Autophagy is a highly regulated catabolic pathway that is potently induced by stressors including starvation and infection. An essential component of the autophagy pathway is an ATG16L1-containing E3-like enzyme, which is responsible for lipidating LC3B and driving autophagosome formation. ATG16L1 polymorphisms have been linked to the development of Crohn’s disease (CD) and phosphorylation of CD-associated ATG16L1 (caATG16L1) has been hypothesized to contribute to cleavage and autophagy dysfunction. Here we show that ULK1 kinase directly phosphorylates ATG16L1 in response to infection and starvation. Moreover, we show that ULK1-mediated phosphorylation drives the destabilization of caATG16L1 in response to stress. Additionally, we found that phosphorylated ATG16L1 was specifically localized to the site of internalized bacteria indicating a role for ATG16L1 in the promotion of anti-bacterial autophagy. Lastly, we show that stable cell lines harbouring a phospho-dead mutant of ATG16L1 have impaired xenophagy. In summary, our results show that ATG16L1 is a novel target of ULK1 kinase and that ULK1-signalling to ATG16L1 is a double-edged sword, enhancing function of the wildtype ATG16L1, but promoting degradation of caATG16L1.

The inflammatory bowel diseases (IBD) are chronic, relapsing and remitting diseases of the gastrointestinal tract. There are two main types of IBD: Crohn’s disease (CD) and ulcerative colitis (UC). The prevalence of IBD is highest in the western world, approximately 100-200 people per 100,000 are affected. In recent years there has been a marked increase in the incidence of CD and UC, in both adults and children (Henderson et al., 2012; Molodecky et al., 2012). This is particularly relevant in Scotland where recent research shows that there has been a 79% increase in the number of cases of paediatric IBD since the 1990’s (Henderson et al., 2012). A yeast 2 hybrid screen identified TLE1as an interacting partner of the known CD susceptibility gene; Nucleotide- binding oligomerisation protein 2 (Nod2). An initial genome wide association study (GWAS) also found an association between the rs6559629 SNP, located in Tle1 and ileal CD (p =3.1 x 10-5) and showed that carriage of the Tle1 risk allele increases the effects of Nod2 mutations in CD. TLE1 functions as a transcriptional co repressor in a variety of different cellular and developmental pathways The work presented in this thesis investigates the potential role of TLE1 in CD. This has been approached using four different strategies: sequencing TLE1 in CD patients and controls, analysing the effects of knocking down TLE1 on genome wide expression, investigating whether the known IBD susceptibility protein XBP1 binds to a predicted binding site in TLE1 and investigating TLE1 levels and localisation in human intestinal samples from CD patients and controls Sequencing TLE1 exons and introns 15/16 and 16/17 in a Scottish cohort of 24 CD patients and healthy controls identified a number of potentially pathogenic exonic and intronic SNPs. Two exonic SNPs and thirteen intronic SNPs were identified and these were further investigated in larger Scottish (203 CD cases, 190 HC) and European cohorts (6,333 CD cases and 15,056 HC) but were not present at statistically significantly different frequencies. Secondly, the effects of TLE1 knock down on genome wide expression were analysed using an Illumina HT12 expression chip. The results showed that TLE1 knock down significantly altered expression of 19 loci (Bonferroni) and 526 loci (FDR). Four of the 19 Bonferroni significant loci are potentially involved in CD: RIOK1 (p=4.3×10-3), SGPL1 (p=4.3×10-3), TUSC3 (p=1.8×10-2) and CCND1 (p=2.7×10-3). Furthermore, expression of SGPL1 and RIOK1 were shown to be differentially expressed at the mRNA level between inflamed patients and controls. The third approach investigates a predicted binding site for the known IBD susceptibility gene, XBP1 in TLE1 which was identified using the Haploreg program. This work shows, using chromatin immunoprecipitation, that exogenous XBP1 does not appear to bind to this predicted binding site. Finally, TLE1 expression was analysed in human intestinal resection samples from patients of known NOD2 status. This work shows that TLE1 and NOD2 are expressed in Paneth cells, however TLE1 expression is not altered in patients carrying CD associated NOD2 variants. In this work TLE1 sequence, expression and potential interacting proteins have been analysed. The results presented suggests multiple mechanisms by which TLE1 may be influencing susceptibility to CD including: the unfolded protein response (TUSC3), S1P signalling and ribosome biogenesis. They also implicate TLE1 in Paneth cell function alongside NOD2. The exact means by which TLE1 may play a role in IBD pathogenesis has yet to be fully elucidated.

