THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2002 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.
Vol. 97, No. 2, 2002 ISSN 0002-9270/02/$22.00 PII S0002-9270(01)04041-2
Antibodies to Soluble Liver Antigen/Liver Pancreas and HLA Risk Factors for Type 1 Autoimmune Hepatitis Albert J. Czaja, M.D., F.A.C.G., Peter T. Donaldson, Ph.D., and Ansgar W. Lohse, M.D. Division of Gastroenterology and Hepatology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota; The Centre for Liver Research, Faculty of Clinical Medical Sciences, The Medical School, University of Newcastle, Newcastle-upon-Tyne, United Kingdom; and Department of Medicine, Johannes Gutenberg University, Mainz, Germany
OBJECTIVE: Antibodies to soluble liver antigen/liver-pancreas are highly specific markers of type 1 autoimmune hepatitis that have been associated with relapse. Our aim was to determine if these antibodies are reflective of a genetic predisposition for recrudescent disease. METHODS: One hundred forty-four white North American patients were evaluated by an enzyme immunoassay and by Western blot using recombinant soluble liver antigen/liverpancreas; 122 were assessed for class II human leukocyte antigens (HLAs). RESULTS: Twenty-two patients (15%) had antibodies to soluble liver antigen/liver-pancreas. These patients were indistinguishable from seronegative patients by clinical, laboratory, and histological features at presentation. Patients with antibodies to soluble liver antigen/liver pancreas had HLA DR3 (79% vs 50%, p ⫽ 0.02) more commonly and HLA DR4 less often (16% vs 47%, p ⫽ 0.02) than patients with smooth muscle antibodies and/or antinuclear antibodies. Seropositivity was associated with DRB1*0301 and seronegativity was associated with DRB1*0401. Relapse after drug withdrawal occurred in all patients with antibodies to soluble liver antigen/liver-pancreas and at a higher frequency than in patients with conventional antibodies (100% vs 78%, p ⫽ 0.05). CONCLUSIONS: Antibodies to soluble liver antigen/liver pancreas are associated with HLA DR3 and the susceptibility allele, DRB1*0301. Antibodies to soluble liver antigen/ liver-pancreas may be surrogate markers of a genetic propensity for recrudescent disease or the target autoantigen. They may be complementary to antinuclear antibodies and smooth muscle antibodies in diagnosis and management. (Am J Gastroenterol 2002;97:413– 419. © 2002 by Am. Coll. of Gastroenterology)
Presented in part at the meeting of the American Association for the Study of Liver Diseases, May 21, 2001, Atlanta, GA.
INTRODUCTION Antinuclear antibodies (ANAs) and smooth muscle antibodies (SMAs) are the serological hallmarks of type 1 autoimmune hepatitis (1–5). These autoantibodies lack disease and organ specificity, and their diagnostic significance depends on the strength of other clinical and histological findings (6 – 8). ANAs are directed against diverse nuclear antigens, including ribonucleoprotein complexes, small ribonucleoproteins, and centromeres (9, 10). These numerous, often concurrent reactivities obscure the principal target of the immune response and suggest that the ANAs are generated in a nonselective fashion by hepatocyte destruction. Similarly, SMAs are directed against actin and nonactin components, including tubulin, vimentin, desmin, and skeletin (3–5). They are also present in a variety of liver and nonliver diseases, and like ANAs, they have a variable expression in individual patients (11). Antibodies to actin have greater diagnostic specificity than SMAs but less sensitivity for autoimmune hepatitis, and a standardized assay for their detection is unavailable (12–14). Antibodies to soluble liver antigen (anti-SLAs) (15) and antibodies to liver pancreas (anti-LPs) (16) have identical reactivities (17), and they are highly specific markers of autoimmune hepatitis (18). We have shown previously that these antibodies, now designated as anti-SLA/LPs, occur in at least 11% of patients with type 1 autoimmune hepatitis (19) and that they are useful in securing the diagnosis in individuals who lack ANAs and SMAs (18, 19). They also identify patients who commonly relapse after corticosteroid therapy (20). A standardized immunoassay for the detection of antiSLA/LPs has been developed based on the recombinant target antigen (20, 21). This assay has greater sensitivity and specificity than the previous inhibition ELISA, which was based on the binding inhibition of an indicator serum positive for anti-SLA/LPs (15, 20, 21). Because anti-SLA/LPs have high specificity for autoimmune hepatitis and are directed against a single antigen, they may reflect a pertinent pathogenic process. The new recombinant assay affords the
Czaja et al.
opportunity to assess the importance of this pathogenic process more directly than the inhibition assay used in our earlier studies. Susceptibility, clinical expression, and outcome of type 1 autoimmune hepatitis in white Northern European and North American patients are associated with human leukocyte antigen (HLA) DR3 and HLA DR4 (22–24). HLA DR3 is associated with early age onset and treatment failure, and HLA DR4 is associated with concurrent immune diseases and remission during corticosteroid therapy (25, 26). High resolution DNA-based techniques have identified DRB1*0301 and DRB1*0401 as the alleles affecting disease behavior (23, 24, 26). Autoantibody expression may also have a genetic predisposition, especially if the autoantibodies are imprints of a pertinent pathogenic pathway (27, 28). The high specificity of anti-SLA/LPs for type 1 autoimmune hepatitis and the association of these autoantibodies with treatment response suggest that they may be independent indices of prognosis or surrogate markers of the principal susceptibility alleles, relevant autoimmune promoters, and/or target autoantigen responsible for the disease (29, 30). As such, anti-SLA/LPs may be probes by which the pathogenic mechanisms of autoimmune hepatitis can be better understood. In this report, we correlate seropositivity for anti-SLA/ LPs by the new recombinant immunoassay with the known HLA risk factors for type 1 autoimmune hepatitis to determine if anti-SLA/LPs are markers of a genetic predisposition that influences disease occurrence and/or outcome.
