Statistical tools for dose individualization of mycophenolic acid and tacrolimus co-administered during the first month after renal transplantation.

AIM: To predict simultaneously the area under the concentration-time curve during one dosing interval [AUC(0,12 h)] for mycophenolic acid (MPA) and tacrolimus (TAC), when concomitantly used during the first month after transplantation, based on common blood samples. METHODS: Data were from two different sources, real patient pharmacokinetic (PK) profiles from 65 renal transplant recipients and 9000 PK profiles simulated from previously published models on MPA or TAC in the first month after transplantation. Multiple linear regression (MLR) and Bayesian estimation using optimal samples were performed to predict MPA and TAC AUC(0,12 h) based on two concentrations. RESULTS: The following models were retained: AUC(0,12 h) = 16.5 + 4.9 × C1.5 + 6.7 × C3.5 (r(2) = 0.82, rRMSE = 9%, with simulations and r(2) = 0.66, rRMSE = 24%, with observed data) and AUC(0,12 h) = 24.3 + 5.9 × C1.5 + 12.2 × C3.5 (r(2) = 0.94, rRMSE = 12.3%, with simulations r(2) = 0.74, rRMSE = 15%, with observed data) for MPA and TAC, respectively. In addition, bayesian estimators were developed including parameter values from final models and values of concentrations at 1.5 and 3.5 h after dose. Good agreement was found between predicted and reference AUC(0,12 h) values: r(2) = 0.90, rRMSE = 13% and r(2) = 0.97, rRMSE = 5% with simulations for MPA and TAC, respectively and r(2) = 0.75, rRMSE = 11% and r(2) = 0.83, rRMSE = 7% with observed data for MPA and TAC, respectively. CONCLUSION: Statistical tools were developed for simultaneous MPA and TAC therapeutic drug monitoring. They can be incorporated in computer programs for patient dose individualization.

Population pharmacokinetic modeling and optimal sampling strategy for Bayesian estimation of amikacin exposure in critically ill septic patients.

Because the sepsis-induced pharmacokinetic (PK) modifications need to be considered in aminoglycoside dosing, the present study aimed to develop a population PK model for amikacin (AMK) in severe sepsis and to subsequently propose an optimal sampling strategy suitable for Bayesian estimation of the drug PK parameters. Concentration-time profiles for AMK were obtained from 88 critically ill septic patients during the first 24 hours of antibiotic treatment. The population PK model was developed using a nonlinear mixed effects modeling approach. Covariate analysis included demographic data, pathophysiological characteristics, and comedication. Optimal sampling times were selected based on a robust Bayesian design criterion. Taking into account clinical constraints, a two-point sampling approach was investigated. A two-compartment model with first-order elimination best fitted the AMK concentrations. Population PK estimates were 19.2 and 9.34 L for the central and peripheral volume of distribution and 4.31 and 2.21 L/h for the intercompartmental and total body clearance. Creatinine clearance estimated using the Cockcroft-Gault equation was retained in the final model. The two optimal sampling times were 1 hour and 6 hours after onset of the drug infusion. Predictive performance of individual Bayes estimates computed using the proposed optimal sampling strategy was reported: mean prediction errors were less than 5% and root mean square errors were less than 30%. The present study confirmed the significant influence of the creatinine clearance on the PK disposition of AMK during the first hours of treatment in critically ill septic patients. Based on the population estimates, an optimal sampling strategy suitable for Bayesian estimation of the drug PK parameters was developed, meeting the need of clinical practice.

Comparative in vitro antimicrobial potency, stability, colouration and dissolution time of generics versus innovator of meropenem in Europe.

Meropenem generics are often imposed on prescribers, however scarce information is available on key properties such as antimicrobial potency, stability and colouration in solution, and dissolution time. This study aimed to generate comparative information for products available in Europe. The originator (ASTRA) and four generics (HOSPIRA, SANDOZ, FRESENIUS and AUROVIT) were compared for: (i) MICs against Pseudomonas aeruginosa clinical isolates (range, 0.125-191 mg/L); (ii) colouration (visual and photometry) and stability of concentrated solutions for prolonged or continuous infusion and maintained at 25-37 °C for up to 8 h (acceptable limit, ≥90% of original concentration); and (iii) dissolution time of concentrated solutions (50 mg/mL [for bolus administration]: turbidimetry and nursing personnel assessment). No significant difference was observed for MICs (except 2/80 isolates). For concentrated solutions storage: (i) SANDOZ produced about two times more yellow-coloured degradation products than the other preparations; (ii) meropenem loss was time-, concentration- and temperature-dependent; (iii) FRESENIUS was the least stable (limit for 1 g/48 mL, ~8 h at 25 °C and 4.5 h at 37 °C); (iv) at 2 g/48 mL, the storage time limit was 5-6 h at 25 °C and ~3 h at 37 °C for all preparations. Complete dissolution (turbidimetry) required 240 s for generics (120 s for ASTRA), and nurses reported longer but highly variable times for generics. Substantial differences between innovator and generics have been identified that could impact on their clinical use and/or make multicentric studies difficult to interpret, requiring suitability studies in the environments of their intended use.

