Hepatocellular adenomas (HCA) are rare benign liver tumors. Most HCA are asymptomatic, yet, they bear a certain risk of bleeding and may progress to hepatocellular carcinoma (HCC). According to the Bordeaux classification, they are categorized in subgroups using (immuno)- histological and molecular features. These subtypes are known to have disparate risks. HNF-1alpha mutated HCA (H-HCA), which are characterized by steatosis, are very stable lesions; β-catenin mutated HCA (B-HCA) may exhibit nuclear atypia and have an increased risk for malignant transformation, inflammatory HCA (I-HCA), that are usually distinguished by dilated sinusoids, have increased risk of haemorrhage. I-HCA with β-catenin-mutations (B-I-HCA) harbour a greater probability of malignancy. In addition, a new HCA subtype (ASS1-HCA) with increased risk of bleeding was recently discovered.
Steatosis, the accumulation of lipid droplets (LD) in hepatocytes is well known to be associated with metabolic disorders. Also in tumorigenesis, tumor-associated
steatosis may occur. On the surface of LDs, amphiphilic proteins of the perilipin-family play a key role in lipid homeostasis by controlling fatty acid storage and release.
In order to determine whether the different subgroups of HCA have specific expression patterns of LDs and associated perilipins, we performed (immuno)-histological analysis in a collective of 51 HCA (including 3 B-HCA, 20 I-HCA and 9 H-HCA) previously diagnosed in our Institute, as well as in the respective non-neoplastic liver parenchyma.
Our histological examinations provide evidence that not only H-HCA but also one third of the I-HCA (n = 7/20) show significant steatosis in over 50% of the parenchymal sectional area (and up to 90%). Contrariwise, B-HCA constantly contained less than 5% of LDs. While in the vast majority of patients with B-HCA and H-HCA, the surrounding liver parenchyma contained no or few LDs (max. 10% of parenchyma,), in I-HCA patients, we noted 40% steatosis or more in almost one third (n = 6/20) of the cases, especially in 4 of the 7 cases with steatotic I-HCA. Immunohistology for perilipins 1 – 5 demonstrated microvesicular LD-accumulation in the different subgroups, even in B-HCA, which were not seen by conventional histology. Principally, all HCA-subtypes showed expression of perilipin 1 – 5, at least focally. Interestingly, we noted that while in I-HCA with steatosis (n = 7) perilipin 2 marked at least 50% of macrovesicular LDs and over 80% in the majority of the cases (n = 4/7), perilipin 2 was not as strongly expressed on contours of macrovesicular LDs in H-HCA: in average only 40% of macrovesicular LDs were marked and in 4 of the 9 cases under 30%.
In summary, our investigations show that I-HCA, classically not associated with increased LD storage, may show significant steatosis. This is highly relevant because current radiological attempts to determine the HCA-subtype and thus estimate the risk profile are partially based on lipid content of the lesion. Moreover, it would be interesting to investigate in further studies whether the lipid storage profile of I-HCA may correlate with the risk of β-catenin activation and therefore the risk of malignant transformation.
This project contains data of an unfinished doctoral thesis by Diane Duret.