No description available.

Ulcerative colitis (UC) and Crohn’s disease (CD) are the two common forms of inflammatory bowel disease (IBD). They share similar clinical and demographic features as well as harbor key differences in tissue damage and prognosis. Previous studies indicated that they contributed to the increased rick to Colorectal cancer (CRC). However, whether UC and CD share inflammatory signatures still remains controversial. In addition, no inflammatory signatures have been reported on CRC. To answer these questions, a comprehensive study has been conducted on collected microarray datasets. Our analysis suggests that although CD and UC share common inflammatory pathways, they also present difference. Especially, CD patients are likely to have type I response, while UC patients are inclined to undergo type II response. Pathway enrichment analysis on CRC uncovered two potential CRC-specific inflammatory pathways.

There is an unmet need for novel biomarker discovery in Inflammatory Bowel Diseases (IBD) to aid clinical management in several clinical settings including diagnosis and prognosis. With an ever-advancing repertoire of biological therapies on the horizon, it is important to personalise treatments at an early stage. The aim of this thesis is to explore the clinical utility of novel blood-based biomarkers in diagnosis, disease classification and prognosis in 2 cohorts: newly diagnosed IBD and acute severe colitis. Investigating the circulating methylome, 290 probes exhibited Holm significant IBD-associated methylation differences, including VMP1/MIR21 (p=7.5×10-14) and RPS6KA2 (1.1×10-19) and were consistent within the European cohort. 11 Differentially methylated positions (DMPs) predicted treatment escalation after Holm adjustment (top probe p=0.003). A panel of 6 probes identified 2 patient subgroups that have significantly different disease courses (Hazard Ratio (HR) 10.5, 95%CI: 4.3-25.6; logrank p=1.5×10-24). The 6 probe marker outperformed conventional biomarkers in predicting treatment escalation (hsCRP > 4mg/L, HR 3.2(1.7-5.8), logrank p=0.0004 and Alb < 36g/L, HR 2.9(1.5-5.6), p=0.0001). Within the same cohort, a novel proximity extension assay (PEA) was then utilised to identify novel diagnostic and prognostic protein markers. 61 proteins were significantly associated with IBD including MMP12 (Holm-adjusted p=4.1×10-26). A total of 9 proteins predicted disease course in this cohort. Using a panel of 7 randomly selected top prognostic probes, 2 patient groups were identified that had significantly different disease courses: logrank p=2.2×10-10, HR 5.6(2.0-15.6), outperforming conventional biomarkers in predicting treatment escalation (hsCRP > 4mg/L, HR 3.2(1.7- 5.8), logrank p=0.0003 and Alb < 36g/L, HR 2.7(1.4-5.2), p=0.0004). In a subcohort, serum calprotectin (SC) and conventional blood markers were investigated for their utility in diagnosis and prognosis in IBD. SC performed at par with CRP at differentiating IBD from controls with an area under the curve (AUC) of 0.87 (CI 0.81-0.92). For prognostication, both albumin and SC remained significant predictors of treatment escalation in IBD (logrank test p=5.1×10-5). MicroRNAs (miRNA) are small non-coding nucleic acids that have the capacity to modulate gene expression. Using small RNA sequencing in acute severe colitis (ASUC) and healthy controls (HC), 10 serum-based miRNA markers were significantly associated with acute severe colitis, including miR-30a-5p. Validating the findings using qPCR, miR-30a-5p was downregulated in ASUC (p=0.003). Furthermore, miR30a-5p remained a significant predictor of eventual colectomy in acute colitis (logrank test p=0.0014). These data highlight the translational potential for methylation, miRNA and proteomic biomarkers in diagnosing and prognosticating in IBD.