MATERIALS AND METHODS Study Population One hundred forty-four patients who satisfied international criteria for autoimmune hepatitis (31, 32) were evaluated by recombinant immunoassay for anti-SLA/LPs. Forty-six of these patients had been evaluated previously by the original inhibition assay for anti-SLA/LPs and reported on in our earlier study (19). One hundred fifteen (80%) were female, and ages ranged from 13 to 82 yr (mean age ⫽ 46 ⫾ 1). The mean duration of symptoms at presentation was 27 ⫾ 3 months (range ⫽ 1–180), and 39 patients (27%) had cirrhosis at accession. Thirty-three patients (23%) had ANAs, 34 (24%) had SMAs, and 77 patients (53%) had ANAs and SMAs at presentation. These findings justified their designation as type 1 autoimmune hepatitis (6). Two of 124 tested patients were also positive for antibodies to liver/kidney microsome type 1 (2%). Each was retained in the analysis as examples of individuals with mixed autoantibody profiles (33). Our study had been approved by the Institutional Review Board of the Mayo Clinic as part of a program project. Clinical Assessments Each patient underwent a comprehensive clinical review and complete physical examination at presentation and at each follow-up visit by one investigator (A.J.C.). Conven-
AJG – Vol. 97, No. 2, 2002
tional laboratory tests of liver inflammation and function were assessed in a uniform fashion. Concurrent diseases of a presumed autoimmune nature were sought systematically in each patient in accordance with a previously published protocol (25, 26, 28). Fifty-five patients (38%) had concurrent diseases of a presumed autoimmune nature, including 21 with autoimmune thyroiditis; nine with ulcerative colitis; five with Graves’ disease; three with synovitis; two with autoimmune hemolytic anemia, rheumatoid arthritis, systemic lupus erythematosus, or neuritis; and one with Crohn’s disease, dermatitis herpetiformis, erythema nodosum, focal myositis, idiopathic thrombocytopenic purpura, pernicious anemia, sprue, pulmonary fibrosis, or vasculitis. SMAs and ANAs were sought by indirect immunofluorescence on murine tissue sections in all patients at presentation, as described previously (8, 9, 14). A serum titer of 1:40 or higher was considered positive for these markers. Antibodies to liver/kidney microsome type 1 were sought by indirect immunofluorescence on murine tissue sections in 124 patients (86%), and a serum titer of 1:10 or higher was considered positive for these antibodies (8, 33). All patients were screened for antibodies to hepatitis C virus and found to be negative by a second generation ELISA (Ortho Diagnostic Systems, Raritan, NJ). Similarly, all patients were tested and negative for hepatitis B surface antigen by ELISA (Abbott Laboratories, North Chicago, IL). Treatment Regimens One hundred thirty-eight patients (96%) were treated with prednisone in combination with azathioprine (87 patients) or a higher dose of prednisone alone (51 patients). These regimens had been shown previously to be of comparable efficacy (34), and outcomes were analyzed only in individuals treated in this fashion. Two patients received no medication, three patients were treated with investigational drugs, and one patient received azathioprine only. Outcomes in these individuals were not analyzed. Treatment was continued until remission, treatment failure, or drug intolerance (34). Remission connoted absence of symptoms; resolution of laboratory abnormalities, except for a less than 2-fold elevation of the serum AST level; and improvement of histological findings to normal, inactive cirrhosis or nonspecific inflammation. Treatment failure connoted clinical, laboratory, and/or histological worsening despite compliance with treatment. Recrudescence of inflammatory activity after remission and drug withdrawal indicated a relapse, and continued inactivity after remission and drug withdrawal connoted a sustained remission. All patients were observed in a uniform fashion during 128 ⫾ 8 months. ELISA for Anti-SLA/LPs Recombinant SLA/LP antigen was expressed in Escherichia coli using the J-D1 complementary DNA clone that had
AJG – February, 2002
