Population pharmacokinetics-pharmacodynamics of ceftazidime in neonates and young infants: Dosing optimization for neonatal sepsis

Dose optimization of β-lactam antibiotics in children: from population pharmacokinetics to individualized therapy

INTRODUCTION

β-Lactams are the most widely used antibiotics in children. Their optimal dosing is essential to maximize their efficacy, while minimizing the risk for toxicity and the further emergence of antimicrobial resistance. However, most β-lactams were developed and licensed long before regulatory changes mandated pharmacokinetic studies in children. As a result, pediatric dosing practices are poorly harmonized and off-label use remains common today.

AREAS COVERED

β-Lactam pharmacokinetics and dose optimization strategies in pediatrics, including fixed dose regimens, therapeutic drug monitoring, and model-informed precision dosing are reviewed.

EXPERT OPINION/COMMENTARY

Standard pediatric doses can result in subtherapeutic exposure and non-target attainment for specific patient subpopulations (neonates, critically ill children, e.g.). Such patients could benefit greatly from more individualized approaches to dose optimization, beyond a relatively simple dose adaptation based on weight, age, or renal function. In this context, Therapeutic Drug Monitoring (TDM) and Model-Informed Precision Dosing (MIPD) emerge as particularly promising avenues. Obstacles to their implementation include the lack of strong evidence of clinical benefit due to the paucity of randomized clinical trials, of standardized assays for monitoring concentrations, or of adequate markers for renal function. The development of precision medicine tools is urgently needed to individualize therapy in vulnerable pediatric subpopulations.

Optimal use of β-lactams in neonates: machine learning-based clinical decision support system

BACKGROUND

Accurate prediction of the optimal dose for β-lactam antibiotics in neonatal sepsis is challenging. We aimed to evaluate whether a reliable clinical decision support system (CDSS) based on machine learning (ML) can assist clinicians in making optimal dose selections.

METHODS

Five β-lactam antibiotics (amoxicillin, ceftazidime, cefotaxime, meropenem and latamoxef), commonly used to treat neonatal sepsis, were selected. The CDSS was constructed by incorporating the drug, patient, dosage, pharmacodynamic, and microbiological factors. The CatBoost ML algorithm was used to build the CDSS. Real-world studies were used to evaluate the CDSS performance. Virtual trials were used to compare the CDSS-optimized doses with guideline-recommended doses.

FINDINGS

For a specific drug, by entering the patient characteristics and pharmacodynamic (PD) target (50%/70%/100% fraction of time that the free drug concentration is above the minimal inhibitory concentration [fT > MIC]), the CDSS can determine whether the planned dosing regimen will achieve the PD target and suggest an optimal dose. The prediction accuracy of all five drugs was >80.0% in the real-world validation. Compared with the PopPK model, the overall accuracy, precision, recall, and F1-Score improved by 10.7%, 22.1%, 64.2%, and 43.1%, respectively. Using the CDSS-optimized doses, the average probability of target concentration attainment increased by 58.2% compared to the guideline-recommended doses.

INTERPRETATION

An ML-based CDSS was successfully constructed to assist clinicians in selecting optimal β-lactam antibiotic doses.

FUNDING

This work was supported by the National Natural Science Foundation of China; Distinguished Young and Middle-aged Scholar of Shandong University; National Key Research and Development Program of China.

Development and validation of a UPLC-PDA method for quantifying ceftazidime in dried blood spots

Bacterial infection is a leading cause of neonatal death. Ceftazidime, commonly used for neonatal infections, is often used off-label. Blood sampling limits pharmacokinetic (PK) studies in neonatal patients. The dried blood spots (DBS) are a potential matrix for microsampling. Herein, we describe an ultra-performance liquid chromatography with a photodiode array (UPLC-PDA) to determine ceftazidime in DBS from neonatal patients in support of pharmacokinetic studies. The Capitainer® device-based DBS samples containing 10 µL blood were extracted in 70% methanol/water (v/v) with acetaminophen as the internal standard (IS). The extraction process was carried out at 20 °C using a block bath shaker at 1000 rpm for 30 min. The extracted ceftazidime was subsequently eluted through an Acquity UPLC HSS T3 column (2.1 × 50 mm, 1.8 µm). Elution was achieved using a water (containing 0.1% trifluoroacetic acid)/acetonitrile linear gradient at a flow rate of 0.5 mL/min, and the analytical time was 3.2 min. The PDA detection wavelength was set at 259 nm. The method underwent thorough validation following the recommendation of the European Bioanalysis Forum (EBF) and the bioanalytical guideline established by the European Medicines Agency (EMA). No interfering peaks were detected at the retention times of ceftazidime and IS. The ceftazidime exhibited a quantification range spanning from 0.5 to 200 µg/mL, and the assay demonstrated good accuracy (intra/inter-assay ranging from 90.1% to 104.8%) and precision (intra/inter-assay coefficient of variations ranging from 4.8% to 11.7%). The method's applicability was demonstrated by analyzing clinical DBS samples collected from neonatal patients.