What is the optimal loading dose of broad-spectrum ß-lactam antibiotics in septic patients? Results from pharmacokinetic simulation modelling.

Optimal loading doses of ß-lactams to rapidly achieve adequate drug concentrations in critically ill patients are unknown. This was a post-hoc analysis of a prospective study that evaluated broad-spectrum ß-lactams [piperacillin (PIP), ceftazidime (CAZ), cefepime (FEP) and meropenem (MEM)] pharmacokinetics (PKs) in patients with sepsis or septic shock (n = 88). Monte Carlo simulation was performed for 1000 virtual patients using specific sets of covariates for various dosing regimens and different durations of administration. Pharmacodynamic (PD) targets were considered as drug concentrations exceeding at least 50% of time above four times the minimum inhibitory concentration (T>4 × MIC) of Pseudomonas aeruginosa, according to EUCAST criteria, for PIP, 70%T>4 × MIC for CAZ and FEP and 40%T>4 × MIC for MEM. The probability of target attainment (PTA) was derived by calculating the percentage of patients who attained the PK/PD target at each MIC. The optimal loading dose was defined as the one associated with a ≥90% probability to achieve the PD targets. Our simulation model identified an optimal loading dose for PIP of 8 g given as a 3-h infusion (PTA of 96.2%), for CAZ and FEP of 4 g given as a 3-h infusion (PTA of 96.5% and 98.4%, respectively), and for MEM of 2 g given as a 30-min infusion (PTA of 93.4%), with the following antibiotic dose administered 6 h thereafter regardless of the drug. A higher first dose of broad-spectrum ß-lactams should be given to adequately treat less-susceptible pathogens in septic patients. These findings need to be validated in a prospective study.

Determination of optimal loading and maintenance doses for continuous infusion of vancomycin in critically ill patients: Population pharmacokinetic modelling and simulations for improved dosing schemes.

OBJECTIVES: Despite extensive clinical use, limited data are available on optimal loading and maintenance doses of vancomycin in critically ill patients. This study aimed to develop a rational approach for optimised dosage of vancomycin given in a continuous infusion in critically ill patients. METHODS: Vancomycin pharmacokinetic (PK) data (total serum concentrations) were obtained from 55 intensive care unit (ICU) patients (Bach Mai Hospital, Hanoi, Vietnam) receiving a 20 mg/kg loading dose followed by continuous infusion stratified by creatinine clearance (CLCr). Population PK modelling and Monte Carlo simulations were performed using a nonlinear mixed-effects modelling (NONMEM) program for a target of 20-30 mg/L to optimise efficacy and minimise nephrotoxicity. RESULTS: A two-compartment model with first-order elimination best fitted the PK data with central and peripheral volumes of distribution of 1.01 and 2.39 L/kg, respectively (allometric scaling to a 70 kg standard subject). The population total clearance of 3.63 L/h was only explained by renal function in the covariate and final model. The simulations showed that a 25-mg/kg loading dose infused over 90 minutes was optimal to reach the target range. The optimal maintenance dose for low renal function (CLCr < 45 mL/min) was 1000-1500 mg/day. For augmented renal clearance (CLCr > 130 mL/min) the dose should be up to 3500 mg/day or even 4500 mg/day to achieve adequate exposure. These simulated maintenance doses were larger than previously proposed for non-ICU patients. CONCLUSION: Large loading and maintenance doses of vancomycin are generally needed in critically ill patients. Because of high interindividual variability in vancomycin PK, drug monitoring may still be necessary.

Optimizing ß-lactams treatment in critically-ill patients using pharmacokinetics/pharmacodynamics targets: are first conventional doses effective?

