Autoimmune Hepatitis: Emerging Frontiers in Research and Clinical Management

Abstract

Autoimmune hepatitis (AIH) is a chronic immune-mediated liver disorder that, without treatment, can advance to fibrosis and cirrhosis. Although standard regimens with corticosteroids and thiopurines have significantly improved survival, many patients still experience relapses and drug-related toxicity, highlighting the urgent need for alternative strategies. Recent studies underscore AIH’s multifactorial nature, revealing intricate interactions among genetic susceptibility, environmental triggers, and dysregulated immune responses. Next-generation diagnostics, ranging from novel biomarkers to high-resolution imaging, are enhancing early detection and more precise disease classification. At the same time, multi-omics analyses and artificial-intelligence-based models are refining predictions of disease trajectory and therapeutic response. On the treatment horizon, investigational options such as targeted immunomodulators, B-cell–depleting therapies, and cell-based interventions aim to achieve durable remission while minimizing adverse effects. This review critically appraises these advances and explores how integrating epidemiological insights with cutting-edge research in pathogenesis, diagnostics, and therapy could pave the way for more personalized and effective management of AIH.

1. Introduction

Autoimmune hepatitis (AIH) is a chronic, progressive immune-mediated liver disease that, if not treated, confers substantial risks of morbidity and mortality [1]. Once considered uncommon, it now shows rising incidence and prevalence across age groups, sexes, and ethnic backgrounds, with a pronounced female predominance [2]. The clinical spectrum ranges from incidental aminotransferase elevations to fulminant hepatic failure [3]. Although its pathogenesis remains incompletely defined, accumulating evidence points to a multifactorial process involving genetic susceptibility, environmental triggers, and breakdown of immune tolerance, resulting in hepatocyte-directed autoimmunity [4]. Diagnosis relies on an integrated assessment of biochemical, serological, and histological findings with careful exclusion of alternative liver diseases. Serological hallmarks frequently include antinuclear antibodies (ANA), anti–smooth muscle antibodies (ASMA), anti–liver/kidney microsomal antibodies (anti-LKM-1), and elevated serum immunoglobulin G (IgG) [5]. Liver biopsy remains pivotal for confirming the diagnosis and informing prognosis [6].

The natural history of AIH is characteristically relapsing–remitting and, in the absence of therapy, can advance to cirrhosis and end-stage liver disease [7]. Conventional management has centred on corticosteroids, alone or combined with azathioprine, to induce remission and sustain biochemical control [8]. Notably, recent European guidance recognizes mycophenolate mofetil as an equivalent first-line alternative to azathioprine, supported by emerging data indicating comparable efficacy and a favourable safety profile [3]. As insights into AIH pathobiology expand, treatment strategies are increasingly oriented toward personalized, mechanism-based approaches, aiming to enhance outcomes while limiting drug-related toxicity. Beyond first-line therapy, second-line immunosuppressants, including tacrolimus, cyclosporine, and 6-mercaptopurine, have shown utility in patients who are refractory to or intolerant of standard regimens [9]. Moreover, biologics targeting B-cell pathways, such as rituximab and belimumab, are under evaluation in growing, albeit small, clinical cohorts and appear promising for disease control in difficult-to-treat or relapsing AIH [10].

This review aims to provide a comprehensive and up-to-date synthesis of the latest evidence on autoimmune hepatitis, including novel therapeutic approaches and current recommendations for clinical management.

2. Epidemiology

AIH has historically been regarded as uncommon, with marked regional variability in prevalence estimates [1]. Global epidemiological efforts, including multiple systematic reviews and meta-analyses, demonstrate heterogeneity in both incidence and prevalence, shaped by geography, sex, age, and genetic background [7]. A recent meta-analysis encompassing 37 studies from 18 countries across five continents reported a pooled incidence of 1.28 per 100,000 person-years and a pooled prevalence of 15.65 per 100,000 individuals [11].

Although incidence appears broadly similar across Europe, Asia, and the Americas, prevalence is generally lower in Asian populations than in European and American cohorts; women consistently exhibit higher incidence and prevalence than men, and prevalence is notably higher among individuals older than 65 years [1].

Over time, prevalence has risen substantially, from 9.95 per 100,000 in 1970–1999 to 27.91 per 100,000 in 2015–2022, likely reflecting increased awareness, refined diagnostic frameworks, and a wider uptrend in autoimmune disorders [11,12]. AIH frequently co-occurs with extrahepatic autoimmune diseases, including hypothyroidism, ulcerative colitis, type 1 diabetes mellitus, dermatologic autoimmune conditions, rheumatoid arthritis, and coeliac disease [13]. In a nationwide Danish cohort, approximately 20% had extrahepatic autoimmune disease at diagnosis, with an additional 13% developing new or additional conditions within five years [14]. Epidemiological data further suggest higher AIH prevalence in countries with a high Human Development Index (HDI > 0.92) and in populations residing at latitudes above 45°, though this association is not uniform across settings [15].

In Northern Europe, prevalence is estimated at 10–20 per 100,000 with an incidence of about 1.2 per 100,000 person-years and a male-to-female ratio of roughly 1:3.5. By contrast, U.S. data from 2009 to 2018 indicate an incidence of 4 per 100,000 person-years and a prevalence of 26.6 per 100,000, with males, Black, and Hispanic individuals more likely to present with cirrhosis [16].

Sex-related differences likely reflect interplay among hormonal, genetic, environmental, and microbiome determinants. Estrogens can potentiate immune activation by promoting dendritic cell maturation and augmenting T-helper 1 (Th1) responses, whereas androgens tend to bias toward Th2 and regulatory T-cell (Treg) pathways with overall immunomodulatory effects [17].

Genetic susceptibility to AIH shows relevant geographic variation and may partly account for differences in autoantibody profiles across populations [18]. In Italian patients, human leukocyte antigen (HLA)-DR4 was not associated with AIH, whereas the known HLA risk factors occurred similarly in type 1 and type 2 AIH and were overall less frequent than in North American patients. Moreover, B8-DR3-DQ2 represented the predominant phenotype in Italian type 1 AIH, while HLA-DR11 appeared to be a distinctive protective factor against type 1 disease [19]. Clinical outcomes also vary by sex and race: men have approximately a two-fold higher incidence of hepatocellular carcinoma than women [20], and Black and Hispanic patients experience higher rates of severe complications such as acute liver failure and portal hypertension [21].

3. Pathophysiology

AIH pathogenesis reflects a convergence of genetic susceptibility, environmental and infectious exposures, and dysregulation across innate and adaptive immunity [13].

Although the inciting event remains undefined, multiple lines of evidence implicate viral infections, molecular mimicry, and failure of immune regulation. Candidate viral triggers include HAV, HBV, HCV, HEV, EBV, HSV, and measles virus [13,22,23]. In parallel, reactive drug metabolites may prime the immune system and foster hepatocyte-directed autoantibodies years before overt disease manifests [24].

Genetic risk is strongly linked to HLA class II alleles, particularly HLA-DRB10301 and DRB10401, and to non-HLA loci such as SH2B3 and CARD10 [25]. Variation across the HLA region and major histocompatibility complex (MHC) on chromosome 6 shapes antigen presentation and T-cell activation, thereby undermining peripheral tolerance [26].

Within the normally tolerogenic hepatic milieu, constantly exposed to microbial products, xenobiotics, and self-antigens, disruption of regulatory networks permits autoimmune injury to emerge [4].

At the cellular level, disease initiation involves presentation of an autoantigenic peptide by HLA class II on an antigen-presenting cell (APC) to a naïve Th0 lymphocyte [27]. Depending on cytokine context and antigen characteristics, Th0 cells polarize toward Th1, Th2, or Th17 phenotypes [28]. Th2 cells secrete interleukin (IL)-4, IL-13, and IL-21, driving B-cell activation and plasma-cell differentiation. Th1 cells release IL-2 and interferon (IFN)-γ, activating cytotoxic T cells and macrophages and upregulating class II HLA on hepatocytes [29]. Activated macrophages amplify inflammation via IL-1 and tumour necrosis factor (TNF)-α [30].

Tregs, generated from Th0 in the presence of transforming growth factor (TGF)-β, normally constrain these responses [30]; in AIH, Treg number/function is frequently diminished, contributing to loss of self-tolerance [31]. Hepatocellular damage is propagated by cytotoxic T cells, pro-inflammatory cytokines, and antibody-dependent mechanisms involving natural killer (NK) cells [32]. Th17 cells, induced by TGF-β and IL-6, likely participate as well, though their precise contribution remains under investigation [4].

Molecular mimicry is a plausible initiating mechanism, whereby immune responses to exogenous pathogens cross-react with structurally related hepatic autoantigens [33]. In type 2 AIH, the principal target is cytochrome P450 (CYP)2D6, recognized by anti-LKM-1 antibodies [34]; sequence homologies between CYP2D6 and proteins from hepatitis C (HCV), herpes simplex virus (HSV)-1, and cytomegalovirus (CMV) support this model [35]. Additional microsomal proteins, including CYP1A2 and CYP2C11, also share epitopes with microbial antigens [36]. In contrast, the dominant autoantigen in type 1 AIH has not been definitively identified despite multiple candidates [34].

Emerging data highlight a role for the gut microbiome. Through the gut–liver axis, by which ~75% of hepatic blood flow is delivered via the portal vein, the liver is continually exposed to microbial products and metabolites [37]. Dysbiosis can impair barrier integrity, increase permeability, and facilitate translocation of components such as lipopolysaccharides (LPS) to the liver, promoting immune activation via mimicry and other pathways [38]. Such alterations have been observed in AIH and correlate with immune activation and disease activity, raising the prospect of non-invasive microbiome-based biomarkers [13].

Clinically, this immune dysregulation is mirrored by disease-specific autoantibodies, including ASMA, ANA, anti-LKM-1, anti-liver cytosol type 1 (anti-LC1), and anti-soluble liver antigen/liver-pancreas antibodies (anti-SLA/LP), which are integral to diagnosis and reflect the underlying pathophysiology [8].

These mechanistic insights are increasingly relevant to clinical practice because they provide a biological rationale for emerging targeted therapies. In particular, B-cell activation, plasma-cell differentiation, and B-cell activating factor (BAFF) signalling support the use of B-cell–directed strategies such as rituximab, belimumab, and ianalumab in refractory AIH. Likewise, persistent T-cell activation, defective regulatory control, and cytokine-driven inflammation provide the conceptual basis for investigating calcineurin inhibitors, mTOR-directed approaches, and other targeted immunomodulators. Although these pathways have not yet yielded validated predictive biomarkers for treatment selection, they are beginning to inform the development of steroid-sparing and biologic therapies.

4. Clinical Features

AIH is a clinically heterogeneous disease, with presentations ranging from asymptomatic aminotransferase elevations to acute liver failure [3]. Although a chronic, insidious onset is most typical, contemporary epidemiological data indicate that acute-onset disease is more common than previously appreciated, accounting for approximately 22–43% of cases [39]. This variability complicates early recognition and timely diagnosis [39]. Presentation also varies with age: in older adults, AIH more often manifests with attenuated symptoms, a higher likelihood of asymptomatic onset, greater prevalence of cirrhosis at diagnosis, and more frequent extrahepatic autoimmune comorbidities [40]. Despite the milder clinical profile, substantial hepatic injury may already be present, emphasizing the need for heightened vigilance in this population [3].

Acute AIH can be categorized histologically into two principal entities: (i) new-onset acute AIH, defined by the absence of chronic liver disease on biopsy, and (ii) acute exacerbation of underlying AIH, either previously recognized or occult, characterized by histological evidence of chronic injury [39]. Clinically, acute AIH often mimics viral or toxic hepatitis, presenting with jaundice, fatigue, nausea, and marked aminotransferase elevation; in a subset, serum IgG may be normal and autoantibodies initially negative, contributing to diagnostic delay [41]. Severity stratification includes: acute icteric AIH (jaundice without coagulopathy or encephalopathy), acute severe AIH (jaundice with coagulopathy; INR > 1.5, no encephalopathy), and acute severe AIH with acute liver failure (jaundice, coagulopathy, and encephalopathy) [3]. These categories inform urgent management decisions and underscore the need for prompt immunosuppression or expedited referral to a liver transplant centre [42].

Outside the acute setting, approximately one third of patients exhibit an insidious onset with nonspecific features: fatigue, malaise, anorexia, weight loss, nausea, right upper quadrant discomfort, pruritus, intermittent jaundice, amenorrhea, and polyarthralgia (typically of small joints without frank arthritis) [3]. Chronic fatigue is the most frequently reported symptom, affecting up to 85% of patients, whereas pruritus and hyperpigmentation are less characteristic; unintended weight loss should prompt evaluation for serious complications, including malignancy [43]. Physical examination may be normal or reveal hepatomegaly and splenomegaly; in advanced disease, stigmata of cirrhosis such as spider naevi and palmar erythema may be present [3]. With progression, manifestations of portal hypertension can predominate, including ascites, oesophageal varices, portal hypertensive gastropathy, hypersplenism-associated cytopenias, and hepatic encephalopathy [44]. Notably, up to one third of cases are asymptomatic at diagnosis, detected incidentally through unexplained transaminase elevations; if unrecognized, this indolent course can culminate in significant hepatic injury over time.

5. Diagnosis

Autoimmune hepatitis is a clinicopathological entity whose diagnosis relies on the integration of biochemical, serological, and histological findings, along with the exclusion of other causes of liver disease. It should be considered in any patient presenting with elevated liver enzymes or cryptogenic cirrhosis, particularly in the absence of clear risk factors for other hepatic conditions [3].

The biochemical profile typically reveals elevated aspartate and alanine aminotransferases (AST and ALT) and hypergammaglobulinemia, most notably an increase in IgG [6]. In the cirrhotic stage, aminotransferase levels may be normal or only mildly elevated, potentially obscuring the underlying diagnosis [45]. Serologically, AIH is supported by the detection of autoantibodies, such as ANA, SMA and anti-LKM-1. However, these markers lack specificity and may be absent in seronegative variants [43].

Liver histology remains a cornerstone of diagnosis. Given the absence of a single definitive marker, the diagnosis of AIH requires a composite assessment of clinical, laboratory, and histological features, and importantly, the exclusion of alternative etiologies, including viral hepatitis, drug-induced liver injury, Wilson’s disease, alpha-1 antitrypsin deficiency, and hereditary hemochromatosis [46].

Historically, AIH has been serologically classified into type 1 AIH, characterized by ANA and/or SMA positivity, and type 2 AIH, characterized by anti-LKM-1 and/or anti-LC1 positivity. Anti-SLA/LP-positive cases were previously labelled as ‘type 3 AIH’; however, this distinction is no longer considered clinically useful and should be avoided in current practice, as anti-SLA/LP positivity is now interpreted within the broader serological spectrum of AIH rather than as a separate subtype [3,47,48,49].

The diagnosis of AIH may be supported by the scoring systems proposed by the International Autoimmune Hepatitis Group (IAIHG). The revised original score, published in 1999, is more comprehensive and was primarily developed for research standardization and for diagnostically challenging cases. By contrast, the simplified score introduced in 2008 was designed for routine clinical practice and is based on four main domains: the presence of autoantibodies, elevated serum IgG, characteristic histological findings, and the exclusion of viral hepatitis. In the simplified system, each parameter is assigned 1 or 2 points according to the strength of the finding, with a total score of ≥6 indicating probable AIH and ≥7 supporting a definite diagnosis [5,50,51].

The simplified score has also been validated in Italian clinical practice, where it showed good diagnostic performance not only in classical AIH but also in patients with AIH/PBC overlap features. This supports its usefulness in real-life autoimmune liver disease cohorts, although score interpretation should always remain integrated with clinical and histological assessment [52,53].

An important limitation of the simplified score emerges in patients with acute-onset AIH. In this setting, a non-negligible proportion of patients may present with negative conventional autoantibodies and normal IgG levels, thereby reducing the sensitivity of the simplified criteria. In the Italian experience, the revised original IAIHG score performed better than the simplified score in acute-onset AIH and should therefore be considered when AIH is strongly suspected despite a low simplified score [54].

5.1. Autoantibodies

Serological testing is pivotal in diagnosing autoimmune hepatitis. The principal autoantibodies comprise non–organ-specific markers, namely ANA and SMA, and organ-specific markers, including anti-LKM-1, anti-LC1, and anti-SLA/LP [3]. Among white North American adults at presentation, ANA are detected in ~80%, SMA in ~63%, and anti-LKM-1 in ~3%; at least one of these autoantibodies is present in nearly all patients, and diagnostic accuracy improves from ~58% to ~74% when two are simultaneously positive [43]. Overall, ANA and SMA are the most common, each identified in ~60–70% of cases. While either can appear in other liver diseases, such as metabolic dysfunction–associated steatotic liver disease, chronic hepatitis B or C, primary sclerosing cholangitis, and alcohol-related liver disease, their concurrent presence outside AIH is uncommon and, in the appropriate clinical context, strongly supports the diagnosis [43].

Among SMA, anti-actin antibodies deserve particular attention because they appear to be highly specific for type 1 AIH and may be easier to recognize and interpret than conventional tissue-based patterns, thereby offering more robust diagnostic support and potentially improving inter-laboratory reproducibility, especially in less experienced laboratories [50]. Anti-actin reactivity is detected in 86–100% of SMA-positive patients with AIH, and anti-α-actinin antibodies, present in approximately 66% of these patients, have been associated with poorer treatment response [55].

Anti-LKM-1 and anti-LC1 are less frequent in adults but occur in up to 40–50% of pediatric patients and are associated with more severe disease. Anti-LKM-3, a subtype within the LKM family, is found in ~17% of cases, whereas anti-LC1 is detected in ~32% of patients who are anti-LKM-1 positive [56]. Anti-SLA/LP, present in ~10–20%, is highly specific for AIH; it should be assessed using solid-phase assays because it is not detected by indirect immunofluorescence. Its presence correlates with severe disease and an increased risk of relapse after withdrawal of immunosuppression [55].

Atypical perinuclear anti-neutrophil cytoplasmic antibodies (pANCA) are frequently demonstrable by immunofluorescence in AIH. Antibodies to double-stranded DNA (anti-dsDNA) are reported in ~30% of AIH overall and up to ~60% in AIH–primary biliary cholangitis overlap (PBC) [57]. Although indispensable for diagnosis, these autoantibodies neither reliably predict prognosis nor therapeutic response. In seronegative presentations with strong clinical suspicion, repeat testing at lower serum dilutions or evaluation in reference laboratories is recommended, recognizing that autoantibody titers may fluctuate over time [3].

5.2. Histological Findings

Every patient with suspected AIH should undergo a liver biopsy, unless contraindicated [58]. Beyond supporting the diagnosis, histological assessment is indispensable for application of standardized diagnostic scoring systems, and requires an adequate core, ideally containing at least eight portal tracts, to allow meaningful interpretation [59].

The canonical histopathologic picture features a dense lymphoplasmacytic infiltrate centred in portal tracts, frequently breaching the limiting plate to produce interface hepatitis, often with variable degrees of lobular inflammation [60]. Interface activity, defined by lymphocyte and plasma cell infiltration at the portal–parenchymal junction with attendant hepatocellular injury, is a key diagnostic attribute; additional findings may include lobular disarray, spotty necrosis, and occasional bridging necrosis [61]. While none of these features is pathognomonic, their constellation in the appropriate clinical context and in the absence of alternative etiologies is highly suggestive of AIH [62].

Emperipolesis (lymphocytes or plasma cells within hepatocyte cytoplasm) and hepatocyte rosette formation, once regarded as characteristic, are no longer considered specific for AIH; however, their presence may be documented as indirect indicators of disease activity or severity [63].

Pathology reporting should systematically survey portal tracts, lobular parenchyma, sinusoids, and terminal hepatic venules, and provide detailed staging and extent of fibrosis. Based on the aggregate of these findings, together with exclusion of competing diagnoses, biopsies are classified as showing likely, possible, or unlikely AIH within established scoring systems, thereby contributing substantially to overall diagnostic confidence [6].

5.3. Non-Invasive Assessment

Progressive hepatic fibrosis is a key prognostic determinant in AIH and has classically been assessed histologically. Routine surveillance with repeated biopsies, however, is generally discouraged because of patient and clinician concerns. In this context, serial liver stiffness measurements (LSM), for example, through annual transient elastography (TE), provide a safe, non-invasive, and dependable alternative for longitudinal fibrosis monitoring [64,65] TE correlates closely with histological fibrosis stage, with commonly used thresholds of approximately 9 kPa for advanced fibrosis and 12.5–16 kPa for cirrhosis. Notably, most data indicate that TE yields the most accurate fibrosis staging in AIH when performed after at least six months of immunosuppressive therapy, reinforcing its value for long-term follow-up [66,67].

6. Current Treatment Approaches for AIH

The primary objectives of treatment in autoimmune hepatitis are to alleviate clinical symptoms, suppress hepatic inflammation, achieve durable remission, and promote fibrosis regression, all while minimizing the risk of drug-related adverse events [68,69] (Table 1). Unless otherwise specified, the therapeutic approaches discussed in this section primarily refer to adult AIH. Pediatric AIH shares the same general treatment principles but requires some important distinctions, particularly regarding the role of second-line, biologic, and investigational therapies.

All patients with active autoimmune hepatitis are candidates for therapy, including those with advanced fibrosis or compensated cirrhosis. By contrast, initiation of treatment in patients with only mild histological activity remains debated, particularly in older adults, where the risks of immunosuppression must be weighed against the likelihood of progression. In such scenarios, close surveillance without immediate therapy may be reasonable, especially when corticosteroids are relatively contraindicated (e.g., uncontrolled diabetes, severe osteoporosis, or metabolic syndrome with morbid obesity) [3,43].

Complete biochemical remission (CBR) is conventionally defined as normalization of aminotransferases and IgG within six months of therapy. Recent data question the necessity of IgG normalization, as outcomes are favourable when aminotransferases normalize despite persistently elevated IgG; however, the evidence base is not yet sufficient to revise the formal definition [70,71]. Failure to achieve CBR by six months, relevant to both first- and second-line regimens, indicates suboptimal response and carries prognostic weight, though it does not by itself mandate an immediate change in therapy [72]. Large cohort studies demonstrate that attaining CBR within 6–12 months is strongly associated with improved outcomes, and normalization of aminotransferases functions as a robust prognostic marker independent of IgG levels; accordingly, assessing CBR within twelve months is a practical management target [3,73,74]. Beyond this, an important practical issue is that histological remission may lag behind biochemical normalization, and its routine use as a therapeutic target remains controversial. In everyday practice, sustained normalization of aminotransferases remains the most pragmatic treatment goal, whereas persistent biochemical activity should prompt reassessment of adherence, drug exposure, and the need for treatment escalation.

According to the 2025 European guidelines, first-line treatment for adults should begin with predniso(lo)ne at ≥0.5 mg/kg/day, escalated to 1 mg/kg/day in patients with more severe or advanced disease. A relatively rapid taper (approximately 5–10 mg every 1–2 weeks) is desirable but should be individualized to clinical and biochemical response with close monitoring throughout [3].

Azathioprine should be added, typically at 50 mg/day, preferably two weeks after corticosteroid initiation and once serum bilirubin has declined below 6 mg/dL, to reduce diagnostic ambiguity between treatment nonresponse and azathioprine toxicity. The dose is subsequently titrated to 1–2 mg/kg/day according to tolerance and therapeutic effect [74]. Azathioprine must be combined with corticosteroids during induction, as monotherapy in this phase has been associated with increased mortality [75]. Use azathioprine cautiously or avoid it in settings of thiopurine S-methyltransferase (TPMT) deficiency; additional concerns include pregnancy, cytopenias, prior malignancy, acute severe AIH, and decompensated cirrhosis. As an alternative, mycophenolate mofetil (MMF) at 1.5–2 g/day may be employed as a first-line option in lieu of azathioprine [76]. In pediatric AIH, however, standard first-line therapy remains centred on predniso(lo)ne/prednisone and azathioprine, which continue to represent the best-established regimen for induction and maintenance treatment, whereas MMF is more commonly positioned as a second-line option in cases of intolerance or insufficient response [43]. Because MMF is teratogenic, comprehensive reproductive counselling is essential for women of childbearing potential and for male patients, and pregnancy testing is required prior to initiation [3].

The prognosis of AIH is generally favourable when the diagnosis is established early and appropriate immunosuppression is promptly initiated. Without treatment, progression to cirrhosis and liver failure is common and mortality rises substantially. By contrast, corticosteroid-based therapy, alone or combined with azathioprine, markedly improves long-term outcomes, with 10-year survival approaching 80–90%. Complete remission can be achieved in some patients; however, relapse after drug withdrawal is frequent, and many require long-term maintenance therapy [27]. Early treatment response is among the strongest predictors of favourable prognosis [27].

With respect to budesonide, recent European guidance does not recommend its use as first-line therapy and contraindicates it in patients with cirrhosis [77]. A switch to budesonide may be considered for non-cirrhotic patients who are predniso(lo)ne-dependent and experiencing steroid-related adverse effects, though evidence for its role as salvage therapy in predniso(lo)ne-intolerant or -dependent individuals remains limited [78]. Once biochemical remission is achieved, a response-guided approach to ongoing therapy is advised [43].

Given its chronic nature, AIH usually necessitates lifelong maintenance in most patients. Maintenance commonly consists of azathioprine or mycophenolate mofetil monotherapy, often with low-dose corticosteroids; treatment must be individualized according to efficacy and tolerability. A sustained period of complete biochemical remission of at least 24 months predicts a higher likelihood of successful tapering of immunosuppression [3]. Long-term management should aim to maintain remission with the lowest effective corticosteroid exposure, while monitoring for steroid-related, thiopurine-related, and MMF-related toxicity. Serial non-invasive fibrosis assessment and regular reassessment of treatment tolerability are also important components of follow-up.

6.1. Second-Line Treatment

Second-line therapy is reserved for patients who fail to respond, respond inadequately, or are intolerant to first-line regimens [79]. Nonresponse is classically defined as <50% reduction in aminotransferases within four weeks and occurs in ~7–9% of adults. However, a recent retrospective study (discovery n = 370; validation n = 370) showed that an >80% decline in aminotransferases by week 8 predicts AST/ALT normalization at 6 and 12 months and superior clinical outcomes, suggesting the four-week threshold may be premature [72]. Insufficient response denotes failure to achieve complete biochemical normalization by six months, while intolerance refers to adverse effects necessitating drug discontinuation [70].

Available second-line options include MMF, calcineurin inhibitors (cyclosporine A, tacrolimus), agents affecting thiopurine synthesis/metabolism (e.g., allopurinol, 6-mercaptopurine, 6-thioguanine), and methotrexate, among others [80]. 6-mercaptopurine can substitute for azathioprine intolerance. 6-thioguanine—converted directly to active 6-thioguanine nucleotides, thereby bypassing formation of 6-methylmercaptopurine—has more limited experience in AIH, and higher doses have been associated with non-cirrhotic portal hypertension [10]. MMF can be used as first-line or as second-line therapy for patients intolerant of azathioprine and 6-MP [9]. In the open-label CAMARO randomized trial comparing azathioprine versus MMF for induction in treatment-naïve AIH, MMF yielded superior outcomes, prompting re-evaluation of first-line strategies [76]. In the pediatric setting, the best-supported second-line options remain MMF and calcineurin inhibitors, particularly tacrolimus or cyclosporine. By contrast, the role of budesonide, anti-TNF agents, rituximab, belimumab, and other biologic or investigational therapies is far less well established in children and should be interpreted with caution, being generally limited to highly selected refractory cases managed in expert centres [43].

Biologic and novel agents constitute a third-line avenue for truly refractory disease [80]. B-cell–directed therapies are particularly promising. Rituximab (anti-CD20) depletes autoreactive B cells and has induced remission in many refractory cases [9]; in a series of 22 difficult-to-manage patients, rituximab produced significant biochemical improvement, reduced flare frequency (71% relapse-free over two years), and enabled steroid tapering in most [81]. Belimumab, which neutralizes BAFF, has shown favourable signals in small cohorts (responses in 5/8 patients) [82]. Ianalumab (VAY736), an anti-BAFF receptor monoclonal antibody with dual activity (ADCC-mediated B-cell depletion and blockade of BAFF-R signalling), is under evaluation in the AMBER Phase 2/3 trial for incomplete responders (NCT03217422) [83].

