Bayesian estimation in NONMEM.

Bayesian estimation is a powerful but underutilized tool for answering drug development questions. In this tutorial, the principles of Bayesian model development, assessment, and prior selection will be outlined. An example pharmacokinetic (PK) model will be used to demonstrate the implementation of Bayesian modeling using the nonlinear mixed-effects modeling software NONMEM.

Bayesian modeling and simulation to inform rare disease drug development early decision-making: Application to Duchenne muscular dystrophy.

Rare disease clinical trials are constrained to small sample sizes and may lack placebo-control, leading to challenges in drug development. This paper proposes a Bayesian model-based framework for early go/no-go decision making in rare disease drug development, using Duchenne muscular dystrophy (DMD) as an example. Early go/no-go decisions were based on projections of long-term functional outcomes from a Bayesian model-based analysis of short-term trial data informed by prior knowledge based on 6MWT natural history literature data in DMD patients. Frequentist hypothesis tests were also applied as a reference analysis method. A number of combinations of hypothetical trial designs, drug effects and cohort comparison methods were assessed. The proposed Bayesian model-based framework was superior to the frequentist method for making go/no-go decisions across all trial designs and cohort comparison methods in DMD. The average decision accuracy rates across all trial designs for the Bayesian and frequentist analysis methods were 45.8 and 8.98%, respectively. A decision accuracy rate of at least 50% was achieved for 42 and 7% of the trial designs under the Bayesian and frequentist analysis methods, respectively. The frequentist method was limited to the short-term trial data only, while the Bayesian methods were informed with both the short-term data and prior information. The specific results of the DMD case study were limited due to incomplete specification of individual-specific covariates in the natural history literature data and should be reevaluated using a full natural history dataset. These limitations aside, the framework presented provides a proof of concept for the utility of Bayesian model-based methods for decision making in rare disease trials.

Population pharmacokinetics of adalimumab biosimilar adalimumab-adbm and reference product in healthy subjects and patients with rheumatoid arthritis to assess pharmacokinetic similarity.

AIMS: Adalimumab-adbm is a monoclonal antibody developed as a biosimilar to adalimumab (Humira, AbbVie Inc.). The key objectives of this study were using a population pharmacokinetic (PPK) approach to assess pharmacokinetic (PK) similarity between adalimumab-adbm and Humira in patients with active rheumatoid arthritis (RA), to quantify the effects of potential covariates on adalimumab PK and to assess the impact of switching treatment from Humira to adalimumab-adbm on PK. METHODS: A PPK model was firstly developed using intensive PK data from the phase-1 study in healthy subjects (NCT02045979). PPK models were developed separately for phase-3 base study (NCT02137226) and its extension study (NCT02640612) in patients with active RA. RESULTS: PPK models were developed for adalimumab from adalimumab-adbm and Humira treatment in healthy subjects and RA patients. Weight and anti-drug antibodies were found to be important predictors of adalimumab clearance. Adalimumab PK was similar between adalimumab-adbm and Humira. The estimated effect of Humira on clearance, relative to the adalimumab-adbm, was 1.02 (i.e., Humira has 0.02 greater clearance). Similarly, the effect of treatment arms (switching) on clearance was estimated to be 1.00 and 0.997 for Humira:Humira:BI and Humira:BI:BI arms, respectively, relative to the BI:BI:BI arm (BI refers to adalimumab-adbm) in the phase-3 extension study. CONCLUSION: PK similarity between adalimumab-adbm and Humira in patients with active RA was demonstrated using PPK approach. Adalimumab PK was also similar when switching treatment from Humira to adalimumab-adbm at either week 24 or 48.

Latent process model of the 6-minute walk test in Duchenne muscular dystrophy : A Bayesian approach to quantifying rare disease progression.

