Autoimmune hepatitis-is histology conclusive?

Liver biopsy is an essential and necessary element in the diagnosis and management of autoimmune hepatitis, and is of very special importance in the scoring system for diagnosis. Histopathology shows moderate to severe inflammatory infiltrates with abundant plasma cells in the enlarged portal tracts with interface hepatitis and moderate to severe necroinflammatory lesions in the lobules with lymphoplasmacytic reaction. Regeneration develops with rosette formation and regenerative nodules. One important issue is the differentiation between acute onset of autoimmune hepatitis and a flare up of chronic disease; this cannot be diagnosed in some cases clinically and therefore requires a biopsy to evaluate the stage of the disease. There are some variants of the disease with cholestatic features such as autoantibody negative autoimmune hepatitis and giant cell hepatitis as well as overlap syndromes with primary biliary cholangitis and primary sclerosing cholangitis. Clinically, three types of autoimmune hepatitis are differentiated according to autoantibody formation and the clinical picture, however, histopathologically there is no difference between these three types. Differential diagnosis of autoimmune hepatitis includes drug-induced liver injury with minocycline, alpha methyldopa, nitrofurantoin and checkpoint inhibitors such as infliximab. Wilson´s disease is also an important differential diagnosis especially in young adults. A liver biopsy is mandatory to confirm the diagnosis of autoimmune hepatitis but histopathology alone is not conclusive. 2021 Annals of Translational Medicine. All rights reserved.

Autoimmune hepatitis (AIH) is a relatively rare and heterogeneous disease and may thus render the diagnosis difficult. Liver biopsy is an essential and mandatory element in the diagnosis and management of autoimmune hepatitis and is, owing to its special importance, included in the scoring system for diagnosis (1-6). Since individual, serological and clinical features are not specific for the diagnosis of AIH, evaluation of the liver biopsy helps to exclude other potential causes of liver disease and to identify variant syndromes. Additionally, the biopsy gives information about disease severity and the stage of fibrosis, and assists in the evaluation of the response to therapy. A normal liver histology is a prerequisite for the withdrawal of therapy. Furthermore, histology is necessary to differentiate between a flare up in chronic disease and acute onset AIH; differentiation can be very difficult clinically.

Histopathology of AIH

According to the AASLD and EASL (4,6) the consensus of histopathology of AIH includes several components: (I) moderate to severe inflammatory infiltrations, which are mostly lymphoplasmacytic, in the enlarged portal tracts with moderate to severe interface hepatitis; (II) moderate to severe necroinflammatory lesions in the lobules, sometimes accompanied by confluent necrosis (7,8) and a quantitative extent of plasma cells.

Ad 1: The portal tracts exhibit dense lymphoplasmacytic infiltrates (). Plasma cells are often seen in clusters with more than five cells. Neutrophilic or eosinophilic granulocytes may occur. About 25% of cases show features of a reactive cholangitis that are mostly focal, but are almost never associated with a destruction of the bile ducts as in primary biliary cholangitis (PBC). The infiltrates extend into the lobules with moderate to severe interface hepatitis () and with necroses and loss of hepatocytes displaying ballooning and apoptosis. Fibrosis starts in the portal tracts and may progress to formation of septa () and finally to cirrhosis when there is no response to therapy.

Figure 1

Histopathologic features of autoimmune hepatitis. (A, hematoxylin and eosin) Enlarged, densely infiltrated portal tract with interface hepatitis. The infiltrate consists mainly of lymphoplasmacytic cells; there are many hepatocellular necroses (arrows) and emperipolesis with lymphocytes (circles). (B, chromotrope aniline blue) Fibrosis starts with formation of collagen fibers in the portal tracts as is shown by staining with chromotrope aniline blue. (C, hematoxylin and eosin) Focal necrosis in the lobule close to zone 3 with several clusters of plasma cells. (D, hematoxylin and eosin) Regeneration of parenchyma with rosetting of hepatocytes in the periportal area (circles).

Ad 2: In the lobules, moderate to severe necroinflammatory lesions () occur mainly in the periportal area with intense lymphoplasmacytic reaction and a conspicuous activation of Kupffer cells. The Kupffer cells proliferate and are activated and may contain typical Kupffer cell hyaline globules (9). Emperipolesis of lymphocytes () is a characteristic but not specific phenomenon of AIH, that is that lymphocytes penetrate hepatocytes, i.e., T-effector cells form an immunologic synapsis with hepatocytes as their targets (10). The lymphocytes that are encroached on the cytoplasm of the hepatocytes may kill those. In severe acute AIH or an acute flare up of chronic disease confluent necrosis in zone 3 of the lobule may occur (), rendering the differentiation from ischemic or drug-induced lesions difficult. As a reaction to the necroses and a sign of the regenerative process, so called rosettes () will form, especially in the periportal area where a ductular reaction and regenerative nodules may develop. However, neither emperipolesis nor rosettes are specific for AIH, because they can also be found in other types of hepatitis. In the areas of necroinflammatory lesions a typical endotheliitis of the central veins can be observed. According to some authors (11), results of copper and cytokeratin 7 staining can help to confirm the diagnosis.

Fibrosis and new collagen formation usually starts in the portal tracts (). With progression of the disease, septa may develop—mostly portal to portal, but occasionally also portal to central septa can be seen. In the further course of the disease, regenerative nodules with distortion of the architecture will develop, resulting in cirrhosis in cases where the disease cannot be treated effectively. The presence or absence of fibrosis does not specifically help in the diagnosis of AIH, but it provides significant information on the disease course. It is important to assess the amount of the fibrosis as it determines the stage of the disease which should be scored according to Ishak’s staging (0-no fibrosis to 6-complete cirrhosis) (12).

