Semiparametric Distributions With Estimated Shape Parameters

Opioid use in treated and untreated obstructive sleep apnoea: remifentanil pharmacokinetics and pharmacodynamics in adult volunteers

BACKGROUND

Patients with obstructive sleep apnoea (OSA) are considered more sensitive to opioids and at increased risk of opioid-induced respiratory depression. Nonetheless, whether OSA treatment (continuous positive airway pressure, CPAP; or bilevel positive airway pressure, BIPAP) modifies this risk remains unknown. Greater opioid sensitivity can arise from altered pharmacokinetics or pharmacodynamics. This preplanned analysis of a previous cohort study of remifentanil clinical effects in OSA tested the null hypothesis that the pharmacokinetics, pharmacodynamics, or both of remifentanil, a representative μ-opioid agonist, are not altered in adults with treated or untreated OSA.

METHODS

A single-centre, prospective, open-label, cohort study administered a stepped-dose, target-controlled remifentanil infusion (target effect-site concentrations 0.5, 1, 2, 3, 4 ng ml-1) to awake adult volunteers (median age 52 yr, range 23-70) without OSA (n=20), with untreated OSA (n=33), or with treated OSA (n=21). Type III (in-home) polysomnography verified OSA. Remifentanil plasma concentrations, end-expired CO2, thermal heat tolerance, and pupil diameter (miosis) were assessed. Population pharmacokinetic (clearance, volume of distribution) and pharmacodynamic (miosis, thermal heat tolerance, end-expired CO2) models were developed.

RESULTS

Remifentanil clearance (median) was 147, 143, and 155 L h-1 (P=0.472), and volume of distribution was 19.6, 15.5, and 17.7 L (P=0.473) for subjects without OSA, untreated OSA, or treated OSA, respectively. Total body weight was an influential covariate on both remifentanil clearance and central volume of distribution. There were no statistically or clinically significant differences between the three groups in miosis EC50 or Emax, or the slopes of thermal heat tolerance or end-expired CO2vs remifentanil concentration. At a plasma remifentanil concentration of 4 ng ml-1, in participants without OSA, with untreated OSA, or with treated OSA, respectively, model-estimated pupil area (12%, 13%, and 17% of baseline, P=0.086), thermal heat tolerance (50°C, 51°C, and 51°C, P=0.218), and end-expired CO2 (6.3 kPa, 6.4 kPa, and 6.7 kPa, P=0.257) were not statistically different between groups.

CONCLUSIONS

OSA (untreated or treated) did not influence remifentanil pharmacokinetics or pharmacodynamics (miosis, analgesia, respiratory depression). Results support the null hypothesis that neither pharmacokinetics nor pharmacodynamics of remifentanil, a representative μ-opioid, are altered in adults with treated or untreated OSA. These findings provide a mechanistic explanation for the lack of influence of OSA or OSA treatment on the clinical miotic, sedative, analgesic, or respiratory depressant response to remifentanil in awake adults. The conventional notion that OSA alters sensitivity to the effects of opioids in awake adults is not supported by our findings, such that opioid dosing might not need adjustment for pharmacokinetic or pharmacodynamic considerations.

CLINICAL TRIAL REGISTRATION

ClinicalTrials.gov, NCT02898792, https://clinicaltrials.gov/ct2/show/NCT02898792. First Posted: September 13, 2016.

Development of a Joint Tumor Size-Overall Survival Modeling and Simulation Framework Supporting Oncology Development Decision-Making

Tumor size-overall survival (TS-OS) models can support decision-making in oncology drug development by predicting long-term OS based on TS data from early data cuts and baseline patient factors. The current work describes the development of a TS-OS framework capable of predicting OS across a variety of treatment modalities and mechanisms of action in patients with non-small cell lung cancer from seven clinical studies. The presented framework jointly models TS with a bi-exponential Stein model and OS with an accelerated failure time log-normal survival model. In the corresponding link function between TS and OS, the most significant predictor of OS was the tumor growth rate (kg), applied via an Emax function. Time to tumor growth and baseline TS were additional TS predictors informing OS. Albumin, total protein, and neutrophil-to-lymphocyte ratio were selected from the tested baseline factors as the most significant predictors of OS. Significant baseline covariates for the TS model included number of target lesions on baseline TS, tumor PD-L1 expression on tumor shrinkage rate, and lactate dehydrogenase levels on kg. The TS-OS framework model adequately describes the OS distributions within this specific set of treatment modalities-chemotherapies, immuno-oncology treatments, and combinations thereof-using a single treatment-independent link function, supporting the use of the framework to support evaluation and design of future studies. Our findings contribute to a body of literature exploring and qualifying TS-OS modeling as a methodology capable of supporting and accelerating oncology drug development.

Population pharmacokinetic analysis in children with different diseases treated with mycophenolate mofetil-Integrated analysis of clinical trials and real-world clinical data

OBJECTIVES

This study aims to develop a population pharmacokinetic (PK) model of mycophenolic acid (MPA) in pediatric patients with different diseases.

