Potential benefit of Bayesian forecasting for therapeutic drug monitoring in neonates

Pharmacokinetic and Pharmacodynamic Considerations of Antibiotic Use in Neonates

The selection of an appropriate dose of a given antibiotic for a neonate not only requires knowledge of the drug's basic pharmacokinetic (PK) and pharmacodynamic (PD) properties but also the profound effects that organ development might have on the volume of distribution and clearance, both of which may affect the PK/PD of a drug. Interest has grown in alternative antibiotic dosing strategies that are better aligned with the antibiotic's PK and PD properties. These strategies should be used in conjunction with minimum inhibitory concentration measurements and therapeutic drug monitoring to measure their potential success. They can also guide the clinician in tailoring the delivery of antibiotics to suit an individual patient's needs. Model-informed precision dosing, such as Bayesian forecasting dosing software (which incorporates PK/PD population models), may be utilized to optimize antibiotic exposure in neonatal populations. Consequently, optimizing the antibiotic dose and exposure in each newborn requires expertise in different fields. It drives the collaboration of physicians together with lab technicians and quantitative clinical pharmacologists.

Use of a common spreadsheet program to demonstrate the ability of Bayesian forecasting to estimate the pharmacokinetic parameters of antibiotics

OBJECTIVES

Recent guidelines for vancomycin have incorporated the use of Bayesian forecasting, reinforcing the need to inform students in pharmacy and clinical pharmacology of its use in therapeutic drug monitoring. The goal was to devise a PharmD research project that could demonstrate to students through simulation and data generation the utility of the Bayesian approach in estimating the pharmacokinetics of gentamicin and vancomycin.

METHODS

A series of steps were devised using Microsoft Excel to simulate patient data based on study-derived means and variances, pharmacokinetic modelling, random selection of sparse blood samples, introduce random error into the selected concentrations based on assay variability measure, and finally, inputting of the information into an add-in computer program to find the pharmacokinetic estimates using Bayesian forecasting.

KEY FINDINGS

Excellent correlations were seen between Bayesian estimates and true clearances. Lower assay variability tended to provide better estimates than larger assay variability for gentamicin, and for vancomycin, selecting a sample during the distribution phase and near the trough values tended to provide estimates with less bias and greater precision.

CONCLUSIONS

The approach used was able to demonstrate all aspects involved in Bayesian forecasting, and the results supported its use for these antibiotics.

Optimizing ganciclovir and valganciclovir dosing regimens in pediatric patients with cytomegalovirus infection: a spotlight on therapeutic drug monitoring

INTRODUCTION

Infants and immunocompromised children with cytomegalovirus (CMV) infection have significant morbidity and mortality. Ganciclovir (GCV) and its oral prodrug valganciclovir (VGCV) are the major antiviral options of choice for the prophylaxis and treatment of CMV infection. However, with the currently recommended dosing regimens used in pediatric patients, large intra- and inter-individual variability of pharmacokinetic (PK) parameters and exposure are observed.

AREAS COVERED

This review describes the PK and pharmacodynamic (PD) characteristics of GCV and VGCV in pediatrics. Moreover, the role of therapeutic drug monitoring (TDM) and current clinical practice for GCV and VGCV dosing regimens optimization in pediatrics are discussed.

EXPERT OPINION

GCV/VGCV TDM has shown the potential value to improve the benefit/risk ratio in pediatrics when using the therapeutic ranges derived from adults. However, well-designed studies are required to evaluate the relationship of TDM with clinical outcomes. Furthermore, studies to explore the children-specific dose-response-effect relationships will be helpful to facilitate the TDM practice. In the clinical setting, optimal sampling methods such as limited sampling strategies for pediatrics can be used in TDM and intracellular ganciclovir triphosphate may be used as an alternative TDM marker.

Voriconazole: an audit of hospital-based dosing and monitoring and evaluation of the predictive performance of a dose-prediction software package

BACKGROUND

Therapeutic drug monitoring (TDM) is recommended to guide voriconazole therapy.

