ESTRO 2020 Abstract Book

S262 ESTRO 2020

Purpose or Objective The Elekta Unity (Elekta AB, Stockholm, Sweden) combines a linear accelerator with a 1.5T magnetic resonance (MR) scanner enabling daily online adaptive MR-guided radiotherapy. We describe the first UK experience of this technology for the treatment of muscle invasive bladder cancer (MIBC). Material and Methods To date, 3 male patients with MIBC deemed unfit for daily radical radiotherapy, have completed treatment in the ongoing PERMIT trial (Prospective Evaluation of Radiotherapy Using Magnetic Resonance Image Guided Treatment, NCT03727698) receiving hypofractionated radiotherapy. The patients underwent a radiotherapy (RT) planning CT with an empty bladder. RT was planned using Monaco version 5.40 (Elekta) to a dose of 36Gy in 6 once weekly fractions to the whole bladder (CTV) using 7-field IMRT. The PTV was defined as CTV plus an anisotropic margin of 1.5cm superiorly/anteriorly, 1cm posteriorly and 0.5cm laterally/inferiorly. The PTV was covered by 95% of prescription dose with OAR dose constraints as per institutional guidelines. For each fraction, a daily T2W session MRI (sMRI) was acquired followed by an ‘adapt to shape’ (ATS) workflow. Under this workflow, the CTV is re-contoured and a new treatment plan optimised using the daily anatomy. Pre ‘beam on’ verification (vMRI) and post-treatment MRI (pMRI) were acquired to enable position verification and CTV dose coverage assessment respectively. CTV coverage assessment was performed by offline contouring of the CTV and OARs on the vMRIs and pMRIs and re-calculating the estimated dose delivered using Monaco. Plan conformity index (CI RTOG ), the proportion of total volume receiving >95% dose compared to CTV was used as a surrogate for the dose received by normal-tissue such that the higher the CI the more normal tissue receives >95% and a CI RTOG = 1 means no normal tissue received this dose. Results In total 18 fractions were delivered. Table 1 summarises the time taken for key adaptive workflow stages.

treatment sessions. Gated SABR was delivered during repeated breath-holds, using a planning target volume (PTV) margin of 5 mm, and an isotropic boundary of 3 mm around the gross tumor volume (GTV) as a gating window. Visual patient feedback was provided during delivery using an in-room monitor (Fig. 1, A), depicting the continuously tracked GTV (green) and the gating window (red) in a sagittal MR plane (Fig. 1, B-C). The acquired real-time MR images were analyzed to determine precision and efficiency of gated delivery as described previously (van Sörnsen de Koste JR, Int J Radiat Oncol Biol Phys 2018).

Results Fourteen patients underwent MR-simulation but 6 were considered unsuitable, mostly due to poor tracking of small tumors. Subsequently, eight suitable patients (median age 75 years; range 65-80) underwent a single fraction of 34 Gy on the MR Linac (Table 1). Treatment indications were a primary lung cancer (n = 6) or lung metastasis (n = 2). The median GTV and PTV were 2.9 cm 3 (range, 1.8-4.3) and 10.1 cm 3 (7.5-13.9), respectively. The total in-room procedure required a median of 111 minutes (range, 74-150), with irradiation comprising 37 minutes (28-59). A geometric analysis of SABR delivery in the first 5 patients revealed a mean duty cycle efficiency of 60.6% (range, 33.9-85.3), and mean GTV coverage by the PTV during beam-on was 99.6% (99.0-100.0). Pre-treatment online plan reoptimization improved PTV coverage (V 34Gy ) by an average of 5.5% (from 89.5% using baseline plans, to 95.0% in reoptimized ones), although this had only limited impact on GTV and organ at risk doses in this cohort. Conclusion Single-fraction lung SABR using MR-guidance is feasible, and it allows high-precision delivery. However, not all evaluated tumors were suitable for SABR delivery on the MR Linac, indicating that further improvements in imaging are needed to ensure tumor tracking in all patients. OC-0469 MR-guided online adaptive radiotherapy for muscle invasive bladder cancer: First UK experience A. Hunt 1 , I. Hanson 2 , A. Dunlop 2 , L. Bower 1 , H. Barnes 3 , J. Chick 2 , T. Herbert 3 , R. Lawes 3 , H. McNair 1 , A. Mitchell 2 , J. Mohajer 2 , T. Morgan 1 , G. Smith 3 , S. Nill 2 , U. Oelfke 2 , R. Huddart 1 , S. Hafeez 1 1 Institute of Cancer Research- The Royal Marsden NHS Foundation Trust, Radiotherapy and Imaging, London, United Kingdom ; 2 Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, The Joint Department of Physics, London, United Kingdom ; 3 The Royal Marsden NHS Foundation Trust, Radiotherapy, London, United Kingdom

Patient/treatment parameters are detailed in Table 2. Mean intra-fraction CTV change (surrogate for bladder filling) ranged between 9-60cc. Post treatment CTV coverage (>95% CTV receiving >95% dose) was maintained for 17/18 fractions; for 1/18 fraction dose was 94.6%. In 17/18 fractions, the delivered plan resulted in estimated dose to OARs within the mandatory dose constraints (based on the anatomy seen on vMRI and pMRI). For non- catheterised patients, the mean CI RTOG improved between