Pancreatic Autoantibodies and Autoantibodies Against Goblet Cells in Pediatric Patients With Inflammatory Bowel Disease

Kovacs, Marta*; Lakatos, Peter Laszlo; Papp, Maria; Jacobsen, Silvia§; Nemes, Eva||; Polgar, Marianne; Solyom, Eniko#; Bodi, Piroska**; Horvath, Agnes††; Muller, Katalin Eszter‡‡; Molnar, Kriszta‡‡; Szabo, Doloresz‡‡; Cseh, Aron‡‡; Dezsofi, Antal‡‡; Arato, Andras‡‡; Veres, Gabor‡‡

Journal of Pediatric Gastroenterology & Nutrition:
doi: 10.1097/MPG.0b013e318256b516

Background: Significance of pancreatic autoantibodies determined by using exocrine pancreas (PAB) and antibodies against recombinant pancreas antigen (rPAB), as well as the importance of autoantibodies against goblet cells (GAB), is not known in pediatric patients with inflammatory bowel disease (IBD). Our aim was to determine the complex analysis of PAB, rPAB, GAB, antibodies against Saccharomyces cerevisiae, and perinuclear components of neutrophils in pediatric patients with IBD. Moreover, association with NOD2/CARD15 and disease phenotype was determined.

Methods: A total of 152 pediatric patients (median age 13.9 years) with IBD (103 patients with Crohn disease [CD] and 49 patients with ulcerative colitis [UC]) and 104 controls were included. Serum autoantibodies were determined by indirect immunofluorescence assay. NOD2/CARD15 variants were tested by polymerase chain reaction/restriction fragment length polymorphism.

Results: The presence of PAB and rPAB was significantly higher in CD (34% and 35.9%) and in UC (20.4% and 24.5%) compared with pediatric control cohort (0% and 0%, P < 0.0001). In addition, GAB positivity was significantly increased in patients with UC in comparison with CD and controls, respectively (UC, 12.2%; CD, 1.9%; controls, 1.9%; P = 0.02). Specificity of PAB and rPAB was 100%; however, sensitivity was low. The combination of PAB and/or antibodies against Saccharomyces cerevisiae/perinuclear components of neutrophils improved the sensitivity of serological markers in CD (87.4%) and in UC (79.6%); specificities were 89.3% and 93.2%, respectively. Pancreatic autoantibodies (PAB, rPAB) and GAB were not related to clinical presentation, medical therapy, or need for surgery in CD or in UC.

Conclusions: Pancreatic autoantibodies and GAB were specific for IBD, but the sensitivity was limited as well because there was lack of correlation with clinical phenotype. Combinations of these antibodies have shown increased sensitivity; therefore, it may be recommended in the diagnostic procedure of IBD.

Author Information

*Department of Pediatrics, Petz Aladár County and Teaching Hospital, Győr

1st Department of Medicine, Semmelweis University, Budapest

2nd Department of Medicine, Medical and Health Science Center, University of Debrecen, Debrecen, Hungary

§Euroimmun Medizinische Labordiagnostika AG, Luebeck, Germany

||Department of Pediatrics, Medical and Health Science Center, University of Debrecen, Debrecen

Madarász Children's Hospital, Budapest

#Pediatric Health Centre, Borsod-A-Z County and University Teaching Hospital, Miskolc

**Department of Pediatrics, Pándy Kálmán Hospital, Gyula

††Department of Pediatrics, Cholnoky Ferenc County Hospital, Veszprém

‡‡1st Department of Pediatrics, Semmelweis University, Budapest, Hungary.

Address correspondence and reprint requests to Gabor Veres, MD, PhD, 1st Department of Pediatrics, Semmelweis University, 53 Bókay St, 1083 Budapest, Hungary (e-mail:

Received 1 March, 2012

Accepted 20 March, 2012

G.V. holds the János Bolyai Research Grant; this article was supported by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences.

The authors report no conflicts of interest.

