Background and Objectives Despite routine inclusion of lymphovascular invasion (LVI) status in pathologic reports of resected pancreatic ductal adenocarcinomas (PDA), the clinical implications of LVI have not been well characterized. Methods This study is a retrospective review of 2640 patients who underwent a pancreatectomy for PDA at Thomas Jefferson University Hospital, Massachusetts General Hospital, or Johns Hopkins Hospital (2003-2014). Clinical and pathologic records were extracted from institutional databases. Results The median post-resection survival for the total cohort was 19.2 months with a 5-year survival rate of 15.2%. In a multivariate Cox proportional hazards model including conventional pathologic features, LVI was an independent predictor of survival (HR = 1.14, P = 0.017). In a stratified Kaplan-Meier survival analysis, patients with N0, LVI- PDA had a significantly improved overall survival compared to those with N0, LVI+ PDA (median 31 vs 24 mo, P = 0.020). Similarly, patients with N1, LVI- PDA had superior survival to patients with N1, LVI+ disease (18.6 vs 16.5 mo, P = 0.001). Conclusions As the first large scale study focused on the clinical impact of LVI status in PDA, these data indicate that this routinely reported pathologic feature is a bona fide and independent adverse prognostic factor.

The prognostic impact of extensive lymph node spread after resection for PDA plateaus above a specific threshold (LNC ≥ 4 or LNR ≥ 0.4), supporting the new 8th edition AJCC criteria for N2 disease. Clinically, lymph node spread above this threshold seems to correlate with occult systemic disease (elevated postoperative CA 19-9 and systemic pattern of failure).

In this cohort, common hepatic arterial abutment or encasement and residual lymph node disease at resection portended a particularly poor outcome in patients with localized, unresectable PDA.