Magnetic Resonance Guided Stereotactic Ablative Radiation Therapy vs. Chemoradiation for Borderline and Locally Advanced Pancreatic Cancer: Single Institution Overall Survival Comparison

Document Type

Conference Proceeding

Publication Date


Publication Title

Int J Radiat Oncol Biol Phys


Purpose/Objective(s): Several academic institutions have investigated stereotactic MR guided adaptive radiation therapy (SMART) to safely dose escalate for locally advanced and borderline resectable pancreatic cancer with initial favorable toxicity and survival outcomes. In 2018 our institution began to evaluate SMART on trial in borderline and locally advanced patients. We previously presented toxicity outcomes for standard fractionated chemoradiation (chemoRT) and SMART groups with acute grade 3+ GI toxicity in 28% vs 11% (P = 0.18) and late toxicity 43% vs 36% (P = 0.77). The purpose of this abstract was to compare overall survival (OS) between chemoRT and SMART.

Materials/Methods: In this IRB approved analysis, we retrospectively reviewed 115 consecutive patients from 2017-2020 with locally advanced or borderline resectable pancreatic cancer who were treated with neoadjuvant radiation therapy. Initially all patients received chemoRT to a dose of 50.4 Gy in 28 fractions. In September 2018 we began to investigate SMART (50Gy in 5 fractions) for these patients. OS was evaluated by Kaplan-Meier and log-rank test. Univariate and multivariate analysis was also performed on multiple treatment variables.

Results: Of the patients included, 30 received chemoRT and 85 received SMART. Median follow up for the chemoRT group was 32.8 months and for SMART was 14.9 months from last day of RT. Groups did not have significant differences in age, gender, tumor location, or initial CA 19-9. Pancreatectomy was performed in 13.3% vs 18.8% of patients in chemoRT and SMART groups. Per NCCN.1.2021 staging in the chemoRT group 33.3% were borderline (BL), 50% were locally advanced (LA), and 16.7% medically unresectable (MU) as compared to 24.7% BL, 49.4% LA, and 25.9% MU in the SMART group. Mean months of neoadjuvant chemo was slightly higher in the SMART group at 3.6 vs 2.3 months. Patients in the chemoRT arm were 36.7% African American vs 15.3% in the SMART group. When evaluated using Kaplan-Meier and log-rank test there was no difference in OS between groups (P = 0.95). Median OS from last day of RT in chemoRT and SMART groups was 10.7 vs 12.1 months. On univariate and multivariate analyses pancreatectomy was associated with improved OS, and both N2 disease at diagnosis and poor performance status (ECOG 2+) were associated with worse OS.

Conclusion: Dose escalated SMART for locally advanced, borderline, and medically inoperable pancreatic cancer shows similar OS to standard fractionated chemoRT at a median 14.9 month follow up in our single institution analysis. With similar toxicity and OS between treatment modalities, SMART may be preferred due to the single week treatment course in a disease with overall very poor prognosis.





First Page


Last Page