Article Outline

Inflammatory bowel diseases (IBDs), Crohn disease (CD), and ulcerative colitis (UC) are chronic relapsing and remitting disorders of the gastrointestinal tract. The pathogenesis of IBD is complex and multifactorial. Present evidence suggests that IBD results from an aberrant immune response and loss of tolerance to the normal intestinal flora, leading to chronic inflammation of the gut in a genetically susceptible host (1). This hypothesis is supported by the occurrence of antibodies directed to microbial antigens and by the identification of NOD2/CARD15 as susceptibility genes to CD (2). Besides genetic predisposition and environmental factors, autoimmune mechanisms are suggested to play a vital part in the pathogenesis of IBD. The search for the underlying trigger of the abnormal intestinal inflammation characteristics of IBD has led to the discovery of antibodies present specifically in the blood of patients with CD and/or UC. Several autoantibodies have been described in IBD.

Relatively low prevalence of exocrine pancreas antibodies (PAB) was detected in adult patients with CD (27%–39%) using indirect immunofluorescence (IIF) (3,4); however, increased prevalence of PAB has been found in unaffected first-degree relatives of patients with IBD, suggesting a genetic origin of these antibodies (5). The determination of autoantibodies against exocrine pancreas by IIF using human cells transfected with the recently identified proteoglycans CUZD1 and GP2 as recombinant target antigens (recombinant pancreas antigen 1 and 2: rPAg1 and rPAg2) represents a new dimension in the serological diagnosis of IBD (6,7). The studies conducted in adult patients provided conflicting data regarding association between PAB and CD phenotype (4,5,8,9).

Autoantibodies against intestinal goblet cells (GAB) have previously been described in UC with a prevalence of 28% to 30% and in 20% of first-degree relatives to patients with IBD (3,10). Recent studies, however, suggested a much lower prevalence in both diseases (4,11). Moreover, PAB, antibodies against recombinant pancreas antigen (rPAB), and GAB have not been examined in a large cohort of pediatric IBD, and data on the specificity and sensitivity are contradictory in adult IBD.

The 2 most intensively studied conservative antibodies are autoantibodies to neutrophils (perinuclear anti-neutrophil cytoplasmic antibodies [pANCA]), primarily associated with UC, and anti-Saccharomyces cerevisiae antibodies (ASCA), primarily associated with CD (2,12). In pediatric IBD, the sensitivity/specificity of pANCA in UC ranged between 57% to 83% and 65% to 97%, respectively, whereas in CD, ASCA showed a sensitivity/specificity in the range of 44% to 76% and 88% to 95%, respectively (13,14). ASCA positivity or high titers are associated with a complicated disease behavior (penetrating or stenosing disease) and could be useful markers for predicting need for surgery also in pediatric patients (15–17). pANCA is noted for its association with “UC-like” CD phenotype (18,19).

The clinical importance of PAB and GAB is understudied in pediatric patients with IBD. The aim of our study was to determine the prevalence of PAB, rPAB, GAB, ASCA, and pANCA in patients with pediatric-onset IBD. In addition, we assessed the association between antibody profile and NOD2/CARD15 status, clinical presentation, response to treatment, and extraintestinal manifestations (EIMs).

Back to Top | Article Outline



Our study included 103 consecutive patients with pediatric-onset CD (male/female [m/f] ratio 63/40, median age 13.9 years [range 5.3–19.6 years]), 49 patients with pediatric-onset UC (m/f ratio 22/27, median age 12.5 years [range 6–19.7 years]), and 104 age- and sex-matched controls. The diagnosis of IBD was based on clinical, radiologic, endoscopic, and histological criteria. Disease activity was evaluated according to the Pediatric Crohn Disease Activity Index in children with CD and according to the Pediatric Ulcerative Colitis Index for the UC group (20,21). Activity index >30 is defined as a moderate to severe disease, the index between 11 and 30 indicates a mild disease, and the index ≤10 refers to an inactive disease.