Crohn’s disease is a chronic inflammatory bowel disease with unknown origin. This study investigates the combined use of surgery and immuno modulation in Crohn’s disease. The outcome of medication and surgery in 371 operations on 237 patients between 1989 and 2006 were evaluated. Moreover the effects of prednisolone, azathioprine and infliximab on the healing of colo-colonic anastomosis in 84 mice with or without colitis were evaluated. The use of thiopurines after abdominal surgery in selected cases of severe Crohn’s disease was found to prolong the time to clinical relapse of the disease from 24 to 53 months. Patients on postoperative maintenance therapy with azathioprine had a decreased symptomatic load over time and needed fewer steroid courses. The use of thiopurines was found to be a risk factor of anastomotic complications in abdominal surgery for Crohn’s disease together with pre-operative intra-abdominal sepsis and colo-colonic anastomosis. The risk for anastomotic complications increased from 4 % in those without any of these risk factors to 13 % in those with any one and 24 % if two or three risk factors were present. In patients with two or more of these, or previously established, risk factors prior to surgery one should consider refraining from anastomosis or doing a proximal diverting stoma. Another possibility is to use a split stoma in which both ends of a future delayed anastomosis are brought out in the same ostomy hole of the abdominal wall. This method was found to significantly decrease the number of risk factors prior to the actual anastomosis as well as decreasing the risk of anastomotic complications, without increasing the number of operations or the time spent in hospital. In the animal model all three medications had an ameliorating effect on the colitis compared with placebo. Only prednisolone was found to interfere with the healing of the colo-colonic anastomoses with significantly decreased bursting pressure compared with placebo as well as azathioprine and infliximab. The association between azathioprine therapy and anastomotic complications may be due to a subgroup of patients with a more severe form of the disease who have an increased risk of such complications and also are more prone to receive intense pharmacological therapy. / Crohns sjukdom är en kronisk inflammatorisk tarmsjukdom av oklar orsak. Huvudsyftet med denna avhandling var att undersöka den kombinerade behandlingen med kirurgi och immunhämmare vid Crohns sjukdom. Utfallet av medicinsk och kirurgisk behandling vid 371 operationer på 237 patienter mellan 1989 och 2006 utvärderades. Därutöver studerades effekterna av kortison, immunhämmare och behandling med inflammationsdämpande antikroppar på läkning av tjocktarms-skarv på 84 möss med eller utan inflammation i tarmen. Vid utvalda fall med svårare form av Crohns sjukdom visade sig förebyggande behandling med immunhämmare efter kirurgi förlänga tiden till återfall av symptom från 24 till 53 månader. Patienter med immunhämmare som underhållsbehandling hade också minskade symptom under uppföljningstiden med ett minskat behov av kortison. Immunhämmande behandling inför kirurgi visade sig, liksom pågående infektion i bukhålan och sydd skarv på tjocktarmen, vara en riskfaktor för att drabbas av komplikationer vid bukkirurgi på grund av Crohns sjukdom. Risken för infektionskomplikationer i bukhålan ökade från 4 % hos dem utan någon av dessa riskfaktorer till 13 % hos dem med någon och 24 % hos dem med två eller tre riskfaktorer inför operationen. Hos patienter med två eller fler kända riskfaktorer bör man överväga att avstå från att sy en skarv på tarmen vid kirurgi eller möjligen skydda skarven med en avlastande stomi. Ett alternativ till detta är att anlägga en delad stomi där bägge ändarna av den framtida skarven tas ut genom en och samma stomiöppning i bukväggen. Denna metod med en fördröjd skarv på tarmen visade sig minska antalet kirurgiska riskfaktorer inför själva skarvningen och dessutom minska risken för tidiga infektiösa komplikationer i bukhålan, utan att vare sig öka antalet kirurgiska ingrepp eller förlänga vårdtiden på sjukhus. I en djurmodell visade sig alla tre läkemedlen ha en lindrande effekt på tarminflammation jämfört med placebo. Endast kortison visade sig påverka läkningen negativt med en sänkning av bristningstrycket i den sydda skarven på tjocktarmen, jämfört med placebo såväl som med immunhämmare och antikropps-behandling. Kopplingen mellan immunhämmare och komplikationer efter sydda skarvar på tarmen behöver alltså inte vara en direkt läkemedelseffekt. Orsaken kan istället vara att en undergrupp av Crohnpatienter har en svårare sjukdomsform som ger både ökad komplikationsrisk och större behov av intensiv medicinsk behandling.

