Aminoglycosides in the Intensive Care Unit: What Is New in Population PK Modeling?

We aimed to compare the performance of renal function and age as predictors of inter-individual variability (IIV) in clearance of amikacin in neonates through parallel development of population pharmacokinetic (PK) models and their associated impact on optimal dosing regimens. Amikacin concentrations were retrospectively collected for 149 neonates receiving amikacin (post-natal age (PNA) between 4-89 days). Two population PK models were developed in parallel, considering at least as predictors current body weight (WT), in combination with either creatinine clearance (CLcr ) or age descriptors. Using stochastic simulations for both renal function or age-based dosing, we identified optimal dosing strategies that were based on attainment of optimal peak- (PCC) and trough target concentration coverage (TCC) windows associated with efficacy and toxicity. The CLcr and age-based population PK models both included current body weight (WT) on CL, central distribution volume and intercompartmental clearance, in combination with either CLcr or PNA as predictors for IIV of clearance (CL). The WT-CLcr model explained 6.9% more IIV in CL compared with the WT-PNA model. Both models successfully described an external dataset (n=53) of amikacin PK. The simulation analysis of optimal dose regimens suggested similar performance of either CLcr or PNA based dosing. CLcr predicted more IIV in CL, but did not translate into clinically relevant improvements of target concentrations. Our optimized dose regimens can be considered for further evaluation to optimize initial treatment with amikacin.

Dear Editor, We have read with great interest the paper of Sherwin et al. on individualized dosing of amikacin based on a population pharmacokinetic and -dynamic (PKPD) study in 80 neonates [1]. To the very best of our knowledge, this is the first PD study (outcome indicator sepsis) of aminoglycosides in neonates. We fully support the clinical need to evaluate both PK and PD of drugs, including aminoglycosides in neonates. The recent review on aminoglycosides in neonates in this journal hereby illustrates that clinical pharmacologists are aware of and interested in the specific needs and characteristics of this patient population [2]. We are, however, intrigued by the dosing suggestions formulated by the authors: 15 mg/kg at 36-h intervals for neonates with a postmenstrual age (PMA) ≤ 28 weeks, 14 mg/kg at 24-h intervals for neonates with a PMA between 29 and 36 weeks, and 15 mg/kg at 24-h intervals for neonates with a PMA ≥ 37 weeks [1]. Firstly, if we apply the clearance model proposed to the range of neonates included (weight 0.44−4.4 kg; PMA 24−41 weeks), clearance ranges from 0.0155 to 0.428 L/h (30-fold interindividual variability). This extensive interindividual variability in clearance is not reflected in the dosing recommendations formulated. As aminoglycoside clearance reflects glomerular filtration rate (GFR), it is to be anticipated that maturational clearance displays a continuum, instead of the dichotomous pattern: 50% reduction in amikacin administration ≤ 28 weeks, and only a very limited additional interindividual variability (<10%) between 29-weeks and full-term neonates seems not to be in line with developmental maturation of renal clearance capacity. Although it is likely that the current dosing suggestions fit best with the available data set, it is a likely that the current suggestions do not yet fully reflect the interindividual variability in clearance of aminoglycosides in neonates [2–4]. Secondly, based on the fact that interindividual variability in any aminoglycoside clearance reflects interindividual variability in GFR, we urgently call for prospective validation of the current and other clearance models suggested. As clearance of aminoglycoside reflects the interindividual variability in GFR, cross-validation studies should include not only different cohorts but also different aminoglycosides. We hereby would like refer to the netilmicin study in neonates recently published by the same group, where the clearance model seems to be based on the same covariates model but with different exponential factors [5]. Besides collaboration between different centers and groups, such cross-validation studies necessitate robust data sets that include other relevant covariates in addition to PMA or weight alone. We miss information on the existence of a patent ductus arteriosus (PDA), on the exposure to indomethacin or ibuprofen to induce closure of the PDA, or on the presence of growth restriction. as it has been clearly demonstrated that these covariates in part explain the interindividual variability in aminoglycoside clearance in neonates [6–8]. Similarly, we would like to mention the relevance of the technique used to determine creatinine level in the blood. For decades we have known that the Jaffe method introduces additional errors when used in neonates because of interferences by, e.g., ketoacids, bilirubin, or cephalosporins. It really makes a difference to use an enzymatic method in this age group [9]. In conclusion, we would like to encourage the authors and other research groups interested to critically re-evaluate their single cohort aminoglycoside clearance model and recommendations. In addition to single population studies, cross-validation studies, including extrapolation of a given clearance model to data sets with other aminoglycosides, are urgently needed to unveil the underlying maturational GFR and its covariates. This will necessitate collaborative efforts and can only be based on robust data sets including covariates of potential relevance and should not be limited to weight or PMA alone.

