ESTRO 2021 Abstract Book

S384

ESTRO 2021

SBRT+/- CHT was significantly correlated with improved LC. Furthermore, we developed a predictive model for LC. This model requires validation using an independent dataset.

PH-0501 Plan Adaptation in Patients with Abdominopelvic Tumors Treated with Stereotactic MR-Guided Adaptive Radiotherapy R. Kotecha 1 , R. Herrera 1 , M. Chuong 1 , M. Rubens 1 , K. Mittauer 1 , T. Romaguera 1 , D. Alavrez 1 , J. McCulloch 1 , D. Doty 1 , M. Hall 1 , J. Contreras 1 , A. Gutierrez 1 , M. Mehta 1 1 Miami Cancer Institute, Baptist Health South Florida, Radiation Oncology, Miami, USA Purpose or Objective Stereotactic body radiation therapy (SBRT) is increasingly used as definitive treatment for primary malignancies of the abdomen and pelvis and in those with oligometastatic and oligoprogressive disease. Ablative dose prescription may not be safe with conventional CT-based approaches given the inferior soft tissue visualization, limited gating options, and potential for significant interfractional changes in gastrointestinal (GI) anatomy. The objective of this study was to analyze the indications for plan adaptation in patients treated with stereotactic MR-guided adaptive radiation therapy (SMART). Materials and Methods Patients with primary and metastatic tumors in the abdomen and pelvis treated using respiratory-gated SMART on a 0.35T-MR Linac from 2018 to 2020 were evaluated. Breath-hold 3D-MRI was acquired prior to each fraction in treatment position to visualize the anatomy-of-the-day and predicted dose (i.e., calculation of initial plan on the anatomy-of-the-day) was performed. If adaptation was indicated (critical organ-at-risk [OAR] dose constraints exceeded and/or desired target coverage not achieved), the plan was reoptimized using the original beam parameters with adjustment of individual beam fluence. Comparisons between the predicted and adaptive dose were performed to evaluate adaptive replanning decision-making based on institutionally-defined criteria. Results 100 patients (95 abdominal, 5 pelvic) received 508 fractions to a median dose of 50 Gy (24–52 Gy) in 5 fractions (range 3-6 fractions). 67 patients were treated to primary tumor sites and 33 to metastases (15 adrenal, 8 abdomen lymph nodes, 5 liver, and 5 pelvic lymph nodes). Plan adaptation was performed for at least 1 fraction with a median of 5 fractions in 96% of patients due to pre-defined GI OAR dose constraints being exceeded using plan prediction on the current fractional anatomy. For patients treated to abdominal targets, predicted plans exceeded pre-defined planning constraints in one OAR in 30% of patients; strikingly, in 65% of patients, ≥2 OARs were exceeded . In patients treated to pelvic targets, ≥2 OARs were exceeded in 4% of patients and one OAR was exceeded in 1% of patients . Across all patients, the most commonly exceeded GI OARs necessitating plan adaptation were the duodenum and stomach in 74% and 71% of patients, respectively. Indications for plan adaptation were to improve target coverage in 30/508 fractions (6%), reduce dose to one critical GI OAR in 178 fractions (35%), reduce dose to ≥2 critical GI OARs in 245 fractions (49%), and for both target and OAR rationales in 27 fractions (5%). Conclusion SMART enabled OAR sparing in response to changes in the anatomy-of-the-day for patients with abdominopelvic malignancies. Significant changes in interfractional motion of GI targets resulted in a significant proportion requiring plan adaptation and in more than one OAR. Long-term outcomes and cumulative dose summations are needed to further explore the use of SMART in this patient population. OC-0502 Real-time cardiorespiratory motion management for MRI-guided stereotactic arrythmia radioablation O. Akdag 1 , P. Borman 2 , P. Uijtewaal 2 , P. Woodhead 2 , J. Kok 2 , B. Van Asselen 2 , B. Raaymakers 2 , M. Fast 2 1 University Medical Center, Radiotherapy, Utrecht, The Netherlands; 2 University Medical Center Utrecht, Radiotherapy, Utrecht, The Netherlands Purpose or Objective Stereotactic arrhythmia radioablation (STAR) serves as a novel alternative approach for the treatment of refractory ventricular tachycardia (VT). Early evidence demonstrates impressive reductions (>90%) in VT burden with a single 25 Gy fraction. Clearly, toxicity is a major concern and should be proactively minimised by cardiorespiratory motion management. While respiratory motion mitigation is widely applied in stereotactic radiotherapy, cardiac motion presents a unique challenge due to its rapid periodicity. In this study, we provide the first experimental demonstration of real-time adaptive MRI-guided STAR. Materials and Methods An MRI-guided real-time adaptive STAR workflow was developed on the 1.5 T Unity MR-linac (Elekta AB, Stockholm, SE) in research mode (Fig. 1). The Quasar MRI4D phantom was used with a cardiorespiratory motion pattern consisting of a respiratory (sine, 12 bpm, 20 mm peak-to-peak) and a cardiac (cos 4 , 60 bpm, 10 mm peak-to-peak) component. Embedded in the water-filled body oval was a moveable film dosimetry insert with a spherical 3 cm target. An IMRT treatment plan for Unity (1x25 Gy) was created in Monaco v.5.40.01 (Elekta AB). MR-guidance was performed by using 2D cine-MRI (bSSFP, 13 Hz, TR/TE=3/1.48 ms, SENSE=1.5, FOV = 400x207 mm 2 , voxel size = 3x3x15 mm, PF factor = 0.65). The detected cardiorespiratory motion was deconvoluted by prospective linear Kalman filtering. The respiratory motion was then used as input for MLC tracking, while the cardiac motion was used as input for beam gating (“track+gate”). Alternative delivery scenarios were created either without any motion (“static”), or by ignoring cardiac motion (“track-only”), or by gating independently on both cardiac and respiratory motion (“dual-gate”). To fit the dynamic range of the Gafchromic EBT3 film, MUs were quartered before the delivery. For cardiac motion, a gating window of 3.8 mm was defined to simulate delivery in diastole. For dual-gating, the exhale gating window was set to 10 mm Proffered papers: Proffered papers 29: New technologies