Evaluating and Improving Neonatal Gentamicin Pharmacokinetic Models Using Aggregated Routine Clinical Care Data

Abstract

Model-informed precision dosing (MIPD) can aid dose decision-making for drugs such as gentamicin that have high inter-individual variability, a narrow therapeutic window, and a high risk of exposure-related adverse events. However, MIPD in neonates is challenging due to their dynamic development and maturation and by the need to minimize blood sampling due to low blood volume. Here, we investigate the ability of six published neonatal gentamicin population pharmacokinetic models to predict gentamicin concentrations in routine therapeutic drug monitoring from nine sites in the United State (n = 475 patients). We find that four out of six models predicted with acceptable levels of error and bias for clinical use. These models included known important covariates for gentamicin PK, showed little bias in prediction residuals over covariate ranges, and were developed on patient populations with similar covariate distributions as the one assessed here. These four models were refit using the published parameters as informative Bayesian priors or without priors in a continuous learning process. We find that refit models generally reduce error and bias on a held-out validation data set, but that informative prior use is not uniformly advantageous. Our work informs clinicians implementing MIPD of gentamicin in neonates, as well as pharmacometricians developing or improving PK models for use in MIPD.

External validation of population pharmacokinetic models of gentamicin in paediatric population from preterm newborns to adolescents

“…As expected, for most literature and tweaked models, the bias and imprecision, expressed as MdPE% and MdAPE%, were lower for a priori predictions than for a posteriori predictions, and re-estimation of the models improved their predictive performance. This is consistent with other studies reporting that adopting literature models with a prior approach improves their predictive ability (24,25,30,31). For a priori predictions, the range of MdPE% for the literature and the tweaked models was -96.2 to 300.8 % and -29.6 to 286.0 %, respectively.…”

“…Adopting literature models with a prior approach improves their predictive ability (24,25,30,31). Namely, the PRIOR subroutine in NONMEM allows estimation of some or all parameters using values from literature models, as an alternative to fixing them or adding data to the sparse dataset (32).…”

“…Re-estimation of literature models using the prior approach is not widely used. It has already been used for population PK models of gentamicin, but either in adult ICU patients or neonates only (31,33). Therefore, the present study is the first to use this approach to model gentamicin PK in the whole paediatric population, from preterm newborns to adolescents.…”

“…Considering the median observed concentration in our dataset (0.9 mg L -1 ), an absolute PE% of 25 % would correspond to a PE of ± 0.2 mg L -1 , which could still be acceptable. In a study by Tong et al (31) trough predictions were considered accurate if they were within 0.5 mg L -1 of the actual value. Interestingly, all tweaked models underestimated the predicted concentrations of gentamicin, which could have consequences for the safety of the treatment.…”

Predictive Performance of a Gentamicin Pharmacokinetic Model in Term Neonates with Perinatal Asphyxia Undergoing Controlled Therapeutic Hypothermia

“…In clinical practice, allowable error margins are typically set at approximately 0.5 mg/L for trough levels, which are generally indicative of low gentamicin concentrations, and 2 mg/L for peak levels. 26 However, it is worth noting that imprecision when estimating a priori trough levels using gentamicin models in neonates is not uncommon. 26 In a recent study that compared model-to-model predictions using 6 previously published neonatal gentamicin models, 4 of these models performed reasonably well a priori of bias and precision.…”