Pancreatic Neuroendocrine Neoplasms: Computed Tomography Features And The Correlation To Pathological Tumor Grade

Feng Jin,Kai Wang,Tingting Qin,Xin Li,Feng Guo,Qiqi Jiang,Yan Zou,Ping Han
2019-01-01
Abstract:Objectives: To analyze the computed tomography (CT) features of pancreatic neuroendocrine neoplasms (pNENs) and to identify useful CT findings for differentiating tumors with different pathological grades. Materials and methods: Fifty patients with pNENs confirmed by postoperative pathology in our hospital from September 2012 to July 2016 who underwent preoperative plain and contrast-enhanced CT scans were included. CT features including tumor location, tumor size, shape, margins, growth pattern, pattern, CT attenuation values, enhancement patterns, degenerative changes, duct dilatation, peripancreatic tissue or vascular invasion, lymphadenopathy and metastases were evaluated. One-way ANOVA, the Kruskal-Wallis test, Fisher's exact test and ROC analysis were used to compare the CT features among different World Health Organization (WHO) tumor pathological grades grades. Results: The following pathological diagnoses were obtained in the 50 pNENs patients: 24 patients had G1 neuroendocrine tumors (NETs), 5 patients had G2 NETs, and 21 had G3 neuroendocrine carcinomas (NECs). The majority of G2 and G3 tumors were poorly defined and lobular in shape and were larger than G1 tumors. Cystic necrosis, pancreatic duct dilatation, bile duct dilatation, peripancreatic tissue or vascular invasion, lymphadenopathy and distant metastasis were absent in most G1 tumors. Tumor size, TNM stage (III), poorly defined margins, and heterogeneous density on contrast-enhanced CT had the highest sensitivity of 95.2% for distinguishing between G3 tumors and G1/2 tumors, while irregular shape had the highest specificity of 89.7%, with a sensitivity of 85.7%. Conclusions: Several CT features were verified as predictors of the pathological tumor classification of pNENs with a high sensitivity and specificity.
What problem does this paper attempt to address?