been described previously (20, 21). Affinity purified recombinant antigen (Diarect, Freiburg, Germany) was used to establish a standardized ELISA that is commercially available (Euroimmun, Gross Groenau bei Luebeck, Germany). Briefly, the recombinant protein was coated on maxisorp ELISA plates (Nunc, Roskilde, Denmark) at an antigen concentration of 0.02 g/well and then incubated with 100 l of patient’s serum (diluted 1:100 in buffer) for 30 min at room temperature. After repeated washings, the plates were incubated with 100 l of peroxidase-conjugated, goat-derived antibody to human IgG for 30 min, washed, and then treated with 100 l of tetramethylbenzidine/hydrogen peroxide to induce a change from colorless to green in the anti-SLA/LP–positive wells. The reaction was then stopped by 100 l of stop solution (1-N phosphoric acid), and the plates were read at ⫽ 450 nm (and a reference wavelength of ⬎ 620 nm) in a Spectra Mini (Tecan, Crailsheim, Germany). Calibration was done in relative units/ml, and a cutoff value of 20 relative units/ml was considered positive, as reported previously (20, 21). All sera testing positive or borderline by ELISA were assessed by Western blot. Positive results required confirmation by this method. Only one serum sample from each patient was tested for anti-SLA/ LPs. Each specimen had been obtained at diagnosis (47 samples), during treatment (79 samples), or during follow-up after treatment (18 samples). Specimens were stored frozen (⫺80°C) in a dedicated serum storage bank before testing. HLA Determinations One hundred twenty-two patients (85%) were assessed for class II (DR locus) HLA by a standard microlymphocytotoxicity technique, restriction fragment length polymorphism, and/or polymerase chain reaction with sequence specific primers (PCR-SSP), as described previously (24, 25). HLA DR3 was present in 67 of the 122 tested patients (55%), and HLA DR4 was detected in 51 of these same patients (42%), including 14 patients who had both HLA DR3 and HLA DR4. One hundred seven patients (74%) were tested for DRB1*0301, and 109 patients (76%) were tested for the 10 most common of the 26 possible alleles associated with HLA DR4 by PCR-SSP (24, 26). These included DRB1*0401, *0402, *0403, *0404, *0405, *0406, *0407, *0408, *0409, and *0410. All individuals were white North Americans. DRB1*0301 was sought in 57 of the 67 patients with HLA DR3, and it was detected in each (100%). DRB1*0401 was sought in 47 of the 51 patients with HLA DR4, and it was detected in 40. Seven patients with HLA DR4 who were tested by PCR-SSP did not have DRB1*0401, and one other patient with DRB1*0401 also had a non-DRB1*0401 allele. Three patients had DRB1*0404, two patients had DRB1*0405, and one patient each had DRB1*0407 or DRB1*0409. One patient with DRB1*0401 also had DRB1*0404. Homozygosity and heterozygosity for HLA
Anti-SLA/LP in Autoimmune Hepatitis
Table 1. Clinical Features at Presentation of Patients With and Without Anti-SLA/LPs
Clinical Features Age (yr) Duration of symptoms (mo) Female:male Concurrent immune diseases AST (nl, ⱕ31 U/L) Bilirubin (nl, ⱕ1.1 mg/dl) ␥-Globulin (nl, 0.7–1.7 g/dl) IgG (nl, 700–1500 mg/dl) SMA only ⱖ1:40 ANA only ⱖ1:40 SMA and ANA ⱖ1:40 Antibodies to liver/kidney microsome type 1 ⬎1:10 Cirrhosis at entry
Anti-SLA/LP Positive (N ⫽ 22)
Anti-SLA/LP Negative (N ⫽ 122)
43 ⫾ 2 40 ⫾ 11 17:5 9 (41) 438 ⫾ 81 4.6 ⫾ 1.2 3.2 ⫾ 0.2 2753 ⫾ 292 7 (32) 6 (27) 9 (41) 0/20 (0)
46 ⫾ 2 25 ⫾ 3 98:24 46 (38) 507 ⫾ 37 3.8 ⫾ 0.4 3.1 ⫾ 0.1 2933 ⫾ 121 27 (22) 27 (22) 68 (56) 2/104 (2)
Numbers in parentheses are percentages.
DR3 and HLA DR4 were determined in each of the 122 patients. Statistical Analyses The Fisher exact probability test was used to compare dichomotous variables, and the unpaired t test was used to compare differences in the means of continuous variables. Nonparametric variables in independent samples were compared by the Mann-Whitney test. For the purposes of this study, only the frequencies of the known HLA susceptibility markers for type 1 autoimmune hepatitis were analyzed, including HLA DR3, HLA DR4, and their alleles. Because the variables for comparison had been formulated a priori and then assessed systematically in the study group, an unadjusted p of 0.05 was used to determine statistical significance. Data are presented as the mean ⫾ SEM in tables and text.
RESULTS Frequency and Clinical Associations of Anti-SLA/LPs Twenty-two patients (15%) were seropositive for anti-SLA/ LPs by ELISA and by Western blot, including seven who had SMAs only, six who had ANAs only, and nine who had both SMAs and ANAs at presentation (Table 1). Patients with anti-SLA/LPs had the same frequencies of SMAs (32% vs 22%, p ⫽ 0.4), ANAs (27% vs 22%, p ⫽ 0.6), or both SMAs and ANAs (41% vs 56%, p ⫽ 0.2) at presentation as patients without anti-SLA/LPs. Furthermore, none of the 20 tested patients with anti-SLA/LPs had antibodies to liver/ kidney microsome type 1. Individuals with anti-SLA/LPs were indistinguishable from seronegative patients with SMAs and/or ANAs by gender, duration of illness, frequency of concurrent immune diseases, laboratory findings, and occurrence of cirrhosis at presentation (Table 1). Frequencies of anti-SLA/LPs were similar in those serum spec-
Czaja et al.