A series of investigations examining the importance of genetic factors in the development of the inflammatory bowel diseases (IBD) namely Crohn’s disease (CD), Ulcerative Colitis (UC) and Indeterminate Colitis (IC) has been undertaken in Scottish children. This has been performed by collection of clinical details and DNA from children with IBD, then analysing the contribution of various candidate genes to both disease susceptibility and disease phenotype. In order to carry out these studies the presenting features of a large cohort of children from across Scotland with IBD diagnosed at less than 16 years were collected, both by examination of hospital case records and by patient interview and questionnaire. For each patient a detailed analysis was made of disease phenotype at presentation including detailed examination of disease location, disease behaviour and growth parameters. A repository of clinical material (DNA, plasma and lymphocytes) was collected from children to accompany the detailed clinical parameters allowing genotype-phenotype analysis at a later stage. Additionally, DNA was also collected from parents where possible to facilitate family based association analysis of candidate genes by transmission disequilibrium testing. A previous DNA repository of healthy Scottish controls had been collected previously and the data generated was available for use in this study. The phenotypic data was collected using an established phenotypic classification (the Vienna classification) used in adult studies as well as a personally devised paediatric phenotypic classification designed for use in this thesis. Firstly, the contribution of the three common mutations within the NOD2/CARD15 gene (R702W, G908R and Leu1007finsC) was analysed in 247 children with IBD. The Leu1007finsC variant was associated with Crohn’s disease by case-control (p = 0.01) and TDT analysis (p = 0.006). Genotype phenotype analysis demonstrated NOD2/CARD15 variants were strongly associated with several markers of disease severity in CD most notably with an increased need for surgery on multifactorial analysis. Then to examine the further contribution of other mutations within the whole NOD2/CARD15 gene, the 12 exons of the gene were sequenced in 24 paediatric CD patients, to identify any additional SNPs that may have conferred an increased susceptibility to CD. Two mutations (V955I, M863V) identified in xii sequencing were genotyped in a large patient cohort, but were not found to confer increased disease susceptibility. Next, the contribution of IBD5 locus was analysed in 299 children with IBD studying 5 SNPs, including mutations in the proposed candidate genes OCTN 1 and 2. Allele frequencies of OCTN1/2 variants were significantly higher in IBD/CD cases (p<0.04). The homozygous mutant OCTN1/2 haplotype was increased in IBD and UC patients (p = 0.02 for both) compared to healthy controls. OCTN1/2 variants however were not independent of the background IBD5 risk haplotype in conferring disease susceptibility. Genotype- phenotype analysis demonstrated association of the risk haplotype with both lower weight and body mass index centiles at diagnosis as analysed by multifactorial analysis. The contribution of the 113 G/A mutation within the discs, large homolog 5 (DLG5) gene was examined in 296 children with IBD. TDT analysis demonstrated a significant association with IBD (p<0.05). Genotype-phenotype analysis demonstrated associations with higher social class, male sex and taller children. Finally, the Anti-Saccharomyces cerevisiae antibodies (ASCA) status of 301 IBD patients was determined. CD patients had a higher prevalence of ASCA antibodies compared to UC patients and healthy controls (p<0.001 for both). A positive ASCA antibody was more common in CD patients with markers of more severe disease and on multifactorial analysis in patients with CD involvement of the oral cavity (p = 0.001). In summary, the candidate genes examined thus far in children with IBD in Scotland have demonstrated a relatively minor contribution to disease susceptibility but have been demonstrated to be associated with specific disease phenotypes in patients with Crohn’s disease. The use of a novel paediatric phenotypic classification in this thesis has allowed description of these novel genotype-phenotype associations.