METHODS

To develop the PK model, electronic medical records (EMR) of pediatric patients with different diseases who received mycophenolic acid mofetil (MMF) at the National Center for Child Health and Development from February 2013 to May 2022 and the results of the MMF pharmacokinetic study of pediatric complicated frequently relapsing steroid-dependent nephrotic syndrome under the Advanced Medical Care B protocol (JSKDC09 study) from October 2015 to June 2019 were integrated and analyzed. Non-linear mixed-effects modeling was used to 1) develop a cross-disease population PK model and 2) estimate PK parameters (apparent clearance CL/F and Q/F; apparent volume of distribution Vc/F and Vp/F; absorption rate constant Ka) by post-hoc Bayesian methods. Patient backgrounds, concomitant medications, and laboratory data were included in the covariate analysis. Visual predictive checks with bootstrap and predictive correction were performed to evaluate the final model.

RESULTS

A total of 249 patients with 1495 measurements were used in the analysis, including 68 with post-liver transplant, 65 with nephrotic syndrome, 28 with post-renal transplant, 31 with autoimmune diseases (17 with lupus nephritis and 14 with other collagen diseases), 13 with hematopoietic stem cell transplantation, and 5 with others from EMR and 39 with nephrotic syndrome from the JSKDC09 study. A two-compartment model with a first-order rate absorption process best explained the PK of MPA. A nonlinear relationship between dose and MPA exposure was observed and described by the power function of the model. The final population mean PK parameter estimates (95 % confidence interval) for non-renal transplant patients and average albumin levels were CL/F 15.2 (13.3, 18.0) L/h/70 kg, Vc/F 15.4 (14.4, 17.2) L, Vp/F 246.8 (159.1, 332.1) L, Q/F 7.0 (5.0, 9.0) L/h, and Ka (without proton pump inhibitor [PPI]) 4.4 (2.4, 4.9) h-1. Weight, disease (post-renal transplant), and serum albumin level were significant covariates for CL/F, and serum albumin level for Vc/F. The use of PPIs also affected Ka.

CONCLUSIONS

An integrated population PK model of MPA was developed in children with the following conditions: post-solid organ transplantation, post-HSCTx, autoimmune disease, and nephrotic syndrome. Estimated parameters and parameter covariates were similar to those previously reported. This model is expected to provide useful information for using MPA in various diseases in the Japanese population.

Translational pharmacokinetic and pharmacodynamic modelling of the anti-ADAMTS-5 NANOBODY® (M6495) using the neo-epitope ARGS as a biomarker

M6495 is a first-in-class NANOBODY® molecule and an inhibitor of ADAMTS-5, with the potential to be a disease modifying osteoarthritis drug. In order to investigate the PK/PD (pharmacokinetic and pharmacodynamic) properties of M6495, a single dose study was performed in cynomolgus monkeys with doses up to 6 mg/kg, with the goal of understanding the PK/PD properties of M6495. The neo-epitope ARGS (Alanine-Arginine-Glycine-Serine) generated by cleavage of aggrecan by ADAMTS-5 was used as a target-engagement biomarker. A long-lasting dose-dependent decrease in serum ARGS could be observed after a single dose of M6495 in cynomolgus monkeys. The serum biomarker ARGS decreased to levels below the limit of quantification of the assay in animals which received doses of M6495 of 6 mg/kg and higher, indicating a strong inhibition of ADAMTS-5. Data from the single-dose PK/PD study was combined with data from a multiple dose study, and a non-linear mixed effects model was used to explore the relationship between plasma concentrations of M6495 and the reduction of serum ARGS. The model was subsequently used to inform the clinical phase 1 study design and was successful in predicting the human clinical pharmacokinetics and pharmacodynamics of M6495. In addition to having enabled a Phase 1 trial with M6495, this is the first PK/PD model describing the pharmacodynamics of the neo-epitope ARGS after ADAMTS5 inhibition. It is expected that in the future, this model can be used or adapted to explore the PK/PD relationship between M6495 serum concentrations and the ARGS serum biomarker.

Quantification of biochemical PSA dynamics after radioligand therapy with [177Lu]Lu-PSMA-I&T using a population pharmacokinetic/pharmacodynamic model

BACKGROUND

There is an unmet need for prediction of treatment outcome or patient selection for [177Lu]Lu-PSMA therapy in patients with metastatic castration-resistant prostate cancer (mCRPC). Quantification of the tumor exposure-response relationship is pivotal for further treatment optimization. Therefore, a population pharmacokinetic (PK) model was developed for [177Lu]Lu-PSMA-I&T using SPECT/CT data and, subsequently, related to prostate-specific antigen (PSA) dynamics after therapy in patients with mCRPC using a pharmacokinetic/pharmacodynamic (PKPD) modelling approach.

METHODS

A population PK model was developed using quantitative SPECT/CT data (406 scans) of 76 patients who received multiple cycles [177Lu]Lu-PSMA-I&T (± 7.4 GBq with either two- or six-week interval). The PK model consisted of five compartments; central, salivary glands, kidneys, tumors and combined remaining tissues. Covariates (tumor volume, renal function and cycle number) were tested to explain inter-individual variability on uptake into organs and tumors. The final PK model was expanded with a PD compartment (sequential fitting approach) representing PSA dynamics during and after treatment. To explore the presence of a exposure-response relationship, individually estimated [177Lu]Lu-PSMA-I&T tumor concentrations were related to PSA changes over time.

