Exploration of the Exact Prognostic Significance of Lymphatic Metastasis in Jejunoileal Neuroendocrine Tumors

Background. How to evaluate the prognostic significance of lymphatic metastasis in patients with small bowel (jejunoileal) neuroendocrine tumors (SBNETs) is still not conclusive. Methods. Data for patients with SBNETs, but without distant metastasis, were retrieved from the Surveillance, Epidemiology, and End Results database. Recursive partitioning analysis (RPA) was used for classification development by combining examined lymph nodes (ELNs) and lymph node ratio (LNR). Results. Overall, 1925 patient records were retrieved. Patients with N0 and N1 disease (based on the definition of the European Neuroendocrine Tumor Society [ENETS] staging classification) did not have different OS (p = 0.7867), nor did patients with N0, N1 (< 12 positive nodes), and N2 (>= 12 positive nodes) disease based on the definition of American Joint Committee on Cancer (AJCC) 8th edition staging classification (p = 0.5276). However, Cox regression analysis indicated that both ELNs (hazard ratio [HR] 0.968, 95% confidence interval [CI] 0.949-0.987; p = 0.0013) and LNR (HR 2.288, 95% CI 1.122-3.682; p = 0.0006) were prognostic factors. Using RPA, we combined ELNs and LNR, and patients were reclassified into three groups (group 1: ELNs >= 12, any LNR; group 2: ELNs < 12, LNR < 0.35; group 3: ELNs < 12, LNR >= 0.35). Survival analysis and multivariate Cox regression showed that groups 1, 2, and 3 had progressively worse survival. Furthermore, we found that ELNs >= 12 could remarkably improve patient survival (p < 0.001). Conclusions. The current definition of lymphatic metastasis could not help predict patient survival. Our newly proposed classification of lymphatic metastasis is better than the ENETS and AJCC 8th edition staging classifications in evaluating the prognostic significance of lymphatic metastasis in SBNETs. Systematic resection of lymph nodes (>= 12) could help improve patient survival.