A benign neoplasm composed of hepatocytes. Synonyms include hepatocellular adenoma, liver adenoma, and liver cell adenoma.

Hepatic adenomas are rare, affecting 3–4/100,000 individuals in North America and Europe [27]. Hepatic adenomas are a multifactorial disease in which gender (female), hormones (estrogen and androgens), and genetic predisposition all play important roles [28–30]. The vast majority of hepatic adenomas (approximately 85 %) occur in young to middle-aged women with a history of oral contraceptive use (usually taken for longer than 2 years) [31, 32]. Exogenous androgens and anabolic steroids are also a recognized risk factor in the development of hepatic adenomas [33]. Furthermore, a number of inherited/metabolic disorders are known risk factors for hepatic adenomas, including glycogen storage disease (especially types I and III) and maturity onset diabetes of the young (MODY 3) [22, 34–38]. The adenomas in the setting of glycogen storage disease are more common in males than females [35]. Finally, hepatic adenomas have been reported in individuals with a variety of other conditions including familial adenomatous polyposis, polycystic ovary syndrome, Peutz-Jeghers syndrome, and individuals using the antiepileptic medications carmabazepine and valproate [39, 40]. Clinical associations of hepatic adenomas are summarized in Table 4.1. Pediatric hepatic adenomas are also discussed further in Chap. 13.

The clinical presentation of hepatic adenoma varies, ranging from asymptomatic mass lesions found during evaluation for other processes, to acute presentation with tumor rupture. In most cases, individuals present with mild nonspecific abdominal pain or discomfort [41]. Hepatic adenomas are often solitary, but recent reports describe multifocality in up to 50 % of patients [41]. When the number of hepatic adenomas exceeds 10, the term “hepatic adenomatosis” is used [42]. Hepatic adenomatosis is seen in 10–25 % of individuals with adenomas and tends to show less association with oral contraceptive use and is more commonly seen in individuals with inherited disorders [40, 43, 44].

Unlike focal nodular hyperplasia, hepatic adenomas carry a risk of bleeding and malignant transformation. Bleeding occurs in 20–25 % of adenomas and can result in tumor rupture and subsequent hemoperitoneum [29, 36, 38, 41, 44]. The risk of malignant transformation into hepatocellular carcinoma is 5–10 %, based on resection specimens [45]. The risk of both bleeding and malignancy increases with tumor size; for instance, in most cases of malignant transformation, the adenomas are 5 cm or more in diameter [41].

Current management guidelines for hepatic adenomas are based mainly on tumor size, patient gender, and to a less degree on histologic subtype (see below, Sect. 4.2.7). Surgical resection is generally recommended for adenomas greater than 5 cm in size. These adenomas are managed more aggressively because of the increased risk of malignancy and bleeding in larger tumors. Hepatic adenomas in males are also generally referred for surgery because of a higher risk for malignancy (up to 50 % risk in some studies). Inflammatory adenomas are often treated by surgery because of modestly increased risk of bleeding. However, much of this increased risk for bleeding appears to be a function of tumor size, and tumors larger than 5 cm would be typically managed by surgery anyway.

Studies also suggest that beta-catenin-activated adenomas have an increased risk for malignancy [30, 36], although not all studies have replicated this finding [41]. The risk of malignancy is not strongly related to the number of lesions, so liver transplantation is not indicated solely on the basis of multiple adenomas [41]. Tumors that do not meet the above mentioned criteria for surgical treatment are managed by observation after discontinuation of oral contraceptive use [41].

The background liver is non-cirrhotic. Tumors can be solitary or multiple (Figs. 4.19 and 4.20). Although the majority of tumors that undergo resection measure 5–15 cm, some may measure up to 30 cm in diameter [18]. Tumors are well-demarcated, but not encapsulated. Their consistency ranges from soft to firm and the colors range from yellow-tan to brown. The cut surface often has a variegated appearance. Cystic change and hemorrhage can occur, often in the center of the lesion. Although focal scarring in areas of previous necrosis can be seen, prominent central scars of the type usually seen in focal nodular hyperplasia are typically absent.

