two-photon photoluminescence without a conjugated fluorophore. We have synthesized gold nanorods with an aspect ratio of 2.5 and functionalized the surface with different ratios of methoxy and amine PEG for modulating the charge and imparting the functional groups. The cellular uptake behavior in prostate cancer cell line PC3 was studied and the results indicated a robust uptake of the amine functionalized nanorods. The quantitative estimation of the nanorods uptake was performed with BCA assay as amount of nanorods present in per microgram of protein. Two photon photoluminescence imaging also confirmed a robust cellular uptake of the amine functionalized nanorods. The radiation damage enhancement of the gold nanorods was confirmed using PC3 cells. The cells treated with gold nanorods were irradiated with kilovoltage X-rays. A γH2AX assay was used for quantitation of the DNA damage with and without nanorods as controls. The results indicated an increased highly efficient DNA damage in cells treated with gold nanorods. The outcome of this research will enable further application of targeted nanorods in advanced animal models as efficient radiation dose enhancement agents wherein two-photon photoluminescence imaging will provide a real time assessment of the therapeutic response and disease progression. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 1472. doi:1538-7445.AM2012-1472

Purpose: Imaging of lungcancer with 4D PET has recently received much attention in oncology diagnosis and staging. Independently studies have also shown the feasibility of dosimetric evaluation of delivery efficacy of lung stereotactic body radiotherapy(SBRT) using an EPID in cine mode. We propose a novel method to evaluate treatment delivery by quantifying dose deposited in the residual tumor using 4D PET and in‐treatment imaging. Methods: A two‐step dosimetric evaluation of treatment delivery for lungSBRT is proposed. First, delivered dose is calculated by motion‐corrected fluence. Megavoltage images are recorded throughout treatment delivery, and tumor positions are tracked using a multi‐template tracking algorithm for all fields and fractions. The delivered fluence is obtained by convolving the tracked tumor positions with the planned fluence. Utilizing the convolved fluence, the 3D delivered dose is obtained. Second, treatment delivery is evaluated utilizing a post‐therapy 4D PET scan. High FDG uptake regions are segmented as sites of residual disease at each phase of 4D PETimages. Residual structures are then fused with the planning CT using regional image registration. Relying on the pre‐computed delivered dose,dose volume histograms (DVHs) are obtained for the residual structures. Results: We validated the method using clinical data acquired from lungSBRT treatments. The planned and delivered DVHs were compared and verified on the target volume. The delivered dose to the residual tumor was calculated for each phase of 4D PET. For a patient with large tumor motion (1 cm), 10% deviation from prescription dose is observed at inhale phase of breathing. Conclusions: Dose delivered to residual disease provides critical evidence for evaluation of the efficacy of treatment delivery. 4D PET has a potential impact on dosimetric evaluation for lungcancer treatment.

Purpose: Gated (4D) PET/CT scans have been shown to be effective in recovering the signal lost due to respiratory motion. Ungated (3D) and 4D FDG‐PET scans obtained pre and post radiation therapy (RT) for lungcancer were compared to examine the role of 4D PET in assessing FDG uptake changes due to therapy. Methods: Sequential 3D and 4D FDG‐PET/CT scans were acquired approximately two weeks before (pre‐RT) and one month after (post‐RT) treatment for patients receiving definitive RT for lungcancer. Targeted lesions were identified for each patient, and a region of interest (ROI) was defined for each lesion on the 3D and each of the five phases of the 4D pre‐RT scans. The ROIs were generated using a 40% of the maximum standardized uptake value (SUV) threshold, and the maximum SUV for each lesion from the pre—RT scans was recorded. Pre‐RT ROIs were then used to locate the areas of the original FDG uptake on the post‐RT scans using rigid registration. A post‐RT ROI was generated, and the maximum post‐RT SUV was recorded. Results: Sixteen lesions were identified and analyzed for twelve patients. Comparison of pre‐RT to post‐RT 3D PET scans showed a median SUV decrease of 62% (range 36–89%), while comparison of pre‐RT to post‐RT 4D PET scans showed a median SUV decrease of 67% (range 30–89%). The mean difference between the SUV change on 3D versus 4D scans was 4.9% (range 0–15%, p=.07), and the highest quartile had an absolute difference greater than 9% in SUV change.Conclusion: When assessing SUV changes with PET scans done pre and post‐RT, a substantial proportion of lesions had showed differences of over 9% in measurement of SUV change when comparing 3D PET to 4D PET. Such differences could potentially influence classification of PET‐based response assessment.