Age, age at onset, presence of EIM, arthritis, ocular manifestations, skin lesions, and hepatic manifestations, frequency of flare-ups (frequent flare-up >1/year), therapeutic effectiveness (eg, need for steroid and/or immunosuppressive therapy), steroid resistance, need for surgery (resection), and the presence of familial IBD were collected by the clinical investigator reviewing the medical charts and completing a questionnaire. The disease phenotype (age at onset, duration, location, behavior) was determined according to the Montreal classification. In the present study, complicated disease behavior in patients with CD was defined as stricturing or penetrating behavior during follow-up. Only patients with a confirmed diagnosis for >1 year were enrolled.

Blood samples were obtained prospectively for PAB, rPAB, GAB, ASCA, and pANCA. At the time of taking blood samples, patients had a clinical assessment, including calculation of clinical disease activity scores. Sera for serological markers determination were coded to maintain blinding. After serum separation, blood samples were stored at −80°C until further analysis.

The study protocol was approved by the ethical and science committee of the Semmelweis University. Each parent of the children was informed about the nature of the study and signed the informed consent form.

Back to Top | Article Outline
Antibody Assays for PAB, rPAB, GAB, ASCA, and ANCA

Presence of PAB, rPAB, GAB, ASCA, and anti-neutrophil cytoplasmic antibodies (ANCA) was determined in a commercially available IIF assay (EUROIMMUN AG, Luebeck, Germany) according to the manufacturers’ instructions. For IIF, coated cover glasses with several biological substrates were cut into millimeter-sized fragments (BIOCHIPs, EUROIMMUN AG) and used side by side in the same reaction field, giving the opportunity to investigate one serum on several tissues or prepared cells: PAB: monkey pancreas, recombinant pancreas antigen 1 (rPAg1); CUZD1, and recombinant pancreas antigen 2 (rPAg2); GP2: human transfected cells, nontransfected cells (control); GAB: human intestinal goblet cell culture; ASCA: Saccharomyces cerevisiae fungal smear; ANCA: ethanol (EOH)-fixed and formalin (HCHO)-fixed human granulocytes. Sera were incubated at a 1:10 dilution (ASCA IgG 1:1000, ASCA IgA 1:100) in phosphate-buffered saline/Tween for 30 minutes. After washing with the buffer, slides were then incubated with fluorescein-labeled goat antihuman IgG or IgA antibodies for another 30 minutes. After further washing, evaluation and classification of the patterns were performed under ultraviolet light using fluorescence microscope (EUROIMMUN LED, EUROStar Bluelight, EUROIMMUN AG, Luebeck, Germany). Sera that showed specific fluorescence patterns were classified as follows:

1. Two relevant patterns against pancreas acinus cells can be found: a reticuloglanular (type 1) and a drop-like fluorescence (type 2). Antibodies against rPAg1 (CUZD1) resulted in a reticulogranular pattern, whereas antibodies against rPAg2 (GP2) resulted in a drop-like pattern.

2. GAB positivity can be detected as a cloudy fluorescence with indistinct boundary on intestinal goblet cells.

3. If ASCA were present, then the smeared globular yeasts on the reaction fields fluoresced clearly in their entirety.

4. Using ethanol-fixed granulocytes, 2 basic ANCA patterns were detectable: a granular fluorescence, which is distributed evenly over the entire cytoplasm, leaving the cell nuclei-free (cytoplasmic type, cANCA), or a fluorescence around the cell nuclei (perinuclear type, pANCA). Interference by anti-nuclear antibodies, which may mimic the pANCA pattern, was excluded by evaluation on HEp-2 cells, liver tissue of primates, and formalin-fixed granulocytes. All of the positive sera were titrated to endpoint.