The purpose of this study was to describe the population pharmacokinetics of gentamicin in patients with cancer, to identify possible relationships between clinical covariates and population pharmacokinetic parameter estimates and to examine the relevance of existing dosage nomograms in light of the population model developed in these patients. Data were collected prospectively from 210 patients with cancer and were analysed with package NONMEM. Data were split into two sets: a population data set and an evaluation set. Creatinine clearance was estimated using measured creatinine concentrations and using 'low' creatinines set to a minimum of 60 micromol l(-1), 70 micromol l(-1) or 88.4 micromol l(-1) A two compartment model was fitted to the concentration-time curve. Two best models were obtained, one that related clearance to estimated creatinine clearance (minimum creatinine value 60 micromol l(-1)) and the other that related clearance to age, creatinine concentration and body surface area. Volume of the central compartment was influenced by body surface area and albumin concentration. For both models 90% of measured concentrations lay within the 95% confidence interval of the simulated concentrations and the mean prediction errors were -7.2% and -6.6%, respectively. A final analysis performed in all patients identified the following relationship CL (1 h(-1))=0.88 x (1 + 0.043 x creatinine clearance) and central volume of distribution V1 (1)=8.59 x body surface area x (albumin/34)(-0.39). The mean population estimate of intercompartmental clearance (Q) was 1.301 h(-1) and peripheral volume of distribution (V2) was 9.801. Coefficient of variation was 18.5% on clearance and 28.2% on Q. Residual error expressed as a standard deviation was 0.36 mg l(-1) at 1.0 mg l(-1) and 1.32 mg l(-1) at 8.0 mg l(-1). The mean population estimate of clearance was 4.21 h(-1) and volume of distribution (Vss) was 24.61 (0.381 kg(-1)). The mean population estimates of half-lives were 1.8 h and 8.0 h. In the context of published nomograms this analysis indicated that both the traditional approach and the new, 'once daily' approach should achieve satisfactory concentrations in cancer patients although serum concentration monitoring is required to confirm optimal dosing in individual patients.

Irinotecan (CPT-11) is an anticancer agent widely used to treat adult solid tumours. Large interindividual variability in the clearance of irinotecan and SN-38, its active and toxic metabolite, results in highly unpredictable toxicity. In 217 cancer patients treated with intravenous irinotecan single agent or in combination, germline DNA was used to interrogate the variation in 84 genes by next-generation sequencing. A stepwise analytical framework including a population pharmacokinetic model with SNP- and gene-based testing was used to identify demographic/clinical/genetic factors that influence the clearance of irinotecan and SN-38. Irinotecan clearance was influenced by rs4149057 in SLCO1B1, body surface area, and co-administration of 5-fluorouracil/leucovorin/bevacizumab. SN-38 clearance was influenced by rs887829 in UGT1A1, pre-treatment total bilirubin, and EGFR rare variant burden. Within each UGT1A1 genotype group, elevated pre-treatment total bilirubin and/or presence of at least one rare variant in EGFR resulted in significantly lower SN-38 clearance. The model reduced the interindividual variability in irinotecan clearance from 38 to 34% and SN-38 clearance from 49 to 32%. This new model significantly reduced the interindividual variability in the clearance of irinotecan and SN-38. New genetic factors of variability in clearance have been identified.