Duchenne muscular dystrophy (DMD) is a rare X-linked genetic pediatric disease characterized by a lack of functional dystrophin production in the body, resulting in muscle deterioration. Lower body muscle weakness progresses to non-ambulation typically by early teenage years, followed by upper body muscle deterioration and ultimately death by the late twenties. The objective of this study was to enhance the quantitative understanding of DMD disease progression through nonlinear mixed effects modeling of the population mean and variability of the 6-min walk test (6MWT) clinical endpoint. An indirect response model with a latent process was fit to digitized literature data using full Bayesian estimation. The modeling data set consisted of 22 healthy controls and 218 DMD patients from one interventional and four observational trials. The model reasonably described the central tendency and population variability of the 6MWT in healthy subjects and DMD patients. An exploratory categorical covariate analysis indicated that there was no apparent effect of corticosteroid administration on DMD disease progression. The population predicted 6MWT began to rise at 1.32 years of age, plateauing at 654 meters (m) at 17.2 years of age for the healthy population. The DMD trajectory reached a maximum of 411 m at 8.90 years before declining and falling below 1 m at age 18.0. The model has potential to be used as a Bayesian estimation and posterior simulation tool to make informed model-based drug development decisions that incorporate prior knowledge with new data.

Population Pharmacokinetic- Pharmacodynamic Analysis to Characterize the Effect of Empagliflozin on Renal Glucose Threshold in Patients With Type 1 Diabetes Mellitus.

Sodium glucose cotransporter 2 inhibitors increase urinary glucose excretion (UGE) by lowering the renal threshold for glucose (RTG ). We aimed to quantify the effect of the sodium glucose cotransporter inhibitor empagliflozin on renal glucose reabsorption in patients with type 1 diabetes mellitus (T1DM) using a mechanistic population pharmacokinetic-pharmacodynamic (PK-PD) model and to compare results with analyses in patients with type 2 diabetes mellitus (T2DM). The PK-PD model was developed using data from a randomized phase 2 study in which patients with T1DM received oral once-daily empagliflozin 2.5 mg, empagliflozin 10 mg, empagliflozin 25 mg, or placebo as an adjunct to insulin. The model assumed that UGE was dependent on plasma glucose and renal function and that empagliflozin lowered RTG . The final model was evaluated using visual predictive checks and found to be consistent with observed data. Calculated RTG with placebo was 181 mg/dL, and with empagliflozin (steady state) 1 mg and 2.5 mg was 53.4 mg/dL and 12.5 mg/dL, respectively. Empagliflozin 10 mg and 25 mg yielded negative RTG values, implying RTG was reduced to a negligible value. Although estimated PK-PD parameters were generally comparable between patients with T1DM and patients with T2DM, slight differences were evident, leading to lower RTG and higher UGE in patients with T1DM compared with patients with T2DM. In conclusion, the model provided a reasonable description of UGE in response to administration of empagliflozin and placebo in patients with T1DM.

Efficacy Projection of Obiltoxaximab for Treatment of Inhalational Anthrax across a Range of Disease Severity.

Inhalational anthrax has high mortality even with antibiotic treatment, and antitoxins are now recommended as an adjunct to standard antimicrobial regimens. The efficacy of obiltoxaximab, a monoclonal antibody against anthrax protective antigen (PA), was examined in multiple studies conducted in two animal models of inhalational anthrax. A single intravenous bolus of 1 to 32 mg/kg of body weight obiltoxaximab or placebo was administered to New Zealand White rabbits (two studies) and cynomolgus macaques (4 studies) at disease onset (significant body temperature increase or detection of serum PA) following lethal challenge with aerosolized Bacillus anthracis spores. The primary endpoint was survival. The relationship between efficacy and disease severity, defined by pretreatment bacteremia and toxemia levels, was explored. In rabbits, single doses of 1 to 16 mg/kg obiltoxaximab led to 17 to 93% survival. In two studies, survival following 16 mg/kg obiltoxaximab was 93% and 62% compared to 0% and 0% for placebo (P = 0.0010 and P = 0.0013, respectively). Across four macaque studies, survival was 6.3% to 78.6% following 4 to 32 mg/kg obiltoxaximab. In two macaque studies, 16 mg/kg obiltoxaximab reduced toxemia and led to survival rates of 31%, 35%, and 47% versus 0%, 0%, and 6.3% with placebo (P = 0.0085, P = 0.0053, P = 0.0068). Pretreatment bacteremia and toxemia levels inversely correlated with survival. Overall, obiltoxaximab monotherapy neutralized PA and increased survival across the range of disease severity, indicating clinical benefit of toxin neutralization with obiltoxaximab in both early and late stages of inhalational anthrax.

