Overview of autoimmune liver disease: Prevalence, risk factors, and role of autoantibodies

Abstract

Answer questions and earn CME Content available: Author Interview and Audio Recording Pediatric autoimmune liver disease encompasses several entities. Autoimmune hepatitis (AIH), primary sclerosing cholangitis (PSC), and AIH/PSC overlap syndrome, also known as autoimmune sclerosing cholangitis (ASC), are the most common pediatric autoimmune liver diseases, while primary biliary cholangitis (PBC) and immunoglobulin G4 (IgG4)-related cholangitis are much rarer in the pediatric population. We will describe the epidemiology of pediatric autoimmune liver disease with a specific focus on distribution, risk factors, and associated autoantibodies. AIH is an inflammatory liver disease that is characterized by variable presentations, including elevated transaminases and interface hepatitis on histology. Most patients are diagnosed before 18 years of age, with peak incidence occurring before puberty.1 Although incidence is increasing worldwide, the epidemiology of AIH varies despite occurring in all ethnicities.2 Alaskan natives have a high frequency of jaundice at presentation, Hispanics more commonly present with cirrhosis, and African Americans typically have accelerated progress of disease and a higher rate of AIH recurrence after liver transplantation. AIH has a female predominance in both adults (71%–95%) and children (60%–76%).3 Worldwide AIH prevalence has been reported to be 17.44 cases per 100,000 individuals. The yearly incidence of pediatric AIH is 0.4 cases per 100,000 individuals in the United States, 0.56 per 100,000 in Argentina, and 0.23 per 100,000 in Canada, with type 1 AIH occurring more frequently than type 2 AIH.4 PSC is characterized by intrahepatic and/or extrahepatic bile duct inflammation, fibrosis, and eventual destruction.5 PSC is a heterogenous disease, including subgroups of large-duct PSC, small-duct PSC, and overlap syndrome or ASC.6 PSC is more common in males (65%–70%), with an estimated incidence of 0.91 to 1.3 cases per 100,000 individuals in Europe. Although relatively infrequent in pediatrics with an incidence of less than 20% of adults, pediatric PSC remains an important cause of morbidity and mortality in children, accounting for 2% of pediatric liver transplants in the United States.6 For pediatrics, about 155 children are diagnosed with PSC annually, with about 1200 currently living with the disease.7 The reported prevalence rate of ASC in children with AIH is 20% to 49%, which is considerably higher than the 1.7% to 10% prevalence rate of overlap syndrome in adults with AIH. However, using magnetic resonance cholangiography as a diagnostic measure, one prospective case–control study showed a pediatric prevalence rate of 10.5% for ASC in AIH, similar to adults.8 PBC, the most common autoimmune liver disease in adults, is characterized by progressive destruction of intrahepatic bile ducts, and its prevalence is increasing and estimated at 430 cases per 1 million women and 110 cases per 1 million men, with a higher prevalence in the Asia-Pacific.9, 10 As of 2019, there were only four documented cases of pediatric primary biliary cirrhosis in the literature, all in females.11 IgG4-related disease is a systemic fibroinflammatory condition with unclear pathogenesis. In the liver, it most commonly presents as IgG4-related cholangitis affecting mainly extrahepatic and perihilar bile ducts but can also affect intrahepatic bile ducts as well.12 There is insufficient data on the prevalence of IgG4-related cholangitis, although a study in Japan reports an estimated adult prevalence of 2.18 cases per 100,000 individuals.13 In 2016, Karim et al.14 performed a systematic review of IgG4-related disease in pediatrics describing 25 cases, of which only two had liver manifestations. Although the etiology of AIH is obscure, genetic, epigenetic, and environmental risk factors are thought to be involved in its pathogenesis.15 AIH develops in genetically predisposed individuals after exposure to environmental factors, leading to loss of self-tolerance and immune-mediated injury. Studies have confirmed the role of human leukocyte antigen (HLA) alleles, mainly HLA-DR3 and HLA DR-4, in AIH occurrence and outcome.16 Both AIH type 1 and type 2 have isolated partial deficiency of the HLA class III complement component C4.2 HLA genes in AIH vary among different ethnicities and geographic regions. Genetic predisposition is conferred mainly by polymorphisms in HLA alleles such as DRB1 and single-nucleotide variants (SNVs) as non-HLA genes associated with AIH.2, 15 Epigenetic modifications such as microRNA (miRNA) and DNA methylation that influence the expression of genes along with their function without affecting the sequence of nucleotides have also been implicated in the development of AIH.16 Concordance of AIH in twins and family accumulation of AIH further solidify a genetic component to the disease.2 Female sex is a risk factor