A case of small hepatocellular carcinoma with malignant ductular reaction

Herein reported is the unique case of a small hepatocellular carcinoma (HCC) with several foci of a minor (10% in area) component of "malignant ductular reactions". The patient was 51-year-old man who was a drinker. HBV/HCV were negative. The tumor was small (12×10×11 mm), solid, expansile and reddish-brown, and contained fibrous septa. The background was cirrhotic without alcoholic features. Histologically, the tumor was well differentiated HCC, and, besides the HCC, it contained several small foci consisting of the following four biliary epithelial elements: clusters of small cells (CSC), ductules (D), ductular hepatocytes (DH), and bile ducts (BD). The proportion of area was as follows: HCC 90%, CSC 3%, D 3%, DH 2%, and BD 2%. These non-HCC elements were intimately admixed and formed several foci that were characteristically located in the fibrous septa (FS), except for CSC which were situated among HCC cells close to FS. There were gradual merges between HCC and CSC, CSC and D, D and DH, and D and BD, respectively. Cells of CSC and D resembled rat oval cells. Cells of these four elements had atypical features regarded as malignant. Immunohistochemically (IHC), HCC were positive for arginase, HepPar1, and less frequently CK7. CSC were positive for CK7. D were positive for arginase, HepPar1, CK7, CK19, EMA, and EpCAM. DH were positive for arginase, HepPar1, and CK7. BD were positive for CK7, CK19, EMA, EpCAM and mucin. Although such tumors as this have been termed stem cell-related cancers, our case lacked definite evidence for stem cell origin in histology as well as in the IHC that showed negativity for KIT, CD34, and OCT3/4. The above findings suggest that CSC, D, DH and BD are analogous to the ductular reaction seen in hepatic inflammation. Therefore, we termed the phenomenon "malignant ductular reaction". It is suggested in the present tumor that at first only HCC developed, and then HCC cells in the interface with FS transformed to CSC, like a fetal ductal plate. Then, the CSC gave rise to D, which in turn led to DH and BD in FS, all findings of which are most likely sequential considering embryonic biliary development. The idea that the present tumor was at first D carcinoma and then D developed on one hand into CSC and HCC, and on the other into DH and BD seems possible, but its probability appears low because the vast majority of the present tumor had the phenotype of HCC.