Pancreatic cancer outcome—local treatment with radiation using MRI-LINAC

Introduction: Stereotactic MR-guided on-table adaptive radiotherapy (SMART) allows the precise delivery of high-dose radiation to tumors in great proximity to radiation-sensitive organs. The aim of this study is to evaluate the toxicity and clinical outcome in locally advanced or recurrent pancreatic tumors, with or without prior irradiation, treated with SMART. Methods: Patients were treated for pancreatic cancer (PC) using SMART technology to a prescribed dose of 50 Gy (BED₁₀, 100 Gy) in five fractions, with daily on-table adaptation of treatment plan. Endpoints were acute and late toxicities, local control, local disease-free period, and overall survival. Results: A total of 54 PC patients were treated between August 2019 and September 2022, with a median follow-up of 8.9 months from SMART. The median age was 70.4 (45.2–86.9) years. A total of 40 patients had upfront inoperable PC (55% were locally advanced and 45% metastatic), and 14 had local recurrence following prior pancreatectomy (six patients also had prior adjuvant RT). Of the patients, 87% received at least one chemotherapy regimen (Oxaliplatin based, 72.2%), and 25.9% received ≥2 regimens. Except from lower CA 19-9 serum level at the time of diagnosis and 6 weeks prior to SMART in previously operated patients, there were no significant differences in baseline parameters between prior pancreatectomy and the inoperable group. On-table adaptive replanning was performed for 100% of the fractions. No patient reported grade ≥2 acute GI toxicity. All previously irradiated patients reported only low-grade toxicities during RT. A total of 48 patients (88.9%) were available for evaluation. Complete local control was achieved in 21.7% (10 patients) for a median of 9 months (2.8–28.8); three had later local progression. Eight patients had regional or marginal recurrence. Six- and 12-month OS were 75.0% and 52.1%, respectively. Apart from mild diarrhea 1–3 months after SMART and general fatigue, there were no significant differences in toxicity and outcomes between post-pancreatectomy and inoperable groups. Conclusion: SMART allows safe delivery of an ablative dose of radiotherapy, with minimal treatment-related toxicity, even in previously resected or irradiated patients. In this real-world cohort, local control with complete response was achieved by 20% of the patients. Further studies are needed to evaluate long-term outcome and late toxicity.