Malabsorption blood test: Assessing fat absorption in patients with cystic fibrosis and pancreatic insufficiency.

The malabsorption blood test (MBT), consisting of pentadecanoic acid (PA), a free fatty acid, and triheptadecanoic acid (THA), a triglyceride that requires pancreatic lipase for absorption of the heptadecanoic acid (HA), was developed to assess fat malabsorption in patients with cystic fibrosis (CF) and pancreatic insufficiency (PI). The objective was to construct a population pharmacokinetic (PK) model to describe PA and HA disposition in healthy subjects and CF subjects. A model was simultaneously fit to PA and HA concentrations, consisting of 1-compartment disposition and a transit model to describe absorption. PA bioavailability estimates for CF subjects without pancreatic enzyme administration (1.07 [0.827, 1.42]) and with enzymes (0.88 [0.72, 1.09]) indicated PA absorption comparable to healthy subjects. HA bioavailability in CF without enzyme administration was 0.0292 (0.0192, 0.0459) and with enzymes increased to 0.606 (0.482, 0.823). In CF, compared with taking enzymes with the MBT, HA bioavailability was further decreased by factors of 0.829 (0.664, 0.979) and 0.78 (0.491, 1.13) with enzymes taken 30 and 60 minutes after MBT, respectively. The MBT detected differences in fat absorption in subjects with CF with and without enzyme administration and with changes in enzyme timing. Future studies will address application of the MBT in CF and other malabsorption diagnoses.

Delayed small bowel transit in children with cystic fibrosis and pancreatic insufficiency.

OBJECTIVE: Gastrointestinal disturbances are common in people with cystic fibrosis (CF); however, motility studies in this population have yielded inconsistent results. This study examined gastric emptying (GE) and small bowel transit (SBT) time in children with CF and pancreatic insufficiency compared with a healthy adult reference group. METHODS: Participants consumed an 8-ounce liquid test meal (approximately 550 calories, 32 g of fat) labeled with 300 µCi 99m technetium (Tc) sulfur colloid. Subjects with CF received a standard dose of pancreatic enzymes before consuming the test meal. GE and SBT were measured using a standard nuclear medicine scan. GE was determined after correcting for 99mTc decay in both anterior and posterior images. SBT was determined by following the movement of the tracer from the stomach to the cecum. The percentage arrival of total small bowel activity at the terminal ileum and cecum/ascending colon at 6 hours was used as an index of SBT. A 1-way analysis of covariance was performed for comparisons between groups after adjustment for age, sex, and body mass index. RESULTS: Subjects with CF (n = 16) had similar GE compared with the healthy reference group (n = 12); however, subjects with CF had significantly prolonged SBT time. At 6 hours, 37.2% ± 25.4% (95% CI 23.7-50.7) of the tracer reached the terminal ileum and colon compared with 68.6% ± 13.1% (95% CI 60.2-76.9) for the reference group (P < 0.001). After controlling for sex, age, and body mass index, this difference remained statistically significant (F = 12.06, adjusted R = 0.44, P < 0.002). CONCLUSIONS: Children with CF and pancreatic insufficiency had unaltered GE but delayed SBT time when taking pancreatic enzymes.

Diagnosing malabsorption with systemic lipid profiling: pharmacokinetics of pentadecanoic acid and triheptadecanoic acid following oral administration in healthy subjects and subjects with cystic fibrosis.

