Michael Weldon

Commissioning and comprehensive evaluation of the ArcCHECK cylindrical diode array for VMAT pretreatment delivery QA

We present commissioning and comprehensive evaluation for ArcCHECK as a QA equipment for volumetric‐modulated arc therapy (VMAT), using the 6 MV photon beam with and without the flattening filter, and the SNC patient software (version 6.2). In addition to commissioning involving absolute dose calibration, array calibration, and PMMA density verification, ArcCHECK was evaluated for its response dependency on linac dose rate, instantaneous dose rate, radiation field size, beam angle, and couch insertion. Scatter dose characterization, consistency and symmetry of response, and dosimetry accuracy evaluation for fixed aperture arcs and clinical VMAT patient plans were also investigated. All the evaluation tests were performed with the central plug inserted and the homogeneous PMMA density value. Results of gamma analysis demonstrated an overall agreement between ArcCHECK‐measured and TPS‐calculated reference doses. The diode based field size dependency was found to be within 0.5% of the reference. The dose rate‐based dependency was well within 1% of the TPS reference, and the angular dependency was found to be ± 3% of the reference, as tested for BEV angles, for both beams. Dosimetry of fixed arcs, using both narrow and wide field widths, resulted in clinically acceptable global gamma passing rates on the 3%/3 mm level and 10% threshold. Dosimetry of narrow arcs showed an improvement over published literature. The clinical VMAT cases demonstrated high level of dosimetry accuracy in gamma passing rates. PACS numbers: 87.56.Fc, 87.55.kh, 87.55.Qr

Radiation therapy combined with intracerebral administration of carboplatin for the treatment of brain tumors

BackgroundIn this study we determined if treatment combining radiation therapy (RT) with intracerebral (i.c.) administration of carboplatin to F98 glioma bearing rats could improve survival over that previously reported by us with a 15 Gy dose (5 Gy × 3) of 6 MV photons.MethodsFirst, in order to reduce tumor interstitial pressure, a biodistribution study was carried out to determine if pretreatment with dexamethasone alone or in combination with mannitol and furosemide (DMF) would increase carboplatin uptake following convection enhanced delivery (CED). Next, therapy studies were carried out in rats that had received carboplatin either by CED over 30 min (20 μg) or by Alzet pumps over 7 d (84 μg), followed by RT using a LINAC to deliver either 20 Gy (5 Gy × 4) or 15 Gy (7.5 Gy × 2) dose at 6 or 24 hrs after drug administration. Finally, a study was carried out to determine if efficacy could be improved by decreasing the time interval between drug administration and RT.ResultsTumor carboplatin values for D and DMF-treated rats were 9.4 ±4.4 and 12.4 ±3.2 μg/g, respectively, which were not significantly different (P = 0.14). The best survival data were obtained by combining pump delivery with 5 Gy × 4 of X-irradiation with a mean survival time (MST) of 107.7 d and a 43% cure rate vs. 83.6 d with CED vs. 30-35 d for RT alone and 24.6 d for untreated controls. Treatment-related mortality was observed when RT was initiated 6 h after CED of carboplatin and RT was started 7 d after tumor implantation. Dividing carboplatin into two 10 μg doses and RT into two 7.5 Gy fractions, administered 24 hrs later, yielded survival data (MST 82.1 d with a 25% cure rate) equivalent to that previously reported with 5 Gy × 3 and 20 μg of carboplatin.ConclusionsAlthough the best survival data were obtained by pump delivery, CED was highly effective in combination with 20 Gy, or as previously reported, 15 Gy, and the latter would be preferable since it would produce less late tissue effects.

Tumoricidal activity of low-energy 160-KV versus 6-MV X-rays against platinum-sensitized F98 glioma cells

The purposes of this study were (i) to investigate the differences in effects between 160-kV low-energy and 6-MV high-energy X-rays, both by computational analysis and in vitro studies; (ii) to determine the effects of each on platinum-sensitized F98 rat glioma and murine B16 melanoma cells; and (iii) to describe the in vitro cytotoxicity and in vivo toxicity of a Pt(II) terpyridine platinum (Typ-Pt) complex. Simulations were performed using the Monte Carlo code Geant4 to determine enhancement in absorption of low- versus high-energy X-rays by Pt and to determine dose enhancement factors (DEFs) for a Pt-sensitized tumor phantom. In vitro studies were carried out using Typ-Pt and again with carboplatin due to the unexpected in vivo toxicity of Typ-Pt. Cell survival was determined using clonogenic assays. In agreement with computations and simulations, in vitro data showed up to one log unit reduction in surviving fractions (SFs) of cells treated with 1–4 µg/ml of Typ-Pt and irradiated with 160-kV versus 6-MV X-rays. DEFs showed radiosensitization in the 50–200 keV range, which fell to approximate unity at higher energies, suggesting marginal interactions at MeV energies. Cells sensitized with 1–5 or 7 µg/ml of carboplatin and then irradiated also showed a significant decrease (P < 0.05) in SFs. However, it was unlikely this was due to increased interactions. Theoretical and in vitro studies presented here demonstrated that the tumoricidal activity of low-energy X-rays was greater than that of high-energy X-rays against Pt-sensitized tumor cells. Determining whether radiosensitization is a function of increased interactions will require additional studies.

