Objective: To determine the prognostic value of PLN and LNR based on a large series with standardized lymphadenectomy and pathological workup.
Background: Lymph node (LN) involvement is a major prognostic factor in pancreatic adenocarcinoma. However, the distinction N0/N1 is not sufficient to accurately predict prognosis. To improve prognostic accuracy in N1 tumors, different LN parameters have been tested. Previous studies were based on series with variable numbers of examined lymph nodes (ELN) and came to inconsistent conclusions as to the value of the number of positive lymph nodes (PLN) and the lymph node ratio (LNR).
Methods: 811 patients who underwent pancreatoduodenectomy for pancreatic adenocarcinoma between October 2001 and June 2012 were identified from a prospective database. Clinicopathological parameters included LN status (N0/N1), ELN, PLN, and LNR. Univariate and multivariate survival analyses were performed.
Results: The median number of ELN was 24 (interquartile range: 18–32). By univariate analysis, both PLN and LNR were significantly associated with survival in N1 tumors. However, by multivariate analysis, only the number of PLN was confirmed as independent predictor of survival. Median survival in patients with only 1 PLN was 31.1 months and comparable to the survival in N0 (33.2 months). With increasing numbers of PLN median survival significantly decreased (2–3 PLN: 26.1 months, 4–7 PLN: 21.9 months, ≥8 PLN: 18.3 months, P < 0.0001).
Conclusions: This study demonstrates that, based on high numbers of ELN, PLN is superior to LNR in predicting survival and allows to distinguish several N-categories that improve prognostic accuracy in LN-positive resectable pancreatic adenocarcinoma.
The TNM staging for pancreatic cancer only distinguishes N0/N1, and TNM-based clinical staging does not accurately predict prognosis in lymph node positive tumors. This single-center study in 811 patients demonstrates that, provided sufficient numbers of lymph nodes are examined, the number of positive nodes accurately predicts survival and allows to distinguish several prognostic categories of nodal involvement in resected pancreatic cancer.
*Department of Surgery
†Division of Biostatistics; and
‡Institute of Pathology, University Hospital Heidelberg, Heidelberg, Germany; and
§Department of Surgery, Ludwig-Maximilians University, Munich, Germany.
Reprints: Markus W. Büchler, MD, Department of Surgery, University of Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany. E-mail: firstname.lastname@example.org.
Disclosure: The authors declare that there are no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.annalsofsurgery.com).