Background: Pancreatic cancer (PC) is largely refractory to immune checkpoint blockade (ICB) in genomically unselected patients (pts). Synergistic effects of combined ICB and radiotherapy have been suggested in varied solid tumors. Few studies have evaluated this approach in pts with PC. We sought to evaluate the safety and efficacy of PD-L1 inhibitor durvalumab (D) and SABR after induction chemotherapy, in LA and BR PC. Herein, we present phase II data from two cohorts. Methods: A multi-institutional, non-randomized phase 1/2 trial of SABR and D was conducted at Cedars-Sinai Medical Center (CS) and Memorial Sloan Kettering (MSK). Key eligibility: ECOG 0-1; unresectable PC, with stable/responding disease following 2-3 cycles gemcitabine and nab-paclitaxel (GnP) or 4-6 months (m) FOLFIRINOX (FFX). CS enrolled pts with BR and LA PC; MSK enrolled pts with LA PC only. All pts received SABR between dose 1 and 2 of D, with slight variation in ablative radiation techniques (CS: 33Gy/5#; MSK: MRI adaptive ablative radiation, 50Gy/5#). D dosed on day 1: 750mg x 4 doses Q14 days, and subsequently Q2 or Q4 weeks x 1 year total, or until resection, progression of disease (POD), or limiting toxicity. Primary endpoint: 6-month progression free survival (6 m PFS). Secondary endpoints: rate of downstaging to resectability, objective response by RECIST v1.1, duration of response, PFS, Overall Survival (OS). PFS and OS estimated from consent date using Kaplan-Meier method. Association between surgery and outcomes was analyzed as a time dependent covariate in univariate Cox regression models for OS and PFS. To assess immunogenomic biomarkers of response, pre- and on-treatment tissue, blood and microbiome samples were gathered. Results: Between 08/2017 and 05/2022, N= 36 enrolled. Median age 67.5 years (range 48-79), 39% (14/36) female. Performance Status: N=12 (33%) ECOG 0; N= 24 (67%) ECOG 1. Median duration chemotherapy: 3.5 m (range 1.4, 5.8). Staging: N=31 (86%) LA; N=5 (14%) BR. N = 9 (25%) underwent resection; all R0. Conversion to resection was not associated with PFS (HR 0.45; 95% CI 0.17, 1.17, p=0.1) or OS (HR 0.69; 95% CI 0.26, 1.84, p=0.5). Survival and response data summarized in table. Toxicities of special interest: Grade 3 treatment-related: in N= 7; diarrhea N= 2; elevated AST/ALT N= 2; lipase/amylase elevation, N=1; nausea N=1. Grade 4: amylase elevation N=1. Conclusions: D and SABR after induction chemotherapy in LA and BR PC is safe, had an encouraging 6-m PFS of 69% and favorable R0 resection rate, warranting further evaluation. Immuno-genomic biospecimen analyses underway. Clinical trial information: NCT03245541.