AJG – Vol. 97, No. 2, 2002
Table 2. Association of Anti-SLA/LPs With HLA Risk Factors
Anti-SLA/LP Positive (N ⫽ 22)
Anti-SLA/LP Negative (N ⫽ 122)
DR3 DR3⫹/DR4⫺ DR4⫹ DR4⫹/DR3⫺ DR3⫹/DR4⫹ Homozygous DR3 Homozygous DR4 DRB1*0301 DRB1*0401 Non-DRB1*0401 alleles
15/19 (79)§ 14/19 (74)* 3/19 (16)㛳 2/19 (10)¶ 1/19 (5) 5/19 (26) 0/19 (0) 15/18 (83)† 2/18 (11)‡ 1/18 (6)
52/103 (50)§ 39/103 (38)* 48/103 (47)㛳 35/103 (34)¶ 13/103 (13) 8/103 (8) 4/103 (4) 42/89 (47)† 38/91 (42)‡ 7/91 (8)
Numbers in parentheses are percentages. Significantly different from each other at * p ⫽ 0.005, † p ⫽ 0.008, ‡ p ⫽ 0.01, §㛳 p ⫽ 0.02, and ¶ p ⫽ 0.05.
imens obtained at presentation (11%), during therapy (18%), and during follow-up after treatment (17%). Autoimmune thyroiditis (18% vs 14%, p ⫽ 0.5), inflammatory bowel disease (4% vs 7%, p ⬎ 0.9), and Graves’ disease (4% vs 3%, p ⫽ 0.6) occurred with similar frequencies in patients with and without anti-SLA/LPs. Vasculitis (one patient), pernicious anemia (one patient), and focal myositis (one patient) were present only in patients with anti-SLA/LPs. Synovitis (three patients), autoimmune hemolytic anemia (two patients), rheumatoid arthritis (two patients), systemic lupus erythematosus (two patients), peripheral neuritis (two patients), celiac sprue (one patient), pulmonary fibrosis (one patient), dermatitis herpetiformis (one patient), erythema nodosum (one patient), and idiopathic thrombocytopenic purpura (one patient) were present only in patients without anti-SLA/LPs. Importantly, the size of the cohort without anti-SLA/LPs who were at risk for immune disorders was 5.5-fold greater than the cohort with anti-SLA/LPs. Anti-SLA/LPs and HLA Risk Factors HLA DR3 occurred more commonly in patients with antiSLA/LPs than in patients without anti-SLA/LPs who had SMAs and/or ANAs at presentation (Table 2). Conversely, HLA DR4 occurred more commonly in patients without anti-SLA/LPs who had SMAs and/or ANAs at accession than in patients with anti-SLA/LPs. DRB1*0301 was detected in 83% of patients with anti-SLA/LPs who were tested for this allele, and DRB1*0401 was significantly less common in these patients than in those tested patients with SMAs and/or ANAs but not anti-SLA/LPs (Table 2). Other Autoantibodies and HLA Risk Factors Patients with anti-SLA/LPs had HLA DR3 more commonly than patients with ANAs only or both SMAs and ANAs at presentation (Table 3). They also had HLA DR4 less frequently than patients with both SMAs and ANAs at entry. DRB1*0301 was detected more frequently in patients with anti-SLA/LPs than in patients with ANAs alone or in conjunction with SMAs, whereas DRB1*0401 was less common in the anti-SLA/LP–positive patients (Table 3).
Anti-SLA/LPs and Other Autoantibodies as Prognostic Indices Patients with anti-SLA/LPs invariably relapsed after drug withdrawal (Table 3). The frequency of relapse was significantly higher in these patients than in patients with both SMAs and ANAs at presentation. Four patients with antiSLA/LPs died of liver failure before liver transplantation was an established option. The frequency of death from hepatic failure or requirement for liver transplantation was comparable in all groups. Follow-up was longer in the patients with anti-SLA/LPs, but their outcomes were similar (Table 4).