Purpose: The goal of the Deep Inspiration Breath Hold (DIBH) technique is to avoid direct radiation to the heart while treating the left breast tissue. Accurate and reproducible treatment setup of the patient is very important to successfully achieve the benefits of the DIBH technique. The purpose of this work is to introduce a device named the Real Time Skin Surface Distance (RTSSD) that accurately tracks the patient position during therapy. Methods: Five left‐sided breast cancer patients were used for this study. The RTSSD device consists of a laser distance meter rigidly fixed to the treatment couch. At the time of simulation, the RTSSD device measures chest position at the sternum tattooed location during free‐breathing and DIBH CT scans. Prior to each radiation treatment, free breathing and DIBH distances are measured at the sternum tattoo location using the RTSSD device and compared to the ones measured at the time of the CT scans. If a difference larger than +/− 3mm is found, the patient is repositioned. KV orthogonal images are acquired during DIBH and compared to the reference digital radiographs for internal verification. During irradiation, the therapist monitors the RTSSD measurement and interrupts the treatment if it exceeds the 3 mm tolerance. Cine EPIDimages, acquired during irradiation, are used retrospectively to validate the localization. Results: The RTSSD device tracked the patient accurately during the DIBH treatment as confirmed by in‐treatment imaging. The cine EPIDimages showed agreement with reference DRRs within the +/− 3mm tolerance. Conclusions: The RTSSD device is non‐invasive, easy to operate and well tolerated by the patients. The device tracked accurately the distance from patient's sternum tattooed location during DIBH simulation and delivery. Imaging results confirmed that the device measurements were reliable to use for patient positioning during DIBH treatment.

Purpose: There has been recent interest in goldnanoparticles as vascular disruptive agents when combined with megavoltage x‐ray irradiation. In addition to the therapeutic benefits, the goldnanoparticles may also have a role as an alternative to implanted fiducials for improving localization for IGRT and real‐time tumor tracking. We demonstrate the feasibility of tracking a goldnanoparticlesolution of moderate concentration with kV x‐ray fluoroscopy. Methods: A 2.0 mg/ml solution of polymer coated goldnanoparticles was injected into 0.5 cm clinical bolus material. Additional build up with solid water was used to create a more realistic radiological scenario. A cylindrical gold fiducial (3 mm × 1 mm) was placed nearby to act as a reference surrogate. The phantom was set in sinusoidal motion with 2 cm amplitude and 4.0 s period. Images were acquired at 30 frames/sec with the kV on‐board imaging system of a Varian IX Clinac. The ratio of the image intensity in each target versus background was measured. Target tracking was performed using a single template matching approach and non‐clinical research software. Tracking error was quantified as the difference in location of each track point from the input sine function. Results: The intensity ratio in a single image of a sequence was 1.3 for the gold fiducial and 1.2 for the 2.0 mg/ml goldnanoparticlesolution. Tracking was maintained throughout the 25 seconds of target motion. RMS tracking error was 0.4 mm for the gold fiducial and 0.6 mm for the goldnanoparticles. The tracking score was smaller for the nanoparticles (mean: 0.63 +/− 0.02) than for the fiducial (mean: 0.89 +/− 0.6). Conclusions: Target tracking in kV fluoroscopy can be enabled by goldnanoparticles. A solution of 2.0 mg/ml can be tracked in a phantom with an accuracy similar to tracking a gold fiducial.

