Methotrexate showed efficacy both in Crohn's disease and ulcerative colitis, predictors of surgery were identified in patients initially treated with methotrexate monotherapy

Abstract

Objective: This study aimed to evaluate methotrexate efficacy in patients with Crohn's disease (CD) and ulcerative colitis (UC), and identify predictors of surgery for patients who were initially treated with methotrexate monotherapy. Design: We performed a retrospective analysis of 34,860 patients with inflammatory bowel disease (IBD) in the IBD Bioresource (United Kingdom) prior to 9 November 2021. Logistic regression was used to identify factors associated with methotrexate efficacy. The data were randomly stratified into training and testing sets (7:3). Nomograms were developed based on Cox regression analysis outcomes. The predictive accuracy and discriminative ability were determined using the concordance index (C-index) and calibration curves. Results: Overall, 1,042 patients (CD: 791, UC: 251) were included. Independent factors associated with effective methotrexate monotherapy were younger age at diagnosis, latest therapy period, exclusive upper gastrointestinal tract disease (for CD), and longer duration between diagnosis and methotrexate initiation (for UC). For CD, predictors in the nomogram were gender, treatment era, tolerance, lesion site, perianal involvement, disease behaviour, and biologics requirements (C-index: 0.711 and 0.732 for training and validation cohorts, respectively). For UC, the factors were age at diagnosis and sex (C-index: 0.784 and 0.690 for training and validation cohorts, respectively). Calibration curves demonstrated good agreement between predictions and actual observations.

Keywords: Crohn’s disease; inflammatory bowel disease; methotrexate monotherapy; prognostic analysis; ulcerative colitis.

FIGURE 1

Flowchart of the study cohort depicting the inclusion and exclusion of patients from the analysis.

FIGURE 2

(A) Time interval (years) between CD or UC diagnosis and methotrexate monotherapy; (B) Odds ratio (OR) of methotrexate monotherapy being effective in UC vs. CD stratified by therapy period. The horizontal coordinates are OR values, blue boxes indicate OR values for every era, and black solid lines indicate 95% confidence intervals (CI) for OR values; (C) The Kaplan-Meier curve shows the duration of methotrexate treatment in patients in whom methotrexate monotherapy was considered effective; (D) The Kaplan-Meier curve shows the duration of effective methotrexate monotherapy without requirement for treatment escalation (p < 0.0001 for log-rank test). (CD, Crohn’s disease; UC, ulcerative colitis).

FIGURE 3

The Kaplan-Meier curve shows time interval from methotrexate initiation to surgery occurance between “Able to tolerate” and “Unable to tolerate” groups of CD patients. (p = 0.029 for log-rank test).

FIGURE 4

Nomogram for 1-, 3- and 5- years no surgery probability in patients with CD treated with methotrexate monotherapy. Find the point for each variable, sum the scores achieved for each covariate, and locate this sum on the “Total Points” axis. Draw a straight line to determine the likelihood of no surgery probability of 1, 3 or 5 years. (Period: treatment era stratified by 5 years; Tolerate: if or not can tolerate methotrexate; Colonic: disease location; Perianal: have perianal involvement or not; Behaviour: disease characteristic: stenosing, internal penetrating or inflammatory and others; Biological: biological therapy requirment).

FIGURE 5

Calibration curve of nomogram for predicting no surgery occurrence at (A) 1-, (C) 3-, and (E) 5- years in the training cohort, and at (B) 1-, (D) 3-, and (F) 5- years in the validation cohort. The actual proportion is plotted on the Y-axis and the nomogram-predicted probability is plotted on the X-axis. (CD, Crohn’s disease).

FIGURE 6

The Kaplan-Meier curve shows time interval from methotrexate initiation to surgery occurance between “Able to tolerate” and “Unable to tolerate” groups of UC patients. (p = 0.013 for log-rank test).

FIGURE 7

Nomogram for 1-, 3- and 5- years no surgery probability in UC patients treated with methotrexate monotherapy. Find the point for each variable, sum the scores achieved for each covariate, and locate this sum on the “Total Points” axis. Draw a straight line to determine the likelihood of no surgery probability of 1, 3 or 5 years. (Age: age at UC diagnosis).

FIGURE 8

Calibration curve of nomogram for predicting no surgery occurrence at (A) 1-, (C) 3-, and (E) 5-years in the training cohort, and at (B) 1-, (D) 3-, and (F) 5- years in the validation cohort. The actual proportion is plotted on the Y-axis and the nomogram-predicted probability is plotted on the X-axis. (UC: ulcerative colitis).