Utilizing dried blood spot micro-sampling and population pharmacokinetic modelling and simulation to inform...

Ampicillin, a β-lactam antibiotic frequently prescribed for bacterial infections, is used off-label in neonates. Blood sampling limitations in neonatal pharmacokinetic (PK) studies make dried blood spots (DBS) a promising matrix for micro-sampling. This study aims to develop a population PK (PopPK) model using a DBS-based approach to optimize ampicillin dosing in Chinese neonatal patients.

DBS samples were collected at predefined intervals from neonatal patients after ampicillin dosing. A PopPK model was developed using NONMEM 7.5, followed by model-based simulations to provide dosing recommendations in virtual population. During the simulations, the predicted blood concentrations were converted to unbound plasma concentrations using a blood-to-plasma ratio of 0.56 and an unbound fraction of 0.8. The PK/pharmacodynamic (PD) target was 100% of the time with the unbound drug plasma concentration above the MIC (%fT > MIC), and the risk of toxicity threshold was defined as a steady-state peak plasma concentration exceeding 140 mg/L.

Data from 53 patients with 102 DBS samples were collected, and the ranges of body weight and postmenstrual age (PMA) were 1.91-4.25 kg and 34.3-41.4 weeks, respectively. Ampicillin PK were characterized using a one-compartment model with first-order elimination. An allometric scaling and renal maturation model were integrated into the model to describe the developmental PK in hospitalized neonates. Simulations suggest that the optimal dosing regimen is 25 mg/kg administered intravenously every 6 h across PMA of 32-42 weeks.

We successfully developed a PopPK model of ampicillin using DBS sampling for Chinese neonates and proposed evidence-based dosing recommendations.