Purpose: MOSFETS were employed in this study for in situ assessment of radiation dose delivered by the Small Animal Radiation Research Platform (SARRP) to mouse anatomical target volumes: brain, left kidney, left lung,liver, pancreas and prostate. Method and Materials:MOSFET detectors were calibrated with conversion factors (CF) determined for each MOSFET as a function of depth in tissue equivalent material. Calibrated MOSFET was surgically implanted post‐euthanasia into the mouse target volume, with minimal organ disruption. Using CBCTimaging and robotic motion, the mouse was then positioned with implanted MOSFET at the isocenter of the SARRP. The depth of the MOSFET from the CBCTimage and corresponding CF were used in the SARRP dose calculation program to calculate the target dose for 220 kVp and 13 mA, without heterogeneity correction. Irradiations were conducted using a 0.15 cm copper filter. Three readings per MOSFET were obtained for each organ of three euthanized male mice. The measured dose was determined by averaging the MOSFET readings for that organ. The percentage difference between the measured and calculated dose was determined. Results: The results showed differences between the calculated and measured dose of between: +1.2 to 4.6 %, −3.3 to +4.8 %, +0.9 to +5.1 %, −5.9 to +2.5 %, ‐2.3 to +1.0 %, and −4.3 to +1.4 %, for the left kidney,liver, pancreas, prostate, left lung, and brain, respectively. Conclusions: The study indicates that, for 220 kVp x‐rays, the radiation dose delivered by the SARRP to characteristic target volumes in mice is within 6% of the calculated dose. The results provide a useful assessment of the need for tissue heterogeneity corrections in SARRP dose calculations for clinically relevant tumor model sites. The work also demonstrates the feasibility of using MOSFET dosimeters for in situ assessment of SARRP radiation dose delivery.

Purpose: To evaluate the effect of an adaptive gating system on treatment accuracy and delivery time as compared to gating based only on an external surrogate. Methods: Two studies were designed to simulate the process of treatment planning, patient setup, and treatment delivery for 1) external surrogate‐based; and 2) adaptive gating. External surrogate‐based gating used only chestwall motion to generate the respiratory gating signal, while adaptive gating updates the external/internal correlation periodically based on the location of implanted fiducial markers. Lungtumor trajectories with corresponding chestwall motions acquired from 8 patients with multiple days of radiation treatments. A total of 20,026 seconds of data from 163 treatment beams was used. The average target miss percentage is defined to be the percentage of beam‐on time that the target is thought to be within the gating window but is actually outside of it. The average duty cycle is defined as the beam‐on time divided by the sum of beam‐on and beam‐off time. Results: The average target miss percentage for adaptive gating was 17.6% (range: 8.0–23.6%) and 22.6% (range: 8.1–44.5%) without adaptation. The magnitude of the average miss was 1.7 mm (range: 0.6–4.8 mm) with adaptation and 1.8 mm (range: 0.6–4.3 mm) without adaptation. The average miss percentage for misses of over 5 mm was 0.6 % (range: 0.0–3.3%) for adaptive gating and 0.9 % (range: 0.0–3.7%) without adaptation, suggesting that additional planning margins of 5 mm would mitigate most gating errors with either method. The average duty cycle using adaptive gating was 33% (range: 23–39%), compared to 45% (range: 35–57%) without adaptation. Conclusion: We have shown that adaptive gating can improve the accuracy of gated treatments, at the cost of longer treatment times. However, both adaptive and non‐adaptive techniques perform well if an additional planning margin of 5 mm is used.

Purpose: To compare the effect of respiration on delivered dose for different treatment techniques under realistic clinical conditions. Methods and Materials: Twenty MOSFETs were inserted into a soft resin tumormodel positioned in a anatomical respiring phantom. The phantom motion was programmed using the trajectory of a real patient. A 4DCT was taken, and treatment plans created using different treatment techniques: Conformai plan (Eclipse), step‐and‐shoot IMRT (Pinnacle), step‐and‐shoot IMRT (XiO), dynamic IMRT (Eclipse), complex dynamic IMRT (Eclipse), hybrid IMRT (60% conformai, 40% dynamic IMRT [Eclipse]), VMAT (single arc [Eclipse]), VMAT (double arc [Eclipse]) and complex VMAT (Eclipse). The complex plans were created by artificially pushing the optimizer to give complex MU sequences. Each plan was treated 5 times (10 times for VMAT), each irradiation starting at a random point in the respiratory cycle. The effect of fractionation was calculated by randomly summing the measured doses. The maximum deviation for each point per fraction, and the probability of 95% of the tumor having dose deviations less than 2% and 5% were calculated as a function of number of fractions. Results: After 5 fractions, measured dose deviations were less than 2% with 95% probability for all plans except the complex IMRT, step‐and‐shoot IMRT (XiO), complex VMAT, and single‐arc VMAT. Reducing the dose rate of the complex IMRT plans to 200MU/min reduced the dose deviations to less than 2%. Dose deviations were less than 5% after 5 fractions for all plans except the complex VMAT plan. Conclusions: For most treatment techniques, the dose deviations averaged out after several fractions. Treatments with unusually complicated MLC had larger dose deviations. Dose deviations are reduced for IMRT treatments by reducing the dose rate. Using two arcs instead of one is effective for RapidArc treatments.

