Reassessement of the histological features of autoimmune hepatitis

Autoimmune hepatitis (AIH) is a progressive inflammatory disease characterised by a female preponderance, elevated transaminase and immunoglobulin G (IgG) levels, non-organ-specific autoantibodies, inflammatory histology and a dramatic response to corticosteroids. Its aetiology is unknown; its pathogenesis is believed to involve environmental triggers leading to loss of tolerance to liver auto-antigens consequent to dysregulation of immune mechanisms in genetically susceptible individuals.1 AIH affects all age groups with two peaks, one in childhood/adolescence, and the other in adulthood around the age of 40 years. It is a rare disease, but its prevalence has been increasing over the years both in adults and children.2, 3 If untreated, AIH usually leads to liver failure, cirrhosis and associated complications. Originally, AIH was described as a chronic disease presenting with advanced liver damage mainly in women,4 but over the years it has become apparent that it can present acutely, particularly in children5 but also in adults, and that it affects males as well, though in a lower proportion than females.1, 6 A prompt diagnosis of AIH is essential for lifesaving treatment but may be challenging. It is based on a combination of clinical, biochemical, immunological and histological features and the exclusion of other known causes of liver disease that may share the same clinical and laboratory profile, such as viral hepatitis B, C, and E, Wilson disease, non-alcoholic steatohepatitis and drug-induced liver disease (DILI), which must be excluded by accurate clinical history and appropriate investigations. In the absence of specific diagnostic tests for AIH, the International Autoimmune Hepatitis Group (IAIHG) devised a diagnostic system for comparative and research purposes, which includes several positive and negative scores, the sum of which gives a value indicative of probable or definite AIH.7 The scoring system was revised in 1999.8 A simplified IAIHG scoring system published later is easier for clinical application,9 but less accurate than the revised one,8 particularly in the context of AIH presenting acutely.10 In all three IAIHG publications, correct autoantibody testing and interpretation of liver histology are pivotal diagnostic tools. A rigorous standardisation of autoantibody testing and histological features of AIH, however, is still missing. Liver biopsy is needed for diagnosis and to evaluate the severity of liver damage11, 12 as transaminase and IgG levels do not reflect accurately the degree of tissue inflammation nor indicate the presence or absence of cirrhosis. The histological feature long considered typical of AIH is interface hepatitis, characterised by a dense inflammatory infiltrate composed of lymphocytes and plasma cells that cross the portal limiting plate and invade the surrounding parenchyma. Interface hepatitis is however not exclusive to AIH.13 Though plasma cells are characteristically abundant at the interface, their presence in low number does not exclude the diagnosis of AIH. When AIH presents acutely, and during episodes of relapse, a common histological finding is panlobular hepatitis with centrilobular and bridging necrosis and, if the disease takes a fulminant course, massive necrosis and multilobular collapse.14 Though sampling variation may occur in needle biopsy specimens, particularly in cirrhotic livers, the severity of the histological appearance is usually of prognostic value.15 Besides interface hepatitis, non-specific features, included in all three IAIHG scoring systems, that are deemed to suggest the diagnosis of AIH are emperipolesis (presence of an intact lymphocyte within the cytoplasm of a hepatocyte) and hepatocyte rosetting (small group of hepatocytes arranged around a central lumen),7-9 which in one study were suggested to be even stronger indicators of AIH than interface hepatitis or plasma-cell rich infiltrate.16 In the Consensus Recommendations for Histological Criteria of Autoimmune Hepatitis published in this issue of Liver International,17 a panel of 17 expert liver histopathologists and two hepatologists scholarly confirm observations published over the years and suggest a simplified description for the histological picture of AIH. Following an international workshop, the 17 histopathologists have used a modified Delphi panel approach to establish consensus criteria for the histopathological diagnosis of AIH. The panel agrees that liver biopsy should remain a key element in diagnosing AIH, but that there are no histological features pathognomonic of the disease. While a lymphoplasmacytic infiltrate, at times including plasma cell clusters (foci of ≥5 plasma cells) is confirmed to be a typical finding, emperipolesis and hepatocellular rosettes, previously regarded as characteristic of AIH, should not be considered as diagnostic, but to be non-specific markers of inflammation severity and regeneration.13, 18, 19 In addition to the histological indices contained in the three IAIHG scoring systems7-9 that cover AIH presenting as a chronic disease, the expert panel aims at covering also AIH presenting acutely with predominant histological features of lobular hepatitis. Therefore, distinct standards are separately proposed for cases with predominantly portal hepatitis and cases with predominantly lobular hepatitis. In the work-up for the diagnosis of AIH, the panel proposes to standardise the histological terminology in ‘likely,’ ‘possible’ or ‘unlikely’ based on a number of parameters. For the ‘portal hepatitis pattern’, they include the degree of interface hepatitis, the degree of lobular hepatitis and the presence of histological features suggestive of other liver diseases; for the ‘lobular hepatitis pattern’, they include lymphoplasmacytic infiltrates, interface hepatitis, portal-based fibrosis and histological features suggestive of other liver diseases. However, the final diagnosis of AIH can only be made by matching the histology picture with the clinical and laboratory information provided by the clinician. AIH is considered ‘likely’ when there is a predominantly portal lymphoplasmacytic hepatitis with more than mild interface activity and/or more than mild lobular hepatitis and in the absence of histological features suggesting other liver diseases. AIH is also considered ‘likely’ if there is predominantly lobular hepatitis with or without centrilobular necroinflammation in association to at least one of the following: portal lymphoplasmacytic hepatitis, interface hepatitis or portal-based fibrosis, in the absence of histological features suggestive of other liver diseases. Criteria for lobular hepatitis may be less stringent than those for portal hepatitis, as the differential diagnosis for severe lobular hepatitis besides acute AIH would include DILI and idiopathic hepatitis, for example, seronegative acute liver failure, conditions that often have lymphoplasmacytic infiltration at least focally and may easily meet the criteria of likely AIH. The category of possible AIH in the lobular hepatitis pattern may also be slightly broader than expected. Any cases of lobular hepatitis are classified as at least “possible AIH” unless there are features suggestive of another liver disease. These points would represent difficulties in the histological differentiation between acutely presenting AIH and other forms of acute hepatitis. The panel agreed that a semi-quantitative assessment of the severity of inflammatory activity of AIH is important and suggest that it should be based on the Ishak’s modified Histological Activity Index (mHAI),20 which they define more ‘granular’ than other scoring systems (Metavir, Scheuer, Desmet, Batts-Ludwig), and which also allows a separate and detailed assessment of centrilobular necroinflammatory changes, being more appropriate for grading acute lobular damage. They are, however, stressing that the mHAI, as well as the other scores, has only been validated in the setting of chronic viral hepatitis and will need to be explored in the setting of AIH. Validation in the real-life clinical setting of this review of the histological characteristics of AIH will lead in time to a collegial re-evaluation of the histological diagnosis of AIH within the IAIHG scoring system. The panel show awareness that the proposed classification will need to be validated also in the context of liver diseases that can mimic AIH, such as Wilson disease, DILI and non-alcoholic steatohepatitis, as well as in the context of overlap conditions, such as AIH/primary biliary cholangitis or AIH/primary sclerosing cholangitis. For the latter, a thorough evaluation of the bile duct inflammation and cholestatic features present in a proportion of patients with AIH21, 22 will need to be addressed. Probably the panel’s most important conclusion is that it is essential that the histopathologist does not work in isolation but has access to all relevant clinical and laboratory data through the hepatologist, as optimal patient management depends, as always, on efficient communication. The data that support the findings of this study are openly available in PubMed.