Multimodal monitoring of pan-enteric mucosal healing and biologic persistence in Crohn's disease-the MILESTONE-CD prospective cohort.

BACKGROUND

Mucosal healing (MH) is associated with improved long-term outcomes in Crohn's disease (CD), but longitudinal pan-enteric data are limited.

METHODS

This prospective observational study enrolled adults with CD initiating first-line biologic who were assessed at weeks 0, 14, 26, 52, 78, and 104 using fecal calprotectin (FCP), intestinal ultrasound (IUS), and pan-enteric video capsule endoscopy (VCE). The primary outcome was biologic discontinuation for any reason within 104 weeks. Secondary outcomes included MH (Lewis score < 135 and Eliakim score = 0) and flare-driven biologic discontinuation. Analyses used generalized estimating equations (GEE), Kaplan-Meier, and Cox models.

RESULTS

Among 59 patients (median age-30 [23-43] years, male-52.5%, inflammatory phenotype-79.7%), 35 received anti-TNF, 21 vedolizumab, and 3 ustekinumab. Overall, 29 (49.2%) patients discontinued biologics after 39 (28-52) weeks, including 21 (35.6%) flare-driven discontinuations. With non-responder imputation, pan-enteric MH increased from 3.4% to 16.9%, 18.6%, and 32.2% at weeks 14, 52, and 104 (P < .001), consistent on observed-data GEE. Week-14 log-transformed FCP was most strongly associated with week-52 MH (area under the curve 0.804, 95% CI, 0.685-0.923, P = .002, optimal cutoff 172 mcg/g; lagged GEE odds ratio [OR] 0.59, 0.39-0.89, P = .012), whereas same-visit transmural remission (bowel wall thickness ≤ 3mm) showed the strongest association with MH (OR 4.97, 1.91-12.91, P = .001). Week-14 Lewis score ≥ 350 (Hazard ratio [HR] 3.18, 95% CI, 1.16-8.73, P = .025) and perianal disease (HR 3.93, 95% CI, 1.45-10.62, P = .007) were independently associated with flare-driven discontinuation.

CONCLUSION

Pan-enteric MH accrues incrementally beyond year 1. Integrated FCP, IUS, and VCE assessments may optimize individualized biologic decisions in CD.