OBJECTIVE: A Malabsorption Blood Test (MBT) is proposed as an alternative method to the 72-hour stool and dietary collection for assessing the degree of fat malabsorption in people with pancreatic insufficiency. The MBT consists of a simultaneous oral dose of pentadecanoic acid (PA), a free fatty acid, and triheptadecanoic acid (THA), a triglyceride with three heptadecanoic (HA) saturated fatty acids requiring hydrolysis by pancreatic lipase before HA can be intestinally absorbed. The aim of this study is to demonstrate the ability of MBT to detect fat malabsorption in healthy adult subjects using the pancreatic lipase (PL) inhibitor Orlistat (Xenical®), and in subjects with CF and PI while on and off routine pancreatic enzyme doses. MATERIALS AND METHODS: The MBT with the PA and THA were delivered in a breakfast test meal (2.5 g PA and either 5 g or 8 g THA) to healthy adult subjects (ages 18 - 50 years, BMI 21 - 30) and to subjects with CF (> 12 years, FEV1% predicted > 40%), after a 12-hour fast and 24 hours without dairy foods. Serum levels of PA and HA were assessed by gas-liquid chromatography, from blood samples drawn prior to MBT and then hourly for 8 hours. For healthy subjects, the MBT was administered before and after Orlistat treatment, and in subjects with CF, both with subjects receiving routine pancreatic lipase treatment ("on enzyme") and also "off enzyme" treatment. Treatment groups were compared for baseline (C0) and maximum (Cmax) plasma concentrations of PA and HA over 8 hours: area under the curve (AUC) was calculated using linear trapezoid method. The ratio of HA to PA Cmax and AUC was also calculated and compared. RESULTS: For the healthy subjects (n = 15, 60% female, ages 21 - 49 years), absorption of HA was reduced 71% for Cmax (p < 0.001) and 65% for AUC (p = 0.001) after Orlistat treatment, and absorption of PA was unchanged. For subjects with CF (n = 6, 50% female, ages 13 - 19 years), absorption of HA was minimal with subjects "off enzymes" and increased significantly with subjects "on enzymes" while absorption of PA did not differ between groups. Enzyme administration resulted in increased Cmax HA/ PA ratios from 0.02 to 0.92 and from 0.05 to 0.73 in subjects with CF receiving 5.0 g and 8.0 g of THA, respectively. AUC HA/PA ratios showed similar increases. CONCLUSIONS: In this pilot and feasibility proof-of-concept study, the MBT, utilizing the relative absorption of HA to PA, two odd-chained fatty acids, responds to changes in fat absorption in healthy subjects using a lipase inhibitor and in subjects with CF while on or off enzyme therapy. The MBT holds promise to provide a more accurate, specific and acceptable alternative to the 72-hour stool collection to quantify pancreatic-based fat malabsorption in a variety of clinical and research contexts.

Plasma topiramate concentrations resulting from doses associated with neuroprotection against white matter injury and stroke in two strains of rat pups.

BACKGROUND: Cerebral white matter (WM) injury and stroke are common neuropathological injuries in newborns with congenital heart defects (CHDs) requiring surgery. Previous investigations in Long Evans rat pups subjected to hypoxia-ischemia found that intraperitoneal (i.p.) topiramate (TPM) at 30 mg/kg, but not 50 mg/kg, conferred neuroprotection. In Sprague-Dawley pups, a dose of 30 mg/kg protected against stroke. Concentrations associated with neuroprotective doses were not measured. The aims of this investigation were to determine concentrations associated with neuroprotective doses and to investigate the pharmacokinetics (PK) of i.p. TPM. METHODS: Concentration-time data following administration of 30 and 50 mg/kg doses were analyzed using nonlinear mixed-effect modeling. RESULTS: Mean predicted steady-state maximum and average concentrations following 30 mg/kg TPM were 31.3 and 16.8 µg/ml in Long Evans and 39.9 and 24.4 µg/ml in Sprague-Dawley pups. Mean predicted steady-state maximum and average concentrations following 50 mg/kg TPM were 52.1 and 28.1 µg/ml in Long Evans and 66.5 and 40.6 µg/ml in Sprague-Dawley pups. The apparent clearance (CL/F) and apparent volume of distribution (V/F) were 0.0470 ml/min and 22.2 ml, respectively, for Long Evans and 0.0325 ml/min and 19.7 ml, respectively, for Sprague-Dawley pups. CONCLUSION: TPM concentrations associated with neuroprotective doses were determined. Body size and strain were significant covariates on CL/F and V/F. Results provide targets for future neuroprotection studies.

Population pharmacokinetics of palivizumab, a humanized anti-respiratory syncytial virus monoclonal antibody, in adults and children.

