Immunochemotherapy for Borderline Resectable or Locally Advanced Pancreatic Cancer: A Randomized Controlled Trial

PURPOSE: The optimal treatment strategy for patients with borderline resectable pancreatic cancer (BRPC) and locally advanced pancreatic cancer (LAPC) remains undefined. We evaluated whether the addition of programmed cell death protein 1 blockade to modified FOLFIRINOX improves outcomes compared with modified FOLFIRINOX alone. METHODS: In this open-label, randomized phase II trial, patients with histologically confirmed BRPC or LAPC were randomly assigned (1:1) to receive modified FOLFIRINOX plus toripalimab (immunochemotherapy) or modified FOLFIRINOX alone (chemotherapy). Treatment response was assessed every 2 months, and surgical eligibility was determined by a multidisciplinary team. The primary end point was event-free survival (EFS) in the intention-to-treat population. RESULTS: = .007). The median overall survival (OS) was 20.6 versus 17.1 months (HR, 0.81; 95% CI, 0.63 to 1.03). Objective response rates were 30.1% (59/196) and 18.9% (37/196), respectively. Tumor resection rates were 42.3% (83/196) and 43.9% (86/196), and R0 resection rates were 84.1% (69/82) and 74.7% (62/83), respectively. Prespecified subgroup analyses suggested improved EFS and OS in BRPC but not LAPC. Grade 3 or higher adverse events occurred in 47.9% (93/194) and 49.2% (96/195) of patients receiving immunochemotherapy and chemotherapy, respectively. Grade III or higher postoperative complications occurred in 30.1% (25/83) and 25.6% (22/86) of resected patients. CONCLUSION: The addition of PD-1 blockade to modified FOLFIRINOX significantly improved EFS without increasing treatment-related toxicity or surgical morbidity in patients with BRPC or LAPC.