Objective: The pharmacokinetics (PK) of docetaxel are characterized by large inter-individual variability in systemic drug exposure (AUC) and drug clearance. The PK variability is thought to be largely related to differences in the catalytic function of CYP3A, involved in docetaxel metabolism and elimination. As variability in efficacy and toxicity is associated with variability in docetaxel AUC and clearance, reducing inter-individual PK variability may help improve the risk-benefit ratio of docetaxel therapy. We investigated if high-dose ketoconazole, a potent CYP3A inhibitor, could result in a uniform reduction of docetaxel clearance and reduce the inter-individual variability in docetaxel AUC and clearance.Methods: Seven patients were treated in a randomized-cross over design with intravenous docetaxel (100 mg/m2) followed 3 weeks later by docetaxel (15 mg/m2) given in combination with orally administered ketoconazole (400 mg 3 times daily, up to 47 hours after docetaxel infusion) or vice versa. Docetaxel plasma concentration-time data were described by a three-compartment PK model. Ketoconazole plasma concentration-time data were described by a one-compartment PK model. Results: Docetaxel clearance was reduced by 50% (P = .018) from 32.8 ± 13.7 L/hr to 16.5 ± 8.15 L/hr upon ketoconazole co-administration, albeit with large inter-individual variability (fractional change in clearance, range 0.31 – 0.66). In the presence of ketoconazole, inter-individual variability in clearance and AUC, expressed as coefficient of variation, was increased from 41.6 to 49.5% and from 28.0 to 35.1%, respectively, and not, as we had hypothesized, reduced.Conclusion: Inhibition of CYP3A by concomitant high-dose ketoconazole administration does not result in a uniform reduction of docetaxel clearance and does not reduce the inter-individual variability in docetaxel AUC or clearance. This approach is unsuitable as method to achieve a uniform docetaxel PK profile.

There may be a difference between the determinants of amikacin exposure in emergency department (ED) versus intensive care (ICU) patients, and the peak amikacin concentration varies widely between patients. Moreover, when the first dose of antimicrobials is administered to septic patients admitted to the ED, fluid resuscitation and vasopressors have just been initiated. Nevertheless, population pharmacokinetic modelling data for amikacin in ED patients are unavailable. The aim of this study was to quantify the interindividual variability (IIV) in the pharmacokinetics of amikacin in patients admitted to the ED and to identify the patient characteristics that explain this IIV. Patients presenting at the ED with severe sepsis or septic shock were randomly assigned to receive amikacin 25 mg/kg or 15 mg/kg intravenously. Blood samples were collected at 1, 6 and 24 h after the onset of the first amikacin infusion. Data were analysed using nonlinear mixed-effects modelling. A two-compartment population pharmacokinetic model was developed based on 279 amikacin concentrations from 97 patients. The IIV in clearance (CL) and central distribution volume (V1) were 71% and 26%, respectively. Body mass index (BMI), serum total protein level, serum sodium level, and fluid balance 24 h after amikacin administration explained 30% of the IIV in V1, leaving 18% of the IIV unexplained. BMI and creatinine clearance according to the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation 24 h after amikacin administration explained 46% of the IIV in CL, and 39% remained unexplained. The IIV of amikacin pharmacokinetics in ED patients is large. Higher doses may be considered in patients with low serum sodium levels, low total protein levels, or a high fluid balance. ClinicalTrials.gov ID: NCT02365272.