Although it has been on the market for over a decade, confusion remains regarding the pharmacokinetics (PK) and optimal dosing of palivizumab, a humanized IgG1κ monoclonal antibody indicated for the prevention of serious lower respiratory tract disease caused by respiratory syncytial virus (RSV) in pediatric patients at high risk of RSV disease. The objectives of this analysis were to characterize the population PK of palivizumab in adults and children using nonlinear mixed-effect modeling, quantify the effects of individual covariates on variability in palivizumab disposition, and compare palivizumab exposures for various dosing scenarios. Palivizumab PK data from 22 clinical studies were used for model development. The model was developed using a two-stage approach: (i) a 2-compartment model with first-order absorption after intramuscular administration was fitted to adult data, and (ii) the same structural model was fitted to the sparse pediatric data using the NONMEM $PRIOR subroutine, with informative priors obtained from the adult analysis. Body weight and an age descriptor that combines gestational age and postnatal age (PAGE) using an asymptotic-exponential model best described palivizumab clearance in pediatric patients. Palivizumab clearance increased slightly from 10.2 ml/day to 11.9 ml/day as a function of PAGE ranging from 7 to 18 months. Covariate analysis indicated a 20% higher clearance in children with chronic lung disease and in children with antidrug antibody titer values of ≥80. These covariates did not substantially explain interindividual variability. In the label-indicated pediatric population, body weight was the primary demographic factor affecting palivizumab PK. Body weight-based dosing of 15 mg/kg yields similar palivizumab concentrations in children of different gestational and postnatal ages. Simulations demonstrated that there was little difference in palivizumab PK between healthy term and premature infants. Simulations also demonstrated that the 5 monthly palivizumab doses of 15 mg/kg, consistent with the label and studied in two randomized, clinical trials, provided greater and more prolonged palivizumab exposure than did an abbreviated dosing regimen of 3 monthly doses.

Population pharmacokinetics of intravenous ondansetron in oncology and surgical patients aged 1-48 months.

PURPOSE: Until recently, ondansetron was approved for the prevention of nausea and vomiting only in patients older than 2 years. However, as the use of ondansetron in patients younger than 2 years had been documented, characterization of ondansetron pharmacokinetics in this younger pediatric age group was warranted. METHODS: The pharmacokinetics of intravenously administered ondansetron were evaluated in oncology and surgical patients aged 1-48 months. Pooled data from 124 patients, including 745 pharmacokinetic samples, were analyzed using nonlinear mixed-effects modeling. RESULTS: Ondansetron pharmacokinetics were described by a two-compartment model. Body-size effects on ondansetron disposition were accounted for via standard allometric relationships, normalized to 10.4 kg. A maturation process with a half-life of approximately 4 months was incorporated to describe a decrease in clearance (CL) in infants. Clearance [95% confidence interval (CI)] for a typical patient was 1.53 (1.34-1.78) L/h/kg(0.75) with an interindividual variability of 56.8%. Ondansetron CL was reduced by 31%, 53%, and 76% for the typical 6-month-, 3-month-, and 1-month-old patient, respectively. Simulations showed that an ondansetron dose of 0.1 mg/kg in children younger than 6 months produced exposure similar to a 0.15-mg/kg dose in older children. CONCLUSIONS: The population pharmacokinetic analysis of ondansetron allows for characterization of individual patients based on body weight and age. It is recommended that patients younger than 4 months receiving ondansetron be closely monitored.

Population pharmacokinetics of the selective serotonin 5-HT1A receptor partial agonist piclozotan.