DISCUSSION Antibodies to SLA/LP are highly specific markers of autoimmune hepatitis (15, 16, 18 –21), and they can now be assessed by a standardized immunoassay based on recombinant antigen (18, 20, 21). Because anti-SLA/LPs are directed against a single antigen and they are associated with relapse after corticosteroid withdrawal, they may be more reflective of a pertinent pathogenic process than either SMAs or ANAs. Our findings indicate that seropositivity for anti-SLA/LPs identifies patients with type 1 autoimmune hepatitis who have HLA DR3 more commonly and HLA DR4 less often than patients with SMAs and/or ANAs at presentation. The principal allele associated with susceptibility and disease severity in type 1 autoimmune hepatitis is DRB1*0301 (24, 26, 35), and anti-SLA/LPs are associated with this allele. Conversely, anti-SLA/LPs have a negative association with DRB1*0401, which is a secondary but independent susceptibility allele and a determinant of good outcome (24 –26). Patients with anti-SLA/LPs relapsed more commonly after drug withdrawal than patients with SMAs and ANAs at accession, and this finding suggested that anti-SLA/LPs were better markers of prognosis than SMAs and ANAs, possibly because of their association with DRB1*0301 (35). DRB1*0301 and DRB1*0401 may influence disease occurrence and severity by affecting autoantigen presentation (29, 30, 36, 37) and/or by synergizing with other autoimmune promoters, such as polymorphisms for tumor necrosis factor ␣ (38, 39) and cytotoxic T-lymphocyte antigen 4 (40). Anti-SLA/LPs may be surrogate markers of these synergisms or reflective of the principal antigenic target. The SLA/LP antigen has sequence homology with a short eight–amino acid segment of human asialoglycoprotein receptor (ASGPR), and this segment forms part of a hydrophobic membrane-spanning region (41). Consequently, SLA/LP may be inserted into the hepatocyte membrane and be targeted by immunocytes. Furthermore, its homology with ASGPR may result in immunological cross-reactions through molecular mimicry. Liver injury may expose the SLA/LP peptide to immunocytes and escalate the inflammatory reaction by epitope spread. Lastly, microbial anti-
AJG – February, 2002
Anti-SLA/LP in Autoimmune Hepatitis
Table 3. HLA Associations With Anti-SLA/LPs, SMAs, and ANAs HLA
Anti-SLA/LPs (N ⫽ 22)
SMAs (N ⫽ 27)
ANAs (N ⫽ 27)
SMAs and ANAs (N ⫽ 68)
DR3⫹ DR3⫹/DR4⫺ DR4⫹ DR4⫹/DR3⫺ DR3⫹/DR4⫹ Homozygous DR3 Homozygous DR4 DRB1*0301 DRB1*0401 Non-DRB1*0401 alleles
15/19 (79)#¶ 14/19 (74)* 3/19 (16)† 2/19 (10)** 1/19 (5) 5/19 (26) 0/19 (0) 15/18 (83)§㛳 2/18 (11)‡ 1/18 (6)
12/21 (57) 10/21 (48) 8/21 (38) 6/21 (29) 2/21 (10) 1/21 (5) 1/21 (5) 9/17 (53) 6/17 (35) 1/17 (6)
11/24 (46)¶ 10/24 (42) 8/24 (33) 7/24 (29) 1/24 (4) 2/24 (8) 2/24 (8) 10/22 (45)㛳 7/22 (32) 1/22 (4)
29/58 (50)# 19/58 (33)* 32/58 (55)† 22/58 (38)** 10/58 (17) 5/58 (9) 1/58 (2) 23/50 (46)§ 25/52 (48)‡ 5/52 (10)
Numbers in parentheses are percentages. Significantly different from each other at *† p ⫽ 0.003, ‡ p ⫽ 0.005, § p ⫽ 0.01, 㛳p ⫽ 0.02, ¶# p ⫽ 0.03, and ** p ⫽ 0.04.
gens may trigger the disease and cross-react with the SLA/LP peptide (21, 41). The failure to detect anti-SLA/ LPs in all patients with type 1 autoimmune hepatitis may reflect an etiological difference between patients with and without these antibodies. Indeed, individuals with type 1 autoimmune hepatitis may share clinical features, but have distinctive etiologies and clinical behaviors (21, 41). The genetic predisposition for type 1 autoimmune hepatitis relates to the configuration of the antigen-binding groove of the class II major histocompatibility complex (MHC) encoded by the susceptibility alleles. DRB1*0301 and DRB1*0401 each encode the same amino acid sequence between positions 67 and 72 of the DR␤ polypeptide chain (24, 37). If the SLA/LP peptide is a prime target of the immune reaction, its configuration must favor presentation by class II MHC molecules encoded by DRB1*0301 and not DRB1*0401. There are 14 sites within the antigen-binding groove that are encoded differently by DRB1*0301 and DRB1*0401, and any one may be the region favoring presentation of this peptide (42). Multiple antigens are probably capable of triggering type 1 autoimmune hepatitis (29, 30), and SLA/LP may differ from the others by restriction of its presentation to class II MHC molecules encoded by the DRB1*0301 allele. Alternatively, SLA/LP may be involved in the processing and/or conversion of the principal autoantigen to an immunogenic peptide, and it may be this peptide that is optimally presented by class II MHC molecules encoded by the DRB1*0301 allele. Furthermore, SLA/LP may be exposed