OBJECTIVES

To determine compliance of hospital-based voriconazole dosing and TDM with the Australian national guidelines and evaluate the predictive performance of a one-compartment population pharmacokinetic voriconazole model available in a commercial dose-prediction software package.

METHODS

A retrospective audit of voriconazole therapy at an Australian public hospital (1 January to 31 December 2016) was undertaken. Data collected included patient demographics, dosing history and plasma concentrations. Concordance of dosing and TDM with Australian guidelines was assessed. Observed concentrations were compared with those predicted by dose-prediction software. Measures of bias (mean prediction error) and precision (mean squared prediction error) were calculated.

RESULTS

Adherence to dosing guidelines for 110 courses of therapy (41% for prophylaxis and 59% for invasive fungal infections) was poor, unless oral formulation guidelines recommended a 200 mg dose, the most commonly prescribed dose (56% of prescriptions). Plasma voriconazole concentrations were obtained for 82% (90/110) of courses [median of 3 (range: 1-27) obtained per course]. A minority (27%) of plasma concentrations were trough concentrations [median concentration: 1.5 mg/L (range: <0.1 to >5.0 mg/L)]. Of trough concentrations, 57% (58/101) were therapeutic, 37% (37/101) were subtherapeutic and 6% (6/101) were supratherapeutic. The dose-prediction software performed well, with acceptable bias and precision of 0.09 mg/L (95% CI -0.08 to 0.27) and 1.32 (mg/L)2 (95% CI 0.96-1.67), respectively.

CONCLUSIONS

Voriconazole dosing was suboptimal based on published guidelines and TDM results. Dose-prediction software could enhance TDM-guided therapy.

Simulated Comparison of a Bayesian Clinical Decision Support System Versus Standard of Care For Achieving Gentamicin Pharmacokinetic Targets in Neonates

BACKGROUND

Gentamicin therapy in neonates is optimized through achieving specific peak and trough concentrations. The objective of this study was to compare the ability a Bayesian clinical decision support system (CDSS) with standard of care (SOC) in determining personalized gentamicin therapies for neonates, at regimen initiation and in response to measured drug concentrations.

METHODS

This retrospective review and simulation compared target attainment among 4 arms: historical dosing according to SOC, via nomogram for initial dosing (SOC-initial) and via clinician judgment in response to measured concentrations (SOC-adjusted), and simulated dosing using the CDSS, incorporating a neonatal pharmacokinetic model for initial dosing (CDSS-initial) and incorporating maximum a posteriori-Bayesian analysis in response to measured concentrations (CDSS-adjusted). "True" patient pharmacokinetic parameters and peak and trough concentration predictions were calculated via the CDSS using the entirety of the patient dosing and concentration history. The primary outcome was pharmacokinetic target attainment of desired gentamicin peak and trough concentrations.

RESULTS

The study included 564 gentamicin concentrations among 339 patients. Mean demographics were 35 weeks gestational age (52% premature births) and 2.44 kg dosing weight. Mean PK parameters were 0.0533 L/h/kg clearance, 0.458 L/kg volume of distribution, and 8.66 hours half-life. Peak concentrations in the desired range were achieved in 96% of significantly more often in the CDSS-initial regimens and 94% of CDSS-adjusted regimens versus 86% of SOC-initial regimens and 66% of SOC-adjusted regimens. No difference was found in trough target attainment among study groups.

CONCLUSIONS

In simulation, a Bayesian CDSS showed superiority to SOC in achieving gentamicin pharmacokinetic exposure targets in neonates. Use of a CDSS may improve the safety and efficacy of gentamicin therapy for neonates.

