Abstract:Abstract Background Despite the significant progress in IBD care with biological therapies, the current 'trial-and-error' method remains suboptimal. Previous results from our EPIC-CD study demonstrated that two distinct epigenetic panels of 25 and 68 DNA methylation markers, respectively, were associated with (non-)response to vedolizumab (VDZ) and ustekinumab (USTE) in active Crohn’s disease (CD)1. Data from clinical trials have demonstrated that response rates in patients naïve to biological treatment are higher. We therefore aimed to evaluate the performance of our models in bionaive and anti-TNF exposed subpopulations. Methods We prospectively measured peripheral blood DNA methylation profiles from 184 adult CD patients prior to treatment with VDZ or USTE in a discovery (n=126) and independent validation (n=58) cohort using the Illumina EPIC BeadChip array. Between 26 and 52 weeks, patients were categorized as responders or non-responders based on a combination of endoscopic response (≥50% reduction in SES-CD score), steroid-free clinical response (≥3-point drop in Harvey-Bradshaw Index (HBI) or HBI ≤4 AND no systemic steroids), and/or biochemical response (≥50% reduction in C-reactive protein (CRP) and fecal calprotectin or a CRP ≤5 g/mL and fecal calprotectin ≤250 μg/g). Epigenetic biomarker identification and classification analyses were conducted through stability selection gradient boosting on the discovery cohort. Subsequently, the performance of the identified epigenetic biomarkers was assessed independently on the validation cohort, with patients stratified by prior anti-TNF exposure. Results Out of the 58 patients in the validation cohort, 25 received treatment with VDZ, and 33 with USTE, with 52% (n=13) and 36% (n=12) respectively naïve to anti-TNF treatment. Baseline characteristics between anti-TNF naive and anti-TNF exposed patients showed no significant differences, despite a slight significant age difference in the vedolizumab group, where anti-TNF exposed patients were significantly younger (table 1). In the overall independent validation cohort, the prediction models demonstrated an AUC of 0.75 for both VDZ and USTE. However, after stratification by prior anti-TNF exposure, we observed increased performance of our models among anti-TNF naive patients for both VDZ (AUCnon-exposed=0.85 versus AUCexposed=0.66) and USTE (AUCnon-exposed=0.97 versus AUCexposed=0.63 (figure 1). Conclusion Our findings demonstrate that prior anti-TNF exposure leads to somewhat reduced predictive power of our epigenetic models for Vedolizumab and Ustekinumab response. The predictive models may therefore find its optimal application in bio-naive patients, although they remain significant in anti-TNF exposed patients.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: when using biologics to treat active Crohn's disease (CD), the existing "trial - and - error" treatment strategy has an unsatisfactory effect. Specifically, the study aims to evaluate the predictive ability of two specific epigenetic marker panels (25 and 68 DNA methylation markers respectively) for the treatment responses to vedolizumab (VDZ) and ustekinumab (USTE), especially in the subgroups of patients who are anti - TNF naïve and anti - TNF exposed. ### Background Although biologic therapies have made significant progress in the care of inflammatory bowel disease (IBD), the current "trial - and - error" treatment method is still not optimized. Previous studies have shown that two different epigenetic marker panels (25 and 68 DNA methylation markers) are associated with the treatment responses to vedolizumab (VDZ) and ustekinumab (USTE). Clinical trial data show that the response rate is higher in patients who have not received biologic treatment. Therefore, this study aims to evaluate the performance of these models in patients who are anti - TNF naïve and anti - TNF exposed. ### Methods The researchers prospectively measured the peripheral blood DNA methylation profiles of 184 adult CD patients before receiving VDZ or USTE treatment, divided into a discovery cohort (n = 126) and an independent validation cohort (n = 58), and analyzed them using the Illumina EPIC BeadChip array. According to endoscopic response (SES - CD score reduction ≥ 50%), steroid - free clinical response (Harvey - Bradshaw index reduction ≥ 3 points or HBI ≤ 4 and no systemic steroids), and/or biochemical response (CRP reduction ≥ 50% or CRP ≤ 5 mg/L and fecal calprotectin ≤ 250 µg/g), patients were classified as responders or non - responders within 26 to 52 weeks. Epigenetic biomarker identification and classification analysis were performed in the discovery cohort by the stability selection gradient boosting method, and their performance was evaluated in the independent validation cohort, stratified according to whether they had received anti - TNF treatment. ### Results Among the 58 patients in the independent validation cohort, 25 received VDZ treatment and 33 received USTE treatment, of which 52% (n = 13) and 36% (n = 12) were anti - TNF naïve patients respectively. Overall, the AUCs of the prediction models in the independent validation cohort were 0.75 for VDZ and 0.75 for USTE. However, after stratification according to whether they had received anti - TNF treatment, the performance of the prediction models in anti - TNF naïve patients was significantly improved, with the AUC of VDZ increasing from 0.66 to 0.85 and the AUC of USTE increasing from 0.63 to 0.97. ### Conclusions The study shows that the history of previous anti - TNF treatment reduces the predictive ability of epigenetic models for the treatment responses to vedolizumab and ustekinumab. Therefore, these prediction models may be most suitable for application in patients who have not received biologic treatment, although they are still statistically significant in patients who have received anti - TNF treatment. ### Formula The main formula involved in this paper is the Area Under the Curve (AUC) of the receiver operating characteristic curve, which is used to evaluate the predictive performance of the model: \[ \text{AUC} = \int_0^1 \text{TPR}(FPR) \, dFPR \] where TPR (true positive rate) and FPR (false positive rate) represent the sensitivity and specificity at different thresholds respectively. ### Summary This study provides new ideas and tools for optimizing personalized treatment of Crohn's disease by evaluating the predictive ability of epigenetic markers in different treatment contexts.