Back to Top | Article Outline
Detection of NOD2/CARD15 Mutations

Genomic DNA was isolated from whole blood using the QIAamp DNA Blood Mini Kit (QIAGEN, Hilden, Germany). The 3 NOD2/CARD15 variants, Arg702Trp, Gly908Arg, and Leu1007fs, were typed using polymerase chain reaction/restriction fragment length polymorphism as previously described (22). NOD2/CARD15 variants were detected by denaturing high-performance liquid chromatography (Wave DNA Fragment Analysis System, Transgenomic, Glasgow, UK). Sequence variation, observed in the denaturing high-performance liquid chromatography profile, was sequenced on both strands to confirm the alteration. Sequencing reactions were performed with the ABI BigDye Terminator Cycle Sequencing Kit version 1.1 (Applied Biosystems, Foster City, CA), and samples were sequenced on an ABI Prism 310 Genetic Analyzer (Applied Biosystems). Genotyping was carried out at the National Hematology and Immunology Institute, Budapest, Hungary.

Back to Top | Article Outline
Statistical Analysis

Statistical analysis was carried out using GraphPad Prism 5 (GraphPad, San Diego CA). The presence of ASCA, pANCA, PAB, rPAB, and GAB was compared between the IBD and control cohorts, as well as within subgroups of patients with IBD using the χ2 and Fisher exact tests. The differences of age between subgroups of patients were assessed by Mann-Whitney test. Logistic regression analysis was also performed to assess the complex associations between clinical phenotype and the serology profile. A P value of <0.05 was considered as significant.

Back to Top | Article Outline


Characteristics of Patients With IBD

The clinical phenotype of patients with CD and UC is shown in Table 1. Median age at diagnosis was 13.9 years. Fifty-one percent of patients with CD and 50% of patients with UC had active disease, whereas 49% of patients with CD and 50% of patients UC were in remission at the time of taking blood samples. Forty pediatric patients with CD (38.8%) developed penetrating and/or stricturing disease after a median follow-up of 18 months and 9.7% underwent surgery. Approximately 80% of patients were treated with steroids at some point during their disease and the majority (72.1%) was treated with azathioprine. Twenty percent of patients with CD received infliximab, 40% had remission or fistula closure, and 60% had partial remission or partial fistula closure.

Back to Top | Article Outline
Diagnostic Accuracy of PAB, rPAB, and GAB and Association With Disease Phenotype in CD and UC

Sensitivity and specificity data for PAB, rPAB, and GAB are given in Table 2. The presence of PAB and rPAB (IgA or IgG) antibodies was significantly higher in CD (34% and 35.9%) and UC (20.4% and 24.5%) compared with controls (0% and 0%, P < 0.0001). The combination of PAB and/or ASCA/pANCA improved sensitivity of serological markers in both CD (87.4%) and UC (79.6%). Specificities were 89.3% and 93.2%, respectively. The positive predictive value was 89.1% in CD for the combination of the markers, and negative predictive value was 87.6% (in UC, positive predictive value 93.2%, negative predictive value 82.2%). The accuracy of serologic antibodies is shown in Table 2. A strong association was observed between the reticulogranular pancreatic IIF pattern and the reactivity of CUZD1-transfected cells (CD, 26 cases; UC, 8 cases) as well as between the droplet pattern and GP2 reactivity (CD, 24 cases; UC, 3 cases). One sera showed positive CUZD1 and 3 sera positive GP2 reactions (CD, 1; UC, 2) but no signal with tissue.

GAB positivity was significantly higher in patients with UC compared with CD and controls, respectively (UC, 12.2%; CD, 1.9%; controls, 1.9%; P = 0.02).

The presence of PAB, rPAB, and GAB was not associated to clinical presentation, medical therapy, need for surgery, or EIMs in either CD or UC. The PAB and rPAB positivity was numerically higher in patients with colonic (28.6% and 27%) and ileocolonic (60% and 59.4%) CD than in ileal disease (11.4% and 13.5%) (Table 3); however, the difference was not statistically significant.

