To Örebro University

Background: The disease course of patients with newly diagnosed ulcerative colitis (UC) is highly uncertain, and there is a lack of validated prognostic biomarkers that could aid in clinical decision making.

Methods: Newly diagnosed, mainly treatment naïve patients with UC from three large inception cohorts were used to develop and validate a serum proteomics-based risk score for prognostication of disease course during the first year from diagnosis. In the discovery cohort (n = 161) and validation cohort 1 (n = 186) an aggressive disease course was defined as the presence of any IBD-related surgery, hospital admission for active disease, treatment refractoriness towards targeted therapies (i.e. biologics, JAK-inhibitors or S1P receptor modulators), and >2 courses or high cumulative doses of systemic corticosteroids. In validation cohort 2 (n = 120), an aggressive disease course was defined as the need for a biologic, ciclosporin or surgery. 178 proteins were measured on Olink platforms, and a machine learning algorithm (i.e. regularised regression) was applied to the discovery cohort to develop an UC risk score comprising 23 proteins. The performance of the UC risk score was assessed in the two external validation cohorts. For validation cohort 2, a condensed version of the UC risk score was applied, as only 14 of the original 23 proteins were available. Cox regression estimated hazard ratios (HR) for the association between the UC risk score at diagnosis, time to escalation to targeted therapy (validation cohort I) and time to the defining episode of an aggressive disease course (validation cohort II).

Results: Based on univariate analyses, we identified 59 proteins associated with an aggressive disease course in the discovery cohort (PFDR <0.10; Figure 1A). Twenty could be validated in validation cohort 1, and nine remained in validation cohort 2 (PFDR <0.10; Figure 1B-C).

In the discovery cohort, the machine learning model showed a high predictive capacity, with an area under the curve (AUC) of 0.81 (Figure 1D) and was numerically superior compared with a clinical model comprising sex, age and CRP (AUC = 0.72). Next, the performance of the UC risk score was confirmed in the two external validation cohorts, displaying AUC:s of 0.77 (Figure 1E-F). Patients with a higher UC risk score at diagnosis had increased risk of initiating targeted therapy (HR 4.26, 95% CI 1.91–9.49; Figure 2A). The HR for having an aggressive disease course was 9.12 (95% CI 3.04–27.3; Figure 2B).

Conclusion: We develop a UC risk score for prognostication of disease course and confirmed its high predictive performance by external validation in two independent cohorts. The risk score is a promising tool for quantifying the risk of having an aggressive disease course in UC.

Background: The clinical course of inflammatory bowel disease (IBD) is highly heterogeneous. Previously we identified a protein-based prognostic signature in serum using the proximity extension assay (PEA). To enable clinical translation, further assay development is required, including the establishment of standard curves and clinically actionable cut-offs. Therefore, we aimed to develop, optimise and validate a serum-based assay that quantifies absolute protein concentrations at diagnosis to identify individuals at higher risk of an aggressive course of IBD.

Methods: Proteins were selected from the commercially available Olink inflammation and oncology II panels as well as from five custom-made multiplex panels comprising 460 proteins based on genes in 163 IBD-risk loci (the IBD Character project1). Serum from 71 newly diagnosed IBD patients (Crohn’s disease, n = 29; ulcerative colitis, n = 42) (SIC-IBD) was analysed. Proteins associated with an aggressive course: I) need for biologics, ciclosporin, or surgery for flare after initial induction or II) hospitalisation, refractoriness to targeted therapy, excessive corticosteroid use or surgery within one year, were identified using penalised logistic regression. From these proteins a focused prognostic panel was developed and analytically optimized for absolute quantification. Model performance was evaluated by nested cross-validation (area under the curve [AUC]) and externally validated in 329 treatment-naïve IBD patients from the population-based IBSEN III cohort. Time to the composite outcome was analysed by estimating hazard rations (HR) using Cox regression.