Introduction Inflammatory bowel disease (IBD) is a chronic gastrointestinal disorder that encompasses two major subtypes; Crohn’s Disease (CD) and Ulcerative Colitis (UC). Our knowledge regarding disease pathogesis is rapidly increasing. However, these disease entities provide challenges in diagnosis, monitoring of disease activity and assessing individual response to treatment, because there is a lack of validated clinical biomarkers. Metabolomics involves the study of numerous analytes that have very diverse physical and chemical properties and occur in a wide concentration range. Early evidence suggests there is potential for metabolomic profiling to be used in the differentiation of CD and UC. However, knowledge is limited regarding the metabolic changes seen in relation to disease activity or to medical or surgical treatments. Aims A metabolomics approach was taken to determine whether metabolomic profiles could distinguish between patients with CD or UC and healthy controls. We also aimed to define the relationship between metabolomic profile and disease activity, and to determine the effect of medical (anti-TNFa agents) and surgical treatment on the metabolome. Methods A metabolomics approach was undertaken. Serum and urine sample sets were collected from a total of 41 patients with ulcerative colitis, 43 patients with Crohn’s disease, and 62 healthy controls (HC). In order to allow a comparison of metablomic profile and disease activity, 4 sample sets were taken from the same patient at 3 monthly intervals over the period of one year. Those patients undergoing either surgical or biological treatment had sample sets taken pre and post intervention. Metabolomic analysis using gas chromatography time of flight mass spectrometry (GC-ToF-MS) and ultra-high performance liquid chromatography Fourier Transform mass spectrometry (UHPLC-FTMS) was carried out on both serum and urine. Results Serum and urine GC-ToF-MS and UHPLC-FTMS metabolomic analyses show differentiation between UC, CD and healthy controls, most significantly in urine analyses. No significant differentiation was seen in pre- and post-surgical patients, or pre- and post-biological therapy patients. It was possible to differentiate surgical patients from healthy controls, especially in the urine analyses. Metabolite identification revealed consistently more dietary variation in the healthy controls than in the IBD patients. Significant differences (p < 0.05) were seen between healthy controls and IBD patients in classes of metabolites relating to the citric acid cycle and the uronic acid pathway, as well as amino acids, fatty acids and cholesterols. The behaviour or location of disease, or the disease activity score did not appear to influence the metabolome in either serum or urine analyses using GC-ToF-MS and UHPLC-FTMS. Conclusion Metabolomic profiling of urine and serum in IBD may provide a novel methodology aiding both clinical diagnosis through biomarker development, and advancing knowledge of disease pathogenesis.

As doenças inflamatórias intestinais (DII) são doenças crônicas incapacitantes com significativa morbidade. A Doença de Crohn (DC) e a Retocolite Ulcerativa (RCU) são as maiores representantes. O diagnóstico é baseado na suspeita clínica e complementado por achados endoscópicos, histopatológicos e radiológicos. Como não existe um teste padrão, o diagnóstico destas doenças permanece sendo um problema para gastroenterologistas e patologistas. O desenvolvimento de terapias mais específicas torna cada vez mais importante o diagnóstico preciso para uma escolha terapêutica individualizada. A imunidade inata está ativada de forma semelhante nas duas DII, porém na RCU os linfócitos CD4 TH2 são os principais envolvidos. Estes linfócitos expressam o receptor CD30 na sua membrana plasmática e produzem citocinas que ativam os eosinófilos. Levando em consideração o conhecimento fisiopatológico atual, este estudo tem como principal objetivo avaliar a relevância da expressão do marcador CD30 por imuno-histoquímica, a contagem de eosinófilos e as características histopatológicas no diagnóstico diferencial das DII. Foram avaliados 185 pacientes de um ambulatório especializado em um Hospital Universitário (105 com DC 80 com RCU). O diagnóstico foi estabelecido pelos critérios de Lennard-Jones, revisados por um gastroenterologista especializado e corroborado por cinco anos de seguimento. As biópsias foram realizadas por diversos profissionais na rotina assistencial, porém todas foram analisadas por um patologista com experiência em tubo digestivo. Nenhum paciente estava usando tratamento no momento da coleta das biópsias. As variáveis histopatológicas que demonstraram poder estatístico para auxiliar no diagnóstico diferencial foram a extensão do processo inflamatório para a submucosa, a presença de granuloma, erosões aftóides e variabilidade de acometimento entre os fragmentos. Avaliando estas características em conjunto, foi possível obter uma acurácia de 69.1% para o diagnóstico diferencial entre DC e RCU. Estudando o segmento mais alterado encontrou-se uma mediana de eosinófilos de 42 (25,5 – 63,5), nos pacientes com DC e 