because 75% of patients with AIH are female. In addition, viral infections, such as the hepatitis viruses, along with Epstein-Barr virus, cytomegalovirus, and herpes simplex type 1 virus, have all been associated with the development of AIH. Low vitamin D levels, altered microbiota, exposure to sex hormones, and specific medications are other environmental factors thought to increase the risk for AIH development.17 PSC is also thought to be secondary to both genetic and environmental risk factors. PSC has been associated with genetic variants within the HLA complex on chromosome 6, suggesting a pathogenic role for T cell–mediated injury. Although inflammatory bowel disease (IBD) is strongly associated with PSC, with approximately 70% of those with PSC having concurrent IBD, PSC rarely occurs within the same family. The hazard ratio for a first-degree relative to also have PSC is about 11, suggesting a significant role of environmental factors in disease pathogenesis. There is a geographical distribution of PSC with predominance in Northern Europe.18 Latitude has been proposed as a risk factor for PSC secondary to the link of vitamin D deficiency and risk for autoimmune conditions, along with differences in lifestyle. Microbial exposure may also play a role, especially during childhood. Breaking of tolerance to the biliary microbiome has been hypothesized as a risk factor for the pathogenesis of PSC. Interestingly, smoking has been described as negative risk factor for PSC.19 ASC affects males and females equally, with approximately 45% of patients having concurrent IBD.20 PBC mainly affects older women and is extremely rare in pediatrics. A combination of genetic and environmental risk factors leads to chronic immune-mediated damage to biliary epithelial cells (BECs).9 A high concordance among identical twins (up to 60%) and a strong association with HLA alleles support a genetic predisposition to PBC. Environmental risk factors include urinary tract infections, reproductive hormone replacement, personal history of cholestasis in pregnancy, nail polish, and cigarette smoking. Geographic clustering of patients with PBC suggests further environmental exposures and potential socioeconomic status risk factors as well.10 For example, toxic waste, coal mining, and environmental cadmium have been described as risk factors for PBC.9 PBC has also been noted with other autoimmune conditions, such as CREST (calcinosis, Raynaud's syndrome, esophageal dysmotility, sclerodactyly, telangiectasia [a limited form of scleroderma]) syndrome, scleroderma, and Raynaud's disease.10 In adults, IgG4-related cholangitis typically occurs in men and presents between 50 and 70 years of age with obstructive jaundice and is usually associated with type 1 autoimmune pancreatitis.12 In children, median age of presentation is 13 years with female predominance.14 Genetic risk factors include HLA susceptibility loci, especially because familial clusters have been noted. Environmental factors, such as occupations with exposure to solvents or industrial gases, along with infections such as Staphylococcus aureus and Mycobacterium tuberculosis, are potential risk factors (Table 1 and Figure 1).13 A key diagnostic criterion of AIH is the detection of autoantibodies, which also allows differentiation between type 1 and type 2 AIH. Evaluation of hepatitis and cholestasis of unknown origin should include screening with liver-related autoantibodies. Although some may be nonspecific, the presence of AIH-related antibodies is important for diagnosis because the pattern of autoantibodies can shed light on clinical presentation and prognosis of AIH, helping differentiate type 1 and type 2 AIH. Still, about 10% to 15% of patients with AIH present without positive autoantibodies, otherwise known as seronegative AIH.2, 21 Type 1 AIH is characterized by anti-nuclear antibody (ANA) and smooth muscle antibody (SMA), while type 2 AIH is characterized mainly by liver-kidney microsomal type 1 antibody (LKM-1), but also liver cytosol antibody type 1 (LC-1).1 At presentation of AIH in North Americans, ANA is detected in 80%, SMA in 63%, and LKM-1 in only 3%. About half of patients with AIH have only one autoantibody present at diagnosis, while half can have multiple autoantibodies present. The diagnostic accuracy of AIH improves from 58% to 74% in the presence of two autoantibodies.21 ANA lacks disease specificity and can also be present in other diseases, such as lupus erythematosus, Hashimoto's thyroiditis, systemic sclerosis, or celiac disease. It can also be present in other liver diseases, including the hepatitis viruses, drug-induced liver injury, Wilson's disease, alcohol-induced liver disease, and nonalcoholic fatty liver disease. ANA positivity has also been clearly described in healthy individuals in the general population.2 SMA is associated with inflammatory activity in type 1 AIH because it targets actin in smooth muscles.21 In pediatrics, SMA is positive in 85% of type 1 AIH cases, with