Mark Ruschin

Digital fluoroscopy to quantify lung tumor motion: potential for patient-specific planning target volumes ☆

PURPOSE To apply digital fluoroscopy integrated with CT simulation to measure lung tumor motion and aid in the quantification of individualized planning target volumes. METHODS AND MATERIALS A flat panel digital fluoroscopy unit was modified and integrated with a CT simulator. The stored fluoroscopy images were overlaid with digitally reconstructed radiographs, allowing measurement of the observed lung tumor motion in relation to the corresponding contours on the static digitally reconstructed radiographs. CT simulation and digital fluoroscopy was performed on 10 patients with non-small-cell lung cancer. Actual tumor motion was measured in three dimensions using the overlaid images. RESULTS Combining the dynamic data with digitally reconstructed radiographs allowed the tumor shadow from the fluoroscopy to be tracked in relation to the CT lung tumor contour. For all patients, the extent of tumor motion in three dimensions was unique. The motion was greatest in the superoinferior direction and minimal in the AP and lateral directions. CONCLUSION We have developed a tool that allows CT simulation to be combined with digital fluoroscopy. Quantitative evaluation of the tumor motion in relation to the CT plan allows for customization of the planning target volume. The variability observed clearly demonstrates the need to generate patient-specific internal motion margins.

Investigation of energy dependence of EBT and EBT-2 Gafchromic film

PURPOSE The purpose of this study was to quantify the extent of energy dependence of Gafchromic film to x-ray energies ranging in quality from 105 kVp to 6 MV, and relate this dependency to the films chemical composition and date of production. METHODS Lots of Gafchromic EBT film manufactured in 2004 and 2005 together with more recent batches produced in 2007 were evaluated for energy dependence. Multiple batches of EBT-2 film were also evaluated. Energy dependence was quantified as Rx-the ratio of net optical density (netOD) measured at a given energy x relative to the netOD measured at 6 MV, as measured on a linear accelerator. Rx was evaluated for beam qualities of 105 and 220 kVp on a clinical orthovoltage unit using two separate techniques-a flatbed scanner (Epson) and a real-time fiber-optic readout system. Neutron activation analysis for chlorine and bromine content was performed on all the films to determine whether the composition of the film had changed between batches of film exhibiting different energy dependence responses. RESULTS For batches of EBT manufactured in 2007, R105 kVp was 0.75 and R220 kVp was 0.85, indicating an under-response at orthovoltage energies. These results were confirmed using both the Epson flatbed scanner as well as the real-time readout system. For batches of EBT film manufactured before 2006, Rl05 kVp ranged from 0.9 to 1.0. The results from the neutron activation analysis confirmed a direct relationship between the concentration of chlorine and the magnitude of under-response at orthovoltage energies. EBT-2 film exhibited R105 kVp values ranging from 0.79 (under-response) to 1.20 (over-response) among batches containing varying concentrations of bromine, chlorine, and potassium. CONCLUSIONS The results of this study indicated that differences in energy response of EBT and EBT-2 films were due to differences in the chemical composition and therefore the effective atomic number of the film, which have changed over time. To achieve an energy independent dosimeter over a range of kilovoltage energies, the effective atomic number of the dosimeter must be closely matched to that of water. Small deviations in chemical composition can lead to large deviations in response as a function of energy.

Performance of a Novel Repositioning Head Frame for Gamma Knife Perfexion and Image-Guided Linac-Based Intracranial Stereotactic Radiotherapy

