Background
Retrospective cohort study from MD Anderson Cancer Center defining and analyzing outcomes in borderline resectable pancreatic cancer (BRPC). 160 consecutive patients with BRPC undergoing preoperative therapy and planned pancreatectomy. Defined anatomic criteria distinguishing resectable, borderline resectable, and locally advanced pancreatic cancer based on SMA, celiac, and portal/SMV involvement.
Interventions and follow up
Arm A: Preoperative chemotherapy ± radiation followed by attempted resection (heterogeneous regimens)
Arm B: N/A — single-arm retrospective cohort
Primary endpoint: R0 resection rate and overall survival in BRPC patient
mFollow up: 22 month
Arm B: N/A — single-arm retrospective cohort
Primary endpoint: R0 resection rate and overall survival in BRPC patient
mFollow up: 22 month
Results
Resection rate: 83% (132/160)
R0 resection: 84% of resected patients
mOS (all BRPC patients): 40 months
mOS (resected): 40 months vs 13 months (unresected)
R0 resection: 84% of resected patients
mOS (all BRPC patients): 40 months
mOS (resected): 40 months vs 13 months (unresected)
Adverse events
Main adverse events: Preoperative therapy tolerability data not systematically reported. Perioperative morbidity comparable to standard pancreatectomy series.
Conclusions
Anatomic criteria-based classification of borderline resectable pancreatic cancer and aggressive preoperative therapy followed by surgery can achieve high R0 resection rates (~84%) and promising OS (~40 months) in this subset. The MD Anderson classification system for BRPC became widely adopted.
Key Limitations
Key Limitations: Retrospective, single-institution study with selection bias — motivated patients at a high-volume center. Heterogeneous preoperative regimens. BRPC classification has since been refined by NCCN, AHPBA/SSO/SSAT, and ALLIANCE consensus definitions. No randomized comparison to upfront surgery. Long-term OS benefit of preoperative versus perioperative therapy in BRPC was not established until PREOPANC (Tienhoven 2020) and ongoing ALLIANCE A021806.
Clinical Context
This paper anchored the concept of preoperative therapy as the preferred approach for BRPC and influenced NCCN guidelines to recommend preoperative therapy for borderline resectable disease. Alliance A021101 subsequently provided Phase II prospective data for this approach. Ongoing phase III Alliance A021806 (mFOLFIRINOX neoadjuvant vs adjuvant) will further define the role of preoperative therapy in resectable and borderline resectable pancreatic cancer.
References