The incidence of pancreatic neuroendocrine tumors (PNETs) is increasing over the past few decades. Surgery for low-grade and small PNETs of less than 2cm and N0M0 is still debated. The purpose of this study is to examine the association between surgical resection and survival in patients with low-grade nonfunctioning PNETs.

Patients diagnosed with PNETs between 2004 and 2015 were extracted from SEER. Kaplan–Meier methods and Cox proportional hazard models were used to estimate independent predictors in PNETs patients.

A total of 2637 patients (2147 underwent surgical resection and 490 did not undergo surgery) with histologically confirmed low-grade PNETs in this cohort study. Overall survival (OS) and cancer-specific survival (CSS) of patients with surgery was better than those without surgery (log rank test P_OS<0.001, P_CSS<0.001). Multivariate Cox regression analysis showed that surgical status was an independent prognostic factor associated with OS (HR 3.257, 95%CI: 2.635, 4.026) and CSS (HR 3.546, 95%CI: 2.798, 4.493). Subgroup analysis suggested the patients receiving surgery apparently had better OS and CSS regardless of tumor size (all log rank test P_OS<0.001, all log rank test P_CSS<0.001) and SEER stage (all log rank test P_OS<0.001, all log rank test P_CSS<0.001), compared to patients without removal of the primary tumor.

Surgical resection of primary tumor may have a significant benefit on survival for patients with low-grade nonfunctioning PNETs. To determine the optimal management, grade, stage and tumor size should be considered comprehensively.