BACKGROUND/AIMS: Serotonin (5-HT) and its receptors are known to play important roles in various physiological and pathophysiological processes. The 5-HT1A receptor subtype is thought to be involved in psychiatric disorders, immunomodulation, and in cerebral ischemic conditions. Piclozotan, a selective and potent partial agonist of 5-HT1A, has been shown to be neuroprotective against ischemic neuronal damage in animal models. Its pharmacokinetics (PK), tolerability, and safety have been evaluated in patients with acute ischemic stroke. The aim of the study was to describe piclozotan PK, using population PK modeling. METHODS: A total of 1308 plasma piclozotan concentration measurements from 84 healthy subjects and 412 plasma piclozotan measurements from 74 stroke patients were included in the analysis. Covariates considered during the model building process included disease status, age, weight, sex, smoking status, and alcohol consumption. Data were analyzed using nonlinear mixed-effects modeling with the NONMEM software system. The final model was qualified via predictive check and nonparametric bootstrap procedures. RESULTS: Piclozotan PK was well described using a 3-compartment model with first-order elimination. Parameter estimates (intersubject variability) were V1, central volume: 64.0 L (66.5%) and CL, systemic clearance: 18.0 L/h (31.4%). Peripheral volumes (V2, V3) were related to total body weight, whereas CL was related to ideal body weight. Clearance decreased with advancing age, yielding a decrease of 35%-65% in patients aged 70-90. There was no discernable difference in PK between healthy subjects and stroke patients. CONCLUSIONS: Piclozotan disposition was well described by the population model, and the intersubject variability around the estimated parameters was moderate in magnitude (<60%). The population PK analysis of piclozotan allows for characterization of piclozotan exposure in individual subjects based on their age and body weight. The availability of a population PK model will facilitate dose optimization and further clinical development of piclozotan.

Population pharmacokinetics of ketorolac in neonates and young infants.

Although ketorolac is commonly used as an analgesic in the pediatric population, there is no information on the pharmacokinetics of ketorolac available for children less than 6 months of age. The objective of this analysis was to construct a population pharmacokinetic model to describe ketorolac disposition in young children. Three neonates and 9 infants, median (range) age 0.4-32 weeks, were administered with 0.5 mg/kg of ketorolac. The data were best described by a 2-compartment model, with an allometric expression to describe body weight effects on clearance. Estimated parameters were clearance [2.8 mL x min(-1) x (kg(0.75))(-1)], intercompartmental clearance (11.5 mL/min), volume of distribution of the central compartment (535 mL), and volume of distribution of the peripheral compartment (322 mL). The clearance values in these neonates and younger group of infants are greater than that reported for older children and adults.

Population pharmacokinetics of imatinib mesylate and its metabolite in children and young adults.

BACKGROUND: Imatinib mesylate (Gleevec) is a small molecule tyrosine kinase inhibitor approved for use in the management of chronic myeloid leukemia in adults and children and in gastrointestinal stromal tumors in adults. Population pharmacokinetic (PPK) studies evaluating the effect of population covariates on the pharmacokinetics of imatinib and its active metabolite have been developed in adults with chronic myeloid leukemia (CML) and gastrointestinal stromal tumor (GIST). However, this still remains to be described in children. PURPOSE: The objectives of the analysis were to develop a PPK model of imatinib and its active metabolite, CGP74588, to describe exposure in children and young adults and to identify covariates that are predictors of variability in disposition. METHODS: Plasma concentrations from 26 subjects with Philadelphia (Ph+) leukemia (Phase I study) and 15 subjects with refractory solid tumors (Phase II study), who received oral imatinib at doses ranging from 260 to 570 mg/m(2), were available for the PPK analysis in NONMEM. Blood samples were drawn prior to dosing and over 24-48 h on days 1 and 8 of the studies. Covariates studied included weight, age, albumin, alanine aminotransferase and the study population. RESULTS: The pharmacokinetics of imatinib and CGP 74588 were well described by one and two compartment models, respectively. Total body weight was the only covariate found to significantly affect Cl/F and V/F. The final imatinib-CGP 74588 model is summarized as follows: CL/F (imatinib) (L/h) = 10.8 x (WT/70)(0.75), V/F (imatinib) (L) = 284 x (WT/70) and D1(duration of zero order absorption,imatinib) (h) = 1.67 and CL/F (CGP 74588) (L/h) = 9.65 x (WT/70)(0.75), V1/F (CGP 74588) (L) = 11.6 x (WT/70), Q (CGP 74588) (L/h) = 2.9 x (WT/70)(0.75) and V2/F (CGP 74588) (L) = 256*(WT/70). Model evaluation indicated that the final model was robust and satisfactory. CONCLUSIONS: Current imatinib dosing guidelines in pediatrics is based on the achievement of exposures consistent with doses known to be safe and efficacious in adults. Dose adjustments in children are guided empirically by the observance of drug-related toxicities. While, the pharmacokinetics of imatinib and its active metabolite, CGP 74588 in children are consistent with prior knowledge in adults, the model will form the basis to support the design of future trials, particularly with a view to managing toxicities and exploring dosing in this population.