Back to Top | Article Outline
Diagnostic Accuracy of ASCA and pANCA and Association With Disease Phenotype in CD and in UC

Of the 103 patients with CD studied, 72.8% were ASCA positive (either IgA or IgG). Specifically, 63.1% were positive for ASCA IgA, 66.9% for ASCA IgG, and 60.2% for both ASCA IgA and IgG. The presence of ASCA (either IgA or IgG) was significantly higher in patients with CD (72.8%) compared with patients with UC (26.5%) and control (4.8%).

ASCA positivity was associated with complicated disease behavior (P = 0.0003) (stenosing [P = 0.02] and/or penetrating disease behavior [P = 0.0003]) and perianal complications (P = 0.01) in CD. Association with location in patients with CD was not found (Table 4). The frequency of stenosing and penetrating disease and ileocolonic involvement increased with increasing number of immune responses, whereas inflammatory behavior (B1) showed inverse correlation with the number of ASCA (Table 5). There was no significant difference with regard to the need for surgery between ASCA-positive and ASCA-negative patients (need for surgery 12% vs 3.6%). ASCA was not associated with medical therapy and EIMs.

ASCA positivity was found in 13 patients with UC (26.5%); however, 7 of these patients had progressive sclerosing cholangitis (PSC).

PANCA were detected in 38 (77.5%) of 49 patients with UC and in 34 (33%) of 103 patients with CD. The presence of pANCA was not associated with clinical presentation, medical therapy, need for surgery, or EIMs neither in CD nor in UC.

Back to Top | Article Outline
NOD2/CARD15 Genotype, Serum Antibodies and Phenotype in CD

NOD2/CARD15 genotypes were known for 43 patients with. Mutations of NOD2/CARD15 were detected in 13 (30.2%) patients with CD. There was no association of ASCA, PAB, rPAB, and pANCA antibody status to NOD2/CARD 15 genotype. Three NOD2 carriers were PAB and ASCA negative, and 2 NOD2 carriers were PAB, ASCA, and pANCA negative. NOD2 mutations were not related to age at onset, disease location, and disease behavior. Nevertheless, NOD2 variants were significantly related to steroid refractory disease and administration of infliximab.

Back to Top | Article Outline


This is the first report to assess the diagnostic value of PAB, rPAB, and GAB as well as relevant phenotype–serotype associations in large cohort of patients with pediatric-onset IBD. Furthermore, we investigated the accuracy in combination with ASCA and pANCA, and NOD2/CARD15 mutations. In the present study, PAB positivity in CD (34%) was similar to previous reports from adult patients. The prevalence of rPAB was numerically higher than that of PAB. Prevalence of 27% to 39% of PAB has been found in adult patients with CD, compared with only 0% to 5% in patients with UC (3,9,23). Stöcker et al (3) reported that PAB could be determined only in the serum of patients with CD; however, other studies found a much higher prevalence of PAB in UC (4,5,26) in accordance with our results (20.4%). Even though the specificity of PAB for CD and UC is high, their sensitivity is low; however, sensitivity can be significantly increased with combinations of different antibodies (see below). Recent studies have demonstrated that GP2 is expressed on the apical surface of intestinal membranous cells of the follicle-associated epithelium, and is essential for host–microbial interaction and the initiation of bacteria-specific mucosal immune responses (24,25).

There are conflicting results of association between PAB and CD phenotype in adult cohorts. In a Belgian study, PAB was negatively associated with stricturing disease behavior in CD (5). In contrast, in other European studies, an increased prevalence of PAB was observed in patients with stricturing or penetrating phenotypes (4,9,26). Lakatos et al reported association between PAB positivity and perianal disease and EIMs (4). In a French study, PAB correlated with early onset of disease in CD (23). In the present study, pancreatic autoantibodies were not related to clinical presentation, medical therapy, need for surgery, or EIMs in CD or in UC.