Purpose: Fiducials can be used to aid organ localization during RapidArc delivery. Specifically, the 3‐D tumor position may be found through the tracking of fiducials during beam delivery. A promising technique is the detection of fiducials in 2‐D EPIDimages, and then reconstruction of the 3‐D position using limited views. Methods and Materials: The CIRS dynamic thorax phantom with a custom tumor model is used in the study. Two fiducials are placed on the surface of the tumor at clinically relevant distances apart from each other. Clinical RapidArc plans are delivered to the phantom based on a single 360° arc. The fiducial detection algorithms are developed based on an image block matching filter and anisotropicdiffusion techniques. A wavelet‐based block matching filter is developed to suppress the noise and texture of images while preserving the edges of fiducials. Furthermore, the anisotropicdiffusion technique removes redundant edges and trivial structures of the results of the previous filtering. Experiments show that the algorithms are robust to image contrast and aperture variations during the RapidArc delivery. With detected 2‐D locations, 3‐D spatial locations of fiducials are reconstructed using the inverse 3‐D rotation matrix. Results: The position accuracy in 2D EPID is 0.77+0.58 mm by comparing with the manual outlines. The 3D fiducial positions are reconstructed using the 2D EPIDimages. The displacement from the references in the 3D CBCT image is within one millimeter. Conclusion: Novel filtering and detection algorithms are developed to identify the fiducials in 2D EPIDimages. The results show that limited angles of full rotation can rigorously reconstruct the locations of each of fiducials in 3D to within one millimeter. Conflict of Interest: Varian Medical Systems, Inc.

Purpose: To evaluate an anatomical breathing phantom and modellungtumor for IGRT and 4D applications. Method and Materials: The shape of a tumor was extracted from a patient's CTimage, and a replica tumor created in a flexible resin using rapid‐prototyping techniques. A 5mm grid of fine holes was included in the modeltumor to allow insertion of up to 30 micro‐mosfets. The physical characteristics of this tumor were compared with those of real patient tumors. This tumor was inserted into a commercial anatomic breathing phantom (RSD Inc.). This phantom includes spine, and a flexible ribcage. The tumor and phantom's chest motion were programmed using real patient traces. CTimages (4D and free breathing), kilo‐voltage setup images (including fluoroscopy), and mega‐voltage portal images were compared with those from real patients. Twenty mosfets were inserted into the modeltumor, and the phantom was irradiated using a step‐and‐shoot IMRT plan. The expected dose distribution was calculated using a previously validated 4D Monte Carlo technique. Results: The appearance of the phantom/tumor in kV and MV images was very similar to real patient images. The average CT number of the modeltumor was 76 (at 120kVp), compared with 30 for the real patient (an effective difference in electron density of 2%). Its physical density and hardness was 1.3g cm−3 and 27durometers (range for real tumors: 0–60), respectively. The CT numbers of the bone and soft tissue regions of the breathing phantom were in the range found for real patients. The CT number of the lung material was the same as air. Agreement between the 4D Monte Carlo and experimental DVH was better than 3%. Conclusion: Rapidprototyping techniques can be used to produce realistic low cost modeltumors that, in combination with an anatomical breathing phantom, can be used for IGRT and 4D applications.

Purpose: In lungcancerradiotherapy biologically image guided dose painting is a promising new technique. Correct localization of tumor sub‐volumes remains a major challenge. The purpose of this work was to investigate tumor deformations as observed with 4D PET.Method and Materials: A total of 17 lesions were identified on the 4D FDG‐PET/CT scans from 13 patients. The data were binned in five phase bins using the AZ‐733V gating system. On each one of the five phases of the 4D scan a classification algorithm was applied to obtain the region of highest uptake and segment the tumor volume. We looked at the shape deformation of both the high uptake sub‐volumes and the corresponding segmented tumor volumes. We examined the dependence of shape deformation on the amplitude of motion and the variation in the volume estimate. The effect of motion was minimized by registering the volumes of phases 20–80% with the 0% phase (inhale). The Overlap (O) of the volumes was measured and the shape deformation was defined as 1‐O. The variation in the volume estimate between the different phases was measured by the standard deviation of the % difference from the average volume.Results: The shape deformation of the high uptake sub‐volumes was on average 169% larger than that of the full tumor volumes; a statistically significant difference. The shape deformation of the high uptake sub‐volumes was moderately correlated with that of the tumor (0.75) and the amplitude of motion (0.55). The inhale‐to‐exhale phase (20%) resulted in the largest shape deformation.Conclusions: There is a complex relationship between the behavior of tumor volumes and high uptake sub‐volumes. In cases where advanced therapy techniques such as dose painting based on biological sub‐volumes, are being employed, a close examination of a 4D PET scan should be considered.

