|
|
Back to 2016 Annual Meeting
The Prognostic Value of Lymph Node Status and Extent of Lymphadenectomy in Pancreatic Neuroendocrine Tumors Confined To and Extending Beyond the Pancreas
Onur Kutlu*2, Jeffrey E. Lee1, Jean-Nicolas Vauthey1, David Adams2, Michael P. Kim1, Jason B. Fleming1, Matthew H. Katz1, Claudius Conrad1
1Surgical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX; 2Division of Surgical Oncology, Medical University of South Carolina, Charleston, SC
Background: Evidence regarding the impact of lymph node status and the value of extended lymphadenectomy (LA) in predicting survival is controversial in pancreatic neuroendocrine tumors (pNET). We aim to identify the impact of tumor extension, grade and location on nodal metastasis, disease specific (DSS) and overall survival (OS).
Methods: The SEER database was queried for patients with histologically proven pNET who underwent surgery from 1998-2012. Patients with unknown grade, T-, M-stage; or with M1-status, multifocal pNET, < 1-month follow-up, or no survival data were excluded. Binary logistic regression was performed for factors associated with nodal status. Kaplan Meier survival analyses were performed to assess the impact of T-stage, grade and nodal status on DSS and OS, and Cox analyses performed for independent predictors of DSS and OS. Patients without defined lymph node status were categorized as Nx (n=165, 16.8%); extended LA was defined ≥10 lymph nodes harvested (n=406, 41.4%). To assess the impact of N-status on DSS and OS, Nx vs. N0 vs. N1 was compared in two T-stage groups (T1-T2 and T3-T4). Further, for these T-stage groups, patients who had < vs. ≥ 10 nodes dissected were compared for the impact of extended LA.
Results: 981 of 5349 patients fit the inclusion criteria. For T1-T2 tumors, N-status was affected only by tumor size; age, sex, location, and grade did not impact N-status. For T3-T4 tumors, neither age, grade, sex, location nor size impacted N-status. For T1-T2 tumors, Cox analyses showed that N status (p=0.001), grade (p<0.001), age (p=0.001) and sex (p=0.007) were associated with OS, while tumor size (p=0.260) and location (p=0.331) were not. For T3-T4 tumors, grade (p<0.001), sex (p=0.004), tumor size (p=0.013) and age (p=0.007) impacted OS; N-status did not impact OS(p=0.789;Table). Specifically, for T1-T2 tumors, OS(p=0.008) and DSS(p=0.003) were longer for N0 vs. N1 tumors, while N0 vs. Nx patients had similar OS (p=0.59) and DSS (p=0.80). Nx compared to N1 status showed a trend to improved OS(p=0.08) and improved DSS(p=0.04). For T3-T4 patients, N-status did not affect OS(p=0.454) or DSS(p=0.365). For all T-groups and any N-status, extended LA was not associated with an improved survival (Figure).
Conclusion: For pNETs confined to the pancreas (T1-T2), N1-status is a significant predictor of poor OS. The comparable outcome of N0 vs. Nx supports limited resection without LA (including enucleation) for selected T1-T2 tumors without clinical evidence of nodal metastasis. However, the inferior outcome of N1-status highlights the importance of pre-operative assessment of regional nodes with high quality imaging. While identification and removal of involved lymph nodes in T1-T2 tumors via extended LA might be occasionally helpful in providing prognostic information, extended LA is unlikely to be impactful in T3-T4 tumors.
Figure 1
  Figure 1
Table 1
Back to 2016 Annual Meeting
|
