K. N. Fiebig (1,2), L. Blom (1), A. van Rongen (3,4), S.H.P. Simons (1), R.B. Flint (1,5), E.H.J. Krekels (3), C.A.J. Knibbe (1,3,4) , S. Völler( 1,2)
(1) Department of Pediatrics, Division of Neonatology, Erasmus MC - Sophia Children's Hospital, Rotterdam, The Netherlands; (2) Division of BioTherapeutics, LACDR, Leiden University, The Netherlands; (3) Department of Systems Biomedicine and Pharmacology, LACDR, Leiden University, The Netherlands; (4) Department of Clinical Pharmacy, St. Antonius Hospital, Nieuwegein, The Netherlands; (5) Department of Pharmacy, Erasmus MC, Rotterdam, The Netherlands
Objectives: One of the most commonly used prophylactic antibiotics at the neonatal intensive care unit (NICU) is gentamicin. It is mainly dosed according to empirical guidelines, after which therapeutic drug monitoring (TDM) and subsequent dose adaptations are applied. With peak concentrations and exposure (AUC) related to efficacy and trough concentrations related to toxicity, optimal dosing of gentamicin is challenging due to high interindividual variability in preterm and term neonates.
Recently, a novel dosing regimen for gentamicin (1) that was based on an extensively internally and externally validated pharmacokinetic (PK) model for multiple aminoglycosides (2), was implemented in the Erasmus MC Sophia, The Netherlands. The new dosing guideline recommends a dose of 4.5 mg/kg in preterm- and term neonates <28 days of postnatal age (PNA) and uses birth weight (BW) and PNA to determine the dosing interval. Compared to earlier dosing guidelines, it recommends an extended dosing interval of 60-72h in neonates with a BW lower than 1kg and/or a PNA lower than 5 days, to reach trough concentrations < 0.5 mg/L in all neonates. Here, we report results on the prospective evaluation of the gentamicin pharmacokinetic model using data obtained upon therapeutic drug monitoring (TDM) in preterm and term neonates dosed according to the novel dosing regimen.
Methods: In total, 163 TDM plasma samples, from 91 neonates with median (range) gestational age of 29 weeks (24–41), BW of 1.09 kg (0.48–4.01), PNA of 4 days (0-28) were collected. Peak (n=43) and trough (n=58) samples were defined as taken within 30 min after dosing and just before the next dose, respectively. In the patients dosed every 60-72h (n=65), 32 samples were taken at 48 h to assess whether earlier re-dosing was possible. To evaluate the descriptive and predictive performance of the previously developed gentamicin model (1), model-based individual parameter values and concentration predictions were obtained for each observation in the prospectively collected dataset, using current bodyweight (covariate for volume), BW, and PNA (covariates for clearance)(NONMEM 7.5). Goodness-of-fit (GOF) was assessed using classical diagnostic plots, as well as GOF plots stratified by quartiles of the included covariates, conditional weighted residuals (CWRES), and/or ETA versus covariate plots and normalised prediction distribution error (NPDE). Target attainment of peak samples (8-12 mg/L) and trough concentrations (≤0.5 mg/L) was analysed
Results: In this prospective model evaluation of a neonatal gentamicin pharmacokinetic model in neonates, it was found that peak concentrations and trough concentrations > 0.5mg/L were adequately predicted by the model. However, concentrations below 0.5 mg/L were overpredicted in all age and weight groups. GOF-plots stratified by covariates and CWRES versus covariate plots did not show any model misfit that could be related to BW, PNA, current weight (CW), or gestational age (GA). NPDE results showed that the variability in the data was well-captured and confirmed the bias for the trough concentrations < 0.5 m/l. While re-estimation of the model parameters based on all data only in part improved the results for the concentrations < 0.5 mg/l, these results illustrate that the current model is not able to predict the concentrations below 0.5 mg/l that were measured upon the prolonged dosing interval and which were not available in the original dataset that was used to develop the model.
Upon the novel gentamicin dosing guideline, peak target attainment was 86%. Trough concentrations were ≤0,5 mg/L in 86% of the samples. In the groups dosed every 60-72h, 38% of the additional samples at 48 h were below 0.5 mg/l.
Conclusions: This prospective model evaluation of the previously developed gentamicin model in neonates shows that the structural and covariate model results in adequate prediction of the concentrations in various birthweight and PNA groups. However, concentrations measured outside the previously studied concentration range (<0.5 mg/L) were overpredicted by the model indicating that caution is warranted when using pediatric models for predictions outside the previously defined dosing, sampling, or concentration range. This study also shows that in the pediatric population next to the structural model, evaluation of the covariate model is needed to ensure a proper fit of the model in all age groups.
References:
[1] TValitalo PAJ, Van Den Anker JN, Allegaert K, et al. Novel model-based dosing guidelines for gentamicin and tobramycin in preterm and term neonates. J Antimicrob Chemother. 2015;70(7):2074-2077. doi:10.1093/jac/dkv052
[2] De Cock RFW, Allegaert K, Sherwin CMT, et al. A Neonatal amikacin covariate model can be used to predict ontogeny of other drugs eliminated through glomerular filtration in neonates. Pharm Res. 2014;31(3):754-767. doi:10.1007/s11095-013-1197-y
Reference: PAGE 29 (2021) Abstr 9776 [www.page-meeting.org/?abstract=9776]
Poster: Drug/Disease Modelling - Paediatrics