Histopathology of AIH may therefore prove difficult to interpret and it is recommended to send the biopsy to an experienced hepatopathologist as an extramural consultant (13).

Variant forms of AIH (14,15)

There is no histological difference between the clinical subtypes of AIH type 1 [autoantibodies positive for antinuclear antibody (ANA) and anti-smooth muscle antibodies (SMA)] and type 2 [positive for liver/kidney microsome (LKM) and liver cytosol antibody type 1 (LC1)] or type 3 [positive for soluble liver antigen/liver pancreas (SLA/LP) antibodies clinically as type 1] (1,9,10). Some of the patients also show antimitochondrial antibodies (AMA), however, this is not associated with PBC-like cholangitis (16).

Syncytial giant cell hepatitis is an uncommon pattern of liver injury defined by giant cell transformation of hepatocytes with each giant cell having more than 5 nuclei and a transformation of hepatocytes to giant cells in more than 10%. In children, this type of hepatitis is not uncommon, but in adults (post-infantile giant cell hepatitis, PIGCH) it is rather infrequent and is supposed to have an autoimmune background with the presence of autoantibodies (17).

There is also a subgroup of patients with AIH (about 15%) who lack autoantibodies, are showing high levels of immunoglobulins, a typical histology of AIH, a typical genetic background in the HLA system and response to therapy. The histology is not different from patients with a typical spectrum of autoantibodies (18).

Acute onset AIH versus flare up in chronic AIH

Acute onset AIH can present as a fulminant hepatitis with extensive hepatocyte necrosis and marked inflammatory changes. The necrosis is often panacinar with zone 3 predominance but can also be quite patchy. Acute onset AIH can be seen when the biopsy is taken within the first three months of onset of symptoms and an elevation of laboratory parameters such as bilirubin and transaminases (19-21). It has to be distinguished from a flare up in a chronic course, which is sometimes difficult to discriminate clinically. In this case, histology proves important: The flare up in a chronic course shows fibrosis to a different degree with features of regeneration in the form of regenerative nodules, and distortion of the lobular architecture. Over time, scars may form in areas of collapse thus yielding a very heterogeneous pattern for the cirrhosis.

Overlap syndromes of AIH with PBC and PSC

Overlap syndrome of AIH and PBC (4,22-24): The diagnosis should be made only when the two concurrent diseases can be established and meet the criteria for each disease both clinically and histologically. According to AASLD guidelines the criteria are the following: serum alkaline phosphatase (AP) level at least two times the upper limit of normal or serum gamma GT at least five times the upper limit of normal, presence of AMA and florid bile duct lesions (two out of three criteria are necessary), moreover, serum alanine aminotransferase (ALT) at least five times the upper limit of normal and IgG at least two times the upper limit of normal or the presence of SMA (6). Histology shows the typical bile duct lesions with the destruction of ducts and granuloma formation on the one hand and a moderate to severe hepatitis in the lobules that exceeds the accompanying hepatitis in PBC, on the other hand.

Overlap of AIH and primary sclerosing cholangitis (PSC) (25,26): This incidence is less frequent than that of AIH and PBC. It may occur in children presenting with a so called autoimmune cholangitis but is more frequent in adults. The criteria in the AASLD guidelines are as follows: typical features of AIH, absence of AMA, evidence of large duct PSC by imaging or evidence of small-duct PSC based on onion skinning periductal fibrosis (6). Usually, AIH precedes PSC which is diagnosed when AIH responds to therapy but a cholestatic syndrome develops. Histology will demonstrate a typical hepatitis pattern and later on the bile ducts will show cholangitis with periductular fibrosis and a thickening of the duct membranes.

Autoimmune sclerosing cholangitis: This disease is more common in children and young adults and may be missed initially because the biopsy may show changes of AIH in the absence of features strongly suggesting the presence of biliary tract disease.

Small duct PSC is defined as liver biopsy findings consistent with PSC but with normal appearance of the bile ducts at cholangiography. A subset of these cases can also show overlapping histologic features with AIH, but require convincing evidence of both diseases.

Differential diagnosis

Viral hepatitis has, of course, to be excluded by serological testing, but especially hepatitis C may concur with AIH. In these cases the patients are positive for HCV but also have autoantibodies. After successful treatment for the virus the patients show characteristic clinical and histologic features of AIH (27,28). Drug induced liver injury may mimic AIH (29), especially with drugs such as minocycline, alpha methyldopa, nitrofurantoin and checkpoint inhibitors such as infliximab when taken for longer periods of time. Therefore, a drug reaction has to be excluded carefully by clinical evaluation. Another disease may present as a hepatitis that is histologically indistinguishable from AIH: Wilson’s disease. Especially in young adults this is an important differential diagnosis. Although it is possible to perform staining for copper in the tissue—rhodanine staining—this technique is not absolutely reliable. Consequently, the absolute copper concentration must be determined either by other techniques in the tissue or clinically by 24-hour urinary copper excretion.

In conclusion

A liver biopsy in AIH is necessary and mandatory to confirm the diagnosis, to determine the activity and severity of the disease and to exclude other diseases. Histopathology of AIH is typical but does not show pathognomonic features and is therefore not conclusive for the disease; thus, a definitive diagnosis of AIH by histology alone is not possible and should be based on the simplified criteria (3) that include presence of autoantibodies, elevation of IgG and histopathology graded as unlikely (0 points), possible (1 point) or likely (2 points).

The article’s supplementary files as