P592 Higher predictive power of epigenetic signatures for response to vedolizumab and ustekinumab in anti-TNF naïve patients with active Crohn’s disease

P181 Combination of EPIC-CD epigenetic signatures accurately identifies non-response to multiple biologics in patients with active Crohn’s disease

OP03 Highly stable epigenome-wide peripheral blood DNA methylation signatures accurately predict endoscopic response to adalimumab, vedolizumab and ustekinumab in Crohn’s disease patients: The EPIC-CD study

Peripheral blood DNA methylation signatures predict response to vedolizumab and ustekinumab in adult patients with Crohn's disease: The EPIC-CD study

P509 Biomarkers of neutrophil activity and extracellular matrix turnover predict long-term response to vedolizumab in patients with Crohn’s disease

P1042 Serum N-glycan Biomarkers Predict Patient Response to Biologics for Crohn’s Disease

P695 Untargeted proteomics analysis of baseline serum samples prior to biologic therapy initiation

Limited predictive value of the gut microbiome and metabolome for response to biological therapy in inflammatory bowel disease

P518 Dietary and non-dietary predictors of treatment response to adalimumab in anti-TNFα-naïve adults with Crohn’s disease

OP05 High-dimensional single-cell analysis identifies cellular signatures associated with response to vedolizumab therapy in ulcerative colitis

P544 Low TNF production and low TREM1 expression as negative predictors for remission under adalimumab therapy in Crohn′s Disease patients

Serological Biomarkers of Extracellular Matrix Turnover and Neutrophil Activity Are Associated with Long-Term Use of Vedolizumab in Patients with Crohn's Disease

P206 Metabolomic predictors of response to vedolizumab in Inflammatory Bowel Disease

P1020 Developing and Validating a Clinical Prediction Model Using Biomarkers to Assess Clinical Response and Severity of Crohn's Disease 12 Weeks After Initiation of Biological Therapy

DOP22 Predictive biomarkers of therapeutic response in Inflammatory Bowel Disease: a step towards personalized medicine

P266 Predictive tissue-based biomarkers in treatment of Inflammatory Bowel Disease with biologicals and small molecules: a systematic review

P803 iGenoMed-MTT: A prospective multiomic study of response to molecularly targeted therapies in IBD

Key outcomes '99: gone west?

P852 Metataxonomic and metabolomic profiling to predict response to anti-TNF therapy in Crohn’s Disease

P915 Multi-Omics Biomarkers for the Prediction of Response to Biologics in Patients with Inflammatory Bowel Disease

Single-cell characterization of peripheral blood mononuclear cells from vedolizumab-treated patients with Crohn's disease identifies response-associated differences among the plasmacytoid dendritic cell and classical monocyte populations.