Therapeutic drug monitoring in neonates

Therapeutic drug monitoring (TDM) aims to integrate drug measurement results into clinical decision making. The basic rules apply when using TDM in neonates (aminoglycosides, vancomycin, phenobarbital, digoxin), but additional factors should also be taken into account. First, due to both pharmacokinetic variability and non-pharmacokinetic factors, the correlation between dosage and concentration is poor in neonates, but can be overcome with the use of more complex, validated dosing regimens. Second, the time to reach steady state is prolonged, especially when no loading dose is used. Consequently, the timing of TDM sampling is important in this population. Third, the target concentration may be uncertain (vancomycin) or depend on specific factors (phenobarbital during whole body cooling). Finally, because of differences in matrix composition (eg, protein, bilirubin), assay-related inaccuracies may be different in neonates. We anticipate that complex validated dosing regimens, with subsequent TDM sampling and Bayesian forecasting, are the next step in tailoring pharmacotherapy to individual neonates.

Population pharmacokinetic study of gentamicin in a large cohort of premature and term neonates

AIM

This study aims to investigate the clinical and demographic factors influencing gentamicin pharmacokinetics in a large cohort of unselected premature and term newborns and to evaluate optimal regimens in this population.

METHODS

All gentamicin concentration data, along with clinical and demographic characteristics, were retrieved from medical charts in a Neonatal Intensive Care Unit over 5 years within the frame of a routine therapeutic drug monitoring programme. Data were described using non-linear mixed-effects regression analysis ( nonmem®).

RESULTS

A total of 3039 gentamicin concentrations collected in 994 preterm and 455 term newborns were included in the analysis. A two compartment model best characterized gentamicin disposition. The average parameter estimates, for a median body weight of 2170 g, were clearance (CL) 0.089 l h(-1) (CV 28%), central volume of distribution (Vc ) 0.908 l (CV 18%), intercompartmental clearance (Q) 0.157 l h(-1) and peripheral volume of distribution (Vp ) 0.560 l. Body weight, gestational age and post-natal age positively influenced CL. Dopamine co-administration had a significant negative effect on CL, whereas the influence of indomethacin and furosemide was not significant. Both body weight and gestational age significantly influenced Vc . Model-based simulations confirmed that, compared with term neonates, preterm infants need higher doses, superior to 4 mg kg(-1) , at extended intervals to achieve adequate concentrations.

CONCLUSIONS

This observational study conducted in a large cohort of newborns confirms the importance of body weight and gestational age for dosage adjustment. The model will serve to set up dosing recommendations and elaborate a Bayesian tool for dosage individualization based on concentration monitoring.

Impact of clinical decision support guidelines on therapeutic drug monitoring of gentamicin in newborns

BACKGROUND

Our institution's gentamicin dosing and therapeutic drug monitoring (TDM) practices for newborns were suspected to be very heterogeneous. Once-daily dosing (ODD) or extended-interval dosing (EID) and trough concentration measurement were recommended. Clinical decision support guidelines were developed and implemented as clinical decision support in the computerized prescriber order entry system. Impact on dosing, TDM practices, and blood sampling were evaluated.

METHODS

A 1-year retrospective historically controlled study before (April 2008-March 2009) and after the implementation of guidelines (January 2010-December 2010) for newborns (<30 days of life) receiving gentamicin. Blood concentrations (% of peak concentrations sampled, % of patients with zero or one concentration sampled, % of trough concentrations ≤1 mg/L) and dose regimen (ODD/EID) were compared between groups. Factors potentially associated with gentamicin concentration were analyzed (multivariate analysis).

RESULTS

One hundred thirty-two (postguidelines) versus 102 (preguidelines) patients were included (median gestational age: 34.3 versus 35.8 weeks, P > 0.05). After implementation of the guidelines, an ODD/EID regimen was almost exclusively used (97.7% versus 61.6%, P < 0.001), the percentage of peak concentrations (0.9% versus 17.2%, P < 0.001) and the number of blood samples per patient (87.1% having 0 or 1 concentration measured versus 48.0, P < 0.001) sharply reduced. A significantly higher percentage of trough concentrations were ≤1 mg/L (68.5% versus 33.0%, P < 0.001). The probability of a trough concentration ≤1 mg/L increased with an ODD/EID regimen (odds ratio, 7.23; 95% confidence interval: 3.48-15.0, P < 0.001) and in the postguidelines group (odds ratio, 2.02; 95% confidence interval: 1.01-4.02, P = 0.045).