Pharmacokinetics of irinotecan and its metabolites in pediatric cancer patients: a report from the children's oncology group.

OBJECTIVE: To develop a population pharmacokinetic model of irinotecan and its major metabolites in children with cancer and to identify covariates that predict variability in disposition. METHODS: A population pharmacokinetic model was developed using plasma concentration data from 82 patients participating in a multicenter Pediatric Oncology Group (POG) single agent phase II clinical trial. Patients between 1 and 21 years of age with solid tumors refractory to standard therapy received irinotecan, 50 mg/m(2), as a 60-min intravenous infusion for 5 consecutive days every 3 weeks. Blood samples were collected and analyzed for irinotecan and three metabolites (SN-38, SN-38G, and APC). The population model was developed with NONMEM. Clearance and volume were scaled allometrically using corrected body weight. Exponential error models were used to describe the interindividual variance in pharmacokinetic parameters, and the residual error was described with a proportional model. Significant covariate effects were identified graphically using S-PLUS and were added to the base-model. The final model was evaluated by simulating data from two other POG trials. RESULTS: The best structural model for irinotecan and its metabolites consisted of six-compartments: two compartments for irinotecan and SN-38, and one each for APC and SN-38G. Age and bilirubin were found to be significant covariates affecting SN-38 clearance. SN-38 clearance was greater in patients less than 10 years of age and lower in patients with a total serum bilirubin >0.6 mg/dL. Simulations revealed that the model was able to predict drug and metabolite exposure (AUC) for patients receiving the same or similar doses (30-65 mg/m(2)) of irinotecan. CONCLUSIONS: This population model accurately describes the pharmacokinetics of irinotecan and its primary metabolites. The model, which includes age and bilirubin as covariate effects on SN-38 clearance, is the first population model to describe the pharmacokinetics of irinotecan and its major metabolites in children.

Integration of modeling and simulation into hospital-based decision support systems guiding pediatric pharmacotherapy.

BACKGROUND: Decision analysis in hospital-based settings is becoming more common place. The application of modeling and simulation approaches has likewise become more prevalent in order to support decision analytics. With respect to clinical decision making at the level of the patient, modeling and simulation approaches have been used to study and forecast treatment options, examine and rate caregiver performance and assign resources (staffing, beds, patient throughput). There us a great need to facilitate pharmacotherapeutic decision making in pediatrics given the often limited data available to guide dosing and manage patient response. We have employed nonlinear mixed effect models and Bayesian forecasting algorithms coupled with data summary and visualization tools to create drug-specific decision support systems that utilize individualized patient data from our electronic medical records systems. METHODS: Pharmacokinetic and pharmacodynamic nonlinear mixed-effect models of specific drugs are generated based on historical data in relevant pediatric populations or from adults when no pediatric data is available. These models are re-executed with individual patient data allowing for patient-specific guidance via a Bayesian forecasting approach. The models are called and executed in an interactive manner through our web-based dashboard environment which interfaces to the hospital's electronic medical records system. RESULTS: The methotrexate dashboard utilizes a two-compartment, population-based, PK mixed-effect model to project patient response to specific dosing events. Projected plasma concentrations are viewable against protocol-specific nomograms to provide dosing guidance for potential rescue therapy with leucovorin. These data are also viewable against common biomarkers used to assess patient safety (e.g., vital signs and plasma creatinine levels). As additional data become available via therapeutic drug monitoring, the model is re-executed and projections are revised. CONCLUSION: The management of pediatric pharmacotherapy can be greatly enhanced via the immediate feedback provided by decision analytics which incorporate the current, best-available knowledge pertaining to dose-exposure and exposure-response relationships, especially for narrow therapeutic agents that are difficult to manage.