The Impact of Spironolactone Co-administration on Cyclosporin Initial Dosage Optimization for Pediatric Refractory Nephrotic Syndrome

OBJECTIVES

Cyclosporin has been used for the treatment of pediatric refractory nephrotic syndrome (PRNS). However, the narrow therapeutic window and large pharmacokinetic variability make it difficult to individualize cyclosporin administration. Meanwhile, spironolactone has been reported to affect cyclosporin metabolism in PRNS patients. This study aims to explore the initial dosage optimization of cyclosporin in PRNS based on the impact of spironolactone co-administration.

METHODS

Monte Carlo simulation based on a previously established cyclosporin population pharmacokinetic model for PRNS was used to design cyclosporin dosing regimen.

RESULTS

In this study, the probability of drug concentration reaching the target and the convenience of times of administration were considered comprehensively. The optimal administration regimen in PRNS without spironolactone was 6, 5, 4 and 3 mg/kg cyclosporin split into two doses for the body weight of 5-8, 8-18, 18-46 and 46-70 kg, respectively. The optimal administration regimen in PRNS with spironolactone was 4, 3, 2 mg/kg cyclosporin split into two doses for body weight of 5-14, 14-65, and 65-70 kg, respectively.

CONCLUSION

The cyclosporin dosing regimen for PRNS based on Monte Carlo simulation was systematically developed and the initial dosage optimization of cyclosporin in PRNS was recommended for the first time.

Population Pharmacokinetics and Dosing Optimization of Ceftazidime in Term Asphyxiated Neonates during Controlled Therapeutic Hypothermia

Ceftazidime is an antibiotic commonly used to treat bacterial infections in term neonates undergoing controlled therapeutic hypothermia (TH) for hypoxic-ischemic encephalopathy after perinatal asphyxia. We aimed to describe the population pharmacokinetics (PK) of ceftazidime in asphyxiated neonates during hypothermia, rewarming, and normothermia and propose a population-based rational dosing regimen with optimal PK/pharmacodynamic (PD) target attainment. Data were collected in the PharmaCool prospective observational multicenter study. A population PK model was constructed, and the probability of target attainment (PTA) was assessed during all phases of controlled TH using targets of 100% of the time that the concentration in the blood exceeds the MIC (T>MIC) (for efficacy purposes and 100% T>4×MIC and 100% T>5×MIC to prevent resistance). A total of 35 patients with 338 ceftazidime concentrations were included. An allometrically scaled one-compartment model with postnatal age and body temperature as covariates on clearance was constructed. For a typical patient receiving the current dose of 100 mg/kg of body weight/day in 2 doses and assuming a worst-case MIC of 8 mg/L for Pseudomonas aeruginosa, the PTA was 99.7% for 100% T>MIC during hypothermia (33.7°C; postnatal age [PNA] of 2 days). The PTA decreased to 87.7% for 100% T>MIC during normothermia (36.7°C; PNA of 5 days). Therefore, a dosing regimen of 100 mg/kg/day in 2 doses during hypothermia and rewarming and 150 mg/kg/day in 3 doses during the following normothermic phase is advised. Higher-dosing regimens (150 mg/kg/day in 3 doses during hypothermia and 200 mg/kg/day in 4 doses during normothermia) could be considered when achievements of 100% T>4×MIC and 100% T>5×MIC are desired.

Optimization of β-Lactam Dosing Regimens in Neonatal Infections: Continuous and Extended Administration versus Intermittent Administration

BACKGROUND AND OBJECTIVE

In neonates, β-Lactam antibiotics are almost exclusively administered by intermittent infusion. However, continuous or prolonged infusion may be more beneficial because of the time-dependent antibacterial activity. In this pharmacokinetic/pharmacodynamic simulation study, we aimed to compare treatment with continuous, extended and intermittent infusion of β-lactam antibiotics for neonates with infectious diseases.