CONCLUSIONS

Guideline implementation generated a sharp reduction in blood sampling. Clinical benefits of better gentamicin dosing and TDM practices were evident. Cost-effectiveness and clinical benefit of reduced blood sampling should be evaluated.

Bumetanide for the treatment of seizures in newborn babies with hypoxic ischaemic encephalopathy (NEMO): an open-label, dose finding, and feasibility phase 1/2 trial

BACKGROUND

Preclinical data suggest that the loop-diuretic bumetanide might be an effective treatment for neonatal seizures. We aimed to assess dose and feasibility of intravenous bumetanide as an add-on to phenobarbital for treatment of neonatal seizures.

METHODS

In this open-label, dose finding, and feasibility phase 1/2 trial, we recruited full-term infants younger than 48 h who had hypoxic ischaemic encephalopathy and electrographic seizures not responding to a loading-dose of phenobarbital from eight neonatal intensive care units across Europe. Newborn babies were allocated to receive an additional dose of phenobarbital and one of four bumetanide dose levels by use of a bivariate Bayesian sequential dose-escalation design to assess safety and efficacy. We assessed adverse events, pharmacokinetics, and seizure burden during 48 h continuous electroencephalogram (EEG) monitoring. The primary efficacy endpoint was a reduction in electrographic seizure burden of more than 80% without the need for rescue antiepileptic drugs in more than 50% of infants. The trial is registered with ClinicalTrials.gov, number NCT01434225.

FINDINGS

Between Sept 1, 2011, and Sept 28, 2013, we screened 30 infants who had electrographic seizures due to hypoxic ischaemic encephalopathy. 14 of these infants (10 boys) were included in the study (dose allocation: 0·05 mg/kg, n=4; 0·1 mg/kg, n=3; 0·2 mg/kg, n=6; 0·3 mg/kg, n=1). All babies received at least one dose of bumetanide with the second dose of phenobarbital; three were withdrawn for reasons unrelated to bumetanide, and one because of dehydration. All but one infant also received aminoglycosides. Five infants met EEG criteria for seizure reduction (one on 0·05 mg/kg, one on 0·1 mg/kg and three on 0·2 mg/kg), and only two did not need rescue antiepileptic drugs (ie, met rescue criteria; one on 0·05 mg/kg and one on 0·3 mg/kg). We recorded no short-term dose-limiting toxic effects, but three of 11 surviving infants had hearing impairment confirmed on auditory testing between 17 and 108 days of age. The most common non-serious adverse reactions were moderate dehydration in one, mild hypotension in seven, and mild to moderate electrolyte disturbances in 12 infants. The trial was stopped early because of serious adverse reactions and limited evidence for seizure reduction.

INTERPRETATION

Our findings suggest that bumetanide as an add-on to phenobarbital does not improve seizure control in newborn infants who have hypoxic ischaemic encephalopathy and might increase the risk of hearing loss, highlighting the risks associated with the off-label use of drugs in newborn infants before safety assessment in controlled trials.

FUNDING

European Community's Seventh Framework Programme.

Challenges in conducting clinical trials in children: approaches for improving performance

Recent legislative changes in both Europe and the USA have increased the responsibility of drug developers to purposefully study the agents they market in children so that specific dosing recommendations can be made to assist clinicians in their use. Typically, clinicians use empiricalor experiential-based rationales for selecting the dose to use in children, generally in a manner that attempts to achieve the same dose-exposure or pharmacokinetic profile in children as in adults. However, whether this approach achieves the necessary dose exposure or exposure effect needed may not be systematically explored during off-label use. This creates the opportunity for under- or over-exposure in children, particularly in very young children (i.e., less than 2 years old) where a combination of factors during development can effect both pharmacokinetics and pharmacodynamics. The ethical, physiological and statistical differences of studying new therapeutic agents in children present economic challenges that may create unintended incentives - both positive and negative - for any individual developer who tries to meet the requirements of new legislation to study pharmaceutical agents in children. There should be a continued emphasis in academic clinical pharmacology programs towards creative methods and approaches to better understand these differences in children compared with adults. The ability to use information from knowledge obtained from adult studies, from preclinical studies, from studies of compounds with similar chemistry or pharmacology, or from known physiological differences between children and adults is essential to choosing a suitable dose for children and achieving these regulatory aims.