Results: Thirteen proteins (CCL19, XPNPEP2, TGFα, VGFA, DLL1, PSGL1, OSM, CSF1, FURIN, IL-18, EpCAM, and TNFRSF9) met the pre-defined criteria and formed the final prognostic signature. In the discovery cohort, the model achieved an AUC of 0.76, with no interaction with age, sex or IBD subtype. External validation in the IBSEN III cohort yielded an AUC of 0.66 (95%CI 0.57-0.75), sensitivity 49% and specificity 70% at the discovery cohort derived optimal cut-off. Among patients classified as high risk, compared to low risk, the HR for the composite endpoint of aggressive disease course was 1.81 (P = 0.0002, Figure 1).

Conclusion: We developed and validated the first PEA-based serum panel enabling absolute quantitation of prognostic proteins for clinical use. At diagnosis, IBD-patients classified as high risk based on the 13-protein signature had an 81% higher relative risk of an aggressive disease course, supporting further evaluation.

Background: Inflammatory bowel diseases (IBDs) are complex conditions marked by chronic inflammation in the gastrointestinal tract. Traditional classification separates IBD into Crohn's disease and ulcerative colitis, but this division may not fully capture disease heterogeneity. Here, we examine whether microbiome-driven subtyping can describe novel clinical IBD phenotypes. To achieve this, we applied unsupervised clustering to fecal microbiota profiles from the population-based Inflammatory Bowel Disease in South-Eastern Norway III (IBSEN III) cohort.

Methods: A Gaussian Mixture Model (GMM) was used to cluster participants with IBD based on microbiome composition and examine associations between clusters and clinical outcomes, including inflammatory markers and disease severity during the first year after inclusion.

Results: Three microbiome-based clusters were identified: CLO (dominated by Clostridia UCG-014), ALF (Agathobacter, Lachnoclostridium, and Faecalibacterium), and RUM (Ruminococcus gnavus). Participants in the RUM cluster had a higher risk of future severe disease than those in the CLO cluster, even among participants with remission-to-mild disease at inclusion (21% vs. 6%, P < 0.00001). This association could not be explained by antibiotic use or baseline disease severity. Cluster membership alone performed comparably to fecal calprotectin in distinguishing severe disease, and a combined model significantly improved accuracy (P < 0.0001).

Conclusion: Our findings demonstrate a connection between microbiome composition and the risk of severe disease development, which is partly independent of inflammation levels at the time of sampling. Microbiome-informed subgrouping could lead to more personalized treatment strategies. Further validation is needed to determine the clinical utility of these clusters.

BACKGROUND AND AIM: Despite the well-established involvement of the gut microbiome in inflammatory bowel disease (IBD), less is known about how the gut microbiome changes over time and how it varies with clinical disease activity and fecal calprotectin (f-calprotectin). To address this gap, we utilized samples from the population-based inception cohort of the Inflammatory Bowel Disease in South-Eastern Norway III (IBSEN III) study.

METHODS: Data and stool samples from study participants with IBD and symptomatic controls were collected at diagnosis and after 3, 6, and 12 months. Microbiome profiling of stool samples was performed targeting the V3-V4 region of the 16S rRNA gene, and a consensus-based approach of mixed models was employed for the longitudinal microbiome analysis.

RESULTS: We included 1251 samples from 744 patients with ulcerative colitis, 618 samples from 356 patients with Crohn' s disease and 266 samples from 164 symptomatic non-IBD controls. In the IBD population, we observed that levels of f-calprotectin decreased over time, as did the patient-reported disease activity (P < .001). Distinct changes in the gut microbiome of IBD patients were observed throughout the first year, such as increased alpha diversity (P < .001) and significant taxonomic changes.Notably, there was no covariation between the changes in alpha diversity and f-calprotectin or symptom score.

CONCLUSION: The gut microbiome during the first year after IBD diagnosis showed changes that paralleled inflammation and clinical disease activity, albeit without covariation, suggesting that there may be a disease-driving impact of gut microbiome independent of inflammation and inflammation-driven symptoms.

