Real-Time Online Adaptive Magnetic Resonance Image Guided Stereotactic Body Radiation Therapy with or without Elective Nodal Irradiation for Inoperable Pancreas Cancer Patients

Abstract

<h3>Purpose/Objective(s)</h3> Stereotactic body radiation therapy (SBRT) to patients with inoperable pancreas cancer has shown promising outcomes. Dose escalated SBRT is usually challenging given the associated risks to nearby organs. This study aims to report treatment outcomes and adverse events of Real-time Online Adaptive Magnetic Resonance-Guided SBRT (MRgSBRT). <h3>Materials/Methods</h3> This IRB approved retrospective study included inoperable pancreas cancer patients who were treated consecutively with MRgSBRT from 2020 to 2021. Most patients (89%) received 50Gy prescribed to gross tumor volume (GTV), two patients (11%) received 40 Gy to GTV. Ten patients (56%) received elective nodal irradiation (ENI) to celiac and super mesenteric lymph nodes, & 8 patients (44%) did not receive ENI. Planning Target Volume (PTV) received a median dose of 35 Gy (range 25-50 Gy) and 40 Gy (range 40-50 Gy), in patients with, and without ENI respectively. PTV was defined as 3 mm expansion of either GTV, or a clinical target volume encompassing GTV and ENI areas when treated. Real-time Online Adaptive MRgSBRT was utilized in 59% of the treated fractions (53 of 90 fractions), while gated MRgSBRT was utilized in all fractions (100%). Real-time Online Adaptive MRgSBRT was utilized whenever the predicted radiation plan did not meet the organs at risk (OARs) constraints based on daily anatomic variations. Reoptimized plans were generated after re-contouring OARs within a 3 cm ring, & verifying GTV & PTV, & always met OARs constraints. All patients were treated twice weekly. <h3>Results</h3> The study included 18 eligible patients with median age of 72.5 years (range 56-85). Patients had local-regional advanced (44%), borderline (33%) & metastatic (22%) diseases. At a median follow up of 10 months from MRgSBRT (range 2-17 months), in patients with & without ENI local-regional progression was 20% versus 67% (p 0.12 Fisher's Exact Test), distant progression was 30% versus 67% (p 0.3), & overall survival was 80% and 37.5% (p 0.14) respectively. Of those who developed disease progression, first site of progression was distant metastases (75%), versus regional nodal progression (75%) in patients with & without ENI respectively. In the non-ENI group, 50%, 12.5%, 12.5%, & 25% were dead of disease (DOD), dead without disease progression (DWODP), alive with disease (AWD), & alive without disease progression (AWODP) respectively. In the ENI group, 10%, 10%, 60%, & 20% were DOD, DWODP, AWD, & AWODP respectively. Median follow up from diagnosis was 19 months (range 6-103 months). Acute & chronic toxicities were uncommon, & included grade 1-2 nausea (22%), fatigue (17%), & abdominal discomfort (11%). No grades 3-5 toxicities were reported. <h3>Conclusion</h3> Real-time Online Adaptive MRgSBRT is a feasible, & safe approach with minimal treatment related toxicities & promising local-regional control in a cohort of advanced pancreas cancer patients. ENI is safe & shows a trend for improved local-regional control. Larger controlled prospective trials are recommended.