Abstract and Introduction
Objectives: β-Catenin–activated hepatocellular adenomas have an elevated risk of harboring foci of hepatocellular carcinoma. Inflammatory adenomas also have an increased propensity for malignant transformation and are associated with a systemic inflammatory syndrome. Patients with these two adenoma subtypes benefit from excision. We assessed whether β-catenin–activated and inflammatory adenomas could be identified using a limited immunohistochemical panel.
Methods: Forty-six adenomas were assessed by morphology and β-catenin, serum amyloid A, and glutamine synthetase immunostains.
Results: Morphologic examination produced a morphologic working diagnosis of inflammatory adenoma in 25 (54%) of 46 cases, β-catenin–activated adenoma in three (7%) of 46 cases, and 18 (39%) of 46 cases of other adenomas. After immunohistochemical staining, the morphologic diagnosis was confirmed in 15 (33%) of 46 and changed in 20 (43%) of 46, for a final distribution of 16 (35%) of 46 inflammatory adenomas, four (9%) of 46 β-catenin–activated adenomas, seven (15%) of 46 β-catenin–activated inflammatory adenomas, and 19 (41%) of 46 other adenomas.
Conclusions: Inflammatory and β-catenin–activated adenomas were readily identified by immunostaining patterns. These findings reinforce the necessity of immunohistochemistry in classifying adenomas, as assessing morphology alone often provided inaccurate subclassification. β-Catenin–activated and inflammatory adenomas can be accurately diagnosed using only a limited panel of widely available immunostains.
Hepatocellular adenomas (HCAs) can be divided into four subtypes based on molecular techniques. β-Catenin–activated HCA (b-HCA) has a notably increased risk of harboring foci of hepatocellular carcinoma (HCC).[2,3] Inflammatory HCA (IHCA) has abnormalities of the interleukin 6/JAK/STAT pathway.[1,4] Some IHCAs also have β-catenin activation (b-IHCAs). Hepatocyte nuclear factor 1 α-mutated HCA (H-HCA) is characterized by biallelic loss of HNF1A and a loss of expression of liver fatty acid binding protein (L-FABP).[5,6] A fourth subtype of HCA, the unclassified HCA (UHCA), is defined by a lack of known molecular abnormalities.
With the discovery of these molecular subtypes has come a proliferation of immunohistochemical markers for their diagnosis, including β-catenin and glutamine synthetase (GS) for b-HCA,[6,7] L-FABP for H-HCA, and serum amyloid A (SAA) and C-reactive protein (CRP) for IHCA.[7,8] Studies have used up to seven markers to classify hepatocellular lesions. Although there is some overlap, the different subtypes also have different morphologic features, including macrovesicular steatosis in H-HCA; inflammatory infiltrates, dilated sinusoids, ductular reaction, and "pseudo-portal tracts" of arterioles embedded in dense collagen patches in IHCA; and pseudoglandular formations and nuclear atypia in b-HCA.[5,6,8]
Overall, 4.2% of HCAs may undergo malignant transformation. However, several factors have been identified that may dramatically increase this risk. HCAs in males and lesions over 5 cm in size harbor an increased risk of malignancy, and excision is recommended.[3,11,12] Up to 46% of b-HCAs may contain malignant foci, and excision is again recommended.[2,3,5,11] IHCA alone and with concomitant β-catenin mutations (b-IHCAs) also have an independent risk of malignant transformation.[4,12,13] IHCA has also been associated with a systemic inflammatory syndrome that may be cured by excision.[6,14,15]
On the other hand, the pathologic diagnosis of H-HCA and UHCA has not been shown to be of immediate therapeutic impact. H-HCAs are associated with rare familial conditions, including familial adenomatosis and maturity-onset diabetes of the young.[16,17] Therefore, they could potentially require additional clinical investigation. However, available literature on IHCA and b-HCA strongly suggests surgical excision is the treatment of choice, making these two adenoma subtypes necessary to identify at the time of tissue examination.
Therefore, the current study was undertaken to specifically investigate whether IHCA and b-HCA could be diagnosed using only a limited but broadly available immunohistochemical panel that could be used in routine clinical practice. Secondarily, the correlation of clinical and morphologic features was investigated by adenoma subtype.
Am J Clin Pathol. 2017;147(6):557-570. © 2017 American Society for Clinical Pathology