its titers correlating with disease activity, and thus is included in the definition of disease remission.2 Interestingly, basically all patients with ASC are positive for ANA and/or SMA with elevated serum IgG levels as well.20 The molecular targets of LKM-1 and LC-1 are the cytochrome P450 2D6 (CYP2DC) and formiminotransferase cyclodeaminase, respectively. Although LC-1 can present alone in type 2 AIH, it is present together with LKM-1 32% of the time. Like ANA, LKM-1 is not completely specific to type 2 AIH because 13% of patients with hepatitis C can also be positive. However, LC-1 is rarely detected in hepatitis C.2 Other autoantibodies have also been described in AIH. Soluble liver antigen (SLA) is associated with more aggressive disease and, if present, has a specificity of 98.9% for AIH with a cutoff of 1:40. However, it is present in only 7% to 22% of patients with type 1 AIH. Studies report that anti-SLA-positive AIH patients more often require lifelong immunosuppression, take longer to achieve disease remission, and less frequently can be weaned from corticosteroid treatment as compared with anti-SLA-negative patients. Perinuclear anti-neutrophil cytoplasmic antibody (pANCA), anti-actin, and anti-asialoglycoprotein receptor antibody (anti-ASGPR) can also point to a diagnosis of AIH.2 pANCA is present more often in ASC (up to 75%) than AIH (45% in type 1 AIH and 10% in type 2 AIH).20 Thus, in general, first-line testing for AIH should include ANA, SMA, and LKM-1. However, in cryptogenic hepatitis with absent ANA, SMA, and LKM-1, second-line testing should be considered because positivity of anti-actin, SLA, LC-1, or atypical pANCA may suggest a diagnosis of AIH, especially because ANA and SMA may be expressed only later in the disease course.2 The diagnosis of PSC is based on an abnormal cholangiogram showing bile duct injury with multifocal strictures and segmental dilatations of the biliary system or via liver biopsy with histological features of periductal fibrosis with obliteration of small bile ducts when the cholangiogram is normal. Still, a wide range of autoantibodies can be detected in PSC, but most are present at a low prevalence and low titers. Autoantibodies typically have a limited role in the diagnosis of PSC, but common ones include pANCA in 93% of patients, ANA in 77%, and SMA in 0% to 83%.5 Few studies have found antibodies of different subtypes directed against BECs in PSC. In addition, levels of IgA antibodies directed against BECs correlate with adverse patient outcomes in PSC.21 Anti-Saccharomyces cerevisiae antibody (ASCA) is an antibody against baker's yeast (microbial antigens). It was first described in Crohn's disease, and ASCA was reported as positive in 44% of patients with PSC. Anti-phospholipid antibodies are reported anywhere from 4% to 63% of patients with PSC. Anti-endothelial cell antibodies are directed against antigens in endothelial cells and have been reported in 35% of patients with PSC but are generally nonspecific.22 IgA class antibodies against glycoprotein 2 (anti-GP2 IgA), which were previously linked to severe Crohn's disease, have also been detected in PSC at a prevalence rate of 46% to 71%. The prevalence of anti-GP2 IgA was strongly associated with large-duct PSC and increased mortality, suggesting its role as a biomarker in risk stratification for this patient population.21 IgG4-related cholangitis can display a similar pattern of bile duct injury as PSC, and thus one must consider evaluation of serum IgG4 levels in these patients.5 Anti-mitochondrial antibody (AMA) is localized to the apical surface of BECs, targets lipoic acid on the inner mitochondrial membrane, and is associated with apoptosis of those cells. AMA is positive in 95% of patients with PBC but can rarely be absent in true disease. If negative, immunofluorescence of ANA can aid in diagnosis (multinuclear dot, perinuclear rim, and centromere activity).9 Anti-kelch like 12 (35%) and anti-hexokinase (22%) antibodies can also be found in AMA-negative PBC but are not widely available.10 Anti-glycoprotein 210 and anti-sp100 antibodies are specific for PBC.9 ANA and anti-SMA can be found in 50% of PBCs. IgM is usually elevated, but IgG is usually normal.10 Assessment of IgG4 antibody is integral to the diagnosis of IgG4-related disease but lacks sensitivity and specificity.12 The diagnostic value of serum IgG4 increases when it is more than four times the upper limit of normal and thus, because the disease is characterized by infiltration of IgG4 plasma cells into tissues, many diagnostic algorithms include serum IgG4 >135 mg/dL and >10 IgG4+ cells per high-power field (hpf) within tissue. Greater than 50 IgG4+ cells per hpf in tissue is more specific for IgG4-related disease.13 More recent studies suggest that class-switched IgG4+ clones in the B cell receptor, circulating plasmablasts, and CC-chemokine ligand 18 could serve as better biomarkers for IgG4-related disease (Table 2).12 Nothing to report.

Topics & Concepts