Development and internal validation of a clinical prediction model for individualized dosing of BAY 81-8973, A full-length recombinant factor VIII, in pediatric patients with haemophilia A

Objectives: Nadroparin, a low-molecular-weight-heparin is commonly used off-label in neonates and infants for thromboembolic events prevention. However, the recommended dosing regimen often fails to achieve therapeutic target ranges. This study aimed to develop a population pharmacokinetic (PK) model of nadroparin to determine an appropriate dosing regimen for neonates and infants less than 8 months. Methods: A retrospective chart review was conducted on patients treated with nadroparin at Children's Hospital of Fudan University between July 2021 and December 2023. A population PK model was developed using anti-Xa levels, and its predictive performance was evaluated internally. Monte Carlo simulations were performed to design an initial dosing schedule targeting anti-Xa levels between 0.5 and 1IU/mL. Results: A total of 40 neonates and infants aged less than 8 months with gestational age ranging from 25 to 41 weeks treated with nadroparin were enrolled in the study for analysis. A one-compartment PK model with first order absorption and elimination was adequately fitted to the data. Creatinine clearance was identified as a significant factor contributing to inter-individual variability in clearance. The typical population parameter estimates of clearance, distribution volume and absorption rate in this population were 0.211L/h, 1.55L and 0.495h-1, respectively. Our findings suggest that current therapeutic doses of nadroparin (150-200IU/kg q12h) may result in subtherapeutic exposure, thus higher doses might be required. Conclusion: The present study offers the first estimation of PK parameters for nadroparin in preterm or term neonates and infants less than 8 months utilizing the model. Our findings have potential implications for recommending initial personalized dosages, particularly among patient populations exhibiting similar characteristics.

To assess the appropriateness of the body weight or fixed dosing regimen, a population pharmacokinetic (PopPK) model of kukoamine B has been built in sepsis patients. Plasma concentrations of kukoamine B and the covariates information were taken from 30 sepsis patients assigned into 0.06 mg/kg, 0.12 mg/kg and 0.24 mg/kg groups in a Phase IIa clinical trial. The PopPK model was built using a nonlinear mixed-effect (NLME) modelling approach. Based on the final model, PK profiles were respectively simulated 500 times applying the body weight and renal function information of 12 sepsis patients from the 0.24 mg/kg group on the body weight or the fixed dosing regimen. For each dosing regimen, PK profiles of 6000 virtual patients were obtained. Statistical analyses for Cmax and Cmin were performed. If the biases of Cmax and Cmin can all meet the criteria of ±15%, the fixed dosing regimen can substitute for the body weight dosing regimen. The PopPK model was successfully developed using the NLME approach. A bi-compartmental model was selected as the basic model. Renal function was identified as a statistically significant covariate of systemic clearance with the objective function value (OFV) decreasing 8.6, resulting in a 5.2% decrease in inter-individual variability (IIV) of systemic clearance. Body weight was not identified as a statistically significant covariate. Simulation results demonstrated two methods had a bias of 8.1% for Cmax , and 8.6% for Cmin . Furthermore, PK variability was lower on the fixed dosing regimen than the body weight regimen. Based on the simulation results, a fixed dosing regimen was recommended in the subsequent clinical trials.

Caffeine has been shown to maintain or improve the performance of individuals, but its pharmacokinetic profile for Asians has not been well characterized. In this study, a population pharmacokinetic model for describing the pharmacokinetics of caffeine in Singapore males was developed. The data were also analysed using non-compartmental models. Data gathered from 59 male volunteers, who each ingested a single caffeine capsule in two clinical trials (3 or 5 mg/kg), were analysed via non-linear mixed-effects modelling. The participants' covariates, including age, body weight, and regularity of caffeinated-beverage consumption or smoking, were analysed in a stepwise fashion to identify their potential influence on caffeine pharmacokinetics. The final pharmacostatistical model was then subjected to stochastic simulation to predict the plasma concentrations of caffeine after oral (204, 340 and 476 mg) dosing regimens (repeated dosing every 6, 8 or 12 h) over a hypothetical 3-day period. The data were best described by a one-compartmental model with first-order absorption and first-order elimination. Smoking status was an influential covariate for clearance: clearance (mL/min) = 110*SMOKE + 114, where SMOKE was 0 and 1 for the non-smoker and the smoker respectively. Interoccasion variability was smaller compared to interindividual variability in clearance, volume and absorption rate (27% vs. 33%, 10% vs. 15% and 23% vs. 51% respectively). The extrapolated elimination half-lives of caffeine in the non-smokers and the smokers were 4.3 +/- 1.5 and 3.0 +/- 0.7 h respectively. Dosing simulations indicated that dosing regimens of 340 mg (repeated every 8 h) and 476 mg (repeated every 6 h) should achieve population-averaged caffeine concentrations within the reported beneficial range (4.5-9 microg/mL) in the non-smokers and the smokers respectively over 72 h. The population pharmacokinetic model satisfactorily described the disposition and variability of caffeine in the data. Mixed-effects modelling showed that the dose of caffeine depended on cigarette smoking status.

