Background and aims
Inflammatory bowel disease (IBD) features diverse clinical presentations and progressions, impacting IFX exposure. Understanding IFX serum concentration changes is crucial for tailored monitoring in specific patient groups. The main objective of this study was to analyze ITL trajectories in patients with IBD to identify distinct groups and subgroups, revealing heterogeneity in treatment responses.
Methods
A retrospective cohort study was conducted involving IBD patients treated with infliximab in a regional reference hospital in Spain. Latent class linear mixed models were applied to identify subgroups based on serum infliximab measurements over time. To analyze the factors associated with IFX discontinuation, we employed both logistic regression and Cox regression models.
Results
The study included 165 IBD patients, and a total of 799 ITL samples were analyzed. The selected model included three clusters, with a random intercept and a random effect on both time and natural cubic spline time in the linear mixed model. Cluster 1 (20.6%) had lower IFX exposure, with 93.9% experiencing treatment discontinuation, compared to 45.1% in Cluster 2 (43.0%) and 43.3% in Cluster 3 (36.4%) (p < 0.001). Treatment discontinuation was observed in 91 individuals (55.2%) out of the total patients. In the multivariate analysis, the presence of cluster 1 was a significant predictor (OR: 7.25, 95% CI: 1.45–36.12). Bayesian dose adjustment was found to significantly reduce the risk of IFX discontinuation (OR: 0.19, 95% CI: 0.46–1.96).
Conclusions
The lack of TDM during induction and a lower proportion of adjustments made through Bayesian methods were associated with a subgroup demonstrating suboptimal pharmacokinetic profiles and reduced drug persistence. These findings highlight the clinical relevance of model-informed TDM in optimizing IFX exposure and minimizing treatment discontinuation in IBD.
扫码关注我们
求助内容:
应助结果提醒方式:
