(T-077) Comparison of Population Pharmacokinetic Platforms, Monolix and Phoenix for Cephalexin in Infants

Zixuan Wei – University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences; Andrew Haynes – Department of Pediatrics, Section of Pediatric Infectious Diseases – Children’s Hospital Colorado; Kristina Brooks – University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences; Douglas Fish – University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences; Peter Anderson – University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences

Objectives: Cephalexin is a commonly used antibiotic for pediatric populations, however, data to guide effective oral dosing in young infants are limited. Population pharmacokinetic (popPK) models of cephalexin were developed to predict pharmacodynamic (PD) target attainment, but different modeling platforms may yield varying estimates and precision. In this study, we aimed to compare two platforms, Monolix and Phoenix, in evaluating cephalexin PK and PD target attainment in infants aged 0–60 days.

Methods: Non-linear mixed effects PK modeling was performed using Monolix (Lixoft Inc) and Phoenix (Certara Inc.). Up to 5 plasma samples were collected from infants receiving cephalexin either as part of routine clinical care or following a single 25 mg/kg dose.1 Maximum likelihood estimates were obtained for absorption rate constant (Ka), apparent oral clearance (CL/F) and apparent volume of distribution (Vd/F). Parameter estimation in Monolix was performed using the stochastic approximation expectation-maximization (SAEM) algorithm, while Phoenix used the first-order conditional estimation with extended least squares (FOCE-ELS). Screened covariates included age, weight, creatinine, estimated glomerular filtration rate (eGFR), gestational age (GA) at birth, and postmenstrual age (PMA) using the same forward inclusion (p < 0.05) and backward elimination (p < 0.01) criteria. A virtual population (n=10,000), generated from PK-Sim and stratified by age and GA (7-28 days old [30-34 weeks GA; ≥35 weeks GA]; 29-60 days old [30-34 weeks GA; ≥35 weeks GA]) was used to estimate the probability of PD target attainment for the standard 25mg/kg, every 6h dosing regimen. PD target was defined as the percentage of time that free concentrations remained above minimal inhibitory concentration (fT>MIC) for at least 50% of dosing interval.2 MICs corresponding to cefazolin-susceptible Enterobacterales and methicillin-susceptible Staphylococcus aureus (MSSA) were applied.3,4

Results: In total 144 observations from 33 participants were analyzed. Median weight was 3.4 kg, median age was 31 days (range 9-56 days), median GA at birth was 37.3 weeks, 64% of infants were full-term, and median PMA was 42.2 weeks. A one-compartment with first-order absorption, lag time (Tlag), combined error model, and removing of random effect on Vd/F was selected by both platforms. Allometric scaling to a 70 kg reference was applied to Vd/F and CL/F, with PMA incorporated on CL/F. Additionally, only Monolix identified age as a significant covariate on Ka. Mean (95% CI) parameter estimates for Monolix vs. Phoenix were CL/F=6.33 (5.8-6.9) vs. 6.75 (6.0-7.5) L/h, Vd/F=30.6 (27.4-34.2) vs. 34.2 (29.7-38.8) L, Ka=1.32 (0.9-2.0) vs. 1.39 (0.8-1.9) h-1, and Tlag=0.61 (0.4-0.9) vs. 0.78 (0.6-1.0) h. Both models predicted that over 90% of virtual patients in each group achieved the PD target at MICs ranging from 1 to 8 mg/L.

Conclusions: Monolix and Phoenix generated comparable theta estimates and identified similar covariates from the same dataset except age on Ka, likely due to differences in parameter estimation algorithms. Nonetheless, both platforms produced comparable predictions of PD target attainment for cephalexin in infants.

Citations: [1] University of Colorado, Denver. Oral Amoxicillin and Cephalexin Pharmacokinetics/Pharmacodynamics (PK/PD) in the Neonatal Intensive Care Unit. clinicaltrials.gov; 2022. Accessed December 31, 2022. https://clinicaltrials.gov/study/NCT04916951
[2] Regazzi M, Berardi A, Picone S, Tzialla C. Pharmacokinetic and Pharmacodynamic Considerations of Antibiotic Use in Neonates. Antibiotics (Basel). 2023;12(12):1747. doi:10.3390/antibiotics12121747
[3] Schuetz AN, Brasso WB, Crandon JL, et al. Cefazolin as a class representative for oral cephalosporins and uncomplicated urinary tract infections caused by indicated Enterobacteriaceae. Diagn Microbiol Infect Dis. 2013;77(4):381-382. doi:10.1016/j.diagmicrobio.2013.08.024
[4] Haynes AS, Prinzi A, Silveira LJ, et al. Cefadroxil Comparable to Cephalexin: Minimum Inhibitory Concentrations among Methicillin-Susceptible Staphylococcus aureus Isolates from Pediatric Musculoskeletal Infections. Microbiol Spectr. 2022;10(4):e0103922. doi:10.1128/spectrum.01039-22

Keywords: cephalosporin, population pharmacokinetic, infant