PURPOSE To evaluate the geometric positioning and immobilization performance of a vacuum bite-block repositioning head frame (RHF) system for Perfexion (PFX-SRT) and linac-based intracranial image-guided stereotactic radiotherapy (SRT). METHODS AND MATERIALS Patients with intracranial tumors received linac-based image-guided SRT using the RHF for setup and immobilization. Three hundred thirty-three fractions of radiation were delivered in 12 patients. The accuracy of the RHF was estimated for linac-based SRT with online cone-beam CT (CBCT) and for PFX-SRT with a repositioning check tool (RCT) and offline CBCT. The RCTs ability to act as a surrogate for anatomic position was estimated through comparison to CBCT image matching. Immobilization performance was evaluated daily with pre- and postdose delivery CBCT scans and RCT measurements. RESULTS The correlation coefficient between RCT- and CBCT-reported displacements was 0.59, 0.75, 0.79 (Right, Superior, and Anterior, respectively). For image-guided linac-based SRT, the mean three-dimensional (3D) setup error was 0.8 mm with interpatient (Sigma) and interfraction (sigma) variations of 0.1 and 0.4 mm, respectively. For PFX-SRT, the initial, uncorrected mean 3D positioning displacement in stereotactic coordinates was 2.0 mm, with Sigma = 1.1 mm and sigma = 0.8 mm. Considering only RCT setups <1mm (PFX action level) the mean 3D positioning displacement reduced to 1.3 mm, with Sigma = 0.9 mm and sigma = 0.4 mm. The largest contributing systematic uncertainty was in the superior-inferior direction (mean displacement = -0.5 mm; Sigma = 0.9 mm). The largest mean rotation was 0.6 degrees in pitch. The mean 3D intrafraction motion was 0.4 +/- 0.3 mm. CONCLUSION The RHF provides excellent immobilization for intracranial SRT and PFX-SRT. Some small systematic uncertainties in stereotactic positioning exist and must be considered when generating PFX-SRT treatment plans. The RCT provides reasonable surrogacy for internal anatomic displacement.

Cone beam computed tomography image guidance system for a dedicated intracranial radiosurgery treatment unit.

PURPOSE Image guidance has improved the precision of fractionated radiation treatment delivery on linear accelerators. Precise radiation delivery is particularly critical when high doses are delivered to complex shapes with steep dose gradients near critical structures, as is the case for intracranial radiosurgery. To reduce potential geometric uncertainties, a cone beam computed tomography (CT) image guidance system was developed in-house to generate high-resolution images of the head at the time of treatment, using a dedicated radiosurgery unit. The performance and initial clinical use of this imaging system are described. METHODS AND MATERIALS A kilovoltage cone beam CT system was integrated with a Leksell Gamma Knife Perfexion radiosurgery unit. The X-ray tube and flat-panel detector are mounted on a translational arm, which is parked above the treatment unit when not in use. Upon descent, a rotational axis provides 210° of rotation for cone beam CT scans. Mechanical integrity of the system was evaluated over a 6-month period. Subsequent clinical commissioning included end-to-end testing of targeting performance and subjective image quality performance in phantoms. The system has been used to image 2 patients, 1 of whom received single-fraction radiosurgery and 1 who received 3 fractions, using a relocatable head frame. RESULTS Images of phantoms demonstrated soft tissue contrast visibility and submillimeter spatial resolution. A contrast difference of 35 HU was easily detected at a calibration dose of 1.2 cGy (center of head phantom). The shape of the mechanical flex vs scan angle was highly reproducible and exhibited <0.2 mm peak-to-peak variation. With a 0.5-mm voxel pitch, the maximum targeting error was 0.4 mm. Images of 2 patients were analyzed offline and submillimeter agreement was confirmed with conventional frame. CONCLUSIONS A cone beam CT image guidance system was successfully adapted to a radiosurgery unit. The system is capable of producing high-resolution images of bone and soft tissue. The system is in clinical use and provides excellent image guidance without invasive frames.

Automated treatment planning for a dedicated multi-source intracranial radiosurgery treatment unit using projected gradient and grassfire algorithms

PURPOSE The purpose of this work is to develop a framework to the inverse problem for radiosurgery treatment planning on the Gamma Knife(®) Perfexion™ (PFX) for intracranial targets. METHODS The approach taken in the present study consists of two parts. First, a hybrid grassfire and sphere-packing algorithm is used to obtain shot positions (isocenters) based on the geometry of the target to be treated. For the selected isocenters, a sector duration optimization (SDO) model is used to optimize the duration of radiation delivery from each collimator size from each individual source bank. The SDO model is solved using a projected gradient algorithm. This approach has been retrospectively tested on seven manually planned clinical cases (comprising 11 lesions) including acoustic neuromas and brain metastases. RESULTS In terms of conformity and organ-at-risk (OAR) sparing, the quality of plans achieved with the inverse planning approach were, on average, improved compared to the manually generated plans. The mean difference in conformity index between inverse and forward plans was -0.12 (range: -0.27 to +0.03) and +0.08 (range: 0.00-0.17) for classic and Paddick definitions, respectively, favoring the inverse plans. The mean difference in volume receiving the prescribed dose (V(100)) between forward and inverse plans was 0.2% (range: -2.4% to +2.0%). After plan renormalization for equivalent coverage (i.e., V(100)), the mean difference in dose to 1 mm(3) of brainstem between forward and inverse plans was -0.24 Gy (range: -2.40 to +2.02 Gy) favoring the inverse plans. Beam-on time varied with the number of isocenters but for the most optimal plans was on average 33 min longer than manual plans (range: -17 to +91 min) when normalized to a calibration dose rate of 3.5 Gy/min. In terms of algorithm performance, the isocenter selection for all the presented plans was performed in less than 3 s, while the SDO was performed in an average of 215 min. CONCLUSIONS PFX inverse planning can be performed using geometric isocenter selection and mathematical modeling and optimization techniques. The obtained treatment plans all meet or exceed clinical guidelines while displaying high conformity.

Integration of digital fluoroscopy with CT-based radiation therapy planning of lung tumors

Radiation dose escalation may be a means to increase the local control rate of inoperable lung tumors. Treatment plans involve the creation of a uniform planning target volume (PTV) to ensure proper coverage despite patient breathing and setup error. This may lead to unnecessary radiation of normal tissue in shallow breathers or target underdosing for patients with excess internal motion. Therefore, the nature of tumor motion for each patient should be measured in 3D, something that cannot be done with CT alone. We have developed a method that acquires 2D real-time fluoroscopic images (loops) and coregisters them with 2D digitally reconstructed radiographs (DRR) formed from the CT scan. The limitations of CT to encompass motion can be overcome by merging the two modalities together. The accuracy of the coregistration method is tested with a stationary grid of radio-opaque markers at various spatial positions. The in-plane (at-depth) displacement between markers on the fluoroscopic image versus the DRR varies with position across the image due to slight misalignments between the x-ray source used in fluoroscopy and the virtual source used for the DRR relative to the test object. At clinically relevant positions, the maximum, measured in-plane displacement, is 1.1 mm. The method is applied to the thorax of an anthropomorphic phantom and a good fit is observed between the appearances of the bony anatomical structures on the coregistered image. Finally, a series of motion measurements are carried out on two oscillating cylindrical objects. The degree of motion as measured by fluoroscopy is accurate to within 1.0 mm, whereas the DRR is inconsistent in predicting motion. The coregistration of fluoroscopic loops with the DRR shows at what point within the oscillation the DRR fails to encompass motion. For any treatment site involving target motion, this real-time imaging is a useful asset in the planning stage.

Stereotactic Body Radiotherapy for Spinal Metastases: Practice Guidelines, Outcomes, and Risks.

AbstractSpine metastases can be a debilitating and difficult therapeutic challenge for a significant number of cancer patients. Surgical management of spine metastases is often limited because of the complexity, risks, and recovery delays associated with open invasive surgical procedures. Conventional palliative external beam radiation therapy is the most common treatment modality. However, it is associated with limited palliative efficacy and local tumor control, including in the postoperative setting. In the era of improving systemic disease control, spine stereotactic body radiotherapy is fast emerging as the therapeutic modality of choice for selected de novo, postoperative, and salvage reirradiation spine metastases patients. Considerable expertise, multidisciplinary collaboration, and rigid adherence to quality metrics are required for the safe application of this highly conformal ablative therapy. This review highlights the current state of the evidence, understanding of the late effects, and technological requirements for spine stereotactic body radiotherapy specific to spinal metastases.

Measurement of mean cardiac dose for various breast irradiation techniques and corresponding risk of major cardiovascular event

After breast conserving surgery, early stage breast cancer patients are currently treated with a wide range of radiation techniques including whole breast irradiation (WBI), accelerated partial breast irradiation (APBI) using high-dose rate (HDR) brachytherapy, or 3D-conformal radiotherapy (3D-CRT). This study compares the mean heart’s doses for a left breast irradiated with different breast techniques. An anthropomorphic Rando phantom was modified with gelatin-based breast of different sizes and tumors located medially or laterally. The breasts were treated with WBI, 3D-CRT, or HDR APBI. The heart’s mean doses were measured with Gafchromic films and controlled with optically stimulated luminescent dosimeters. Following the model reported by Darby (1), major cardiac were estimated assuming a linear risk increase with the mean dose to the heart of 7.4% per gray. WBI lead to the highest mean heart dose (2.99 Gy) compared to 3D-CRT APBI (0.51 Gy), multicatheter (1.58 Gy), and balloon HDR (2.17 Gy) for a medially located tumor. This translated into long-term coronary event increases of 22, 3.8, 11.7, and 16% respectively. The sensitivity analysis showed that the tumor location had almost no effect on the mean heart dose for 3D-CRT APBI and a minimal impact for HDR APBI. In case of WBI large breast size and set-up errors lead to sharp increases of the mean heart dose. Its value reached 10.79 Gy for women with large breast and a set-up error of 1.5 cm. Such a high value could increase the risk of having long-term coronary events by 80%. Comparison among different irradiation techniques demonstrates that 3D-CRT APBI appears to be the safest one with less probability of having cardiovascular events in the future. A sensitivity analysis showed that WBI is the most challenging technique for patients with large breasts or when significant set-up errors are anticipated. In those cases, additional heart shielding techniques are required.