Population pharmacokinetic investigation of actinomycin-D in children and young adults.

Actinomycin-D is an antineoplastic agent that inhibits RNA synthesis by binding to guanine residues and inhibiting DNA-dependent RNA polymerase. Although actinomycin-D has been used to treat rhabdomyosarcoma and Wilms tumor for more than 40 years, the dose/exposure relationship is not well characterized. The objective of this study was to develop an initial population pharmacokinetic model to describe actinomycin-D disposition in children and young adults from which a prospective study could be designed. A total of 165 actinomycin-D plasma concentration measurements from 33 patients, aged 1.6 to 20.3 years, were used for the analysis. The data were analyzed using nonlinear mixed-effects modeling with the NONMEM software system. Age, weight, and gender were examined as covariates for the ability to explain interindividual variability in actinomycin-D pharmacokinetics. The final model was qualified via predictive check and nonparametric bootstrap procedures. A 3-compartment model with first-order elimination was chosen as the structural model. Allometric expressions incorporating weight were used to describe the effects of body size on actinomycin-D pharmacokinetics. Age and gender had no discernible effects on actinomycin-D pharmacokinetics in the population studied. The predictive check showed that the developed model was able to simulate data in close agreement with the actual study observations. The availability of an initial population pharmacokinetic model to describe actinomycin-D pharmacokinetics will facilitate the development of a large-scale clinical trial to study the actinomycin-D dose/exposure relationship in pediatric patients with rhabdomyosarcoma and Wilms tumor. The covariate analysis described by the current data set suggests that indices of body size captured via allometric expressions improve the partition of variation in actinomycin-D pharmacokinetics from this pilot data set. Relationships between pharmacokinetics and toxicity will be examined in future prospective studies in which children less than 1 year old will be enrolled.

An informatics approach to assess pediatric pharmacotherapy: design and implementation of a hospital drug utilization system.

Drug utilization in the inpatient setting can provide a mechanism to assess drug prescribing trends, efficiency, and cost-effectiveness of hospital formularies and examine subpopulations for which prescribing habits may be different. Such data can be used to correlate trends with time-dependent or seasonal changes in clinical event rates or the introduction of new pharmaceuticals. It is now possible to provide a robust, dynamic analysis of drug utilization in a large pediatric inpatient setting through the creation of a Web-based hospital drug utilization system that retrieves source data from our accounting database. The production implementation provides a dynamic and historical account of drug utilization at the authors' institution. The existing application can easily be extended to accommodate a multi-institution environment. The creation of a national or even global drug utilization network would facilitate the examination of geographical and/or socioeconomic influences in drug utilization and prescribing practices in general.

Model-based drug development applied to oncology.

Model-based drug development (MBDD) is an approach that is used to organize the vast and complex data streams that feed the drug development pipelines of small molecule and biotechnology sponsors. Such data streams are ultimately reviewed by the global regulatory community as evidence of a drug's potential to treat and/or harm patients. Some of this information is captured in the scientific literature and prescribing compendiums forming the basis of how new and existing agents will ultimately be administered and further evaluated in the broader patient community. As this data stream evolves, the details of data qualification, the assumptions and/or critical decisions based on these data are lost under conventional drug development paradigms. MBDD relies on the construction of quantitative relationships to connect data from discrete experiments conducted along the drug development pathway. These relationships are then used to ask questions relevant at critical development stages, hopefully, with the understanding that the various scenarios explored represent a path to optimal decision making. Oncology, as a therapeutic area, presents a unique set of challenges and perhaps a different development paradigm as opposed to other disease targets. The poor attrition of development compounds in the recent past attests to these difficulties and provides an incentive for a different approach. In addition, given the reliance on multimodal therapy, oncological disease targets are often treated with both new and older agents spanning several drug classes. As MBDD becomes more integrated into the pharmaceutical research community, a more rational explanation for decisions regarding the development of new oncology agents as well as the proposed treatment regimens that incorporate both new and existing agents can be expected. Hopefully, the end result is a more focussed clinical development programme, which ultimately provides a means to optimize individual patient care.