Abstract Background Antibodies against tumor necrosis factor-alpha (TNF-α) such as infliximab (IFX) are important in the treatment of patients with inflammatory bowel disease (IBD), but many patients lose drug response over time or do not respond at all. Loss-of-response and non-response can at least be partially attributed to suboptimal dosing due to individual differences in pharmacokinetics (PK)1. In this real-world study, we aimed to identify clinical, biochemical and endoscopic variables influencing IFX PK in patients with IBD. Methods Patients diagnosed with IBD aged ≥18 years who were treated with IFX were included based on available IFX trough levels from a hospital-based pharmacy registry. Information on endoscopic, clinical, and biochemical parameters was collected. Univariable and multivariable linear mixed-effects models were used to model associations between individual IFX trough level trajectories and identified factors to examine their relative contributions to IFX PK. In addition, a systematic literature study on IFX PK in adults with IBD and other immune-mediated inflammatory diseases was performed to better contextualize the obtained findings. Results A total of 200 patients were included (Crohn’s disease [CD] (n=163) and ulcerative colitis [UC] (n=37)) with 452 IFX trough levels available for analysis (Figure 1). Patients with antibodies (ATI) to IFX (unstandardized β=4.35, p<0.001) and those treated with longer dosing intervals (β=-1.89, p=<0.001) exhibited lower trough levels (Figure 2). Higher levels of aspartate aminotransferase (AST) (β=0.04, p=0.027), hemoglobin (β=0.76, p=0.022) and albumin (β=0.15, p=0.029) were positively associated with IFX trough levels. Among clinical factors, Simple Clinical Colitis Activity Index (SCCAI) scores in patients with UC, both as continuous (β=3.31, p=0.03) and as categorical variable (β=13.25, p=0.033), were positively associated with higher IFX trough levels. The presence of endoscopic inflammation, regardless of its severity, was also positively correlated with IFX trough levels (β=2.60, p=0.045). Conclusion This data demonstrates a role of AST, albumin, hemoglobin, SCCAI, and endoscopic inflammation as covariates potentially relevant to IFX PK in IBD. Findings of albumin, endoscopic inflammation, dosage interval and ATI were supported by our systematic review confirming them as previously identified covariates. AST, hemoglobin, and SCCAI were newly identified covariates. These covariates could optimize treatment strategies and should therefore be considered for integration in population-PK/PD models and PK dashboards in order to make patient-specific prediction models for IFX treatment.
Read full abstract