Purpose: Several methods exist for tracking tumors on EPIDimages acquired during treatment. Utilizing the tumor tracks, the planned fluence can be averaged to create a “superfluence” that reflects the delivered dose. We investigated the feasibility of 3D dose verification using the superfluence. Method and Materials: The CIRS dynamic phantom was used with 15 MOSFETs inserted in a soft resin tumor model created from a real patient tumor segmented on a CT scan. A real patient breathing pattern with 1.5cm amplitude was set and a 4D CT was acquired. The PTV was obtained by expanding the ITV uniformly by 5mm as typically done in lungSBRT treatments. A 7‐field plan was created with the same coverage as is clinically required. The phantom was irradiated with the tumor static and moving with several amplitudes and set up errors. EPIDimages were recorded throughout these deliveries and the tumor was tracked post‐treatment with in‐house developed software. The planned fluence was blurred according to the motion tracks to obtain a superfluence from which the delivered dose was calculated. The estimated delivered dose was compared to the dose directly measured by the MOSFETs.Results: The calculation of delivered dose with the superfluence method was in good agreement with the MOSFET measurements. The difference varied on average from 0.8% to 8.3% depending on the proximity of a dose gradient.Conclusions: A novel method is presented for 3D dose verification based on the superfluence concept. The only requirements are the acquisition of EPIDimages during treatment and a tracking algorithm to obtain the tumor motion. The excellent agreement between the measured and calculated dose indicates the validity of the method.

Purpose: To maximize the potential benefit of respiratory gating for lungtumors, a method of image‐guided verification is necessary. We propose using the MV EPID in cine mode during respiratory‐gated radiotherapy to verify the location of the tumor without the aid of implanted fiducials. Method and Materials: The verification procedure to be studied is as follows: prior to each fraction of gated radiation therapy, a gated radiograph is acquired for each of the therapy beam angles. This reference image is then compared to cine EPIDimages acquired during respiratory‐gated radiotherapy. The target can be visualized in the images and the deviation from the reference position calculated. Phantom studies were performed using a dynamic phantom and internal respiratory traces derived from patients undergoing lungSBRT treatments. Setup images were taken when the external surrogate signal was in the reference (exhale) position and gated radiotherapy was simulated using clinically realistic fields. The reference and treatment images were loaded into offline research software and the location of the target relative to the reference was calculated using a template matching algorithm. Clinical patient data has also been used to demonstrate the feasibility of the technique for an oblique treatment beam. Results: Preliminary phantom and clinical studies demonstrate that the target can be tracked through a series of MV cine EPIDimages. From both studies, accuracy was assessed to be better than 1 mm. Conclusion: A procedure for confirming the target location during lungcancer respiratory‐gated radiotherapy without the aid of internal fiducials has been defined. Although the proposed procedure is demonstrated retrospectively, in principle, in could be performed in real‐time. Conflict of interest: Research funded, in part, by Varian Medical Systems, Inc.

