Early mucosal healing predicts favorable outcomes in patients with moderate to severe ulcerative colitis treated with golimumab: Data from the real-life BE-SMART cohort

Abstract

Background Golimumab (GOL) is registered for moderate to severely active ulcerative colitis (UC). Data on the use of GOL in daily clinical practice are limited. Currently, it is unclear which factors are predictive of a favorable outcome. The goals of this study were to evaluate the mid-term outcome of GOL (week 26) in patients with moderate to severe UC and to determine predictors of favorable outcome. Methods Patients included in the SMART study (NCT02155335) were evaluated for their mid-term outcome. Demographic data, disease characteristics, and medical history were recorded retrospectively. Data on disease activity based on total Mayo score, previous and concomitant medication, GOL dosing, mucosal healing (Mayo 0 or 1), adverse events (colectomy, hospitalization), and biomarkers (C-reactive protein, fecal calprotectin, hemoglobin, and albumin) were collected at baseline and weeks 2, 6, 14, 26, and 52. GOL was dosed at 200 and 100 mg at weeks 0 and 2, respectively, and 50 mg (<80 kg body weight) or 100 mg (≥80 kg body weight) every 4 weeks thereafter. The primary end point was steroid-free GOL continuation at week 26. Results From the 91 evaluable patients (42% female; median age, 42 years; median disease duration, 5 years), 4% were active smokers, 25% had extensive colitis, and 38% had an endoscopic Mayo score of 3 at baseline. The median (interquartile range [IQR]) baseline Mayo score was 9 (8-10). Although 75% of patients had previously failed immunomodulators (IMMs), the majority (87%) were anti-tumor necrosis factor (TNF) naïve. GOL was started in combination with IMM in 40% and steroids in 64%. The median (IQR) duration of GOL therapy during follow-up was 35.7 (11.4-105.7) weeks. Twenty-six weeks after GOL induction, 37 patients (41%) were steroid-free and still on GOL, of whom 8 (21.6%) required GOL dose optimization. Short-term mucosal healing (STMH) at week 14 was evaluated in 60% of the patients. Considering the whole cohort, only 40% achieved STMH. No predictors could be retained of short-term treatment outcome. In multivariate analysis, STMH was predictive of steroid-free GOL continuation at week 26 (odds ratio [OR], 5.56; 95% confidence interval [CI], 1.90-16.29; P = 0.002) and week 52 (OR, 9.38; 95% CI, 2.68-32.84; P < 0.001). In patients continuing GOL after week 14, STMH was predictive of intervention-free survival (OR, 2.05; 95% CI, 1.09-3.86; P = 0.026) and discontinuation-free survival (OR, 3.47; 95% CI, 1.58-7.58; P = 0.002). During follow-up, 78% needed an intervention, 68% discontinued GOL, and 3 patients needed a colectomy. Conclusions Real-life data confirm the moderate effectiveness of GOL on the mid-term in active UC, but therapeutic interventions are frequently needed. Short-term mucosal healing predicts a favorable outcome.