ESTRO 2020 Abstract Book

S878 ESTRO 2020

Purpose or Objective Treatment session times on an MR-linac are generally longer compared to session times on a conventional linac. Bladder filling during this prolonged session time could impact treatment volumes for potential future bladder cancer radiotherapy on the MR-linac. The purpose of this study was to quantify bladder filling during MR-linac treatment sessions and to investigate consistency in filling patterns between fractions. Material and Methods MRI scans acquired during MR-linac treatment sessions of 25 patients were retrospectively analyzed. Patients were treated in 5 fractions for non-bladder cancer related malignancies, i.e. stereotactic body radiotherapy for pelvic lymph node metastases. No drinking or voiding protocols were applied. During each MR-linac treatment session, MRI scans were made at the start of the session for treatment planning (MRI pre ), after dose calculation for position verification (MRI pv ) and after irradiation for post- treatment validation (MRI post ). The bladder was contoured on each MRI scan. Bladder filling rate was calculated as [Δ volume in milliliters / Δ time in minutes]. Regression analysis was used to investigate the relationship between filling rate and bladder volume at baseline. Furthermore, filling rates were compared between fractions. Results 375 MRI scans were analyzed. The median time between MRI pre and MRI post was 30 minutes (range 21-57). Median time between MRI pre and MRI pv was 15 minutes (range 8-25) and between MRI pv and MRI post 15 minutes (range 10-33). Median filling rate for the total session duration (MRI pre – MRI post ) was 1.6 mL/minute (range 0.3-12.0). Figure 1A shows the total session filling rate per patient for each treatment session. Filling rate was not associated with bladder volume at the start of the treatment session (r 2 =0.13, p<0.05) (figure 1B). Filling rates between the first (MRI pre – MRI pv ) and the second (MRI pv – MRI post ) part of the treatment session differed significantly, with 1.3 mL/minute and 1.8 mL/minute, but were related to each other (r 2 =0.74, p<0.05) (Figure 2). Overall, there was a weak relation in bladder filling rates between fractions (r 2 =0.42). However, if the filling rate at the first fraction was below 2 mL/minute, the filling rate was strongly related to the filling rate at the following fractions (r 2 =0.90). Conclusion In healthy bladders, bladder filling during MR-linac treatment sessions varied greatly between patients and treatment sessions. In patients with a filling rate below 2 mL/minute, filling rates appeared more consistent. This dataset will be further analyzed to quantify bladder wall motion, which will allow for (individual) PTV margin calculations for potential future treatment of bladder cancer patients on an MR-linac. Alternatively, bladder filling data might be used for PRV calculation of the bladder for other treatment sites in the pelvis. This will enable comparison of treatment volumes between conventional linac and MR-linac treatments.

PO-1613 Reliability of ITV approach to varying treatment fraction time: considerations based on 2D cine MRI D. Cusumano 1 , J. Dhont 2 , L. Boldrini 1 , S. Longo 1 , G. Chiloiro 1 , L. Placidi 1 , A. Romano 1 , L. Azario 1 , M. De Spirito 1 , D. Verellen 3 , V. Valentini 1 1 Fondazione Policlinico Universitario A.Gemelli IRCCS, Dipartimento di Diagnostica per immagini- Radioterapia Oncologica ed Ematologia, Roma, Italy ; 2 Vrije Universiteit Brussels, Department of Electronics and Informatics ETRO, Brussels, Belgium ; 3 Iridium Kankernetwerk- Antwerp- Belgium, Faculty of Medicine and Health Sciences, Antwerp, Belgium Purpose or Objective Internal Target Volume (ITV) estimated from 4DCT is one of the most common strategies to passively manage tumor motion in Radiotherapy (RT). The reliability of this approach is based on the assumption that the tumor motion estimated during 4DCT acquisition is representative of that one that will happen during the