107 (67 – 123) nos pacientes com RCU (p< 0.001). Assumindo como ponto de corte um número V a 70 eosinófilos, a sensibilidade foi de 78,3% e a especificidade de 71% favorecendo o diagnóstico de RCU, a área sob a curva ROC foi de 0,767 (IC 95%: 0,696–0,838). A imuno-histoquímica com CD30 demonstrou uma mediana de 3 células CD30+ (2-6) na DC e 33 (24-52) na RCU, demonstrando uma diferença estatística significativa para o diagnóstico diferencial entre as duas doenças (p<0,001). Além disso, nos pacientes com RCU as células CD30+ estavam distribuídas mais frequentemente em agrupamentos nos centros dos folículos linfoides. O ponto de corte determinado pela curva ROC foi de 15 células marcadas (S = 97,5%, E = 94,3%, RV + = 17,1; RV = 0.03, área sob a curva: 0.967, IC 95%: 0.941 - 0.993). Todos estes parâmetros estudados tiveram capacidade discriminatória para o diagnóstico de DC e RCU. Designando um valor para cada variável, baseado no poder estatístico de cada uma e de forma a obter uma soma de 10 pontos no total, foi construído um escore histopatológico para o diagnóstico da DC. Considerando-se a soma das variáveis V 5 obteve-se uma especificidade de 100% e uma sensibilidade de 86,8%. Considerando 4 como ponto de corte do escore, a sensibilidade aumenta para 95,3%, e a especificidade reduz de 100% para 94,9%. Concluindo, o uso rotineiro da avaliação das características histopatológicas descritas em associação com a contagem de eosinófilos e células CD30+ proporciona uma alta acurácia no diagnóstico diferencial entre DC e RCU. Todos os parâmetros estudados são de fácil avaliação tanto por patologistas especialistas quanto generalistas. Sugere-se a realização de um estudo prospectivo para validação deste escore. / Inflammatory bowel diseases (IBD) are chronic disabling diseases with significant morbidity, being Crohn’s disease (DC) and Ulcerative colitis (UC) their greatest representatives. Diagnosis is based on clinical suspicious and complemented by endoscopic, histopathological and radiological findings. However, there is not a gold standard test, so it remained a problem for gastroenterologists and pathologists. The development of more targeted therapies makes it even more important to establish an accurate diagnosis for a better individualized therapeutic approach. Innate immune response is activated in both IBD, but in UC the lymphocytes CD4 TH2-like are mainly engaged, this kind of lymphocytes has CD30 expressed in their plasma membrane and produces cytokines that activate eosinophils. Considering the current pathophysiological knowledge, this study aimed to evaluate the relevance of CD30 expression by immunohistochemical, eosinophil count and histopathological features in differential diagnosis of IBD. A total of 185 patients were evaluated (105 CD/ 80 UC). Patients were followed at a specialized clinic of a university hospital, diagnosed according to Lennard- Jones criteria reviewed by a gastroenterologist expert and corroborated by five years of follow up. Biopsy samples were taken by different professionals in routine care, but analyzed by an experienced gastrointestinal pathologist. None of the patients were using treatment at the time of biopsy. Of all the pathological variables assessed, those that had statistical capacity to assist in the differential diagnosis were extension of the inflammatory process to submucosa, granuloma, aphthous erosion and variability of involvement between fragments. Evaluating these variables together, an accuracy of 69.1% in the differential diagnosis between CD and UC was found. Assessing the most altered sample the median of eosinophils was 42 (25.5 – 63.5) in CD patients and 107 (67 – 123) in UC patients (p< 0.001). Assuming a cutoff V 70 eosinophils, the sensitivity was 78,3% and specificity of 71% favoring the UC diagnosis, the area under the ROC curve was 0.767 (CI 95%: 0,696–0,838). Immunohistochemical CD30+ cells presented with a median of 3 cells (2-6) in CD and 33 cells (24-52) in UC, demonstrating a highly significant statistical difference between the two diseases (p<0.001). Besides, CD30+ cells were distributed most clustered in the center of lymphoid follicles in UC patients. The cutoff determined by ROC curve was 15 (S = 97.5%, E = 94.3%, LR + = 17.1;-RV = 0.03, AUC: 0.967, 95% CI: 0.941 - 0.993). All these parameters studied had discriminatory capacity for diagnosis of CD and UC. A value was assigned to each variable based on the statistical power of each, making a total sum of 10 points to build a histopathological score for the CD diagnosis. Setting the cutoff as V 5 we found a specificity of 100% and a sensitivity of 86.8%. When we consider as cutoff as V 4 points the score sensitivity comes to 95.3%, and reduces the specificity of 100% to 94.9%. In conclusion, the use of routine assessment of the histopathological features described previously in association with the eosinophils and CD30+ cells count provides a high accuracy for CD and UC differential diagnosis. All parameters assessed here are easily performed by pathologists specialists and generalists. The next step seems to be the validation of this score in a prospective study.