Purpose: To evaluate the quality of 4D‐CBCT images acquired during a RapidARC treatment. Methods and Materials: A RapidARC plan (879 MUs, treatment time: 1.5min) was delivered to a 2 cm moving target in a lung phantom. Simultaneous kilovoltage projection images were acquired during the RadpidARC delivery. Each image was identified with the respiratory phase using information from the RPM system, and the kilovoltage projections were sorted by breathing phase (10 bins), filtered with a Shepp‐Logan filter and reconstructed into CBCTimage sets using an FDK algorithm. The target sphere and a fiducial were located in all 10 4D‐CBCT sets using an edge‐detection software tool, and the detected target motion was compared with the known motion. These experiments were repeated using target motion following a breathing trace obtained from a real patient (14 cm amplitude, 5 sec period), and four different simulated traces with target amplitudes ranging from 5 mm to 20 mm, 4.5 second period and sin6 motion. For the real patient trace, the motion was in three dimensions. For the other four cases, the motion was constrained to the cranial‐caudal direction. Results: The motion of the fiducial and the target sphere detected in the 4D CBCTimages were found to have an accuracy of 0.156 mm ± 1.03 mm and 0.4 mm ± 0.22 mm for the sphere and the fiducial, respectively. The diameter of the sphere was measured in each phase for each experiment with an accuracy of 0.46 mm ± 0.78 mm. Conclusions: A 4D‐CBCT acquired during a RapidARC treatment can detect target shifts and breathing amplitudes, providing four‐dimensional markerless information of a target during treatment.

Introduction: In order to account for internal tumor motion and setup uncertainty, 3D‐conformal treatment (3D‐CRT) for upper gastrointestinal malignancies requires the placement of margins around the clinical target volume (CTV). These larger margins can result in excessive dose to normal tissue. To spare critical structures while elevating the dose to the tumor, we have investigated the dosimetric benefits of combinations of respiratory‐gating, image‐guidance and intensity modulated radiation therapy (gated‐IG‐IMRT) for pancreatic cancertreatment.Materials and Methods: Both 4D‐CT and 3D‐CT were acquired for each of the cases under study. For respiratory‐gated treatment, contours for the gross tumor volume (GTV) and critical structures on the end‐of‐exhale 4D‐CT were generated from the original contours on 3D‐CT using the MIMvista deformable registration algorithm. A reduced setup margin for image‐guided plans was used to create the planning target volume (PTV). IMRT plans were optimized based on dose volume histograms and published normal tissue constraints. Six plans for each patient were created and the PTV coverage and the dose distributions in critical structures were compared. Results: The six combinations of treatment modalities provide clinically acceptable PTV coverage and normal tissuedoses that meet all dose constraints for organs at risk. IMRT and image‐guided plans provided equivalent or lower mean doses to normal tissues than 3D‐CRT and non‐image‐guided plans, respectively. Respiratory‐gated treatment plans provided slightly lower doses to liver, stomach, right kidney, spinal cord, and spleen, but significantly reduced dose (60–70%) to left kidney when compared with non‐gated plans. Conclusions: By combining technologies and decreasing the margin, dose escalation will be possible in the treatment of pancreatic cancer without compromising the dose to normal tissue. This work was partially supported by Technology and Innovation Grant at Dana‐Farber/Brigham and Women's Cancer Center.

Purpose: We have investigated the dosimetric errors caused by the interplay between the motions of the LINAC and the tumor.Materials and Methods: RapidArc plans were created to treat a 5cm diameter target in a lung phantom (200cGy fractions). The complexity of each field was controlled by adjusting the MU (300–950MU, ∼50MU steps, giving 13 RapidArc plans). This covers the range of MUs found clinically (Mean: 540±220, range: 249–999MU). Each plan was delivered to a phantom, and a movie of the delivered dose (0.2sec frames) taken using a 2D ion chamber array. Patient motion was modeled by shifting each dose frame according to a respiratory trace, starting the motion at different phases. The expected dose distribution was calculated by blurring the static dose distribution with the target motion. The dose error due to the interplay effect was calculated by comparing the delivered dose with the expected dose distribution. Peak‐to‐peak motion of 0.5cm, 1.0cm, and 2.0cm were evaluated. Supporting data was taken using RapidArc plans for lung patients (348–966MU). Results: The daily dose error due to the interplay effect was less than 10% for 99% of all pixels in the target for all plans investigated. The percentage of pixels for which the daily dose error could be larger than 5% increased with increasing plan complexity (field MU), but was less than 15% for all plans if the motion was 1cm or less. For 2cm motion the dose error could be larger than 5% for 40% of pixels, but was less than 5% for more than 80% of pixels for MU