Teicoplanin is a glycopeptide antibiotic used for the treatment of methicillin-resistant Staphylococcus aureus infections. To ensure successful target attainment, therapeutic drug monitoring-informed dosage adjustment is recommended. However, it relies on the experience of the clinician and the frequency of drug measurements. This study aimed to design a new optimal dosing regimen of teicoplanin with a maintenance dosing strategy for neonates and children based on their physiological characteristics. Data from teicoplanin-treated patients (n = 214) were collected from electronic medical records. Covariate analyses were performed using population pharmacokinetic (PK) modeling with 399 serum teicoplanin concentrations from 48 neonates and 166 children. Multiple PK simulations were conducted to explore optimal dosing regimens that would allow control of the trough concentration to the target of 15-30 mg/L quicker than the current standard regimen. Allometrically scaled body weight, postmenstrual age (PMA), renal function, and serum albumin were implemented as substantial covariates for teicoplanin clearance in a two-compartment PK model. Covariate analyses and comprehensive simulation assessments recommended the following modifications to the current regimen: (1) decreased dose for premature babies (PMA ≤28 weeks), (2) decreased dose for children with renal dysfunction, and (3) increased dose for children (0.5-11 years) with an estimated glomerular filtration rate of ≥90 mL/min/1.73 m2. This study leverages real-world clinical information and proposes new optimal dosing regimens for teicoplanin in neonates and children through PK modeling and simulation analyses, taking into account the age, including PMA, and renal function of patients.

Antibiotic Pharmacokinetic and Pharmacodynamic Considerations in Patients With Kidney Disease

Amikacin is an aminoglycoside commonly used in intensive care units for the treatment of patients with life-threatening Gram-negative infections. Although aminoglycosides are extensively used, the accurate determination of their optimal dosage is complicated by marked intra- and interindividual variability in intensive care unit patients. Amikacin pharmacokinetics have been described in numerous studies over the past 25 years. This review presents a synthesis of the population pharmacokinetic models for amikacin described in critically ill patients. The objective was to determine whether there was a consensus on a structural model and which covariates had been identified. A literature search was conducted from the PubMed database, from its inception up until December 2015, using the following terms: 'amikacin', 'pharmacokinetic(s)', 'population', 'model(ling)' and 'nonlinear mixed effect'. Articles were excluded if they were not pertinent. The reference lists of all selected articles were also evaluated. Ten articles were included in this review: pharmacokinetics of amikacin were described by a one-compartment or a two-compartment model. Various covariates were tested, but only two (creatinine clearance and total body weight) were included in almost all of the described models. After inclusion of these covariates, the interindividual variability (range) in clearance and the volume of distribution were 44.4 % (28.2-69.4 %) and 31.3 % (8.1-44.7 %), respectively. The residual variability (range) was around 21.0% (9.0-31.0 %), using a proportional model, and for a combined model (proportional/additive), the median (range) values were 0.615 mg/L (0.2-1.03 mg/L) and 29.2 % (26.8-31.6 %). This review highlights the different population pharmacokinetic models for amikacin developed in critically ill patients over the past decades and proposes relevant information for clinicians and researchers. To optimize amikacin dosage, this review points out the relevant covariates according to the target population. In a population of critically ill patients, dose optimization mainly depends on creatinine clearance and total body weight. New pharmacokinetic population studies could be considered, with new covariates of interest to be tested in model building and to further explain variability. Another future perspective could be external evaluation of previously published models.

