Search in the Abstract Database

Abstracts Search 2019

P342 A population pharmacokinetic model to support therapeutic drug monitoring during vedolizumab therapy

E. Dreesen*1, B. Verstockt2,3, S. Vermeire2,3, M. Ferrante2,3, A. Gils1

1University of Leuven, Department of Pharmaceutical and Pharmacological Sciences, Leuven, Belgium, 2University of Leuven, Department of Chronic Diseases, Metabolism and Ageing, Leuven, Belgium, 3University Hospitals Leuven, Department of Gastroenterology and Hepatology, Leuven, Belgium

Background

Patients with ulcerative colitis (UC) and Crohn’s disease (CD) starting vedolizumab (VDZ) therapy can benefit from therapeutic drug monitoring (TDM).1 A population pharmacokinetic (popPK) model may support dose optimisation to improve attainment of the predefined trough concentration (TC) targets.

Methods

A total of 939 trough samples (from week (w) 2 to w30) of 178 patients (66 UC, 112 CD; excluding one patient with antibodies to VDZ) was used to develop a popPK model. Data were analysed under a known two-compartment model with parallel linear and nonlinear clearance by using prior distributions from the GEMINI popPK model to support estimation of PK parameters that were poorly informed by the current data (NONMEM 7.4 with $PRIOR).1 Simulations were performed using Berkeley-Madonna 8.3.

Results

Our model with fully data-driven estimation of the linear clearance (CLL; 0.207 L/day [3%], typical value [relative standard error]) and volume of distribution in the central compartment (Vc; 4.62 L [9%]) showed good predictive capacity. Linear terminal elimination half-life of VDZ was 15.5 days. Lower albumin, mean platelet volume and haemoglobin, and higher C-reactive protein and fat-free mass were associated with higher CLL, thus predicting lower VDZ exposure. Prior anti-TNF therapy did not impact CLL. Furthermore, CLL was not different between patients with UC and CD. Still, 28% and 40% of the interindividual variability (IIV) on CLL and Vc, respectively, remained unexplained. Patients with Mayo endoscopic sub-score (MES) ≤1 at w14 had a lower VDZ CLL already at w2 (p = 0.009) (Figure 1A). VDZ CLL slightly decreased with time (p = 0.028). In addition, the cumulative area under the VDZ concentration-time curve (AUC) from w0 to w14 was higher in patients with MES ≤1 at w14 (p = 0.001) (Figure 1B). Although VDZ is characterised by nonlinear CL, this only appeared to be relevant in the sub-therapeutic concentration range (<10.0 mg/l), providing additional motivation to target patients above the predefined ~14.0 mg/l TC threshold during maintenance therapy (Figure 2).1

Conclusion

Our model demonstrates good predictive capacity and may be implemented in a TDM software tool to improve attainment of the exposure targets (TC and AUC) in individual patients with inflammatory bowel diseases.

(A) The estimated vedolizumab linear clearance (linear mixed effects model) and (B) the area under the vedolizumab concentration-time curve from week 0 to Week 14 (Wilcoxon Rank-Sum test) of patients with Mayo endoscopic sub-score at Week 14 >/≤1.

Simulated profiles from the vedolizumab popPK model (n = 1,000; median covariate values). The red line indicates the critical 10.0 mg/l concentration below which concentrations drop more rapidly due to an increasing contribution of nonlinear clearance.

References

1. Dreesen E, Verstockt B, Bian S, et al. Evidence to support monitoring of vedolizumab trough concentrations in patients with inflammatory bowel diseases. Clin Gastroenterol Hepatol. 2018. doi:10.1016/j.cgh.2018.04.040 [Epub ahead of print]

2. Rosario M, Dirks NL, Gastonguay MR, et al. Population pharmacokinetics-pharmacodynamics of vedolizumab in patients with ulcerative colitis and Crohn's disease. Aliment Pharmacol Ther 2015;42:188–202.