As doenças inflamatórias intestinais (DII) são doenças crônicas incapacitantes com significativa morbidade. A Doença de Crohn (DC) e a Retocolite Ulcerativa (RCU) são as maiores representantes. O diagnóstico é baseado na suspeita clínica e complementado por achados endoscópicos, histopatológicos e radiológicos. Como não existe um teste padrão, o diagnóstico destas doenças permanece sendo um problema para gastroenterologistas e patologistas. O desenvolvimento de terapias mais específicas torna cada vez mais importante o diagnóstico preciso para uma escolha terapêutica individualizada. A imunidade inata está ativada de forma semelhante nas duas DII, porém na RCU os linfócitos CD4 TH2 são os principais envolvidos. Estes linfócitos expressam o receptor CD30 na sua membrana plasmática e produzem citocinas que ativam os eosinófilos. Levando em consideração o conhecimento fisiopatológico atual, este estudo tem como principal objetivo avaliar a relevância da expressão do marcador CD30 por imuno-histoquímica, a contagem de eosinófilos e as características histopatológicas no diagnóstico diferencial das DII. Foram avaliados 185 pacientes de um ambulatório especializado em um Hospital Universitário (105 com DC 80 com RCU). O diagnóstico foi estabelecido pelos critérios de Lennard-Jones, revisados por um gastroenterologista especializado e corroborado por cinco anos de seguimento. As biópsias foram realizadas por diversos profissionais na rotina assistencial, porém todas foram analisadas por um patologista com experiência em tubo digestivo. Nenhum paciente estava usando tratamento no momento da coleta das biópsias. As variáveis histopatológicas que demonstraram poder estatístico para auxiliar no diagnóstico diferencial foram a extensão do processo inflamatório para a submucosa, a presença de granuloma, erosões aftóides e variabilidade de acometimento entre os fragmentos. Avaliando estas características em conjunto, foi possível obter uma acurácia de 69.1% para o diagnóstico diferencial entre DC e RCU. Estudando o segmento mais alterado encontrou-se uma mediana de eosinófilos de 42 (25,5 – 63,5), nos pacientes com DC e 107 (67 – 123) nos pacientes com RCU (p< 0.001). Assumindo como ponto de corte um número V a 70 eosinófilos, a sensibilidade foi de 78,3% e a especificidade de 71% favorecendo o diagnóstico de RCU, a área sob a curva ROC foi de 0,767 (IC 95%: 0,696–0,838). A imuno-histoquímica com CD30 demonstrou uma mediana de 3 células CD30+ (2-6) na DC e 33 (24-52) na RCU, demonstrando uma diferença estatística significativa para o diagnóstico diferencial entre as duas doenças (p<0,001). Além disso, nos pacientes com RCU as células CD30+ estavam distribuídas mais frequentemente em agrupamentos nos centros dos folículos linfoides. O ponto de corte determinado pela curva ROC foi de 15 células marcadas (S = 97,5%, E = 94,3%, RV + = 17,1; RV = 0.03, área sob a curva: 0.967, IC 95%: 0.941 - 0.993). Todos estes parâmetros estudados tiveram capacidade discriminatória para o diagnóstico de DC e RCU. Designando um valor para cada variável, baseado no poder estatístico de cada uma e de forma a obter uma soma de 10 pontos no total, foi construído um escore histopatológico para o diagnóstico da DC. Considerando-se a soma das variáveis V 5 obteve-se uma especificidade de 100% e uma sensibilidade de 86,8%. Considerando 4 como ponto de corte do escore, a sensibilidade aumenta para 95,3%, e a especificidade reduz de 100% para 94,9%. Concluindo, o uso rotineiro da avaliação das características histopatológicas descritas em associação com a contagem de eosinófilos e células CD30+ proporciona uma alta acurácia no diagnóstico