Background and ObjectivePrevious pharmacokinetic (PK) studies of ciprofloxacin in intensive care (ICU) patients have shown large differences in estimated PK parameters, suggesting that further investigation is needed for this population. Hence, we performed a pooled population PK analysis of ciprofloxacin after intravenous administration using individual patient data from three studies. Additionally, we studied the PK differences between these studies through a post-hoc analysis.MethodsIndividual patient data from three studies (study 1, 2, and 3) were pooled. The pooled data set consisted of 1094 ciprofloxacin concentration–time data points from 140 ICU patients. Nonlinear mixed-effects modeling was used to develop a population PK model. Covariates were selected following a stepwise covariate modeling procedure. To analyze PK differences between the three original studies, random samples were drawn from the posterior distribution of individual PK parameters. These samples were used for a simulation study comparing PK exposure and the percentage of target attainment between patients of these studies.ResultsA two-compartment model with first-order elimination best described the data. Inter-individual variability was added to the clearance, central volume, and peripheral volume. Inter-occasion variability was added to clearance only. Body weight was added to all parameters allometrically. Estimated glomerular filtration rate on ciprofloxacin clearance was identified as the only covariate relationship resulting in a drop in inter-individual variability of clearance from 58.7 to 47.2%. In the post-hoc analysis, clearance showed the highest deviation between the three studies with a coefficient of variation of 14.3% for posterior mean and 24.1% for posterior inter-individual variability. The simulation study showed that following the same dose regimen of 400 mg three times daily, the area under the concentration–time curve of study 3 was the highest with a mean area under the concentration–time curve at 24 h of 58 mg·h/L compared with that of 47.7 mg·h/L for study 1 and 47.6 mg·h/L for study 2. Similar differences were also observed in the percentage of target attainment, defined as the ratio of area under the concentration–time curve at 24 h and the minimum inhibitory concentration. At the epidemiological cut-off minimum inhibitory concentration of Pseudomonas aeruginosa of 0.5 mg/L, percentage of target attainment was only 21%, 18%, and 38% for study 1, 2, and 3, respectively.ConclusionsWe developed a population PK model of ciprofloxacin in ICU patients using pooled data of individual patients from three studies. A simple ciprofloxacin dose recommendation for the entire ICU population remains challenging owing to the PK differences within ICU patients, hence dose individualization may be needed for the optimization of ciprofloxacin treatment.

Imatinib, a tyrosine kinase inhibitor, is the drug of choice for the treatment of chronic myeloid leukemia in Nigeria. Several studies have established interindividual and interpopulation variations in imatinib disposition although no pharmacokinetic study have been conducted in an African population since the introduction of the drug. This study explored a population pharmacokinetic approach to investigate the disposition of imatinib in Nigerians and examined the involvement of some covariates including genetic factors in the variability of the drug disposition with a view to optimize the use of the drug in this population. A total of 250 plasma concentrations from 126 chronic myeloid leukemia patients were quantified using a validated method. A population pharmacokinetic model was fitted to the data using NONMEM VII software, and the influences of 12 covariates were investigated. The mean population-derived apparent steady-state clearance, elimination half-life, area under the concentration-time curve over 24 hours, and volume of distribution were 17.2 ± 1.8 L/h., 12.05 ± 2.1 hours, 23.26 ± 0.6 μg·h/mL, and 299 ± 20.4 L, respectively. Whole blood count, ethnicity, CYP3A5*3, and ABCB1 C3435T were found to have significant influence on the apparent clearance, while the interindividual variability in clearance and interoccasion variability in bioavailability were 17.4% and 20.4%, respectively. There was a wide variability in apparent clearance and area under the curve compared to those reported in other populations. Thus, treatment with a standard dose of imatinib in this population may not produce the desired effect in most of the patients, whereas continuous exposure to a low drug concentration could lead to pharmacokinetic-derived resistance. The authors suggest the need for therapeutic drug monitoring-guided dose individualization in this population.