Hepatic adenomas can be sharply delimited from the adjacent non-tumor liver (Fig. 4.21) or can blend in with the background liver (Fig. 4.22). In some cases, the adenoma and non-tumoral liver have an irregular border, with intertwined tumor and non-tumor (Figs. 4.23 and 4.24). Hepatic adenomas are typically not encapsulated (see Figs. 4.21, 4.22, and 4.23), but some cases may have a thin fibrous rim of compressed vessels and connective tissue (Fig. 4.25). The background liver usually shows no significant pathology or fibrosis. Exceptions include individuals with glycogen storage disease, in which case some degree of fibrosis can be seen. In addition, individuals with inflammatory type adenomas commonly have an elevated body mass index and their background liver can show fatty change.

Hepatic adenomas consist of benign hepatocytes; they differ from the background liver by the lack of portal tracts. Instead of portal tracts, aberrant naked arteries are typically scattered throughout the lesion (Fig. 4.26). The hepatocytes within the tumor are fundamentally indistinguishable from normal hepatocytes at high power magnification (Fig. 4.27). They typically have no cytological atypia and essentially no mitotic activity. One exception is androgen-related adenomas, which can demonstrate mild cytological atypia (Fig. 4.28). Hepatic adenomas only rarely have cholestasis (Fig. 4.29) or a pseudoacinar growth pattern (Fig. 4.30); these features are considered atypical and also occur more frequently in hepatic adenomas arising in association with androgen use. Fatty change can occur in hepatic adenomas (Fig. 4.31), but this tends to be characterized by steatosis only, without Mallory hyaline, ballooning, or perisinusoidal fibrosis; finding a true steatohepatitis pattern is atypical for hepatic adenoma and should raise consideration for a steatohepatitic variant of hepatocellular carcinoma. Hepatic adenomas can also demonstrate focal areas of prominent congestion (Fig. 4.32), areas of hemorrhage, and foci of tumor necrosis (Fig. 4.33). Finally, rare adenomas can show areas with striking sinusoidal fibrosis (Fig. 4.34). It is unclear if this represents a degenerative change or the presence of unique genetic changes.

Hepatic adenomas show an intact reticulin meshwork, similar to that of normal livers (Fig. 4.35). As a caveat, adenomas with fatty change can show occasional foci of reticulin loss that can mimic hepatocellular carcinoma (Fig. 4.36). In addition, there can be areas of reticulin loss adjacent to areas of intratumoral hemorrhage or necrosis (Figs. 4.37 and 4.38). Focal reticulin disruption in this setting should not be interpreted as evidence of malignancy.

Hepatic adenomas are strongly and diffusely positive for HepPar1 and Arginase-1 and are negative for Glypican 3. A CD34 typically shows patchy sinusoidal staining, in contrast to the diffuse sinusoidal staining typical of hepatocellular carcinoma. This marker, however, does not reliably distinguish adenomas from hepatocellular carcinoma in isolation. To illustrate, one study found that 35 % of inflammatory hepatic adenomas showed diffuse CD34 staining [46]. Ki-67 shows a very low proliferative index (<1–2 %) (Fig. 4.39). Inflammatory cells, however, will often show Ki-67 labeling and should be distinguished from tumor cells when evaluating proliferation (Fig. 4.40). One exception is tumor cells adjacent to areas of tumor necrosis, which can focally demonstrate higher proliferative rates (Fig. 4.41).

Because biopsy specimens usually provide only partial representation of the entire lesion, they should be interpreted with care. If the clinical history and the histological features are typical, a diagnosis of hepatic adenoma can be comfortably rendered on biopsy. On the other hand, if the clinical findings (e.g., older age or male gender) or histologic features (cytologic atypia, architectural atypia, or focal reticulin loss) are unusual, it is best to use the term “well-differentiated hepatocellular neoplasm,” and explain in a note that the differential diagnosis includes an atypical hepatic adenoma versus a well-differentiated hepatocellular carcinoma. In fact, definitive classification of a small subset of cases cannot be accomplished with certainty on needle biopsy.

Hepatic adenomas have been subdivided into four groups based on morphology, molecular findings, and immunohistochemical correlates [30, 36, 44, 47, 48]. Subtyping a hepatic adenoma should only be attempted once the diagnosis of hepatic adenoma has been made; if the diagnosis is uncertain (e.g., adenoma versus carcinoma), immunostains for hepatic adenoma subtyping are not useful. Both resection specimens and biopsies of hepatic adenomas can be successfully classified using immunostains. Although most adenomas are readily classified using this schema, sometimes individual stains can be difficult to interpret and may need to be repeated. The main subtypes of hepatic adenoma are described below and summarized in Table 4.2.