Autoantibodies against different colonic antigens have been found in patients with UC, for example GAB. In previous studies, GAB have been detected in adult patients with UC; the prevalence was 28% to 30% (3,10). In contrast, other studies suggested a much lower prevalence in both diseases (4,11). Probably these differences are due to methodological differences, such as in enzyme-linked immunosorbent assay antigen substrates and in evaluation of fluorescence patterns. GAB produce mucin that has multiple functions: it serves as a lubricant, provides nonspecific protection against unwanted microbial agents, and hosts the normal bacterial flora. Through complicated and strictly regulated glycosylation, mucins act as a decoy in binding a range of different microbes, thereby maintaining a normal intestinal flora. The significance of these antibodies, however, has not been established and it remains unclear.

In agreement with the results published by Lakatos et al (4) and Lawrance et al (11), the prevalence of GAB in our UC group was lower (12.2%) than previously reported (3,10). One explanation for the lower prevalence of GAB in our study may be the younger age of the patients. In concordance with earlier data, we could not demonstrate any association between the presence of GAB and clinical presentation, medical therapy, need for surgery, or EIMs in patients with UC. Consequently, the clinical utility of GAB for diagnostic purposes in pediatric IBD is limited.

In the present study, the prevalence of ASCA in CD (72.8%) was similar to that in previous pediatric reports (44%–76%) (13–15,27). In concordance with published data, we observed association between ASCA positivity and stenosing and penetrating behavior (16,23,28,29).

ASCA positivity in our patients with UC was higher compared with earlier reports, perhaps due to the relatively higher proportion of patients with PSC. Of note, in an Italian study, ASCA reactivity was detected in 44% of patients with PSC (30). In the present study, the presence of ASCA in patients with UC without PSC was found similar to other studies (31,32).

The prevalence of pANCA in our patients with UC and CD (UC 77.5%, CD 33%) was comparable with earlier reports (57%–83%) (13,14,32). Although pANCA has been established as an UC-specific marker, approximately 25% of all of the patients with CD also express pANCA (18). Several studies have suggested that pANCA expression is significantly higher in colonic CD (17,19,33), but others have been unable to confirm this finding (34). In the present study, we did not observe any correlation between pANCA positivity and colonic CD.

In addition, we evaluated the diagnostic accuracy of combining markers. In concordance with previous data (11) in the present study, the specificity of PAB was 100%; however, sensitivity was low (CD, 34%, UC, 20.4%). The sensitivity increased for combinations (eg, PAB, ASCA, and pANCA for CD [87.4%] as well as PAB and pANCA for UC [79.6%]). In combinations the specificity for CD was 89.3% and for UC it was 94.2%.

Associations between NOD2 variants and complicated disease course, earlier time to surgery in children with CD, were reported (35,36). In the present study, we could not demonstrate any association of NOD2/CARD15 genotype, serum antibodies, and phenotype in CD; however, in accordance with the results of Roesler et al (37), we found an association between NOD2 variants and need for more intensive therapy (steroid refractory disease and infliximab use) in our CD cohort.

In conclusion, pancreatic autoantibodies (PAB, rPAB) and GAB were specific for IBD, but the sensitivity was limited; also, they were not associated with clinical phenotype. Combinations of these antibodies with conventional serology markers are associated with increased sensitivity; therefore, the use of combinations may be recommended in the diagnostic workup of selected cases with IBD.