T-cell–targeted approaches have also been explored: basiliximab (anti-IL-2R) in isolated cases and sirolimus (mTOR inhibition) in occasional refractory patients, though evidence remains sparse [84].

TNF-α inhibitors (infliximab, adalimumab) have been used as salvage therapy with enzyme improvements reported [85]. In a multicentre retrospective cohort of 42 patients across 21 centres in 12 countries, infliximab achieved or maintained biochemical remission in 78%, and among those with active disease at initiation (n = 26), 65% reached complete remission, including 55% after failure of standard/second/third-line regimens, with generally acceptable safety (three discontinuations: one severe allergy, two anti-drug antibodies) [86]. Their use nevertheless warrants caution due to infection risk and reports of anti-TNF–induced AIH-like hepatitis in susceptible individuals [87]. Notably, no single rescue regimen is established; the evidence base is dominated by case series and retrospective studies, guidelines do not endorse a preferred second-line agent, personalized trial-and-error is often required, and a subset will ultimately progress to liver transplantation despite optimal medical therapy.

Several emerging drugs, the mechanisms of which are distinct and have different targets of immunomodulation, are in development. Zetomipzomib, a first-in-class small-molecule selective immunoproteasome inhibitor, is being investigated in a double-blind phase IIa RCT in patients with hard-to-treat AIH (insufficient response even after 3 months of first-line treatment or relapse after biochemical remission) (PORTOLA; NCT05569759).

Novel small molecules are also under investigation: JKB-122, a toll-like receptor 4 antagonist, completed a Phase II trial in steroid-refractory AIH (NCT02556372) with the aim of reducing proinflammatory cytokine release from innate immune activation [9]. Cell-based therapy is being explored through the Merlin trial (Phase I/II), which is administering mesenchymal stromal cells (ORBCEL-C™) to induce immune tolerance in autoimmune liver diseases including AIH (NCT02997878) [88]. Beyond these, several repurposed immunotherapies are in early studies or compassionate use for AIH: IL-6 pathway blockade (e.g., tocilizumab), IL-12/23 inhibition (e.g., ustekinumab), and JAK-STAT pathway inhibitors (such as tofacitinib) have shown promise in isolated cases or preclinical models [80]. These cutting-edge approaches, alongside more rigorous clinical trials, herald a future in which difficult AIH cases may be managed with mechanism-based, steroid-sparing therapies. The hope is that by targeting specific immune pathways (B cells, T cells, cytokines, etc.), next-generation treatments will achieve higher remission rates in refractory AIH while minimizing the systemic toxicity of prolonged corticosteroid use.

Table 1. Current and emerging therapies for autoimmune hepatitis: clinical role and key considerations.

Therapeutic Step	Therapy	Current Role/Positioning	Key Considerations	Key References
First-line induction therapy	Predniso(lo)ne + azathioprine (AZA)	Standard first-line treatment in adult AIH. Remains the best-established regimen for induction of remission.	Predniso(lo)ne is usually started at ≥0.5 mg/kg/day, up to 1 mg/kg/day in more severe disease, with response-guided tapering. AZA is generally introduced after corticosteroid initiation and titrated according to tolerance and response.	[3,43,75]
First-line induction therapy (alternative in selected adults)	Mycophenolate mofetil (MMF)	First-line alternative to AZA in selected adult patients, particularly when AZA is contraindicated or poorly tolerated.	Usually given at 1.5–2 g/day. Teratogenicity and reproductive counselling must be considered. In pediatric AIH, MMF is generally positioned more often as second-line rather than standard first-line therapy.	[3,9,76]
Maintenance therapy	AZA monotherapy ± low-dose corticosteroids	Standard maintenance strategy after remission induction.	Long-term therapy is required in most patients. The aim is sustained biochemical remission with the lowest effective steroid exposure.	[3,43]
Maintenance therapy	MMF monotherapy ± low-dose corticosteroids	Alternative maintenance strategy in patients intolerant of AZA or already controlled on MMF-based treatment.	Useful in selected patients, but long-term toxicity and reproductive issues must be considered.	[3,43,76]
Maintenance/steroid-sparing option in selected patients	Budesonide	Not a standard first-line option; may be considered in selected non-cirrhotic patients with steroid-related adverse effects or steroid dependency.	Contraindicated in cirrhosis. Current European guidance does not recommend it as routine first-line therapy. Evidence in salvage settings remains limited.	[3,77,78]
Second-line conventional therapy	MMF	Preferred second-line option in many patients with AZA intolerance and in selected cases of insufficient response to standard therapy.	Evidence is strongest in AZA intolerance; efficacy in truly refractory disease is less robust.	[9,79,80]
Second-line conventional therapy	Calcineurin inhibitors (tacrolimus, cyclosporine A)	Important second-line options when first-line therapy fails or is not tolerated, especially if MMF is ineffective, contraindicated, or not tolerated.	Require individualized use and close monitoring for nephrotoxicity, hypertension, neurotoxicity, and other drug-specific adverse events.	[9,79,80]
Second-line conventional therapy	6-mercaptopurine (6-MP)	Alternative thiopurine option in selected patients intolerant to AZA.	Clinical experience is more limited than with AZA or MMF.	[9,10]
Second-line/highly selected therapy	6-thioguanine (6-TG)	Highly selected option in difficult cases.	Experience in AIH is limited; higher doses have been associated with non-cirrhotic portal hypertension.	[10]
Third-line/biologic rescue therapy	Rituximab (anti-CD20)	Rescue therapy for difficult-to-manage or refractory AIH after failure of conventional second-line approaches.	Available evidence suggests biochemical improvement and steroid-sparing effects, but data come mainly from small cohorts and retrospective series.	[81]
Third-line/biologic rescue therapy	Belimumab (anti-BAFF)	Investigational/rescue biologic for highly selected refractory patients.	Preliminary results are encouraging, but its role remains to be defined in larger prospective studies.	[82]
Third-line/biologic rescue therapy	Ianalumab/VAY736 (anti-BAFF receptor)	Emerging biologic currently under formal clinical evaluation in incomplete responders.	Not established in routine clinical practice; positioning depends on trial results.	[83]
Third-line/biologic rescue therapy	Anti-TNF agents (infliximab, adalimumab)	Exceptional salvage option in refractory AIH.	Some patients achieve biochemical remission, but caution is needed because anti-TNF agents may also induce AIH-like liver injury.	[85,86,87]
Third-line/biologic rescue therapy	Basiliximab (anti-IL-2 receptor)	Anecdotal rescue therapy in isolated cases.	Evidence is limited to case reports or highly selected settings, including immune checkpoint inhibitor-associated AIH.	[84]
Third-line/rescue therapy	Sirolimus (mTOR inhibitor)	Possible rescue option in highly selected refractory patients.	Evidence remains sparse and largely anecdotal.	[80]
Investigational therapy	Zetomipzomib (selective immunoproteasome inhibitor)	Experimental therapy under clinical investigation for difficult-to-treat AIH.	Currently being studied in the PORTOLA phase IIa randomized trial.	[80]
Investigational therapy	JKB-122 (TLR4 antagonist)	Experimental strategy aimed at reducing innate immune activation.	Studied in steroid-refractory AIH; not part of routine management.	[9,80]
Investigational therapy	Cell therapy (mesenchymal stromal cells; MERLIN trial)	Experimental immune-tolerance approach.	Promising but still preliminary; currently limited to clinical trial settings.	[88]
Definitive rescue therapy	Liver transplantation	Definitive rescue option for fulminant hepatic failure with encephalopathy or acute-severe AIH without adequate response to corticosteroids.	Early transplant evaluation is essential in these settings.	[3,42]

