Cdx2 expression in pancreatic tumors: Relationship with prognosis of invasive ductal carcinomas

Kakuya Matsumoto, Tsutomu Mizoshita, Tetsuya Tsukamoto, Naotaka Ogasawara, Akihiro Hirata, Yasuhiro Shimizu, Masakazu Haneda, Kenji Yamao, Masae Tatematsu

Research output: Contribution to journalArticlepeer-review

27 Citations (Scopus)


We have previously reported that Caudal-related homeobox gene 2 (Cdx2) is a useful prognostic intestinal phenotypic marker for advanced gastric cancers. In this study, we examined Cdx2 expression and phenotype in pancreatic tumors. We evaluated 19 mucinous cystic tumors (MCTs), 17 intraductal papillary-mucinous tumors (IPMTs), and 41 invasive ductal carcinomas (IDCs) with regard to their gastrointestinal phenotype. The expression of Cdx2 was also assessed immunohistochemically. The lesions were phenotypically divided into 39 gastric (G type), 29 gastric and intestinal mixed (GI type), 3 intestinal (I type), and 6 null (N type) types, independent of the histopathological type. Most of the pancreatic tumors were thus judged to be the positive for gastric phenotypic members. Cdx2 nuclear staining demonstrated a close relation to the intestinal phenotypic expression in all three types (MCTs, IPMTs, and IDCs; p<0.05). In IDCs, Kaplan-Meier analysis of Cdx2 expression showed the Cdx2-positive group to have a significantly better outcome than their negative counterparts (p=0.015). In conclusion, our data suggest that Cdx2 might be necessary for intestinal phenotypic expression even in pancreatic tumor cells. In addition, Cdx2 expression in IDCs may be a novel prognostic marker for patient survival.

Original languageEnglish
Pages (from-to)1239-1243
Number of pages5
JournalOncology reports
Issue number6
Publication statusPublished - 12-2004
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Oncology
  • Cancer Research


Dive into the research topics of 'Cdx2 expression in pancreatic tumors: Relationship with prognosis of invasive ductal carcinomas'. Together they form a unique fingerprint.

Cite this