NP-009 Development and validation of a population pharmacokinetic model of teicoplanin in adult patients with haematologic malignancies

<h3>Background and Importance</h3> Teicoplanin is a widely used antibiotic in patients with haematologic malignancies (HM); however, a population pharmacokinetic (popPK) model for these patients is not available. <h3>Aim and Objectives</h3> To develop a popPK model for teicoplanin in adult patients with HM and validate its predictive capacity for individualising dosages. <h3>Materials and Methods</h3> Prospective and multidisciplinary study conducted from February 2021 to December 2023. Adult patients with HM undergoing teicoplanin treatment were included. Two-thirds of the patients were assigned to the development of the popPK model, and one-third to its validation. All patients received an initial intravenous dose of 600 mg/12 h, and the dose was subsequently optimised through teicoplanin plasma concentrations (TPC) monitoring. The PopPK model was developed using a pharmacostatistical approach based on nonlinear mixed-effects models with NONMEM v7.3 and the FOCEI method. Predictive accuracy was calculated using the mean prediction error (MPE), and precision using the mean absolute prediction error (MAPE). A visual predictive check (VPC) was performed to assess model performance. Results were processed using R software v4.3.2. <h3>Results</h3> A total of 151 patients (65 females) were included, 100 assigned to the development group and 51 to the validation group. Median (range) age was 62 (17–87) years, and total body weight was 68 (41.5–130) kg. A total of 263 samples were analysed, with a mean (SD) TPC of 14.27 (6.81) µg/mL. The characteristics of both groups were well balanced. The pharmacokinetics were characterised using a one-compartment model with first-order elimination. The volume of distribution (Vd) was estimated at 92L, and clearance (CL) was modelled by the following equation: CL(L/h)= 1.28*[1–0.01*(AGE–62)]*[(eGFR/92)^0.4]*[(Adjusted ideal weight/61)^3,2] <h3>Where</h3> Adjusted ideal weight (Kg)= ideal weight+ (total weight– ideal weight)*0.25; AGE in years; eGFR= estimated glomerular filtration rate using CKD-EPI. External validation showed adequate accuracy and precision, with MPE and MAPE of 5.2% (95% CI: −15.2–39.3%) and 13.0% (95% CI: 0.8–39.3%), respectively. VPC graph confirmed the model’s predictive ability. <h3>Conclusion and Relevance</h3> A PopPK model was developed that characterises the kinetic behaviour of teicoplanin in HM patients, which included adjusted ideal weight, age and eGFR as factors influencing its clearance. The model predicts TPC with adequate precision and accuracy, making it a useful tool for optimising teiclopanin dosage.