Standard Error of Empirical Bayes Estimate in NONMEM® VI

The pharmacokinetics/pharmacodynamics analysis software NONMEM® output provides model parameter estimates and associated standard errors. However, the standard error of empirical Bayes estimates of inter-subject variability is not available. A simple and direct method for estimating standard error of the empirical Bayes estimates of inter-subject variability using the NONMEM® VI internal matrix POSTV is developed and applied to several pharmacokinetic models using intensively or sparsely sampled data for demonstration and to evaluate performance. The computed standard error is in general similar to the results from other post-processing methods and the degree of difference, if any, depends on the employed estimation options.

Optimal sampling strategy development methodology using maximum a posteriori Bayesian estimation

Maximum a posteriori Bayesian (MAPB) pharmacokinetic parameter estimation is an accurate and flexible method of estimating individual pharmacokinetic parameters using individual blood concentrations and prior information. In the past decade, many studies have developed optimal sampling strategies to estimate pharmacokinetic parameters as accurately as possible using either multiple regression analysis or MAPB estimation. This has been done for many drugs, especially immunosuppressants and anticancer agents. Methods of development for optimal sampling strategies (OSS) are diverse and heterogeneous. This review provides a comprehensive overview of OSS development methodology using MAPB pharmacokinetic parameter estimation, determines the transferability of published OSSs, and compares sampling strategies determined by MAPB estimation and multiple regression analysis. OSS development has the following components: 1) prior distributions; 2) reference value determination; 3) optimal sampling time identification; and 4) validation of the OSS. Published OSSs often lack all data necessary for the OSS to be clinically transferable. MAPB estimation is similar to multiple regression analysis in terms of predictive performance but superior in flexibility.

Therapeutic drug monitoring of aminoglycosides in neonates

The efficacy and toxicity of aminoglycosides show a strong direct positive relationship with blood drug concentrations, therefore, therapy with aminoglycosides in adults is usually guided by therapeutic drug monitoring. Dosing regimens in adults have evolved from multiple daily dosing to extended-interval dosing. This evolution has also taken place in neonates. Neonates, however, display large interindividual differences in the pharmacokinetics of aminoglycosides due to developmental differences early in life. The volume of distribution of aminoglycosides shows a strong relationship with bodyweight, which tends to be larger (corrected for bodyweight) in more premature infants and those with sepsis. Renal clearance of aminoglycosides increases with gestational age and accelerates immediately after birth. Because of these developmental influences, there is great inter- and intraindividual variability in the volume of distribution and clearance of these drugs, and investigators have established aminoglycoside dosing regimens based on bodyweight and/or gestational age. Widely practised dosing regimens comprise 4-5 mg/kg bodyweight of gentamicin every 24-48 hours as a first dose, followed by dose adjustment based on therapeutic drug monitoring. Although formal toxicity studies are scarce, there is no evidence that aminoglycoside toxicity in neonates differs from that in adults. Monitoring of blood drug concentrations and intelligent reconstruction of individual pharmacokinetic behaviour using a population pharmacokinetic model, optimally chosen blood sampling times and appropriate pharmacokinetic software, help clinicians to quickly optimize aminoglycoside dosing regimens to maximize the clinical effect and minimize the toxicity of these drugs.

