Prediction of Tumor Prognosis of Pancreatic Neuroendocrine Tumors Using Image, Surgical and Pathologic Findings.

Abstract

Objectives: To evaluate the magnetic resonance imaging (MRI) and computed tomography (CT) findings along with other surgical and pathologic features as prognosis predictors in pancreatic neuroendocrine tumors (PNETs).

Methods: In this study, we retrospectively analyzed a clinical data pool of patients with pathologically confirmed PNETs. CT and MRI findings were evaluated as potential prediction parameters of tumor-nodes-metastases (TNM) stage and grade, using Fisher's exact test. Univariate and multivariate logistic regression models were used to estimate the risk factors associated with tumor recurrence after surgery. The Kaplan-Meier method and Cox proportional hazards model were used for recurrence-free survival analysis.

Results: The predictors of higher TNM stages were tumor diameter, tumor boundary, distant metastases, and lymphadenopathy on CT scan. From MRI images, tumor diameter, T2-weighted image, tumor enhancement, and pancreatic duct dilatation showed statistically significant differences among TNM stages. Univariate analysis showed that American Joint Committee on Cancer (AJCC) TNM stage, World Health Organization (WHO) tumor grade, sex, smoking, and drinking were associated with tumor recurrence and disease-free survival (DFS); while tumor and metastasis also affected DFS. Multivariate survival analysis confirmed that AJCC TNM was an independent predictor after adjusting other covariates. Peripancreatic invasion and lymph node metastases as well as blurred boundary detected by CT or MRI may be independent risk factors for TNM stage and clinical outcome of PNETs.

Conclusion: TNM stage is a valuable predictor of prognosis in PNETs. Information from CT and MRI imaging can be used to determine the TNM stage, and to estimate the tumor prognosis, guide the follow-up, and avoid ineffective treatments.