Back to Top | Article Outline


1. Xavier RJ, Podolsky DK. Unravelling the pathogenesis of inflammatory bowel diseases. Nature 2007; 448:427–434.
2. Papp M, Altorjay I, Dotan N, et al. New serological markers for inflammatory bowel disease are associated with earlier age at onset, complicated disease behaviour, risk for surgery, and NOD2/CARD 15 genotype in a Hungarian IBD cohort. Am J Gastroenterol 2008; 103:665–681.
3. Stöcker W, Otte M, Ulrich S, et al. Autoimmunity to pancreatic juice in Crohn's disease. Results of autoantibody screening in patients with chronic inflammatory bowel disease. Scand J Gastroenterol Suppl 1987; 22:41–52.
4. Lakatos PL, Altorjay I, Szamosi T, et al. Pancreatic autoantibodies are associated with reactivity to microbial antibodies, penetrating disease behaviour, perianal disease, and extraintestinal manifestations, but not with NOD2/CARD15 or TLR4 genotype in a Hungarian IBD cohort. Inflamm Bowel Dis 2009; 15:365–374.
5. Joossens S, Vermeire S, Van Steen K, et al. Pancreatic autoantibodies in inflammatory bowel disease. Inflamm Bowel Dis 2004; 10:771–777.
6. Roggenbuck D, Hausdorf G, Martinez-Gamboa L, et al. Identification of GP2, the major zymogen granule membrane glycoprotein, as autoantigen of pancreatic antibodies in Crohn's disease. Gut 2009; 58:1620–1628.
7. Roggenbuck D, Reinhold P, Wex T, et al. Autoantibodies to GP2, the major zymogen granule membran glycoprotein, are new markers in Crohn's disease. Clin Chim Acta 2011; 412:718–724.
8. Bogdanos DP, Rigopoulou EI, Smyk DS, et al. Diagnostic value, clinical utility and pathogenic significance of reactivity to the molecular targets of Crohn's disease specific-pancreatic autoantibodies. Autoimmun Rev 2011; 11:143–148.
9. Klebl FH, Bataille F, Huy C, et al. Association of antibodies to exocrine pancreas with subtypes of Crohn's disease. Eur J Gastroenterol Hepatol 2005; 17:73–77.
10. Hibi T, Ohara M, Kobayashi K, et al. Enzyme linked immunosorbent assay (ELISA) and immunoprecipitation studies on anti-goblet cell antibody using a mucin producing cell line in patients with inflammatory bowel disease. Gut 1994; 35:224–230.
11. Lawrance IC, Hall A, Leong R, et al. A comparative study of globlet cell and pancreatic exocrine autoantibodies combined with ASCA and pANCA in Chinese and Caucasian patients with IBD. Inflamm Bowel Dis 2005; 11:890–897.
12. Dubinsky MC, Ofman JJ, Urman M, et al. Clinical utility of serodiagnostic testing in suspected pediatric inflammatory bowel disease. Am J Gastroenterol 2001; 96:758–765.
13. Zholudev A, Zurakowski D, Young W, et al. Serologic testing with ANCA, ASCA, and anti-OmpC in children and young adults with Crohn's disease and ulcerative colitis: diagnostic value and correlation with disease phenotype. Am J Gastroenterol 2004; 99:2235–2241.
14. Khan K, Schwarzenberg SJ, Sharp H, et al. Role of serology and routine laboratory tests in childhood inflammatory bowel disease. Inflamm Bowel Dis 2002; 8:325–329.
15. Dubinsky MC, Kugathasan S, Mei L, et al. Increased immune reactivity predicts aggressive complicating Crohn's disease in children. Clin Gastroenterol Hepatol 2008; 10:1105–1111.
16. Amre DK, Lu SE, Costea F, et al. Utility of serological markers in predicting the early occurrence of complications and surgery in pediatric Crohn's disease patients. Am J Gastroenterol 2006; 101:645–652.
17. Vasiliauskas EA, Kam LY, Karp LC, et al. Marker antibody expression stratifies Crohn's disease into immunologically homogeneous subgroups with distinct clinical characteristics. Gut 2000; 47:487–496.
18. Dubinky M. Special issues in pediatric inflammatory bowel disease. World J Gastroenterol 2008; 14:413–420.
