Background and aims: This study evaluated the minimal clinically important short-term improvement in the Modified Multiplier Simple Endoscopic Score for crohn's Disease [MM-SES-CD], a novel modified scoring system of the SES-CD, which reliably predicted 1-year endoscopic remission [ER].
Methods: This post-hoc analysis of two CD clinical trial programmes pooled data of 198 participants with baseline ulcers and SES-CD ≥3, who had baseline, post-induction [8-12 weeks], and 1-year endoscopic assessments. Different cut-off values for endoscopic response were evaluated using receiver operating characteristic [ROC] curves, positive likelihood ratios [PLR], and negative likelihood ratios [NLR]. ER [SES-CD <3] was the binary classifier in all cases. A distribution of cut-offs minimising NLR and maximising PLR was created with 10 000 bootstrapped resamples. An optimal cut-point for low and high probability of 1-year ER was determined based on the maximum Youden Index.
Results: MM-SES-CD ≥40% reduction from baseline was selected as the cut-off maximising PLR and minimising NLR. Among 7.6% [15/198] participants achieving this cut-off post-induction, 1-year ER was 46.7%. One-year ER was 16.9% among those not achieving this cut-off. This threshold predicted 1-year ER with 95.0% (95% confidence interval [CI] 90.4%-97.8%) specificity and a PLR of 3.7 [95% CI 1.4-9.5], which was higher than traditional endoscopic response criteria of SES-CD ≥50% reduction [specificity 62.5%, 95% CI 54.5%-70.0%; PLR 1.9, 95% CI 1.4-2.5]. Lower thresholds of MM-SES-CD reduction also were highly specific for 1-year ER [e.g., MM-SES-CD ≥20% reduction was achieved in 19.7% of patients with 83.1% specificity].
Conclusions: In CD patients, post-induction endoscopic response defined by MM-SES-CD ≥40% reduction from baseline identified patients most likely to achieve 1-year ER.
Keywords: Crohn’s disease; SES-CD; endoscopic remission; inflammatory bowel disease.
© The Author(s) 2021. Published by Oxford University Press on behalf of European Crohn’s and Colitis Organisation. All rights reserved. For permissions, please email: email@example.com.