Background: Limited statistical power has hampered previous estimates of concordance between relatives and heritability in inflammatory bowel diseases (IBD). To examine the genetic component of Crohn’s disease and ulcerative colitis, we established the largest nationwide IBD twin cohort to date and assessed estimates of concordance and heritability.

Methods: We used the Swedish Twin Registry to identify all twins from complete pairs with known zygosity born between 1886 and 2004. The Swedish National Patient Register was used to identify all patients diagnosed with IBD. We calculated proband concordance rates and fitted a model estimating explained variance in diseases due to genetics (i.e., the heritability), environment shared between twins, and environment unique to each twin.

Results: A cohort of 111,080 twins was followed until a median age of 62.2 years, during which 964 individuals were diagnosed with IBD (Table 1). The proband concordance rate for Crohn’s disease was 0.30 in monozygotic pairs and 0.02 in dizygotic pairs (Table 2). The corresponding rates for ulcerative colitis were 0.15 and 0.03. After adjusting for sex and birth year, heritability was estimated to be 0.78 (95% CI: 0.68–0.87) for Crohn’s disease and 0.57 (95% CI: 0.46–0.69) for ulcerative colitis.

Conclusion: In this large population-based twin study, the heritability of Crohn’s disease was 0.78 and 0.57 for ulcerative colitis. These findings highlight the disparity between heritability estimates from twin studies and those inferred from genome-wide association studies, underscoring the need for continued exploration of the genetic basis of IBD.

BACKGROUND AND AIMS: Limited statistical power has hampered previous estimates of concordance between relatives and heritability in inflammatory bowel diseases (IBD). To examine the genetic component of Crohn's disease and ulcerative colitis, we established the largest nationwide IBD twin cohort to date and assessed estimates of concordance and heritability.

METHODS: We used the Swedish Twin Registry to identify all twins from complete pairs with known zygosity born between 1886 and 2004. The Swedish National Patient Register was used to identify all patients diagnosed with IBD. We calculated proband concordance rates and fitted a model estimating explained variance in diseases due to genetics (ie, the heritability), environment shared between twins, and environment unique to each twin.

RESULTS: A cohort of 111 080 twins was followed until a median age of 62.2 years, during which 964 individuals were diagnosed with IBD. The proband concordance rate for Crohn's disease was 0.30 in monozygotic pairs and 0.02 in dizygotic pairs. The corresponding rates for ulcerative colitis were 0.15 and 0.03. After adjusting for sex and birth year, heritability was estimated to be 0.78 (95% CI: 0.68-0.87) for Crohn's disease and 0.57 (95% CI: 0.46-0.69) for ulcerative colitis.

CONCLUSION: In this large population-based twin study, the heritability of Crohn's disease was 0.78 and 0.57 for ulcerative colitis. These findings highlight the disparity between heritability estimates from twin studies and those inferred from genome-wide association studies, underscoring the need for continued exploration of the genetic basis of IBD.

Background: Minimally invasive biomarkers that can reliably distinguish adults with inflammatory bowel disease (IBD) from symptomatic controls are still lacking. We investigated whether serum lipidomics can provide a reproducible diagnostic signature for IBD and how such a signature performs in relation to relevant inflammatory markers.

Methods: Untargeted lipidomic profiling was conducted on serum from the Swedish SIC-IBD inception cohort, comprising adults referred for suspected IBD, symptomatic controls, as well as healthy controls. Candidate diagnostic lipids were identified using supervised machine-learning models in the discovery cohort and then evaluated in the independent, population-based Norwegian IBSEN III inception cohort of treatment-naïve adults. We examined the diagnostic performance of the lipidomic signature in combination with high-sensitivity C-reactive protein (hs-CRP) and fecal calprotectin (FCP).