during a disease process triggered by a particular autoantigen whose presentation is favored by class II MHC molecules encoded by the DRB1*0301 allele. In this instance, anti-SLA/LPs may be collateral manifestations of the immune response. As the pathogenic pathways of type 1 autoimmune hepatitis are clarified, transgenic animal models will be able to assess antigenic targets and develop sitespecific therapies (30, 43), such as T-cell vaccination (44) and oral tolerance (45). Only three other autoantibodies have had prognostic implications in autoimmune hepatitis, and two of these may be barometers of disease activity rather than true reflections of pathogenic mechanisms (7, 8). Antibodies to ASGPR identify patients prone to relapse after corticosteroid withdrawal, but their detection correlates with histological activity (46, 47). Consequently, their presence at the time of drug withdrawal may reflect inadequate disease suppression and a propensity for relapse due to residual hepatocellular inflammation. Similarly, antibodies to liver cytosol type 1 fluctuate with disease activity in patients with type 2 autoimmune hepatitis (48), and they characterize patients with more aggressive liver disease (49, 50). Speculation that these antibodies are associated with a pathogenic factor that influences disease severity is counterbalanced by clinical observations linking their production with the degree of hepatocellular injury (48 –50). Only antibodies to actin have a prognostic connotation independent of the conventional indices of inflammatory activity, and they may be truly reflective of a pathogenic process (7, 8, 14). Like anti-SLA/
Table 4. Treatment Outcomes Associated With Anti-SLAs, SMAs, and ANAs Outcome
Anti-SLA/LPs (N ⫽ 21)
SMAs (N ⫽ 27)
ANAs (N ⫽ 25)
SMAs and ANAs (N ⫽ 65)
Remission Relapse Sustained remission Treatment failure Incomplete response Death from hepatic failure Hepatic death or liver transplantation Duration follow-up (mo)
16 (76) 16/16 (100)‡ 0/16 (0)§ 3 (14) 2 (10) 4 (19) 4 (19) 173 ⫾ 24*†
19 (70) 15/19 (79) 4/19 (21) 3 (11) 5 (18) 1 (4) 8 (30) 144 ⫾ 18
18 (72) 14/18 (78) 4/18 (22) 3 (12) 4 (16) 1 (4) 2 (8) 88 ⫾ 14*
51 (78) 40/51 (78)‡ 11/51 (22)§ 10 (15) 4 (6) 3 (5) 8 (12) 123 ⫾ 10†
Numbers in parentheses are percentages. Significantly different from each other at * p ⫽ 0.003, † p ⫽ 0.03, and ‡§ p ⫽ 0.05.
Czaja et al.
LP, they also occur more commonly in HLA DR3–positive patients (14). Future studies must correlate anti-SLA/LPs with conventional laboratory and histological features of disease activity and, in this fashion, determine the basis for their association with relapse. Autoantibodies may fluctuate in level, disappear and reappear, or emerge later de novo in the course of type 1 autoimmune hepatitis (11). This variability is common with SMAs and ANAs, and similar vagaries of expression may be exhibited by anti-SLA/LPs. Previous studies, however, have indicated that the clinical expression of SMAs and ANAs is rarely suppressed fully by conventional treatment schedules (11), and our studies indicate that the anti-SLA/ LPs were detected as commonly before, during, and after therapy. Other investigations have also demonstrated that the frequency of anti-SLA/LPs is similar in different patient populations regardless of sampling conditions (20). These observations predict that anti-SLAs persist during the disease and its treatment. Future studies must establish this behavior. The associations between anti-SLA/LPs, HLA risk factors, and the class II MHC alleles were evident even when compared to patients without anti-SLA/LPs who were categorized by different patterns of SMAs and/or ANAs at presentation (Table 3). Similarly, anti-SLA/LPs were invariably associated with relapse after drug withdrawal, in contrast to seronegative patients with other patterns of the conventional markers at accession (Table 4). These observations underscored the distinctive features of anti-SLA/LPs and indicated that their clinical connotations could not be matched by the classical markers in any pattern. SMAs and/or ANAs do not define different clinical subgroups of type 1 autoimmune hepatitis, and their patterns of seropositivity can change during the course of the disease (11). Our study precluded the existence of a pattern of conventional markers at presentation associated with greater immunoreactivity and prognostic value than anti-SLA/LPs. The target autoantigen of anti-SLA/LP is a 50-kd cytosolic protein (20, 21). Immunoprecipitation experiments suggest that it colocalizes with the selenocysteine-specific transfer factor (tRNP(Ser)Sec) (51). Antibodies to tRNP(Ser)Sec have been associated with severe autoimmune hepatitis and poor outcome (52). In this respect, antibodies to tRNP(Ser)Sec bear a resemblance to anti-SLA/LPs. Future studies will need to clarify the exact function of the SLA/LP protein and detail similarities/differences between antibodies to tRNP(Ser)Sec and anti-SLA/LPs. In summary, antibodies to SLA/LP are associated with HLA DR3 and the susceptibility allele, DRB1*0301, in contrast to SMAs and ANAs. The association of anti-SLA/ LPs with relapse after corticosteroid withdrawal may be reflective of this allelic association. Antibodies to SLA/LP may be surrogate markers of a genetic propensity for recrudescent disease, the target autoantigen, and/or nonspecific autoimmune promoters. They have genetic and prognostic connotations that differ from those of the conventional
AJG – Vol. 97, No. 2, 2002
markers, and they may be complementary to these markers in the clinical assessment of patients with type 1 autoimmune hepatitis.
ACKNOWLEDGMENTS Maren Baeres is a doctoral student at Johannes Gutenberg University, Mainz, Germany, and she is acknowledged for testing our sera for antibodies to soluble liver antigen/liverpancreas. Linda Grande at the Mayo Clinic, Rochester, MN, is acknowledged for providing secretarial assistance. Reprint requests and correspondence: Albert J. Czaja, M.D., F.A.C.G., Mayo Clinic, 200 First Street Southwest, Rochester, MN 55905. Received Mar. 8, 2001; accepted June 27, 2001.
