Experimental evaluation of an intraoperative-imaging based workflow for electron beam radiotherapy of pancreatic cancer using in situ dosimetry.
The aim of this study is to perform an experimental evaluation of an imaging-based intraoperative electron beam radiotherapy (IOERT) and in vivo dose verification workflow for pancreatic cancer on a porcine cadaver.
The Imaging Ring m (ImR) mobile cone-beam computed tomography (CBCT) scanner (medPhoton GmbH, Salzburg), the Radiance (GMV, Tres Cantos, Madrid, Spain) treatment planning system (TPS) and the Mobetron (IntraOp Medical Inc, Sunnyvale, CA, USA) mobile linear accelerator (LINAC) were used. Cylindrical thermoluminescent dosimeters (TLD-100) were employed for in situ dose measurements. ImR calibration data were acquired and imported into Radiance for CT table commissioning. The porcine cadaver was immobilized using standard radiotherapy (RT) equipment and scanned preoperatively with a SOMATOM Go.Open Pro CT scanner (Siemens Healthineers AG, Forchheim, Germany) to obtain a reference abdominal CT image. Subsequently, a surgical procedure was performed to expose the pancreas, and a dedicated TLD-based dosimetry system was secured on its surface. The 5 cm diameter/30°-bevel IOERT plastic applicator was positioned over the dosimeters and intraoperative CBCT images were acquired. Treatment was delivered using a 9 MeV electron beam, prescribing 10 Gy to the distal 90% isodose depth. The intraoperative CBCT images were imported into Radiance, where the applicator was positioned based on imaging, relevant anatomy was contoured and three-dimensional (3D) dose distributions were calculated using a Monte Carlo (MC) algorithm and compared to the TLD measurements.
ImR CBCT calibration scans yielded CBCT numbers consistent with reference data. Image quality was sufficient for clear visualization of the applicator and TLD-based dosimetry systems without significant artifacts; however, soft-tissue contrast was limited for clear determination of pancreatic tissue and important neighboring vessels. Due to observed tissue extension within the applicator, dose calculations were adjusted to begin inside the applicator volume. In situ TLD dose measurements agreed with MC-calculated doses within 3%.
The developed image-guided pancreatic IOERT workflow was successfully simulated under near-clinical conditions using a porcine cadaver model. Agreement between in situ TLD measurements and MC dose calculations was within the accepted tolerance of ±5%, supporting further clinical implementation.
The Imaging Ring m (ImR) mobile cone-beam computed tomography (CBCT) scanner (medPhoton GmbH, Salzburg), the Radiance (GMV, Tres Cantos, Madrid, Spain) treatment planning system (TPS) and the Mobetron (IntraOp Medical Inc, Sunnyvale, CA, USA) mobile linear accelerator (LINAC) were used. Cylindrical thermoluminescent dosimeters (TLD-100) were employed for in situ dose measurements. ImR calibration data were acquired and imported into Radiance for CT table commissioning. The porcine cadaver was immobilized using standard radiotherapy (RT) equipment and scanned preoperatively with a SOMATOM Go.Open Pro CT scanner (Siemens Healthineers AG, Forchheim, Germany) to obtain a reference abdominal CT image. Subsequently, a surgical procedure was performed to expose the pancreas, and a dedicated TLD-based dosimetry system was secured on its surface. The 5 cm diameter/30°-bevel IOERT plastic applicator was positioned over the dosimeters and intraoperative CBCT images were acquired. Treatment was delivered using a 9 MeV electron beam, prescribing 10 Gy to the distal 90% isodose depth. The intraoperative CBCT images were imported into Radiance, where the applicator was positioned based on imaging, relevant anatomy was contoured and three-dimensional (3D) dose distributions were calculated using a Monte Carlo (MC) algorithm and compared to the TLD measurements.
ImR CBCT calibration scans yielded CBCT numbers consistent with reference data. Image quality was sufficient for clear visualization of the applicator and TLD-based dosimetry systems without significant artifacts; however, soft-tissue contrast was limited for clear determination of pancreatic tissue and important neighboring vessels. Due to observed tissue extension within the applicator, dose calculations were adjusted to begin inside the applicator volume. In situ TLD dose measurements agreed with MC-calculated doses within 3%.
The developed image-guided pancreatic IOERT workflow was successfully simulated under near-clinical conditions using a porcine cadaver model. Agreement between in situ TLD measurements and MC dose calculations was within the accepted tolerance of ±5%, supporting further clinical implementation.
Authors
Iliaskou Iliaskou, Gainey Gainey, Kollefrath Kollefrath, Kuhn Kuhn, Boronikolas Boronikolas, Thomsen Thomsen, Ruess Ruess, Grosu Grosu, Baltas Baltas
View on Pubmed