OBJECTIVE: The World Health Organization 2010 classification divides pancreatic neuroendocrine tumors (p-NETs) entity to well-differentiated neuroendocrine tumors (NET) and poorly differentiated neuroendocrine carcinomas (NEC) by Ki-67 index. The aim of this study is elucidate the pathophysiology and tumor biology of p-NETs. METHODS: We assessed the expression of transcription factors sex determining region Y-box 2 (SOX2) and pancreatic and duodenal homeobox 1 (Pdx1)) essential for the normal fetal development of pancreatic neuroendocrine cells in 46 surgically resected p-NETs by immunohistochemistry. The relationship of expression levels of these factors and clinicopathological factors were analyzed. RESULTS: SOX2 was positive in 6 p-NETs (13.0%). Five of 7 NEC patients showed positive for SOX2. SOX2 was highly (sensitivity 71%) and specifically (specificity 97%) expressed in NEC. Patients with SOX2 positive p-NET showed the significantly shorter disease-free and overall survival than patients with SOX2 negative p-NET. High Pdx1 expression was seen in 25 p-NET patients (54.3%). None of the NEC patients showed high Pdx1 expression. There was a significant reverse correlation between SOX2 and Pdx1 expression. CONCLUSIONS: The expression patterns of SOX2 and Pdx1 highly correlated with prognosis of p-NETs. These expression patterns may represent the biological and pathophysiological difference of p-NETs and indicate the origin of tumor.
OBJECTIVE: The World Health Organization 2010 classification divides pancreatic neuroendocrine tumors (p-NETs) entity to well-differentiated neuroendocrine tumors (NET) and poorly differentiated neuroendocrine carcinomas (NEC) by Ki-67 index. The aim of this study is elucidate the pathophysiology and tumor biology of p-NETs. METHODS: We assessed the expression of transcription factors sex determining region Y-box 2 (SOX2) and pancreatic and duodenal homeobox 1 (Pdx1)) essential for the normal fetal development of pancreatic neuroendocrine cells in 46 surgically resected p-NETs by immunohistochemistry. The relationship of expression levels of these factors and clinicopathological factors were analyzed. RESULTS:SOX2 was positive in 6 p-NETs (13.0%). Five of 7 NEC patients showed positive for SOX2. SOX2 was highly (sensitivity 71%) and specifically (specificity 97%) expressed in NEC. Patients with SOX2 positive p-NET showed the significantly shorter disease-free and overall survival than patients with SOX2 negative p-NET. High Pdx1 expression was seen in 25 p-NET patients (54.3%). None of the NEC patients showed high Pdx1 expression. There was a significant reverse correlation between SOX2 and Pdx1 expression. CONCLUSIONS: The expression patterns of SOX2 and Pdx1 highly correlated with prognosis of p-NETs. These expression patterns may represent the biological and pathophysiological difference of p-NETs and indicate the origin of tumor.