REFERENCES 1. Whittingham S, McNeilage LJ. Antinuclear antibodies as molecular and diagnostic probes. Mol Cell Probes 1988;2:1169 – 79. 2. Tan EM. Antinuclear antibodies. Diagnostic markers for autoimmune diseases and probes for cell biology. Adv Immunol 1989;44:93–151. 3. Whittingham S, Irwin J, Mackay IR, et al. Smooth muscle autoantibody in “autoimmune” hepatitis. Gastroenterology 1966;51:499 –505. 4. Bottazzo GF, Florin-Christensen A, Fairfax A, et al. Classification of smooth muscle autoantibodies (SMA) detected by immunofluorescence. J Clin Pathol 1976;29:403–10. 5. Toh B-H. Smooth muscle autoantibody and autoantigens. Clin Exp Immunol 1979;38:621– 8. 6. Czaja AJ, Manns MP. The validity and importance of subtypes of autoimmune hepatitis: A point of view. Am J Gastroenterol 1995;90:1206 –11. 7. Czaja AJ. Autoantibodies. Baillieres Clin Gastroenterol 1995; 9:723– 44. 8. Czaja AJ, Homburger HA. Autoantibodies in liver disease. Gastroenterology 2001;120:239 – 49. 9. Czaja AJ, Nishioka M, Morshed SA, et al. Patterns of nuclear immunofluorescence and reactivities to recombinant nuclear antigens in autoimmune hepatitis. Gastroenterology 1994;107: 200 –7. 10. Parveen S, Morshed SA, Arima K, et al. Antibodies to Ro/La, Cenp-B, and snRNPs antigens in autoimmune hepatitis of North America versus Asia. Dig Dis Sci 1998;43:1322–31. 11. Czaja AJ. Behavior and significance of autoantibodies in type 1 autoimmune hepatitis. J Hepatol 1999;30:394 – 401. 12. Fusconi M, Cassani F, Zauli D, et al. Anti-actin antibodies: A new test for an old problem. J Immunol Methods 1990;130: 1– 8. 13. Cancado ELR, Vilas-Boas LS, Abrantes-Lemos CP, et al. Heat serum inactivation as a mandatory procedure for antiactin antibody detection in cell culture. Hepatology 1996;23:1098 – 104. 14. Czaja AJ, Cassani F, Cataleta M, et al. Frequency and significance of antibodies to actin in type 1 autoimmune hepatitis. Hepatology 1996;24:1068 –73. 15. Manns M, Gerken G, Kyriatsoulis A, et al. Characterization of a new subgroup of autoimmune chronic active hepatitis by autoantibodies against a soluble liver antigen. Lancet 1987;1: 292– 4.
AJG – February, 2002
16. Stechemesser E, Klein R, Berg PA. Characterization and clinical relevance of liver-pancreas antibodies in autoimmune hepatitis. Hepatology 1993;18:1–9. 17. Klein R, Berg PA. Glutathione S-transferase as a major autoantigen in anti-SLA-positive autoimmune hepatitis. Gastroenterology 1999;116:1015–16 (letter to editor). 18. Kanzler S, Weidemann C, Gerken G, et al. Clinical significance of antibodies to soluble liver antigen in autoimmune hepatitis. J Hepatol 1999;31:635– 40. 19. Czaja AJ, Carpenter HA, Manns MP. Antibodies to soluble liver antigen, P450IID6, and mitochondrial complexes in chronic hepatitis. Gastroenterology 1993;105:1522– 8. 20. Baeres M, Wies I, Kanzler S, et al. Establishment of a standardized SLA/LP immunoassay: SLA/LP positive autoimmune hepatitis occurs worldwide. Hepatology 2000;32:166A (abstract). 21. Wies I, Brunner S, Henninger J, et al. Identification of target antigen for SLA/LP autoantibodies in autoimmune hepatitis. Lancet 2000;355:1510 –5. 22. Donaldson PT, Doherty DG, Hayllar KM, et al. Susceptibility to autoimmune chronic active hepatitis: Human leukocyte antigens DR4 and A1-B8-DR3 are independent risk factors. Hepatology 1991;13:701– 6. 23. Doherty DG, Donaldson PT, Underhill JA, et al. Allelic sequence variation in the HLA class II genes and proteins in patients with autoimmune hepatitis. Hepatology 1994;19:609 – 15. 24. Strettell MDJ, Donaldson PT, Thomson LJ, et al. Allelic basis for HLA-encoded susceptibility to type 1 autoimmune hepatitis. Gastroenterology 1997;112:2028 –35. 25. Czaja AJ, Carpenter HA, Santrach PJ, et al. Significance of HLA DR4 in type 1 autoimmune hepatitis. Gastroenterology 1993;105:1502–7. 26. Czaja AJ, Strettell MDJ, Thomson LJ, et al. Associations between alleles of the major histocompatibility complex and type 1 autoimmune hepatitis. Hepatology 1997;25:317–23. 27. Griffing WL, Moore SB, Luthra HS, et al. Associations of antibodies to native DNA with HLA-DRw3: A possible major histocompatibility complex-linked human immune response gene. J Exp Med 1980;152:319s–25s. 28. Czaja AJ, Carpenter HA, Santrach PJ, et al. Genetic predispositions for the immunological features of chronic active hepatitis. Hepatology 1993;18:816 –22. 29. Czaja AJ. Immunopathogenesis of autoimmune-mediated liver damage. In: Moreno-Otero R, Clemente-Ricote G, GarciaMonzon C, eds. Immunology and the liver: Autoimmunity. Madrid: Aran Ediciones, 2000:73– 83. 30. Czaja AJ. Understanding the pathogenesis of autoimmune hepatitis. Am J Gastroenterol 2001;96:1224 –31. 31. Johnson PJ, McFarlane IG, Alvarez F, et al. Meeting report. International Autoimmune Hepatitis Group. Hepatology 1993; 18:998 –1005. 32. Alvarez F, Berg PA, Bianchi FB, et al. International Autoimmune Hepatitis Group report: Review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;31:929 –38. 33. Czaja AJ, Manns MP, Homburger HA. Frequency and significance of antibodies to liver/kidney microsome type 1 in adults with chronic active hepatitis. Gastroenterology 1992;103: 1290 –5. 34. Summerskill WHJ, Korman MG, Ammon HV, et al. Prednisone for chronic active liver disease: Dose titration, standard
Anti-SLA/LP in Autoimmune Hepatitis
36. 37. 38.
