In a previous prospective study on the diagnostic value of serological markers in patients with indeterminate colitis (IC), we demonstrated that those who remained classified as IC during prospective follow up were more often negative for anti‐Saccharomyces cerevisiae antibodies (ASCA) and atypical perinuclear antineutrophil cytoplasmic antibodies (pANCA) compared with patients with an eventual diagnosis of Crohn's disease (CD) or ulcerative colitis (UC).1 New antibodies directed against luminal bacterial components have been reported in inflammatory bowel disease. Anti‐OmpC IgA (directed against outer membrane porin C of E coli) and anti‐I2 (directed against Pseudomonas fluorescens) have both been found in approximately 50% of CD patients2,3,4 and are both associated with ileal disease, increasing disease duration, and more severe disease that requires small bowel surgery.5,6
Table1 Marker combinations in the study population.
| n | CD | UC | IC | |
|---|---|---|---|---|
| Four positive markers | ||||
| pANCA+/ASCA+/OmpC+/I2+ | 1 | 1 | 0 | 0 |
| Three positive markers | ||||
| pANCA+/ASCA+/OmpC+/I2− | 1 | 0 | 0 | 1 |
| pANCA−/ASCA+/OmpC+/I2+ | 3 | 1 | 0 | 2 |
| pANCA+/ASCA−/OmpC+/I2+ | 3 | 1 | 0 | 2 |
| Total | 7 | 2 | 0 | 5 |
| Two positive markers | ||||
| pANCA+/ASCA−/OmpC‐/I2+ | 3 | 1 | 0 | 2 |
| pANCA−/ASCA+/OmpC‐/I2+ | 8 | 2 | 2 | 4 |
| pANCA+/ASCA+/OmpC−/I2− | 2 | 1 | 1 | 0 |
| pANCA−/ASCA−/OmpC+/I2+ | 6 | 1 | 1 | 4 |
| pANCA−/ASCA+/OmpC+/I2− | 1 | 1 | 0 | 0 |
| pANCA+/ASCA−/OmpC+/I2− | 1 | 0 | 1 | 0 |
| Total | 21 | 6 | 5 | 10 |
| One positive marker | ||||
| pANCA+/ASCA−/OmpC−/I2− | 12 | 4 | 6 | 2 |
| pANCA−/ASCA−/OmpC−/I2+ | 12 | 1 | 2 | 9 |
| pANCA−/ASCA+/OmpC−/I2− | 11 | 3 | 2 | 6 |
| Total | 35 | 8 | 10 | 17 |
| No positive markers | ||||
| pANCA−/ASCA−/OmpC−/I2− | 23 | 4 | 2 | 17 |
| Total | 87 | 21 | 17 | 49 |
IC, indeterminate colitis; CD, Crohn's disease; UC, ulcerative colitis; pANCA, perinuclear antineutrophil cytoplasmic antibodies; ASCA, anti Saccharomyces cerevisiae antibodies; anti‐OmpC, antibodies against outer membrane porin C of E coli; anti‐I2, antibodies against Pseudomonas fluorescens.
We investigated if anti‐OmpC and anti‐I2 were additive to ASCA and pANCA in the definitive diagnosis of patients who were included in our initial cohort of IC, and if patients who remained unclassified over time also lacked response to these microbial antigens. For additional testing of anti‐OmpC and anti‐I2, the serum of 93/97 patients included in the initial cohort was available. Further clinical evolution was possible in 90/97 patients and seven were lost to follow up. Additionally, serum from 93 healthy individuals (female/male 54/39; mean age 41.8 years (range 24–77)) and 64 inflammatory controls (female/male 29/35; mean age 57.8 years (range 20–92)), including infectious gastroenteritis (n = 28), diverticulitis (n = 23), ischaemic colitis (n = 4), acute self limiting colitis (n = 1), pseudomembraneous colitis (n = 2), and other non‐specific colitis (n = 6), were tested for anti‐I2 and anti‐OmpC to investigate the specificity of these markers.
Criteria for diagnosis of IC in the present study as well as criteria for change in diagnosis to CD or UC during further clinical follow up have been described in detail previously.1 Recently, however, a proposal towards a novel classification of IC has been made.7 This should be taken into account for future studies in IC. Both anti‐OmpC and anti‐I2 were determined by an ELISA assay. For anti‐OmpC antibody determination, samples were sent to Prometheus Laboratories Inc. (San Diego, California, USA).8 For anti‐I2 antibody determination, samples were sent to Dr Carol Landers (Los Angeles, California, USA).3
Mean disease duration for patients was 11.4 years (range 2.5–40.5). After 2.5 years of prospective follow up, disease in 41/90 (45.6%) patients has evolved towards either CD (n = 23) or UC (n = 18) after a mean disease duration of 5.3 years (range 1 month–26 years). Additional follow up of 1.5 years resulted in six more patients in whom the diagnosis was changed to CD and four more patients to UC. The remaining 49 (54.4%) patients remained classified as IC (mean disease duration 13.9 years (range 2.5–40.5)).
