3127 - Ablative Treatment with CT Guided Adaptive Radiation Therapy for Pancreatic Cancer

T. Zhu¹, L. Habimana-Griffin², B. Kalaghchi², R. Beckert², J. P. Schiff², E. Laugeman², P. Samson², E. Morris², A. Mo², and H. Kim²; ¹Washington University in Saint Louis, Saint Louis, MO, ²Washington University School of Medicine in St. Louis, Department of Radiation Oncology, St. Louis, MO

Purpose/Objective(s): Adaptive radiotherapy with online magnetic resonance imaging guidance improves the therapeutic ratio of SBRT for abdominopelvic malignancies. Online CT guided adaptive radiation therapy (CTgART) is an emerging technology, however oncologic outcomes and toxicities of CTgART for abdominal SBRT have not been previously described. Here we report a single institution’s experience with pancreas-directed ablative CTgART. We hypothesized that CTgART would enable delivery of ablative doses without high toxicity. Materials/

Methods: Patients treated with ablative CTgART (50 Gy in 5 fractions) to the pancreas between 2020 to 2023 were identified from a single institutional database. Patients were imaged for online adaptive treatment with a 17 or 6 second per rotation cone beam CT (CBCT). Adaptive plans were created using the anatomy of the day employing an isotoxic approach and coverage improvement threshold of 5%. Contouring prioritized minimizing toxicity at organ and target interfaces or if air artifact obscured organ contours. Survival analyses were performed with Kaplan-Meier method, and univariate and multivariate analyses by Cox proportional hazards modeling.
Results: Twenty-six patients were treated with CTgART to the pancreas (50 Gy in 5 fractions). Sixteen, 6 and 4 were nonmetastatic, metastatic primary pancreas, and secondary to pancreas, respectively. Of the nonmetastatic patients 4 were borderline resectable, 7 were medically inoperable, 5 were locally advanced. Eleven (68.8%) of the nonmetastatic patients received a median of 5 cycles of neoadjuvant chemotherapy, most commonly nab-paclitaxel. All but three patients were treated using the 17 second per rotation CBCT. Median follow-up was 4.0 months. One-year local control was 61.2%. Median progression free survival in the nonmetastatic patients was 6.26 months (Range 0.3-17.1). Median overall survival in the nonmetastatic patients was 12.7 months (Range 3.7-23.47). Acute and late grade 3 GI toxicity were 3.9% and 7.7%, respectively.
Conclusion: These single institution early data indicate that ablative CTgART is feasible with low toxicity and reasonable local control. Improved image quality with faster CBCT acquisition may improve organ delineation and subsequently target coverage. Overall, these data support ongoing phase I/II trials evaluating CTgART for treatment of borderline and locally advanced pancreatic cancer.