Abbreviations: AIH, autoimmune hepatitis; AZA, azathioprine; MMF, mycophenolate mofetil; 6-MP, 6-mercaptopurine; 6-TG, 6-thioguanine; BAFF, B-cell activating factor; anti-CD20, CD20-directed monoclonal antibody; anti-TNF, tumour necrosis factor inhibitor; anti-IL-2 receptor, interleukin-2 receptor antagonist; mTOR, mechanistic target of rapamycin; TLR4, toll-like receptor 4.

Table 1. Current and emerging therapies for autoimmune hepatitis: clinical role and key considerations.

Therapeutic Step	Therapy	Current Role/Positioning	Key Considerations	Key References
First-line induction therapy	Predniso(lo)ne + azathioprine (AZA)	Standard first-line treatment in adult AIH. Remains the best-established regimen for induction of remission.	Predniso(lo)ne is usually started at ≥0.5 mg/kg/day, up to 1 mg/kg/day in more severe disease, with response-guided tapering. AZA is generally introduced after corticosteroid initiation and titrated according to tolerance and response.	[3,43,75]
First-line induction therapy (alternative in selected adults)	Mycophenolate mofetil (MMF)	First-line alternative to AZA in selected adult patients, particularly when AZA is contraindicated or poorly tolerated.	Usually given at 1.5–2 g/day. Teratogenicity and reproductive counselling must be considered. In pediatric AIH, MMF is generally positioned more often as second-line rather than standard first-line therapy.	[3,9,76]
Maintenance therapy	AZA monotherapy ± low-dose corticosteroids	Standard maintenance strategy after remission induction.	Long-term therapy is required in most patients. The aim is sustained biochemical remission with the lowest effective steroid exposure.	[3,43]
Maintenance therapy	MMF monotherapy ± low-dose corticosteroids	Alternative maintenance strategy in patients intolerant of AZA or already controlled on MMF-based treatment.	Useful in selected patients, but long-term toxicity and reproductive issues must be considered.	[3,43,76]
Maintenance/steroid-sparing option in selected patients	Budesonide	Not a standard first-line option; may be considered in selected non-cirrhotic patients with steroid-related adverse effects or steroid dependency.	Contraindicated in cirrhosis. Current European guidance does not recommend it as routine first-line therapy. Evidence in salvage settings remains limited.	[3,77,78]
Second-line conventional therapy	MMF	Preferred second-line option in many patients with AZA intolerance and in selected cases of insufficient response to standard therapy.	Evidence is strongest in AZA intolerance; efficacy in truly refractory disease is less robust.	[9,79,80]
Second-line conventional therapy	Calcineurin inhibitors (tacrolimus, cyclosporine A)	Important second-line options when first-line therapy fails or is not tolerated, especially if MMF is ineffective, contraindicated, or not tolerated.	Require individualized use and close monitoring for nephrotoxicity, hypertension, neurotoxicity, and other drug-specific adverse events.	[9,79,80]
Second-line conventional therapy	6-mercaptopurine (6-MP)	Alternative thiopurine option in selected patients intolerant to AZA.	Clinical experience is more limited than with AZA or MMF.	[9,10]
Second-line/highly selected therapy	6-thioguanine (6-TG)	Highly selected option in difficult cases.	Experience in AIH is limited; higher doses have been associated with non-cirrhotic portal hypertension.	[10]
Third-line/biologic rescue therapy	Rituximab (anti-CD20)	Rescue therapy for difficult-to-manage or refractory AIH after failure of conventional second-line approaches.	Available evidence suggests biochemical improvement and steroid-sparing effects, but data come mainly from small cohorts and retrospective series.	[81]
Third-line/biologic rescue therapy	Belimumab (anti-BAFF)	Investigational/rescue biologic for highly selected refractory patients.	Preliminary results are encouraging, but its role remains to be defined in larger prospective studies.	[82]
Third-line/biologic rescue therapy	Ianalumab/VAY736 (anti-BAFF receptor)	Emerging biologic currently under formal clinical evaluation in incomplete responders.	Not established in routine clinical practice; positioning depends on trial results.	[83]
Third-line/biologic rescue therapy	Anti-TNF agents (infliximab, adalimumab)	Exceptional salvage option in refractory AIH.	Some patients achieve biochemical remission, but caution is needed because anti-TNF agents may also induce AIH-like liver injury.	[85,86,87]
Third-line/biologic rescue therapy	Basiliximab (anti-IL-2 receptor)	Anecdotal rescue therapy in isolated cases.	Evidence is limited to case reports or highly selected settings, including immune checkpoint inhibitor-associated AIH.	[84]
Third-line/rescue therapy	Sirolimus (mTOR inhibitor)	Possible rescue option in highly selected refractory patients.	Evidence remains sparse and largely anecdotal.	[80]
Investigational therapy	Zetomipzomib (selective immunoproteasome inhibitor)	Experimental therapy under clinical investigation for difficult-to-treat AIH.	Currently being studied in the PORTOLA phase IIa randomized trial.	[80]
Investigational therapy	JKB-122 (TLR4 antagonist)	Experimental strategy aimed at reducing innate immune activation.	Studied in steroid-refractory AIH; not part of routine management.	[9,80]
Investigational therapy	Cell therapy (mesenchymal stromal cells; MERLIN trial)	Experimental immune-tolerance approach.	Promising but still preliminary; currently limited to clinical trial settings.	[88]
Definitive rescue therapy	Liver transplantation	Definitive rescue option for fulminant hepatic failure with encephalopathy or acute-severe AIH without adequate response to corticosteroids.	Early transplant evaluation is essential in these settings.	[3,42]

