AIMS: The histogenesis and biological behaviour of peripheral intrahepatic cholangiocarcinoma (peripheral CC) remain unclarified. The aim of this study was to examine the growth pattern of peripheral CC (24 cases) in comparison with hepatocellular carcinoma (HCC, 27 cases) and metastatic colorectal adenocarcinoma (MCA, 24 cases). METHODS AND RESULTS: Tumour/surrounding liver borders were classified as: (i) fibrous encapsulation, (ii) compressive growth, and (iii) infiltrating replacement. Nineteen of 24 peripheral CCs showed (iii), whereas 23 of 27 HCCs showed (i) and 17 of 24 MCAs showed (ii). In (iii), carcinoma cells infiltrated the surrounding liver without compression, and hepatic supporting vascular structures such as portal tracts were secondarily incorporated into the tumour. In (i) and (ii), the surrounding liver was compressed and no or few portal tracts were incorporated within the tumour. Fifteen of 24 peripheral CCs were composed of carcinoma cells resembling reactive bile ductules and these cells were positive for neural cell adhesion molecule (NCAM), a marker of proliferating bile ductules. The remaining nine peripheral CCs were composed of ordinary adenocarcinoma and negative for NCAM. CONCLUSIONS: A subgroup of peripheral CCs with an infiltrating replacement growth pattern resembles reactive bile ductules and expresses NCAM. 'Bile ductular carcinoma' may be a better term for this subgroup.
AIMS: The histogenesis and biological behaviour of peripheral intrahepatic cholangiocarcinoma (peripheral CC) remain unclarified. The aim of this study was to examine the growth pattern of peripheral CC (24 cases) in comparison with hepatocellular carcinoma (HCC, 27 cases) and metastatic colorectal adenocarcinoma (MCA, 24 cases). METHODS AND RESULTS:Tumour/surrounding liver borders were classified as: (i) fibrous encapsulation, (ii) compressive growth, and (iii) infiltrating replacement. Nineteen of 24 peripheral CCs showed (iii), whereas 23 of 27 HCCs showed (i) and 17 of 24 MCAs showed (ii). In (iii), carcinoma cells infiltrated the surrounding liver without compression, and hepatic supporting vascular structures such as portal tracts were secondarily incorporated into the tumour. In (i) and (ii), the surrounding liver was compressed and no or few portal tracts were incorporated within the tumour. Fifteen of 24 peripheral CCs were composed of carcinoma cells resembling reactive bile ductules and these cells were positive for neural cell adhesion molecule (NCAM), a marker of proliferating bile ductules. The remaining nine peripheral CCs were composed of ordinary adenocarcinoma and negative for NCAM. CONCLUSIONS: A subgroup of peripheral CCs with an infiltrating replacement growth pattern resembles reactive bile ductules and expresses NCAM. 'Bile ductular carcinoma' may be a better term for this subgroup.