SEER-based evaluation of lymph node yield as a prognostic indicator of cancer-specific survival in nonmetastatic pancreatic ductal adenocarcinoma.

Abstract

Background/objectives: Although the American Joint Committee on Cancer (AJCC) 8th edition recommends harvesting at least 12 lymph nodes for optimal staging in pancreatic ductal adenocarcinoma (PDAC), the precise lymph node yield (LNY) needed for accurate prognostication in different treatment settings remains unclear. This study aimed to identify subgroup-specific LNY cutoffs and evaluate their prognostic significance in nonmetastatic PDAC.

Methods: We analyzed 5609 patients with nonmetastatic PDAC from the Surveillance, Epidemiology, and End Results (SEER) database undergoing pancreatectomy. Patients were categorized by nodal status (N0 vs. N+) and receipt of neoadjuvant therapy (NAT) or upfront surgery (UPS). We used maximum selected rank statistics and a conditional inference tree approach to determine optimal LNY cutoffs for each subgroup. Kaplan-Meier curves and Cox proportional hazards models were employed to assess cancer-specific survival (CSS) and identify independent prognostic factors.

Results: Distinct LNY thresholds were identified for N0 (>13) and N+ (>10) cohorts, with the highest cutoffs in N0-NAT subgroups (>27). Across all analyses, patients exceeding these LNY cutoffs demonstrated significantly prolonged CSS. The N0-NAT group with LNY >27 achieved the longest median survival (60 months), whereas N+ patients undergoing UPS with LNY ≤10 had the poorest outcomes (16 months). Multivariate Cox regressions consistently showed that higher LNY was an independent predictor of improved survival.

Conclusions: Higher LNY thresholds than the current AJCC standard of 12 appear beneficial for more accurate staging and improved survival in resected PDAC. Tailoring LNY goals based on nodal status and treatment modality may further refine prognostic stratification and guide more effective therapeutic strategies.