ESTRO 2021 Abstract Book

S1512

ESTRO 2021

In conclusion, dose reduction in areas outside the radiation field is possible, but the dose administration approach must be carefully selected and individually adjusted in according to the treatment objectives.

PO-1786 Computed tomography of the head for therapy planning of pediatric patients – impact of scan- length J. Wulff 1 , R. Schmidt 2 , C. Bäumer 1 , B. Timmermann 1 , K. Zink 2 1 University Hospital Essen, Westdeutsches Protonentherapiezentrum Essen, Essen, Germany; 2 University of Applied Sciences Mittelhessen, Institute of Medical Physics and Radiation Protection, Giessen, Germany Purpose or Objective A computed tomography (CT) is fundamental in the overall radiation therapy process. Following the ALARA principle the scan-length is generally kept short, just imaging the volume to be treated and the surrounding tissue, and set typically to anatomical landmarks plus a safety margin. The latter is required as the exact target is not always known and further immobilization equipment may need to be included in the CT. The total length may thus vary and the impact on the resulting imaging dose is difficult to estimate, especially in case of pediatric patients with variable total body size. This work investigates the impact of scan-length on effective dose as it results in a planning CT of a pediatric head, calculated via Monte Carlo (MC) simulation. Materials and Methods A model of a Philips BigBore Brilliance CT was implemented in the GMctdospp MC framework (Schmidt et al. MedPhys 2015). The CT was experimentally characterized in terms of X-ray half-value layer, beam-shaping filter, patient table attenuation and z-profile distribution. A calibration of MC calculations in terms of KERMA was established by measurements with an ionization chamber free in air on the central axes. The scanner model was validated by CTDI measurements at different locations in a 32 cm CTDI phantom. Five different anthropomorphic phantoms (Norris et al. MedPhys 41(3) 2014, Segars et al. MedPhys 42(8) 2015) (infant, 1, 5, 10 and 15 year old) were used to calculate the effective dose as well as the organ doses according to ICRP103. The clinical protocols were chosen corresponding to the clinical settings of a head-scan matching the age/ sizes of the phantoms. The scan length was varied between +3.2 and -1.6 cm from the reference position, which was defined as the position of cervical vertebra 7 (CV7) in each phantom. Results The scanner simulation model agreed with measurements in terms of CTDI vol (32 cm) at different locations within 5% on average. The largest deviations occurred for the bottom position within the CTDI-phantom which most likely is caused by an imperfect model of the patient table. The effective dose for the default scan length with CV7 as landmark were between 8-4.4 mSv, generally decreasing with age. The change of effective dose as a function of length was 15%/ cm on average. As expected, the largest impact was found for the infant model, where the underlying organ doses close to the scan-field edge, such as thyroid, are significantly impacted by the scan length. Conclusion The scan length is crucial in pediatric CT imaging and should be limited as much as possible. The characterized scanner model in the Monte-Carlo simulation environment allows for further optimization in pediatric imaging protocols. The influence of scan length and age on organ and effective dose has been quantitative tabulated and thus (the general stochastic risk of) variations in CTs could be take into account in radiation planning decisions. PO-1787 The Effect of Reducing CT Dose on Proton Therapy Dose Calculation – Phantom Study M. Elhamiasl 1 , K. Salvo 2 , E. Sterpin 3,4 , J. Nuyts 1 1 KU Leuven, Department of Imaging and Pathology, Nuclear Medicine & Molecular Imaging, Leuven, Belgium; 2 UZ Leuven, Department of Radiotherapy, Leuven, Belgium; 3 KU Leuven, Department of Oncology, Laboratory of Experimental Radiotherapy, Leuven, Belgium; 4 UCLouvain, Institut de Recherche Expérimentale et Clinique, Molecular Imaging Radiotherapy and Oncology Lab, Brussels, Belgium Purpose or Objective Proton treatment plans are sensitive to anatomical changes during the course of a treatment. To minimize range uncertainties due to anatomical variations, daily CT acquisition would be paramount for accurate dose calculation and subsequent plan adaptation. However, the series of CT scans results in a substantial accumulated additional patient dose, especially if a 4D CT scan is required. We hypothesized that the signal- to-noise ratio provided by conventional CT protocols is higher than needed for this application. In this research, we aim to investigate the effect of CT dose reduction on proton therapy dose calculations. Materials and Methods To verify our hypothesis and enable a patient-dependent reduction of the CT dose, a CT dose reduction simulation tool was developed to simulate lower-dose CT scans from an existing standard-dose scan. Several phantom studies with different acquisition protocols on two different CT scanners were performed to evaluate the accuracy of the proposed framework. The results demonstrated excellent accuracy in providing realistic lower-dose images where the texture and magnitude of the simulated noise matched the measurements in both image and projection domains. A realistic head phantom (Proton Therapy Dosimetry Head, Model 731-HN) was scanned with a routine clinical protocol [192 effective mAs, helical, pitch: 0.55] and subsequently, the lower-dose CTs of 20, 47, and 70 mAs were simulated from the standard-dose scan. The brainstem was contoured and the PTV was defined like a horseshoe around the brainstem. The plan was optimized on the PTV using the standard-dose CT scan. The dose distributions were then recomputed for the lower-dose simulated CTs using the optimized plan (OptHigh- DoseLow) and the results compared with that of the standard-dose scan (OptHigh-DoseHigh). To avoid additional variation due to the statistical uncertainty of the Monte Carlo simulation, a pencil beam engine was used for plan optimization. Results