Abbreviations: AIH, autoimmune hepatitis; AZA, azathioprine; MMF, mycophenolate mofetil; 6-MP, 6-mercaptopurine; 6-TG, 6-thioguanine; BAFF, B-cell activating factor; anti-CD20, CD20-directed monoclonal antibody; anti-TNF, tumour necrosis factor inhibitor; anti-IL-2 receptor, interleukin-2 receptor antagonist; mTOR, mechanistic target of rapamycin; TLR4, toll-like receptor 4.

In practical terms, MMF is generally the preferred second-line option, particularly in patients intolerant to azathioprine and in many patients with insufficient response to standard therapy. Calcineurin inhibitors, especially tacrolimus, may be considered when MMF is ineffective, contraindicated, or not tolerated. Biologic agents, such as rituximab or belimumab, should currently be reserved for carefully selected refractory cases after failure of conventional second-line therapies, given the limited evidence base and the absence of head-to-head comparative data. Thus, current management remains largely stepwise and individualized, rather than guided by a validated therapeutic algorithm.

6.2. Treatment of Acute-Severe AIH

Acute-severe or fulminant presentations of AIH pose major management challenges and require early evaluation for liver transplantation. In fulminant hepatic failure with encephalopathy, corticosteroids provide little therapeutic benefit; urgent transplantation is strongly recommended [3]. If encephalopathy is absent, a trial of predniso(lo)ne at 1 mg/kg/day may be undertaken with reassessment after 3–7 days; failure of INR and bilirubin to improve should trigger expedited transplant referral [3]. Although no fixed cutoffs are defined, MELD, MELD-Na, and UKELD scores can aid assessment of disease severity and therapeutic response [3].

6.3. Treatment Withdrawal

A carefully supervised trial of treatment withdrawal should be reserved for highly selected patients who have maintained stable CBR on low-dose monotherapy for at least two years [3]. Prior EASL guidance suggested considering a pre-withdrawal liver biopsy but did not issue a universal recommendation owing to limited prospective evidence [8]. Whether histology is superior to AST/ALT and IgG for confirming remission remains uncertain; nonetheless, residual inflammatory activity on biopsy during complete biochemical remission appears to correlate with a higher risk of relapse after withdrawal, based on imperfect and heterogeneous data [27]. Conversely, treatment withdrawal should generally be avoided in patients with previous relapse, unstable biochemical control, cirrhosis, or a history of difficult-to-treat disease, in whom long-term maintenance is often the safer strategy.

7. Controversies and Future Directions

Several unresolved issues continue to shape the management of AIH. First, it remains debated whether complete biochemical remission should strictly require normalization of IgG in addition to aminotransferases, or whether transaminase normalization alone may be sufficient in selected patients. Second, the optimal strategy for treatment withdrawal remains uncertain, particularly regarding the role of liver biopsy before discontinuation and the identification of patients at highest risk of relapse. Third, the sequencing of MMF, calcineurin inhibitors, and biologic therapies in refractory disease has not been standardized, and current practice still relies largely on expert opinion, small retrospective cohorts, and individualized decision-making.

Future research should prioritize prospective head-to-head comparisons of second-line agents, a clearer positioning of biologic therapies, and the validation of biomarkers capable of predicting treatment response, relapse risk, and safe withdrawal. In parallel, improved integration of pathophysiological mechanisms with therapeutic stratification may help move AIH management toward a more personalized and steroid-sparing approach.

8. Conclusions

Autoimmune hepatitis is a chronic, heterogeneous, immune-mediated liver disease in which timely recognition and a composite diagnostic approach based on biochemistry, serology, and histology, remain pivotal. First-line therapy continues to centre on corticosteroids with azathioprine or mycophenolate mofetil, while second-line agents and emerging B- and T-cell-targeted biologics offer options for refractory or intolerant patients, albeit with evidence largely from small cohorts. Response-guided management, careful steroid tapering, and longitudinal non-invasive fibrosis assessment are essential to optimize outcomes and limit treatment-related toxicity. Future priorities include multicentre randomized trials to define comparative effectiveness and sequencing, development of validated biomarkers (including microbiome-based signatures) for risk stratification and withdrawal decisions, and mechanism-based, steroid-sparing strategies capable of inducing durable remission.