Results: The discovery cohort included 96 IBD patients, 66 non-IBD symptomatic controls, and 48 healthy controls, whereas the validation cohort comprised 349 patients with IBD and 198 symptomatic controls (Table 1). In the discovery cohort, a serum signature of hs-CRP and the two top lipids, lactosyl ceramide (d18:1/16:0) and phosphatidylcholine (O-44:5), demonstrated a high diagnostic performance (area under the curve [AUC] 0.80, 95% CI 0.73-0.87) compared with hs-CRP alone (AUC 0.70, 95% CI 0.63-0.79). When applied to the validation cohort, hs-CRP combined with the serum lipid signature significantly enhanced discrimination between IBD and symptomatic controls (AUC 0.79, 95% CI 0.75-0.83) compared with hs-CRP alone (AUC 0.68, 95% CI 0.64-0.73; P < 0.0001). As illustrated in Figure 1, among patients with available stool samples, combining FCP with hs-CRP and the lipid signature numerically improved diagnostic accuracy (AUC 0.89, 95% CI 0.86-0.92) compared with FCP alone (AUC 0.86, 95% CI 0.83–0.90).

Conclusion: We identified and externally validated a serum lipidomic signature that improves the diagnostic prediction of IBD when combined with hs-CRP. Although the blood-based model of hs-CRP, lactosyl ceramide (d18:1/16:0), and phosphatidylcholine (O-44:5) did not outperform FCP alone, it approached its diagnostic performance and further enhanced accuracy when integrated with FCP. These findings support added value these lipids and may offer insights into lipid-related pathways underlying IBD pathogenesis.

Background: IgG autoantibodies against integrin αvβ6 have been associated with preclinical ulcerative colitis (UC),1,2 but data on their emergence and validation are still scarce. Therefore, we aimed to establish and validate a cutoff of anti-αvβ6 for predicting future UC across two independent preclinical cohorts. Additionally, we examined their development in early-life, assessed the influence of genetic-, environmental risk factors and inflammation on anti-αvβ6.

Methods: Within large population-based cohorts (n > 180,000), prediagnostic blood samples from individuals who later in life were diagnosed with UC were compared with age- and sex-matched controls who remained UC-free during follow-up (Figure 1). To explore early-life presence and genetic and shared environmental influences, we analysed serum from ABIS, a population-based birth cohort (n = 17,055) and the Swedish Twin Registry (n > 97,000). Anti-αvβ6 levels were measured using an in-house automated fluorescence enzyme immunoassay. Relative estimates of proteins were measured on the Olink platform3.

Results: In the discovery cohort (Table 1), anti-αvβ6 antibodies differentiated preclinical UC from controls (area under the curve [AUC]=0.68). Applying the model to the validation cohort yielded an AUC of 0.79, with a sensitivity of 46% and a specificity of 93%, at the optimal cut-off (109 UA/l) derived from the discovery cohort. Predictive capacity increased closer to diagnosis but remained significant for samples collected ≥20 years before UC diagnosis (AUC=0.67). In the birth cohort, anti-αvβ6 levels at birth appeared elevated in children who later developed UC (OR = 2.13; 95%CI, 0.84-5.38), likely reflecting transplacental antibody transfer, but also after 5 years (OR = 2.58; 95%CI, 0.83-8.03) and 8 years (OR = 2.50; 95%CI, 0.30-21.00). When assessing the level of agreement within twin pairs, low intra-class correlation coefficients were observed irrespective of zygosity and disease status, suggesting limited influence of genetic and shared environmental exposures. However, anti-αvβ6 levels differed across smoking categories in the preclinical cohorts, but inclusion of smoking status added little to model performance (AUC=0.82). By calculating a composite proteomic score, a progressive increase in the score was observed from the lowest to highest quartile of αvβ6 levels in preclinical UC, whereas most controls were found in the lowest quartiles for both the proteomic score and autoantibody levels.

Conclusion: Anti-αvβ6 autoantibodies represent a reproducible biomarker for preclinical UC. Their early-life presence, environmental modulation, and correlation with immuno-inflammatory proteins highlight their potential utility for risk stratification.

Background: Paediatric inflammatory bowel disease (IBD), comprising the subtypes Crohn’s disease (CD) and ulcerative colitis (UC), is believed to result from an interplay between genetic predisposition and environmental exposures. Examining exposome metabolome may help to identify environmental exposures, including host-derived metabolites, with potential relevance in paediatric IBD.

