The Utility of ANCA and ASCA in Inflammatory Bowel Disease

Serum autoantibody detection has been used for diagnosis and stratification of immune-mediated diseases for many years. Tests to determine the presence of such autoantibodies are indicated for screening and confirmation of diagnoses for patients with symptoms suggestive of one or more maladies. Autoantibodies are also useful markers for differentiating among various forms of diseases, particularly those with multiple or complex clinical presentations. In addition, determining the presence and levels of certain antibodies may be helpful for prognosticating the clinical course of a disease. The usefulness of autoantibody assessment is illustrated by the example of systemic lupus erythematosus (SLE). The presence of antinuclear antibodies (ANA) has commonly been used in the diagnosis of symptoms suggestive of SLE. The currently available detection method is highly sensitive although the specificity is relatively poor. Patients with ANA positive serum are generally tested further to determine the presence of antibodies to double-stranded DNA to confirm that the ANA detected is specific to SLE. Despite the high sensitivity of ANA, a proportion of SLE patients is ANA serum negative. Juvenile rheumatoid arthritis (JRA) is illustrative of a disease in which autoantibody expression is correlated with variant clinical presentations. The expression of the autoantibody rheumatoid factor (RF) is associated with polyarticular disease. By contrast, ANA-positive JRA patients have a characteristically pauciarticular presentation. Patients with Still's disease express neither ANA nor RF, and do not correlate with any of the known HLA markers of JRA. Similar concepts are being applied to ulcerative colitis (UC) and Crohn's disease (CD), in which cases autoantibody profiles appear to be indicative of distinct inflammatory mechanisms (1). While the goal of early studies may have been to differentiate between UC and CD, and perhaps to define the group of patients with “indeterminate colitis,” further investigation revealed that these markers are likely to be indicative of different cytokine profiles and, therefore, different types of inflammation. While the classical descriptors are still useful, the technology is now available that may lead to specifically targeted immunotherapy for the various inflammatory processes, regardless of traditional clinical classification. The search for autoantibodies that correlate with the inflammatory bowel diseases dates back over 30 years; however, it has only been in the last 10 years that disease-specific markers such as those in rheumatological disorders have begun to emerge. Early reports in inflammatory bowel disease included an association between a subtype of antineutrophil cytoplasmic antibodies (ANCA) and patients with ulcerative colitis. This subtype of ANCA, known as pANCA, does not react to the same antigens as the ANCA associated with Wegener's granulomatosis and other vasculitides. Refinements in technology have led to the discovery that the UC-pANCA reactive antigen is Histone H1. At present, the most sensitive and specific method for detection of ANCA in serum is by a combination of fixed neutrophil enzyme-linked immunosorbent assay (ELISA), in combination with immunofluorescent microscopy and confirmation of loss of the perinuclear pattern following DNase digestion of neutrophils. UC-specific pANCA is detected in the sera of 60% to 70% of patients with UC. Interestingly, pANCA has also been found in 15% of patients diagnosed with CD. Further evaluation revealed that the CD patients with serum pANCA represent a disease subgroup with distinct clinical and subclinical characteristics (2). CD patients with serum p-ANCA have clinical features of left-sided colitis with endoscopic and/or histopathologic features of UC. The label given to this group of patience, for the purposes of reference is “ulcerative colitis-like,” or “UC-like” (2). Recently, considerable attention has been focused on a second autoantibody, antisaccharomyces cerevisiae (ASCA). ASCA, as a serum marker of CD, was first reported in the late 1980s (3,4). Using the originally described ELISA detection method, ASCA was demonstrated in the serum of 50-60% of patients with CD (5,6). Optimization of the ELISA technology, coupled with measurement of immunoglobulins (Ig) G and A, has enhanced the sensitivity and specificity of ASCA as a serum marker for CD to 67% and 92%, respectively. Ongoing study has contributed evidence in support of using the combined results of these two assays as a screen for inflammatory bowel disease (IBD), and as a means for differentiating among the various forms of the disease. Recent studies have demonstrated not only improved sensitivity, but also high specificity of these tests for diagnosing UC and CD. For screening patients presenting with a myriad of symptoms common to a large number of various gastrointestinal disorders, the results of ANCA ELISA and ASCA ELISA, optimized to measure IgG and IgA, are combined and interpreted using a lower threshold for positivity. The sensitivity of this method is >90% among patients with known IBD. The increased sensitivity, however, was accompanied by a decrease in specificity. Thus, a positive result for IBD, using the combination of assays indicates subsequent analysis using ANCA and ASCA ELISAs with a higher threshold for positive (as established for the individual assays), followed by the confirmatory steps described above. The combined assay approach with positive results followed by secondary, more specific assays has been studied prospectively in children presenting with symptoms suggestive of IBD. The ANCA and ASCA combination as a screen was 80% sensitive in this pediatric population and >90% specific when positive screening was followed by the confirmatory tests. For the purposes of comparison, the sensitivity of combined ANCA and ASCA screening is far superior to that of rheumatoid factor (15-20% sensitive) or ANA (24-66% sensitive) for JRA. These preliminary results from studies of combined ANCA and ASCA detection are very supportive of the use of this method for screening patients who present with symptoms of CD or UC with these combinations of tests. Autoantibodies have been used to predict disease course among patients with rheumatoid arthritis. A high titer of rheumatoid factor in conjunction with expression of the homozygote HLA-DRB 1*0401/0404 MHC class II allele is associated with very aggressive, nodular, erosive joint disease (7). Similarly, a recent study demonstrated that the combination of ASCA and ANCA evaluations is predictive of certain manifestations of CD (1). The serum expression of both IgG and IgA, particularly at higher levels, is associated with small bowel involvement as compared to colonic disease. The expression of high levels of DNasesensitive pANCA is associated with colonic disease and other UC-like features. Examinations of homogeneous autoantibody expression and the magnitude of that expression have shown that these parameters further stratify subgroups of disease. The expression of higher levels of both IgG and IgA ASCA and the absence of pANCA is associated with fibrostenotic and perforating disease. In contrast, higher levels of DNase-sensitive pANCA and the absence of ASCA expression is associated with patients who do not have fibrostenotic or fistulizing disease. Among patients with small bowel disease, those with high levels of IgG and IgA ASCA and no pANCA had an increased frequency and number of surgeries as compared to patients with high levels of pANCA and no ASCA has been demonstrated. The year 1999 should see accelerated use of these autoantibodies in the diagnosis and management of patients with IBD. Using the lower threshold for positive, ASCA and ANCA ELISAs detected 90% of diagnosed IBD patients. To generate specificity, positive results from the initial screening should be followed up by the more specific tests. This approach appears to have value in both pediatrics as well as adult populations. These tests can be used as a first line screen for patients who present with symptoms that could be due to IBD and therefore define those patients who would benefit by further evaluation in this vein. This approach will need to be confirmed in prospective and economic analyses of these tests. Employing these tests to predict the natural history of CD and assist in differentiating CD from UC is warranted. In the not-too-distant future, combinations of these tests in association with genetic markers may determine patient response to new therapies. This suggestion is based on the hypothesis that expression of a particular autoantibody is reflective of a specific mucosal cytokine profile. Although this latter hypothesis needs to be directly proven, these markers may well be useful as guide lines for treatment with newer therapeutics such as anti-TNF-α, etc. As antibody reactivity to more specific and potentially cross-reacting antigens is identified, it may be possible to further stratify groups of patients by disease aggressiveness and therapeutic responses. As we enter the next century, this technology will provide a useful diagnostic and prognostic platform that likely will evolve and become even more useful in the future.