Adaptive designs for dose-finding in non-cancer phase II trials: influence of early unexpected outcomes

BACKGROUND

In non-cancer phase II trials, dose-finding trials are usually carried out using fixed designs, in which several doses including a placebo are randomly distributed to patients. However, in certain vulnerable populations, such as neonates or infants, there is an heightened requirement for safety, precluding randomization.

PURPOSE

To estimate the minimum effective dose of a new drug from a non-cancer phase II trial, we propose the use of adaptive designs like the Continual Reassessment Method (CRM). This approach estimates the dose closest to some target response, and has been shown to be unbiased and efficient in cancer phase I trials.

METHODS

Based on a motivating example, we point out the individual influence of first outliers in this setting. A weighted version of the CRM is proposed as a theoretical benchmark to control for these outliers. Using simulations, we illustrate how this approach provides further insight into the behavior of the CRM.

RESULTS

When dealing with low targets like a 10% failure rate, the CRM appears unable to rapidly overcome an early unexpected outcome. This behavior persisted despite changing the inference (Bayesian or likelihood), underlying dose-response model (though slightly improved using the power model), and the number of patients enrolled at each dose level.

LIMITATIONS

The choices for initial guesses of failure rates, the vague prior for the model parameter, and the log-log shape of weights can appear somewhat arbitrary.

CONCLUSIONS

In phase II dose-finding studies in which failure targets are below 20%, the CRM appears quite sensitive to first unexpected outcomes. Using a power model for dose-response improves some behavior if the trial is started at the first dose level and includes at least three to five patients at the starting dose before applying the CRM allocation rule.

Therapeutic drug monitoring

PURPOSE OF REVIEW

Traditionally, therapeutic drug monitoring has been used for the management of epilepsy, cardiac arrhythmias, asthma and depression. This review provides an update, particularly for the newer clinical applications, and how therapeutic drug monitoring (including use of analytical and interpretation tools) can improve clinical outcomes.

RECENT FINDINGS

Improved drug assay methodologies and a greater understanding of pharmacokinetic and pharmacodynamic mechanisms has allowed the use of therapeutic drug monitoring for immunosuppressant drugs in organ transplant recipients, antiretroviral agents for HIV/AIDS and antimetabolite drugs for leukaemia. In addition, new computer software to analyse drug concentrations in complex populations is being developed and introduced into routine clinical applications to allow increasingly patient-specific assay interpretation.

SUMMARY

Therapeutic drug monitoring assists in improving clinical success rates and minimizing toxicity. The use of therapeutic drug monitoring is therefore likely to become more widespread as new modalities are adopted.

Limitations in the Clinical Usefulness of Single-Dose Pharmacokinetic Studies of Drugs and a Bayesian Approach for the Estimation of Kinetic Parameters

This review presents several clinical examples indicating that physiological changes in the body dependent and/or independent of developmental age, genetic polymorphisms, different disease states, acute and/or chronic inflammations, and physicochemical properties of drugs as well as some environmental factors, such as viral infections, may exert a significant effect on the first-time assessment of kinetic parameters of drug absorption, disposition, metabolism, and excretion after a single-dose administration in children and adults. The available pharmacokinetic data in the literature suggest that one must be cautious in interpretation and practical use of pharmacokinetic variables derived from either single-dose studies or bayesian methods, especially in a pediatric population.

A Bayesian approach for the estimation of pharmacokinetic parameters in children

A Bayesian method is proposed for the first-time assessment of pharmacokinetic parameters (area under the curve [AUC], maximum concentration [C(max)], half-life [T(1/2)]) in children after a single oral dose. Pharmacokinetic parameters were compared using sparse sampling (three blood samples) with extensive sampling (nine blood samples). Two different therapeutic classes of drugs were evaluated in this study. The population pharmacokinetic parameters and the blood sampling time points were obtained from adult data. The results indicate that the Bayesian approach can be used to estimate pharmacokinetic parameters in children with fair degree of accuracy. The Bayesian method described here can be used to assess pharmacokinetic parameters in children, provided there is a prior knowledge of the pharmacokinetics of a drug in adult population.