This is the authors' abstract. We don't add key points for this paper.
Biliary adenofibroma (BAF) is an extremely rare benign hepatic tumor. It is characterized by dilated ductular structures lined with bland cuboidal epithelial cells and abundant fibroblastic stroma. The histological components and expression pattern of foregut antigens (D10+ and IF6-) are similar to those of bile duct hamartomas (von Meyenburg complexes); however, BAFs are larger and lack typical von Meyenburg complexes in the background liver.1, 2 Because of their rarity, BAF histogenesis and its histological and biological spectra remain unclarified. Since 1993, there have been several reported cases of BAF with either benign1-6 or suspicious features,3 in situ/microinvasive disease,4 and malignant5 pathology. We report on a case of BAF with features suggestive of malignant transformation. A 37-year-old Caucasian woman presented with a 5-month history of postprandial nausea, vomiting, and epigastric pain. Her past medical history was notable only for chronic constipation, tubal ligation, and liver cancer in her maternal grandmother. Her initial studies included an abdominal computed tomography scan demonstrating a multiloculated hepatic lesion posterior to the gallbladder fossa in segment V, radiographically interpreted as a complex cystic lesion likely cystadenoma. She was subsequently referred for further evaluation and management. Magnetic resonance imaging (MRI) scans of the abdomen were obtained and the mass was 4.9 × 4.8 × 4.6 cm. The mass was T1 hypointense, heterogeneously T2 hyperintense, and displayed varying levels of peripheral enhancement on postcontrast sequences with washout (Fig. 1A,B). The mass was excised with wide local margins. Her postoperative course was uncomplicated and she was discharged on postoperative day 5. Gross examination of the resection specimen revealed a 4.5 × 4.1 × 3.0 cm mass with pushing borders. There was no communication between the lesion and the intrahepatic bile ducts. Histological sections showed a lobulated neoplasm consisting of tubuloglandular and microcystic structures lined by cuboidal to short columnar epithelial cells (Fig. 2A). The lining epithelial cells had rounded to slightly elongated, normochromatic to mildly hyperchromatic nuclei, and moderate amounts of eosinophilic cytoplasm (Fig. 2B). No goblet cells or mucin hypersecretion were observed. Many areas of the lesion included tubulocystic structures with varying amounts of intervening fibrous stroma; however, the tumor showed a heterogeneous “lobule within lobule” appearance with internal areas of architectural complexity (Fig. 2A), often showing back-to-back or cribriform arrangements of tubular and microcystic structures without intervening stroma (Fig. 2C). There were also small foci of low papillary epithelial overgrowth with increased mitotic activity. Whereas the tumor grossly had pushing borders, multiple areas had small islands of tumor cells infiltrating the fibrous stroma (Fig. 2D). Immunohistochemically, the epithelial cells stained positive for cytokeratin (CK) 7 (diffuse), CK19 (>90%), synaptophysin (40%-50%), and CD56 (15%-20%). Staining was negative for chromogranin, CK20, caudal-type homeobox 2, HepPar-1, and p53. Whereas a significant portion of the lesion was positive for Ki67 (10%-15% of epithelial cells; Fig. 2B, inset), the aforementioned areas of cribriform and papillary epithelial overgrowth had a markedly increased Ki-67 index (50%; Fig. 2C, inset). All margins were negative. Based on these findings, we diagnosed a liver epithelial neoplasm, strongly favoring well-differentiated cholangiocarcinoma (CCA) arising in a BAF. BAF is an extremely rare tumor affecting both sexes, with only 10 BAFs and 2 cases of unclassified tumor with BAF features reported in the literature (Table 1). Although it has been reported as a benign biliary neoplasm, 3 cases in the literature showed features of malignant transformation.10-12 Other than frank invasion and distant metastasis, Tsui et al. suggested that dysplastic and papillary epithelial changes might be precursors to malignancy.1 Based on the histological features of this tumor and previously reported cases, the presence of epithelial components with architectural complexity, such as atypical papillary projections and back-to-back or cribriforming glands, which are not typical of BAF, may indicate tumor progression or malignant transformation.9 Ki-67index (%) papillary growth or “back to back” glands BAF consists predominantly of microcystic structures that are lined by bland-appearing cuboidal or low-columnar epithelium and is histologically distinct from ordinary intrahepatic cholangiocarcinoma. The typical growth patterns of intrahepatic CCAs are tubular, cord-like, and papillary, and histologically malignant glands are dispersed in a haphazard fashion in a variably desmoplastic stroma. Some rare subtypes, such as cholangiocellular carcinoma and CCA with a predominant “ductal plate malformation” pattern, tend to be extremely well differentiated and mimic benign biliary structures (e.g., reactive bile ductules and malformed bile ducts).13, 14 However, to date, biliary epithelial tumor with extensive microcystic growth pattern is not considered to be within the histological spectrum of ordinary CCA. Intraductal papillary neoplasm of the bile duct (IPN-B), a biliary counterpart of intraductal papillary mucinous neoplasm of the pancreas, is characterized by intraductal growth of neoplastic biliary epithelium and exhibits an exophytic or polypoid mass within the lumen of the bile duct. IPN-B can be benign, borderline, or malignant. The tumor grows in a complex papillary or papillotubular fashion and can be histologically similar to BAF. However, as the name implies, the presence of a communication between the lesion and the bile duct system is a necessary feature for the diagnosis of IPN-B. In the present case, there was no bile duct communication either grossly or microscopically, and this finding, together with a predominantly microcystic growth pattern, made the diagnosis of IPN-B unlikely. The tumor described in this case had unequivocal positive immunoreactivity for synaptophysin and, to a lesser degree, for CD56, suggesting mixed glandular and neuroendocrine differentiation. It remains unclear whether partial neuroendocrine differentiation is common in BAF because neuroendocrine marker expression was not evaluated in other cases. Whereas the literature potentially under-represents the number of benign BAFs, the rate of invasive malignancy in the literature is 30.8% (4 of 13). We recommend wide local resection of these lesions and imaging surveillance follow-up for potential recurrence. We are grateful to Dr. Anthony Demetris from Department of Pathology at the University of Pittsburgh for his valuable opinion regarding the pathological diagnosis.
No takes yet. Share an insight, caveat, or question.
Kaminsky et al. (2016) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: