Predictive Modeling of Postoperative Crohn’s Recurrence, and Biologic Prophylaxis

May 21, 2026

Issue 10

Clinical Question

Can a clinical model predict early recurrence after ileocolonic resection in Crohn’s disease and can biologics prevent this complication?

Editor’s Bottom Line

This predictive model supports a personalized approach to post surgical prophylaxis and monitoring in patients with Crohn’s disease undergoing ileocolonic resection.

Reference

Allez M, Bak MTJ, Brand S, et al. Development and external validation of a predictive model for postoperative recurrence of Crohn’s disease in the biologic era. Clin Gastroenterol Hepatol. Epub ahead of print March 27, 2026. https://doi.org/10.1016/j.cgh.2026.03.015

Synopsis

Post-ileocolonic resection endoscopic recurrence, defined as a modified Rutgeerts score (mRS) ≥i2b at the neoterminal ileum, precedes clinical relapse and is strongly associated with long-term adverse outcomes. While current guidelines risk-stratify patients using clinical factors to guide prophylactic treatment decisions, the evidence for several established risk factors has been challenged and the impact of prior biologic exposure has not been systematically incorporated into validated predictive tools.

To update a clinical predictive score, researchers at 15 French centres prospectively examined 632 adults with ileal CD undergoing ileocolonic resection between 2010 and 2023. They assessed endoscopic recurrence (mRS ≥i2b) within one year of surgery (median 6.4 months post-operatively). Patients had disease for a median 7.8 years prior to surgery and were a median 34 years of age at the time of surgery. Seventy-eight percent of patients had received a biologic pre-operatively and 44% received postoperative biologic prophylaxis. Early post-operative endoscopic recurrence occurred in 237 patients (37.5%), with 16.9% of patients experiencing severe recurrence (mRS i3 or i4) within a year.

Multivariate logistic regression analysis identified four independent risk factors for ER: male sex (Odds Ratio [OR]: 1.97; 95% Confidence Interval [CI], 1.38–2.83; p <0.001), active smoking (OR: 2.37; 95% CI, 1.56–3.62; p <0.001), prior ICR (OR: 1.78; 95% CI, 1.13–2.82; p=0.014), and prior exposure to one or more biologics (OR for 1 prior biologic: 2.15; 95% CI,1.37–3.42; p=0.001 and OR for more than 1 biologic: 2.41; 95% CI, 1.39–4.22; p=0.002). Three additional variables; age at diagnosis >40 (OR: 1.87; 95% CI, 0.96–3.69; p=0.068), presence of a perioperative temporary stoma (OR: 1.62; 95% CI, 0.96–2.74; p=0.069), and B2 disease phenotype (OR: 0.45; 95% CI, 0.20–1.04; p=0.060), did not reach statistical significance, but were retained in the final model on the basis of their contribution to overall model fit.

Postoperative biologic prophylaxis was also independently associated with a reduction in early endoscopic recurrence risk (OR: 0.31; 95%, CI 0.20–0.45; p <0.001), while post-operative thiopurine prophylaxis did not have a significant effect on recurrence (OR: 0.82; 95% CI, 0.46–1.42). No statistically significant interaction was detected between biologic prophylaxis and any of the prognostic factors, suggesting that biologics reduced recurrence rates across all risk strata, including among non-smokers and biologic-naïve patients.

The prognostic model integrating all eight variables, including post-operative prophylaxis, achieved an Area Under the Curve (AUC) of 0.72 (95% CI, 0.67–0.76). The investigators applied the model to two independent external datasets and yielded AUCs of 0.67 (95% CI, 0.61–0.72) and 0.66 (95% CI 0.59–0.73), or 0.70 (95% CI, 0.65–0.73) when the external cohorts were pooled.

Of 458 patients with long-term follow-up (median 3.4 years), clinically significant recurrence, defined as a composite of severe endoscopic or radiologic inflammation, treatment escalation, hospitalization, or surgical recurrence, occurred in 22.9% at 3 years, 36.5% at 5 years, and 53.2% at 10 years. A multivariate analysis adjusting for early modified Rutgeerts score and preoperative risk factors found that only mRS at six months remained independently associated with clinically significant recurrence. Patients with an mRS i0 had a median event-free survival of over 10 years, compared to a median of 3.6 years among those with an mRS i4.

Details

Study Design: Prospective multicenter cohort study with external validation
Funding: Helmsley Charitable Trust, Association François Aupetit (AFA), Takeda, Johnson & Johnson (Janssen Cilag), MSD
Allocation: Non-randomized
Setting: Multicenter
Level of Evidence: 2b