With the four markers tested, various combinations of responses were found (see table 1).
Calculations were made based on those patients in whom clinical as well as serological data were complete (n = 87). These 87 patients represented 21 CD, 17 UC, and 49 IC patients. Of all marker combinations, pANCA−/ASCA−/anti‐OmpC−/anti‐I2− was most frequently present in 23 of 87 (26.4%) patients. In 17/23 (73.9%) seronegative patients, the diagnosis remained IC. Only six of 23 (26.1%) patients in this seronegative group were given a final diagnosis of CD (n = 4) or UC (n = 2). The proportion of patients with at least one antibody response present that evolved towards a specific diagnosis was significantly higher: 32/64 (50%) patients (p = 0.047). The prevalence of anti‐I2 in the IC cohort, healthy controls, and inflammatory controls was 41.9% (39/93), 17.2% (16/93), and 31.3% (20/64), respectively. Consequently, the respective sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of anti‐I2 in the IC cohort were 41.9%, 76.4%, 48.1%, and 71.6%. The prevalence of anti‐OmpC in this IC cohort, healthy controls, and inflammatory controls was 17.2% (16/93), 2.2% (2/93), and 25% (16/64), respectively. Respective sensitivity, specificity, PPV, and NPV of anti‐OmpC in patients with IC were 17.2%, 88.5%, 47.1% and 64.4%. The prevalence of anti‐I2 and anti‐OmpC in this study cohort was lower than the reported prevalence of these markers in CD.2,3,6 This is not unexpected as both markers are more often found in the presence of ileal CD. However, a large proportion of controls, especially inflammatory controls, also expressed anti‐OmpC and anti‐I2 antibodies.
The results of the present study indicate that by adding the antibodies anti‐OmpC and anti‐I2, the predictive capacity of serological tests only increases marginally and specificity drops significantly. More importantly, our results confirm the presence of a large group of IC patients who are negative for serological markers and may represent a separate phenotype.
Conflict of interest: declared (the declaration can be viewed on the Gut website at http://www.gutjnl.com/supplemental).
Supplementary Material
Footnotes
Conflict of interest: declared (the declaration can be viewed on the Gut website at http://www.gutjnl.com/supplemental).
References
- 1.Joossens S, Reinisch W, Vermeire S.et al The value of anti‐Saccharomyces cerevisiae (ASCA) and perinuclear anti‐neutrophil cytoplasmic antibodies (pANCA) in indeterminate colitis (IC): a prospective follow‐up study. Gastoenterology 20021221242–1247. [DOI] [PubMed] [Google Scholar]
- 2.Landers C, Cohavy O, Misra R.et al Selected loss of tolerance evidenced by Crohn's disease‐associated immune responses to auto‐ and microbial antigens. Gastroenterology 2002123689–699. [DOI] [PubMed] [Google Scholar]
- 3.Sutton C L, Kim J, Yamane A.et al Identification of a novel bacterial sequence associated with Crohn's disease. Gastroenterology 200011923–31. [DOI] [PubMed] [Google Scholar]
- 4.Wei B, Huang T, Dalwali H.et al Pseudomonas fluorescens encodes the Crohn's disease‐associated I2 sequence and T‐cell superantigen. Infect Immun 2002706567–6575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mow W, Vasiliauskas E A, Lin Y C.et al Association of antibody responses to microbial antigens and complications of small bowel Crohn's disease. Gastroenterology 2004126414–424. [DOI] [PubMed] [Google Scholar]
- 6.Arnott I, Landers C, Nimmo E.et al Sero‐reactivity to microbial components in Crohn's disease is associated with disease severity and progression, but not NOD2/CARD15 genotype. Am J Gastroenterol 2004992376–2384. [DOI] [PubMed] [Google Scholar]
- 7.Silverberg M, Satsangi J, Ahmad T.et al Towards an integrated clinical, molecular and serological classification of inflammatory bowel disease: Report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005195–36A. [DOI] [PubMed] [Google Scholar]
- 8.Cohavy O, Bruckner D, Gordon L.et al Colonic bacteria express an ulcerative colitis pANCA‐related protein epitope. Infect Immun 2000681542–1548. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.