METHODS

We selected population pharmacokinetic models of penicillin G, amoxicillin, flucloxacillin, cefotaxime, ceftazidime and meropenem, and performed a Monte Carlo simulation with 30,000 neonates. Four different dosing regimens were simulated: intermittent infusion in 30 min, prolonged infusion in 4 h, continuous infusion, and continuous infusion with a loading dose. The primary endpoint was 90% probability of target attainment (PTA) for 100% ƒT>MIC during the first 48 h of treatment.

RESULTS

For all antibiotics except cefotaxime, continuous infusion with a loading dose resulted in a higher PTA compared with other dosing regimens. Sufficient exposure (PTA >90%) using continuous infusion with a loading dose was reached for amoxicillin (90.3%), penicillin G (PTA 98.4%), flucloxacillin (PTA 94.3%), cefotaxime (PTA 100%), and ceftazidime (PTA 100%). Independent of dosing regimen, higher meropenem (PTA for continuous infusion with a loading dose of 85.5%) doses might be needed to treat severe infections in neonates. Ceftazidime and cefotaxime dose might be unnecessarily high, as even with dose reductions, a PTA > 90% was retained.

CONCLUSIONS

Continuous infusion after a loading dose leads to a higher PTA compared with continuous, intermittent or prolonged infusion, and therefore has the potential to improve treatment with β-lactam antibiotics in neonates.

Beta-Lactams Therapeutic Monitoring in Septic Children-What Target Are We Aiming for? A Scoping Review

The antimicrobial therapy of sepsis and septic shock should be individualized based on pharmacokinetic/pharmacodynamic (PK/PD) parameters to deliver effective and timely treatment of life-threatening infections. We conducted a literature scoping review to identify therapeutic targets of beta-lactam antibiotics in septic pediatric patients and the strategies that have been applied to overcome sepsis-related altered pharmacokinetics and increase target attainment against susceptible pathogens. A systematic search was conducted in the MEDLINE, EMBASE and Web of Science databases to select studies conducted since 2010 with therapeutic monitoring data of beta-lactams in septic children. Last searches were performed on 02 September 2021. Two independent authors selected the studies and extracted the data. A narrative and qualitative approach was used to summarize the findings. Out of the 118 identified articles, 21 met the eligibility criteria. Population pharmacokinetic modeling was performed in 12 studies, while nine studies reported data from bedside monitoring of beta-lactams. Most studies were conducted in the United States of America (n = 9) and France (n = 5) and reported PK/PD data of amoxicillin, ampicillin, azlocillin, aztreonam, cefazolin, cefepime, cefotaxime, ceftaroline, ceftazidime, doripenem, meropenem and piperacillin/tazobactam. Therapeutic targets ranged from to 40% fT> MIC to 100% fT> 6 × MIC. Prolonging the infusion time and frequency were most described strategies to increase target attainment. Monitoring beta-lactam serum concentrations in clinical practice may potentially maximize therapeutic target attainment. Further studies are required to define the therapeutic target associated with the best clinical outcomes.

Drug Clearance in Neonates: A Combination of Population Pharmacokinetic Modelling and Machine Learning Approaches to Improve Individual Prediction

BACKGROUND

Population pharmacokinetic evaluations have been widely used in neonatal pharmacokinetic studies, while machine learning has become a popular approach to solving complex problems in the current era of big data.

OBJECTIVE

The aim of this proof-of-concept study was to evaluate whether combining population pharmacokinetic and machine learning approaches could provide a more accurate prediction of the clearance of renally eliminated drugs in individual neonates.

METHODS

Six drugs that are primarily eliminated by the kidneys were selected (vancomycin, latamoxef, cefepime, azlocillin, ceftazidime, and amoxicillin) as 'proof of concept' compounds. Individual estimates of clearance obtained from population pharmacokinetic models were used as reference clearances, and diverse machine learning methods and nested cross-validation were adopted and evaluated against these reference clearances. The predictive performance of these combined methods was compared with the performance of two other predictive methods: a covariate-based maturation model and a postmenstrual age and body weight scaling model. Relative error was used to evaluate the different methods.

RESULTS

The extra tree regressor was selected as the best-fit machine learning method. Using the combined method, more than 95% of predictions for all six drugs had a relative error of < 50% and the mean relative error was reduced by an average of 44.3% and 71.3% compared with the other two predictive methods.

CONCLUSION

A combined population pharmacokinetic and machine learning approach provided improved predictions of individual clearances of renally cleared drugs in neonates. For a new patient treated in clinical practice, individual clearance can be predicted a priori using our model code combined with demographic data.