Methods: Relative concentrations of metabolites were measured in serum samples from treatment-naïve patients with IBD (N = 80) and symptomatic, non-IBD controls (N = 37) included in the IBSENIII cohort (Table 1). Ultra-high-performance liquid chromatography–mass spectrometry (UHPLC-MS) was performed. Authentic standards or the Human Metabolome Database were used for annotation. Associations with IBD, with CD (N = 53), and with UC including IBD-unclassified (N = 27), were assessed using logistic regression while adjusting for sex, age, and BMI. In addition, correlations with serum proteins (proximity extension assay technology) were examined. A false discovery rate (PFDR) threshold of < 0.1 was applied.

Results: We observed positive association of aryl sulfate (phenol sulfate; PFDR=0.03) with IBD versus symptomatic controls (Figure 1). Within the IBD population, aryl sulfate was positively correlated with several serum proteins, including CCL20 (r = 0.48, PFDR=0.004), MCP-3 (r = 0.47, PFDR=0.004), and TNF-α (r = 0.45, PFDR=0.001). We observed inverse associations of four perfluoroalkyl substances (PFAS); (linear-perfluorooctane sulfonate (PFOS-L), branched-perfluorohexanesulfonate (PFHxS-B), perfluorooctanoate (PFOA), and branched-perfluorooctane sulfonate (PFOS-B)) with IBD, UC, or both. In IBD, PFOA, PFOS-L, PFHxS-B positively correlated with Delta/Notch-like epidermal growth factor (EGF)-related receptor (DNER) protein levels. The secondary bile acid 6-ketholithocholic acid was inversely associated with UC, while the primary bile acids glycochenodeoxycholic acid-3-O-sulfate, cholic acid and glycocholic acid were positively associated with UC (all PFDR<0.1).

Conclusion: Several metabolites including aryl sulfate were associated with paediatric IBD and correlated with inflammation-related proteins. These findings confirm recent results from preclinical Crohn’s disease1 and may point towards a role of aryl sulfate at IBD diagnosis.

BACKGROUND AND AIMS: The diagnostic and prognostic properties of anti-integrin αvβ6 IgG autoantibodies in ulcerative colitis (UC) are poorly understood. We aimed to assess the diagnostic performance of anti-integrin αvβ6 autoantibodies and examine their association with disease outcomes.

METHODS: Serum samples from a Swedish inception cohort of patients with suspected inflammatory bowel disease (IBD, n=473) were analysed using an in-house fluorescence enzyme immunoassay based on EliA™technology. Findings were validated in a Norwegian population-based inception cohort (n=570). Diagnostic performance was assessed by calculating the area under the curve (AUC) with 95% confidence intervals (CIs) and determining sensitivity and specificity. Reclassification was evaluated using the net reclassification index.

RESULTS: In the discovery cohort, patients with UC, IBD-unclassified, or colonic Crohn's disease exhibited higher median autoantibody levels compared to symptomatic and healthy controls. In the validation cohort, the autoantibody demonstrated 79% sensitivity and 94% specificity for UC vs symptomatic controls at a cut-off of 400 UA/l. Its diagnostic performance (AUC=0.92, 95%CI 0.89-0.95) was superior to hs-CRP (AUC=0.65, 95%CI 0.60-0.70, P<0.001) and faecal calprotectin (fcalpro) (AUC=0.88, 95%CI 0.84-0.92, P=0.09). Combining the autoantibody with fcalpro further improved diagnostic accuracy (AUC=0.97, 95%CI 0.95-0.98) and patient reclassification (P<0.001). Autoantibody positivity was associated with a severe phenotype of UC, characterised by increased inflammatory activity and higher IL-17A and granzyme B levels. Higher autoantibody levels were linked to an aggressive disease course, remaining stable in aggressive UC but decreasing in indolent disease (P=0.003).

CONCLUSIONS: Anti-integrin αvβ6 is a reliable diagnostic and prognostic marker for UC, with potential clinical implementation.