19. Ruemmele FM, Targan SR, Levy G, et al. Diagnostic accuracy of serological assay in pediatric inflammatory bowel disease. Gastroenterology 1998; 115:822–829.
20. Kundhal PS, Critch JN, Zachos M, et al. Paediatric Crohn disease activity index: response to short-term change. J Pediatr Gastroenterol Nutr 2003; 36:83–89.
21. Turner D, Otley AR, Mack D, et al. Development, validation, and evaluation of a paediatric ulcerative colitis activity index: a prospective multicenter study. Gastroenterology 2007; 133:423–432.
22. Heliö T, Halme L, Lappalainen M, et al. CARD15/NOD2 gene variants are associated with family occuring and complicated forms of Crohn's disease. Gut 2003; 52:558–562.
23. Desplat-Jego S, Johanet C, Escande A, et al. Update on anti-Saccharomyces cerevisiae antibodies, anti-nuclear associated anti-neutrophil antibodies and antibodies to exocrine pancreas detected by indirect immunfluorescence as biomarkers in chronic inflammatory bowel diseases: results of a multicenter study. World J Gastroenterol 2007; 13:2312–2318.
24. Holzl MA, Hofer J, Kavarik JJ, et al. The zymogen granule protein 2 (GP2) binds to scavanger receptor expressed on endothelial cells1 (SREC-1). Cell Immunol 2011; 267:88–93.
25. Ohno H, Hase K. Glycoprotein 2 (GP2): grabbing the FimH bacteria into M cells for mucosal immunity. Gut Microbes 2010; 1:401–407.
26. Koutroubakis IE, Drygiannakis D, Karmiris K, et al. Pancreatic autoantibodies in Greek patients with inflammatory bowel disease. Dig Dis Sci 2005; 50:2330–2334.
27. Markowitz J, Kugathasan S, Dubinsky M, et al. Age of diagnosis influences serologic responses in children with Crohn's disease: a possible clue to etiology? Inflamm Bowel Dis 2009; 15:714–719.
28. Dubinky MC, Lin YC, Dutridge D, et al. Serum immune responses predict rapid disease progression among children with Crohn's disease: immune responses predict disease progression. Am J Gastroenterol 2006; 101:360–367.
29. Elitsur Y, Lawrence Z, Tolaymat N. The diagnostic accuracy of serologic markers in children with IBD: the West Virginia experience. J Clin Gastroenterol 2005; 39:670–673.
30. Muratori P, Muratori L, Guidi M, et al. Anti-Saccharomyces cerevisiae antibodies (ASCA) and autoimmune liver diseases. Clin Exp Immunol 2003; 132:473–476.
31. Demirsoy H, Ozdil K, Ersoy O, et al. Anti-pancreatic antibody in Turkish patients with inflammatory bowel disease and first-degree relatives. World J Gastroenterol 2010; 16:5732–5738.
32. Bartunkova J, Kolárová I, Sedivá A, et al. Antineutrophil cytoplasmic antibodies, anti-Saccharomyces cerevisiae antibodies, and specific IgE to food allergens in children with inflammatory bowel disease. Clin Immunol 2002; 102:62–68.
33. Arnott ID, Landers CJ, Nimmo EJ, 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 2004; 99:2376–2384.
34. Papp M, Altorjay I, Norman GL, et al. Seroreactivity to microbial components in Crohn's disease is associated with ileal involvement, noninflammatory disease behaviour and NOD2/CARD15 genotype, but not with risk for surgery in a Hungarian cohort of IBD patients. Inflamm Bowel Dis 2007; 13:984–992.
35. Kugathasan S, Collins N, Maresso K, et al. CARD15 gene mutations and risk for early surgery in pediatric-onset Crohn's disease. Clin Gastroenterol Hepatol 2004; 2:1003–1009.
36. Adler J, Rangwalla SC, Dwamena BA, et al. The prognostic power of the NOD2 genotype for complicated Crohn's disease: a meta-analysis. Am J Gastroenterol 2011; 106:699–712.
37. Roesler J, Thüringen A, Sun L, et al. Influence of CARD15 mutations on disease activity and responses to therapy in 65 pediatric Crohn patients from Saxony, Germany. J Pediatr Gastroenterol Nutr 2005; 41:27–32.

GAB; inflammatory bowel disease; PAB; pediatric; rPAB

Copyright 2012 by ESPGHAN and NASPGHAN