Bile Duct–Narrowing Neuroendocrine Tumors Show a PDX1+/CDX2+ Phenotype Resembling Duodenal Rather Than Pancreatic Neuroendocrine Tumors

Le résumé fourni par la source

Neuroendocrine tumors (NETs) associated with the intrapancreatic bile duct are rare and poorly characterized. Their relationship to conventional pancreatic neuroendocrine tumors (PanNETs) and to neuroendocrine cells of the periampullary and peribiliary regions remains unclear. A total of 199 resected NETs from the pancreas were evaluated for anatomical location and intrapancreatic bile duct narrowing. Transcription factor and hormone expression were assessed by whole-slide immunohistochemistry. For comparison, 22 duodenal NETs, 6 ampullary NETs and non-neoplastic duodenal, ampullary, and bile duct tissues were examined. Nineteen tumors (10%) were associated with bile duct narrowing, including 11 lower (periampullary) and 8 upper bile duct lesions. Compared with NETs without bile duct narrowing, these tumors were exclusively non-functioning, occurred more frequently in women and exhibited higher Ki-67 indices. Lower bile duct-narrowing tumors were associated with shorter progression-free survival. All but one bile duct-narrowing tumor expressed PDX1 (18/19, 95%), whereas CDX2 expression was observed in 73% (8/11) of lower bile duct-narrowing tumors. These tumors frequently expressed gastrin (73%) and somatostatin (73%), occasionally serotonin (27%), and lacked glucagon and insulin expression. Their transcription factor and hormone expression profiles closely resembled those of duodenal NETs and neuroendocrine cells of periampullary and peribiliary glands and differed from those of conventional PanNETs. Bile duct-narrowing NETs from the pancreas, particularly those involving the lower bile duct, represent a distinct clinicopathological subgroup characterized by a PDX1-positive, frequently CDX2-positive phenotype and enrichment for gastrin and somatostatin expression. Their resemblance to duodenal NETs and periampullary/peribiliary neuroendocrine cells supports a shared differentiation program and suggests a possible non-islet cell origin.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.