Purpose: To keep safety margins in lung stereotactic body radiation therapy(SBRT) small and provide retrospective calculation of the delivereddose, the tumor motion should be monitored. We propose a tumor tracking algorithm that can estimate the tumor location from portal images taken during the treatment without the help of fiducial markers. Method and Materials: An algorithm based on a normalized mutual information technique was developed for tumor tracking. First a tumor template and a search region are identified on a DRR set reconstructed from a 4DCT acquired prior to the treatment. The set consists of 10 images relating to 10 equally sized breathing phase bins. The template is then used to track the tumor over the sequence of portal images. To estimate the tracking precision a dynamic thorax phantom was employed. Results: The phantom study showed a sub millimeter tracking accuracy in the superior‐ inferior direction for anterior‐posterior and lateral fields. In a preliminary retrospective patient study the algorithm was able to track the tumor motion throughout the whole imagesequence. Manual verification yielded a tracking magnitude error of xy = (3.4 ± 0.8) mm. Furthermore the algorithm's robustness was tested with portal imagesequences from two other patients with different tumor motion amplitude and contrast. The accuaracy was estimated by comparison with manual tracking and yielded xy = (1.7 ± 1.9) mm and xy = (1.2 ± 0.8) mm, respectively. Conclusion: The algorithm has shown great potential for markerless lungtumor tracking. First test results showed that it can perform tumor tracking on portal images and DRRs even if the tracking template was defined in the other modality respectively. Conflict of Interest: Varian Medical Systems, Inc.

Purpose: As many clinics would prefer to use a pair of kV radiographs for gated patient setup, it is important to quantify the possible errors that could occur due to the non‐simultaneity of the acquisitions. Method and Materials: The data set from 11 patients includes internal and external coordinates acquired at 30Hz during multi‐fraction SBRT treatments. We simulated a gated setup procedure which included an AP radiograph, followed by a lateral radiograph. The time interval between the two radiographic acquisitions was set to a minimum of 30 seconds based on our clinical experience. The setup position is derived from a combination of both the initial (AP) and the second (LAT) image in the following way: LR setup = LR initial , SI setup = SI initial + ( SI second − SI initial )/2, AP setup = AP second . The 3D error is then the magnitude of the vector from the initial position to the setup position. The calculation was done for every exhale phase in the data for which there was another one at least 30 seconds later at an amplitude within 2 mm from the first. In all, 1908 data points were analyzed.Results: The 3D error is found to be less than 2 mm for 91% of the data points and less than 3 mm for 97% of the data. The maximum error found was 7.5 mm. 97.5% of the second acquisitions could be completed within 35 seconds of the first. The maximum time interval found was 42 seconds. No correlation was found between the length of the time interval and the 3D error. Conclusion: When acquiring gated radiographs for patient setup, no large setup errors should be expected if those images are not taken simultaneously. Conflict of Interest: This work was sponsored, in part, by a grant from Varian Medical Systems, Inc.

Introduction 4DPET/CT imaging is a powerful technique to assist clinicians in accurate definition of gross‐tumor‐volumes (GTV) in the lung and abdomen. We present the results of phantom studies designed to evaluate the accuracy of 4DPET/CT at reconstructing the true GTV when tumor motion and abdominal motion are reproduced based on pre‐recorded data from real patients. Methods Measurements were conducted using a 4D dynamic phantom control system capable of reproducing a time‐dependent 3D motion (GTV) and a synchronous, decoupled 1D motion (abdominal surface). The tumor and the abdominal motion trajectories programmed into the phantom are based on pre‐recorded patient data (tumor and abdominal motion). A 4ml spherical vial of FDG, representing the tumor, was used within a water‐FDG mixture adjusted to produce a target‐to‐background activity‐ratio of 8:1. The image reconstruction was done using 10 phase bins over the respiratory cycle. The patient selection was based on the amplitude of the internal marker motion (> 1cm). Results and Discussion Based on a 3DPET scan with no target motion, two methodologies were used to determine the activity threshold that recovers the active volume of the target (4ml) relative to: (1) the maximum intensity voxel; (2) the average intensity obtained from the five highest intensity voxels;. The percentage difference between the true and calculated volume was determined for each phase bin of the 4D studies. Errors as large as 45% were observed for specific patients and phases with a statistical bias towards overestimation of the target volume. However, the activity threshold based on the average intensity has shown to reduce the inter‐phase volumetric distortions. Conclusion Our results have quantified volumetric distortions that occur during 4D PET/CT imaging based on previously recorded internal target and abdominal motions in patients. These results are relevant to the accurate application of 4D PET/CT to radiotherapy target definition.