diferencial entre DC e RCU. Todos os parâmetros estudados são de fácil avaliação tanto por patologistas especialistas quanto generalistas. Sugere-se a realização de um estudo prospectivo para validação deste escore. / Inflammatory bowel diseases (IBD) are chronic disabling diseases with significant morbidity, being Crohn’s disease (DC) and Ulcerative colitis (UC) their greatest representatives. Diagnosis is based on clinical suspicious and complemented by endoscopic, histopathological and radiological findings. However, there is not a gold standard test, so it remained a problem for gastroenterologists and pathologists. The development of more targeted therapies makes it even more important to establish an accurate diagnosis for a better individualized therapeutic approach. Innate immune response is activated in both IBD, but in UC the lymphocytes CD4 TH2-like are mainly engaged, this kind of lymphocytes has CD30 expressed in their plasma membrane and produces cytokines that activate eosinophils. Considering the current pathophysiological knowledge, this study aimed to evaluate the relevance of CD30 expression by immunohistochemical, eosinophil count and histopathological features in differential diagnosis of IBD. A total of 185 patients were evaluated (105 CD/ 80 UC). Patients were followed at a specialized clinic of a university hospital, diagnosed according to Lennard- Jones criteria reviewed by a gastroenterologist expert and corroborated by five years of follow up. Biopsy samples were taken by different professionals in routine care, but analyzed by an experienced gastrointestinal pathologist. None of the patients were using treatment at the time of biopsy. Of all the pathological variables assessed, those that had statistical capacity to assist in the differential diagnosis were extension of the inflammatory process to submucosa, granuloma, aphthous erosion and variability of involvement between fragments. Evaluating these variables together, an accuracy of 69.1% in the differential diagnosis between CD and UC was found. Assessing the most altered sample the median of eosinophils was 42 (25.5 – 63.5) in CD patients and 107 (67 – 123) in UC patients (p< 0.001). Assuming a cutoff V 70 eosinophils, the sensitivity was 78,3% and specificity of 71% favoring the UC diagnosis, the area under the ROC curve was 0.767 (CI 95%: 0,696–0,838). Immunohistochemical CD30+ cells presented with a median of 3 cells (2-6) in CD and 33 cells (24-52) in UC, demonstrating a highly significant statistical difference between the two diseases (p<0.001). Besides, CD30+ cells were distributed most clustered in the center of lymphoid follicles in UC patients. The cutoff determined by ROC curve was 15 (S = 97.5%, E = 94.3%, LR + = 17.1;-RV = 0.03, AUC: 0.967, 95% CI: 0.941 - 0.993). All these parameters studied had discriminatory capacity for diagnosis of CD and UC. A value was assigned to each variable based on the statistical power of each, making a total sum of 10 points to build a histopathological score for the CD diagnosis. Setting the cutoff as V 5 we found a specificity of 100% and a sensitivity of 86.8%. When we consider as cutoff as V 4 points the score sensitivity comes to 95.3%, and reduces the specificity of 100% to 94.9%. In conclusion, the use of routine assessment of the histopathological features described previously in association with the eosinophils and CD30+ cells count provides a high accuracy for CD and UC differential diagnosis. All parameters assessed here are easily performed by pathologists specialists and generalists. The next step seems to be the validation of this score in a prospective study.