4CPS-275 Adaptation and simulation of an azathioprine pharmacokinetic model

<h3>Background and Importance</h3> Azathioprine is a non-specific immunosuppressor of narrow therapeutic range (0.157-0.3mg/L), used in inflammatory bowel disease. Its metabolism includes enzymes as TPMT and NUDT15, which polymorphisms can diminish its clearance. A 30–80% dose reduction is recommended in patients with heterozygous mutant genotype. <h3>Aim and Objectives</h3> To perform population-based simulations using an azathioprine pharmacokinetic model to narrow down the dose adjustment in patients with TPMT or NUDT15 polymorphisms. <h3>Material and Methods</h3> A literature search was conducted in PubMed and Embase databases. Only one population pharmacokinetic model was found, published by Lin et al. It included 100 patients and studied different covariates that influenced clearance, finding three that generated statistically significant differences: body weight (BW), TPMT heterozygous variant, and concomitant mesalazine treatment. They found no differences in plasma clearance as a result of NUDT15 mutation. They obtained a monocompartmental pharmacokinetic model with first-order kinetics, interindividual variability (ηCL) 0.342 and residual variability 0.054. The model by Lin et al. was adapted by adding the absorption phase, using a Ka from the literature of 7.12 h⁻¹. Population-based Monte Carlo simulations were performed for 100 patients, and a dose of 50 mg of azathioprine every 24h, obtaining mean plasma concentrations and 5^th-95^th percentiles. R was used for performing the simulations, with mrgsolve language package and ggplot to create the graphics representing azathioprine plasma levels versus time. A Shiny app was also created. <h3>Results</h3> A bicompartmental pharmacokinetic model with first-order pharmacokinetics was obtained, with the differential equation: Dxdt_Cp=7.12 h^(-1) x C_gut-[(11.6 x (BW/70)^0.625 xθ_TPMT xθ_MESALAZINE xe^ηCL)/(809(V))]xCp. Non-mutated TPMT (nm-TPMT)=1, mutated (m-TPMT)= 0.515, no mesalazine treatment (n-MESA)=1 and treated with mesalazine (t-MESA)=0.802. Simulations results are expressed as mean trough concentration in steady state (m-TCSS), (5^th-95^th percentiles): – BW=70 kg, nm–TPMT, n–MESA: 0.16 (0.08–0.32 mg/L) – BW=50 kg, nm–TPMT, n–MESA: 0.15 (0.08–0.28 mg/L) – BW=70 kg, m–TPMT, n–MESA: 0.31 (0.18–0.46 mg/L) – BW=70 kg, nm–TPMT, t–MESA: 0.20 (0.11–0.31 mg/L) – BW=70 kg, m–TPMT, t–MESA: 0.37 (0.22–0.51 mg/L) <h3>Conclusion and Relevance</h3> The mean concentrations obtained with the adapted model can provide a more accurate estimate of the starting dose of azathioprine for patients according to body weight, TPMT polymorphism, and treatment with mesalazine than previous recommendations. However, interindividual variability is substantial, and further studies are required to confirm it. If upheld, azathioprine might be considered for pharmacokinetic monitoring given its narrow therapeutic range. <h3>Conflict of Interest</h3> No conflict of interest