39. 40. 41. 42. 43. 44.
45. 46. 47. 48.
49. 50. 51.
dose, and combination with azathioprine compared. Gut 1975; 16:876 – 83. Czaja AJ, Rakela J, Hay JE, et al. Clinical and prognostic implications of human leukocyte antigen B8 in corticosteroidtreated severe autoimmune chronic active hepatitis. Gastroenterology 1990;98:1587–93. McFarlane IG. Pathogenesis of autoimmune hepatitis. Biomed Pharmacother 1999;53:255– 63. Czaja AJ, Donaldson PT. Genetic susceptibilities for immune expression and liver cell injury in autoimmune hepatitis. Immunol Rev 2000;174:250 –9. Czaja AJ, Cookson S, Constantini PK, et al. Cytokine polymorphisms associated with clinical features and treatment outcome in type 1 autoimmune hepatitis. Gastroenterology 1999;117:645–52. Cookson S, Constantini PK, Clare M, et al. Frequency and nature of cytokine gene polymorphisms in type 1 autoimmune hepatitis. Hepatology 1999;30:851– 6. Agarwal K, Czaja AJ, Jones DEJ, et al. CTLA-4 gene polymorphism and susceptibility to type 1 autoimmune hepatitis. Hepatology 2000;31:49 –53. McFarlane IG. Lessons about antibodies in autoimmune hepatitis. Lancet 355:2000;1475– 6 (editorial). Stern LJ, Brown JH, Jardetzky TS, et al. Crystal structure of the human class II MHC protein HLA-DR1 complexed with an influenza virus peptide. Nature 1994;368:215–21. Czaja AJ, Manns MP, McFarlane IG, et al. Autoimmune hepatitis: The investigational and clinical challenges. Hepatology 2000;31:1194 –200. Lohse AW, Dienes HP, Meyer zum Buschenfelde K-H. Suppression of murine experimental autoimmune hepatitis by Tcell vaccination or immunosuppression. Hepatology 1998;27: 1536 – 43. Wardrop RM, Whitacre CC. Oral tolerance in the treatment of inflammatory autoimmune diseases. Inflamm Res 1999;48: 106 –19. McFarlane IG, Hegarty JE, McSorley CG, et al. Antibodies to liver-specific protein predict outcome of treatment withdrawal in autoimmune chronic active hepatitis. Lancet 1984;2:954 – 6. Czaja AJ, Pfeifer KD, Decker RH, et al. Frequency and significance of antibodies to asialoglycoprotein receptor in type 1 autoimmune hepatitis. Dig Dis Sci 1996;41:1733– 40. Muratori L, Cataleta M, Muratori P, et al. Liver/kidney microsomal antibody type 1 and liver cytosol antibody type 1 concentrations in type 2 autoimmune hepatitis. Gut 1998;42: 721– 6. Martini E, Abuaf N, Cavalli F, et al. Antibody to liver cytosol (anti-LC1) in patients with autoimmune chronic active hepatitis type 2. Hepatology 1988;8:1662– 6. Abuaf N, Johanet C, Chretien P, et al. Characterization of the liver cytosol antigen type 1 reacting with autoantibodies in chronic active hepatitis. Hepatology 1992;16:892– 8. Costa M, Rodriques-Sanchez JL, Czaja AJ, et al. Isolation and characterization of cDNA encoding the antigenic protein of the human tRNA(Ser)Sec complex recognized by autoantibodies from patients with type 1 autoimmune hepatitis. Clin Exp Immunol 2000;121:364 –74. Gelpi C, Sontheimer EJ, Rodriguez-Sanchez JL. Autoantibodies against a serine tRNA-protein complex implicated in cotranslational selenocysteine insertion. Proc Natl Acad Sci U S A 1992;89:9739 – 43.