Non Tumor Perfusion Changes Following Stereotactic Radiosurgery to Brain Metastases.

PURPOSE To evaluate early perfusion changes in normal tissue following stereotactic radiosurgery (SRS). METHODS Nineteen patients harboring twenty-two brain metastases treated with SRS were imaged with dynamic susceptibility magnetic resonance imaging (DSC MRI) at baseline, 1 week and 1 month post SRS. Relative cerebral blood volume and flow (rCBV and rCBF) ratios were evaluated outside of tumor within a combined region of interest (ROI) and separately within gray matter (GM) and white matter (WM) ROIs. Three-dimensional dose distribution from each SRS plan was divided into six regions: (1) <2 Gy; (2) 2-5 Gy; (3) 5-10 Gy; (4) 10-12 Gy; (5) 12-16 Gy; and (6) >16 Gy. rCBV and rCBF ratio differences between baseline, 1 week and 1 month were compared. Best linear fit plots quantified normal tissue dose-dependency. RESULTS Significant rCBV ratio increases were present between baseline and 1 month for all ROIs and dose ranges except for WM ROI receiving <2 Gy. rCBV ratio for all ROIs was maximally increased from baseline to 1 month with the greatest changes occurring within the 5-10 Gy dose range (53.1%). rCBF ratio was maximally increased from baseline to 1 month for all ROIs within the 5-10 Gy dose range (33.9-45.0%). Both rCBV and rCBF ratios were most elevated within GM ROIs. A weak, positive but not significant association between dose, rCBV and rCBF ratio was demonstrated. Progressive rCBV and rCBF ratio increased with dose up to 10 Gy at 1 month. CONCLUSION Normal tissue response following SRS can be characterized by dose, tissue, and time specific increases in rCBV and rCBF ratio.

Investigation of Dose Falloff for Intact Brain Metastases and Surgical Cavities Using Hypofractionated Volumetric Modulated Arc Radiotherapy.

Introduction: Intact brain metastases tend to be small and spherical compared to postsurgery brain cavities, which tend to be large and irregular shaped and, as a result, a challenge with respect to treatment planning. The purpose of the present study is to develop guidelines for normal brain tissue dose and to investigate whether there is a dependence on target type for patients treated with hypofractionated volumetric modulated arc radiotherapy (HF-VMAT). Methods: Treatment plans from a total of 100 patients and 136 targets (55 cavity and 81 intact) were retrospectively reviewed. All targets were treated with HF-VMAT with total doses ranging between 20 and 30 gray (Gy) in 5 fractions. All plans met institutional objectives for organ-at-risk constraints and were clinically delivered. Dose falloff was quantified using gradient index (GI) and distance between the 100% and 50% isodose lines (R50). Additionally, the dose to normal brain tissue (brain contour excluding all gross tumor or clinical target volumes) was assessed using volume receiving specific doses (Vx) where x ranged from 5 to 30 Gy. Best-fit curves using power law relationships of the form y = axb were generated for GI, R50, and Vx (normal brain tissue) versus target volume. Results: There was a statistically significant difference in planning target volume (PTV) for cavities versus intact metastases with mean volumes of 37.8 cm3 and 9.5 cm3, respectively (P < .0001). The GI and R50 were statistically different: 3.4 and 9.8 mm for cavities versus 4.6 and 8.3 mm for intact metastases (P < .0001). The R50 increased with PTV with power law coefficients (a, b) = (6.3, 0.12) and (5.9, 0.15) for cavities and intact, respectively. GI decreased with PTV with coefficients (a, b) = (5.9, −0.18) and (5.7, −0.14) for cavities and intact, respectively. The normal brain tissue Vx also exhibited power law relationships with PTV for x = 20 to 28.8 Gy. In conclusion, target volume is the main predictor of dose falloff. The results of the present study can be used for determining target volume-based thresholds for dose falloff and normal brain tissue dose–volume constraints.