Unique In Vitro and In Vivo Thrombopoietic Activities of Ingenol 3,20 Dibenzoate, A Ca++-Independent Protein Kinase C Isoform Agonist

Thrombopoiesis following severe bone marrow injury frequently is delayed, thereby resulting in life-threatening thrombocytopenia for which there are limited treatment options. The reasons for these delays in recovery are not well understood. Protein kinase C (PKC) agonists promote megakaryocyte differentiation in leukemia cell lines and primary cells. However, little is known about the megakaryopoietic effects of PKC agonists on primary CD34+ cells grown in culture or in vivo. Here we present evidence that the novel PKC isoform-selective agonist 3,20 ingenol dibenzoate (IDB) potently stimulates early megakaryopoiesis of human CD34+ cells. In contrast, broad spectrum PKC agonists failed to do so. In vivo, a single intraperitoneal injection of IDB selectively increased platelets in mice without affecting hemoglobin or white counts. Finally, IDB strongly mitigated radiation-induced thrombocytopenia, even when administered 24 hours after irradiation. Our data demonstrate that novel PKC isoform agonists such as IDB may represent a unique therapeutic strategy for accelerating the recovery of platelet counts following severe marrow injury.

SU‐E‐T‐564: Validation of Photon Dose Calculation Using Mobius3D System Compared to AAA and Acuros XB Systems

PURPOSE To study the Collapsed Cone Convolution/Superposition algorithm in Mobius 3D system for dose calculations in comparison to Acuros XB and AAA dose calculation algorithms in Eclipse treatment planning system. METHODS Dose modeling or correction at density heterogeneity (lung/tissue or bone/tissue interfaces) remains an area of maximum discord amongst treatment planning systems. Thus, four phantoms were constructed and CT scanned with both horizontal and vertical density heterogeneity interfaces. Treatment plans were created with varying field sizes (3×3 cm2 , 5×5 cm2 , and 10×1 0 cm2 ) and energies (6FFF, 6MV, and 15 MV). All plans were created in Eclipse TPS with one single AP field, 100 cm SSD, 1 mm grid size for improved resolution, and 200 MU. Plans were calculated with AAA and Acuros XB algorithms and exported to Mobius3D for recalculation and comparison. Percent depth dose (PDD) and horizontal profiles at multiple depths through density heterogeneity interfaces were compared and analyzed. RPC Lung phantom was also used for complex plan (3DCRT and IMRT) dose comparisons. RESULTS Examination of PDD and horizontal dose profiles were reported graphically and numerically. Highest conformality was noted between AcurosXB and Mobius3D in homogenous sites. CCC in Mobius3D generally matches closer with AcurosXB, especially for large field sizes, compared to AAA. In PDDs, Mobius3D tends to under-predict dose compared to AcurosXB at the tissue-to-water interface up to 3%, while over-predicting dose in and beyond the lung-to-tissue interface compared to AcurosXB up to 7%. In 10×10 profiles with vertical heterogeneity interface, Mobius3D agrees with AcurosXB within 3%/3mm for 6MV and 15MV, but higher difference was seen for 6X-FFF. CONCLUSION The recently released Mobius3D program offers physicist and physician ease in evaluation, rapid plan review, and dose second check to TPS calculations. Future research directions include confirmatory clinical dose calculations and additional evaluation with patient geometry.

Treating Cutaneous T-Cell Lymphoma with Highly Irregular Surfaces with Photon Irradiation Using Rice as Tissue Compensator

Purpose: Cutaneous T-cell lymphoma (CTCL) is known to have an excellent response to radiotherapy, an important treatment modality for this disease. In patients with extremity and digit involvement, the irregular surface and depth variations create difficulty in delivering a homogenous dose using electrons. We sought to evaluate photon irradiation with rice packing as tissue equivalence and determine clinical tolerance and response. Materials and methods: Three consecutive CTCL patients with extensive lower extremity involvement including the digits were treated using external beam photon therapy with rice packing for tissue compensation. The entire foot was treated to 30–40 Gy in 2–3 Gy per fraction using 6 MV photons prescribed to the mid-plane of an indexed box filled with rice in which the foot was placed. Treatment tolerance and response were monitored with clinical evaluation. Results: All patients tolerated the treatment without treatment breaks. Toxicities included grade 3 erythema and desquamation with resolution within 4 weeks. No late toxicities were observed. All patients had a partial response by 4 weeks after therapy with two patients achieving a complete response. Patients reported improved functionality after treatment. No local recurrence has been observed. Conclusion: Tissue compensation with rice packing offers a convenient, inexpensive, and reproducible method for the treatment of CTCL with highly irregular surfaces.