INTRODUCTION: The pharmacokinetic/pharmacodynamic index determining ß-lactam activity is the percentage of the dosing interval (%T) during which their free serum concentration remains above a critical threshold over the minimum inhibitory concentration (MIC). Regrettably, neither the value of %T nor that of the threshold are clearly defined for critically-ill patients. Areas covered: We review and assess the targets proposed for ß-lactams in critical illness by screening the literature since 1997. Depending on the study intention (clinical cure vs. suppression of resistance), targets proposed range from 20%T > 1xMIC to 100%T > 5xMIC. Assessment and comparative analysis of their respective clinical efficacy suggest that a value of 100%T > 4xMIC may be needed. Simulation studies, however, show that this target will not be reached at first dose for the majority of critically-ill patients if using the most commonly recommended doses. Expert commentary: Considering that critically-ill patients are highly vulnerable and likely to experience antibiotic underexposure, and because effective initial treatment is a key determinant of clinical outcome, we support the use of a target of 100%T > 4xMIC, which could not only maximize efficacy but also minimize emergence of resistance. Clinical and microbiological studies are needed to test for the feasibility and effectiveness of reaching such a demanding target.

A simultaneous d-optimal designed study for population pharmacokinetic analyses of mycophenolic Acid and tacrolimus early after renal transplantation.

Mycophenolic acid (MPA) and tacrolimus (TAC) are immunosuppressive agents used in combination with corticosteroids for the prevention of acute rejection after solid organ transplantation. Their pharmacokinetics (PK) show considerable unexplained intraindividual and interindividual variability, particularly in the early period after transplantation. The main objective of the present work was to design a study based on D-optimality to describe the PK of the 2 drugs with good precision and accuracy and to explain their variability by means of patients' demographics, biochemical test results, and physiological characteristics. Pharmacokinetic profiles of MPA and TAC were obtained from 65 stable adult renal allograft recipients on a single occasion (ie, day 15 after transplantation). A sampling schedule was estimated based on the D-optimality criterion with the POPED software, using parameter values from previously published studies on MPA and TAC modeling early after transplantation. Subsequently, a population PK model describing MPA and TAC concentrations was developed using nonlinear mixed-effects modeling. Optimal blood-sampling times for determination of MPA and TAC concentrations were estimated to be at 0 (predose) and at 0.24, 0.64, 0.98, 1.37, 2.38, and 11 hours after oral intake of mycophenolate and TAC. The PK of MPA and TAC were best described by a 2-compartment model with first-order elimination. For MPA, the absorption was best described by a transit compartment model, whereas first-order absorption with a lag time best described TAC transfer from the gastrointestinal tract. Parameters were estimated with good precision and accuracy. While hematocrit levels and CYP3A5 genetic polymorphism significantly influenced TAC clearance, the pharmaceutical formulation and MRP2 genetic polymorphism were retained as significant covariates on MPA absorption and elimination, respectively. The prospective use of the simultaneous D-optimal design approach for MPA and TAC has allowed good estimation of MPA and TAC PK parameters in the early period after transplantation characterized by a very high unexplained variability. The influence of some relevant covariates could be shown.

Population pharmacokinetics of four ß-lactams in critically ill septic patients comedicated with amikacin.

OBJECTIVES: The study aimed to characterize the pharmacokinetics (PK) of four ß-lactams (piperacillin, ceftazidime, cefepime, and meropenem) in patients comedicated with amikacin (AMK), and to confirm the predictive performance of AMK data, obtained from therapeutic drug monitoring (TDM), on these PK, using a population modeling approach. DESIGN AND METHODS: Serum samples were collected in 88 critically ill septic patients. For each ß-lactam, the covariate model was optimized using renal function. Furthermore, predictive performance of AMK concentrations and PK parameters was assessed on ß-lactam PK. RESULTS: A two-compartment model with first-order elimination best fitted the ß-lactam data. Results supported the superiority of AMK concentrations, over renal function and AMK PK parameters, to assess the ß-lactam PK. CONCLUSION: The study confirmed the significant link between the exposure to AMK and to ß-lactams, and presented population models able to guide ß-lactam dosage adjustments using renal biomarkers or TDM-related aminoglycoside data.

Empirical models for dosage optimization of four beta-lactams in critically ill septic patients based on therapeutic drug monitoring of amikacin.

OBJECTIVES: The study aims to develop empirical models able to predict the pharmacokinetics (PK) of four beta-lactams using the amikacin (AMK) therapeutic drug monitoring (TDM), in order to optimize their dosage regimens. DESIGN AND METHODS: 69 critically ill septic patients were included. All received a first dose of AMK combined with piperacillin/tazobactam, ceftazidime, cefepime or meropenem. A multivariate analysis was performed to predict the beta-lactam PK using AMK PK parameters estimated from TDM and using pathophysiological variables. RESULTS: An optimal prediction model was identified for each PK parameter of each beta-lactam. The best predictor of each model was one of the AMK PK parameters estimated from TDM. Other variables included colloid solution, renal and hepatic biomarkers, age and body weight. CONCLUSION: PK of the four beta-lactams could be easily and rapidly predicted in critically ill septic patients using the AMK TDM. These predictions could improve the beta-lactam dosages in clinical practice.