System for dose audit for external radiation therapy based on EPR dosimetry with Lithium Formate

Abstract: Radiation therapy is an important method to treat cancer with the aim to deliver as high doses as reasonably achievable to the tumor while protecting the surrounding healthy tissue and organs at risk, OARs. Therefore, it is essential to have high accuracy in the dose delivered clinically and quality assurances are required. In the meantime, radiation therapy techniques are becoming more advanced and complex, introducing a significant risk of random and systematic errors that needs to be investigated. Hence, the need of independent dose verifications has increased. The purpose of the present work is to design and create a mailed audit system for external evaluation of the dose to water in relevant points in a phantom, including influences from the whole treatment chain, from computed tomography, CT, scanning, to contouring of structures, treatment planning and treatment delivery. The measurements were performed using an anthropomorphic Polymethyl methacrylate, PMMA, phantom designed to be relevant for the head-and-neck region containing inserts corresponding to tumour, salivary glands and medulla made of PMMA and that are easily distinguishable from the surroundings for contouring. Inhomogeneities of both Teflon, corresponding to the spinal cord, and air were also included. Pellet shaped electron paramagnetic resonance, EPR, dosimeters made of lithium formate with a diameter of 4.5 mm and height of 5 mm were made for the measurements. The dosimeters can be placed in various positions in the different structures of the phantom using PMMA tubes and can be analyzed using a spectrometer. In order to test the precision and accuracy of the EPR dosimetry method, measurements with three blind tests were performed simultaneously with an ionization chamber for comparison of absorbed doses. For the audit measurement, the audit phantom was CT scanned twice both with a Siemens CT scan and GE (General Electric)) CT scan for comparison of Hounsfield Units, HU, and dose distributions. The target and the OARs were contoured in the treatment planning system, TPS, (Varian, Eclipse) and a dynamic Intensity modulated radiation therapy, IMRT, treatment plan was created. The treatment plan consisted of seven coplanar 6 MV fields giving the target a dose of 5 Gy delivered with a Varian, Clinac iX accelerator. The absorbed doses to water were determined in seven locations: three points in the target, one in each parotis, one in the medulla and one in the air cavity. The absorbed doses were determined using the signal from the EPR dosimeters and were compared to the planned doses. Also, the measured and reconstructed volumes of the structures were compared. The blind tests doses obtained from the EPR dosimeters agreed with the results obtained from the ionization chamber within 1% and are well below the calculated uncertainties (1 SD) in the EPR measurements. The absorbed doses and the dose distributions were not affected by any spread in HU and the absorbed doses had an agreement within 0.5% in comparison between the Siemens and GE CT studies. The determined doses agreed with planned doses within 4% for all the structures except the air cavity. This deviation is not covered by the calculated standard uncertainty. However, the deviation does fall within two standard deviations, corresponding to a confidence interval of 95%. Also the measured and planned volumes had an agreement within 2.5% for smaller structures and within 5% for larger structures. Repeating the whole measurement chain with other dosimeter batches is required using two or three dosimeters in each measurement point for higher precision. A conclusion can be made that this work showed promising initial results for an audit system for evaluation of the dose to water in relevant points in a phantom, including influences from the whole treatment chain.