Application of Critical Volume-Dose Constraints for Stereotactic Body Radiation Therapy in NRG Radiation Therapy Trials

*Department of Radiation Oncology, Hunter Holmes McGuire VA Medical Center, Richmond, Virginia; yDepartment of Radiation Oncology, University of Michigan, Ann Arbor, Michigan; zDepartment of Radiation Oncology, Henry Ford Health System, Detroit, Michigan; xDepartment of Radiation Oncology, Duke University, Durham, North Carolina; kDepartment of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas; {Department of Radiation Oncology, The Ohio State University, Columbus, Ohio; Department of Radiation Oncology, Washington University, St. Louis, Missouri; **Department of Radiation Oncology, University of Pennsylvania, Philadelphia, Pennsylvania; yyNRG Radiation Oncology Committee and/or NRG Medical Physics Subcommittee

SU-G-TeP2-09: Evaluation of the MaxFOV Extended Field of View (EFOV) Reconstruction Algorithm On a GE RT590 CT Scanner

PURPOSE This study compares the newly released MaxFOV Revision 1 EFOV reconstruction algorithm for GE RT590 to the older WideView EFOV algorithm. Two radiotherapy overlays from Q-fix and Diacor, are included in our analysis. Hounsfield Units (HU) generated with the WideView algorithm varied in the extended field (beyond 50cm) and the scanned objects border varied from slice to slice. A validation of HU consistency between the two reconstruction algorithms is performed. METHODS A CatPhan 504 and CIRS062 Electron Density Phantom were scanned on a GE RT590 CT-Simulator. The phantoms were positioned in multiple locations within the scan field of view so some of the density plugs were outside the 50cm reconstruction circle. Images were reconstructed using both the WideView and MaxFOV algorithms. The HU for each scan were characterized both in average over a volume and in profile. RESULTS HU values are consistent between the two algorithms. Low-density material will have a slight increase in HU value and high-density material will have a slight decrease in HU value as the distance from the sweet spot increases. Border inconsistencies and shading artifacts are still present with the MaxFOV reconstruction on the Q-fix overlay but not the Diacor overlay (It should be noted that the Q-fix overlay is not currently GE-certified). HU values for water outside the 50cm FOV are within 40HU of reconstructions at the sweet spot of the scanner. CatPhan HU profiles show improvement with the MaxFOV algorithm as it approaches the scanner edge. CONCLUSION The new MaxFOV algorithm improves the contour border for objects outside of the standard FOV when using a GE-approved tabletop. Air cavities outside of the standard FOV create inconsistent object borders. HU consistency is within GE specifications and the accuracy of the phantom edge improves. Further adjustments to the algorithm are being investigated by GE.

SU-E-T-624: Portal Dosimetry Commissioning of Multiple (6) Varian TrueBeam Linacs Equipped with PortalVision DMI MV Imager

Purpose: To show that a single model for Portal Domisetry (PD) can be established for beam-matched TrueBeam™ linacs that are equipped with the DMI imager (43×43cm effective area). Methods: Our department acquired 6 new TrueBeam™s, 4 “Slim” and 2 “Edge” models. The Slims were equipped with 6 and 10MV photons, and the Edges with 6MV. MLCs differed between the Slims and Edges (Millennium 120 vs HD-MLC respectively). PD model was created from data acquired using a single linac (Slim). This includes maximum field size profile, as well as output factors and acquired measured fluence using the DMI imager. All identical linacs were beam-matched, profiles were within 1% at maximum field size at a variety of depths. The profile correction file was generated from 40×40 profile acquired at 5cm depth, 95cm SSD, and was adjusted for deviation at the field edges and corners. The PD model and profile correction was applied to all six TrueBeam™s and imagers. A variety of jaw only and sliding window (SW) MLC test fields, as well as TG-119 and clinical SW and VMAT plans were run on each linac to validate the model. Results: For 6X and 10X, field by field comparison using 3mm/3% absolute gamma criteria passed 90% or better for all cases. This was also true for composite comparisons of TG-199 and clinical plans, matching our current department criteria. Conclusion: Using a single model per photon energy for PD for the TrueBeam™ equipped with a DMI imager can produce clinically acceptable results across multiple identical and matched linacs. It is also possible to use the same PD model despite different MLCs. This can save time during commissioning and software updates.

Scalp uniform bolus application (SCUBA) technique for homogenous scalp and regional nodal irradiation

Angiosarcoma is a rare and aggressive malignant neoplasm of unknown etiology, representing only 1% of all sarcomas, with half occurring in the head and neck region.1 It is a subtype of vascular sarcoma, and usually affects patients in the sixth or seventh decade of life. Treatment includes resection with wide margins, with postoperative scalp radiation therapy frequently recommended, given the local failure rate of approximately 50%. Increasingly, systemic or targeted agents are utilized with the aim of improving the 5-year survival rate, which is usually less than 20%.2 Metastatic disease is common, and usually involves lungs and liver. In the head and neck, surgery can be difficult because of the infiltrating nature of the disease and the anatomic constraints of the head and neck that make wide surgical margins difficult to achieve. In such cases, definitive radiation therapy with (neoadjuvant or concurrent) chemotherapy is commonly utilized. Conventionally, total scalp irradiation is delivered using megavoltage photons and electrons, usually 6 MV