RESULTS

The population PK model adequately described observed data in all compartments (based on visual inspection of goodness-of-fit plots) with adequate precision of parameters estimates (< 36.1% relative standard error (RSE)). A significant declining uptake in tumors (k14) during later cycles was identified (uptake decreased to 73%, 50% and 44% in cycle 2, 3 and 4-7, respectively, compared to cycle 1). Tumor growth (defined by PSA increase) was described with an exponential growth rate (0.000408 h-1 (14.2% RSE)). Therapy-induced PSA decrease was related to estimated tumor concentrations (MBq/L) using both a direct and delayed drug effect. The final model adequately captured individual PSA concentrations after treatment (based on goodness-of-fit plots). Simulation based on the final PKPD model showed no evident differences in response for the two different dosing regimens currently used.

CONCLUSIONS

Our population PK model accurately described observed [177Lu]Lu-PSMA-I&T uptake in salivary glands, kidneys and tumors and revealed a clear declining tumor uptake over treatment cycles. The PKPD model adequately captured individual PSA observations and identified population response rates for the two dosing regimens. Hence, a PKPD modelling approach can guide prediction of treatment response and thus identify patients in whom radioligand therapy is likely to fail.

Population pharmacokinetic-pharmacodynamic modeling of clopidogrel for dose regimen optimization based on CYP2C19 phenotypes: A proof of concept study

Clopidogrel is an antiplatelet drug used to reduce the risk of acute coronary syndrome and stroke. It is converted by CYP2C19 to its active metabolite; therefore, poor metabolizers (PMs) of CYP2C19 exhibit diminished antiplatelet effects. Herein, we conducted a proof-of-concept study for using population pharmacokinetic-pharmacodynamic (PK-PD) modeling to recommend a personalized clopidogrel dosing regimen for individuals with varying CYP2C19 phenotypes and baseline P2Y12 reaction unit (PRU) levels. Data from a prospective phase I clinical trial involving 36 healthy male participants were used to develop the population PK-PD model predicting the concentrations of clopidogrel, clopidogrel H4, and clopidogrel carboxylic acid, and linking clopidogrel H4 concentrations to changes in PRU levels. A two-compartment model effectively described the PKs of both clopidogrel and clopidogrel carboxylic acid, and a one-compartment model of those of clopidogrel H4. The CYP2C19 phenotype was identified as a significant covariate influencing the metabolic conversion of the parent drug to its metabolites. A PD submodel of clopidogrel H4 that stimulated the fractional turnover rate of PRU levels showed the best performance. Monte Carlo simulations suggested that PMs require three to four times higher doses than extensive metabolizers to reach the target PRU level. Individuals within the top 20th percentile of baseline PRU levels were shown to require 2.5-3 times higher doses than those in the bottom 20th percentile. We successfully developed a population PK-PD model for clopidogrel considering the impact of CYP2C19 phenotypes and baseline PRU levels. Further studies are necessary to confirm actual dosing recommendations for clopidogrel.

Longitudinal nonlinear mixed effects modeling of EGFR mutations in ctDNA as predictor of disease progression in treatment of EGFR-mutant non-small cell lung cancer

Correlations between increasing concentrations of circulating tumor DNA (ctDNA) in plasma and disease progression have been shown. A nonlinear mixed effects model to describe the dynamics of epidermal growth factor receptor (EGFR) ctDNA data from patients with non-small cell lung cancer (NSCLC) combined with a parametric survival model were developed to evaluate the ability of these modeling techniques to describe ctDNA data. Repeated ctDNA measurements on L858R, exon19del, and T790M mutants were available from 54 patients with EGFR mutated NSCLC treated with erlotinib or gefitinib. Different dynamic models were tested to describe the longitudinal ctDNA concentrations of the driver and resistance mutations. Subsequently, a parametric time-to-event model for progression-free survival (PFS) was developed. Predicted L858R, exon19del, and T790M concentrations were used to evaluate their value as predictor for disease progression. The ctDNA dynamics were best described by a model consisting of a zero-order increase and first-order elimination (19.7/day, 95% confidence interval [CI] 14.9-23.6/day) of ctDNA concentrations. In addition, time-dependent development of resistance (5.0 × 10^-4 , 95% CI 2.0 × 10^-4 -7.0 × 10^-4 /day) was included in the final model. Relative change in L858R and exon19del concentrations from baseline was identified as most significant predictor of disease progression (p = 0.001). The dynamic model for L858R, exon19del, and T790M concentrations in ctDNA and time-to-event model adequately described the observed concentrations and PFS data in our clinical cohort. In addition, it was shown that nonlinear mixed effects modeling is a valuable method for the analysis of longitudinal and heterogeneous biomarker datasets obtained from clinical practice.

Prediction of Maternal and Fetal Doravirine Exposure by Integrating Physiologically Based Pharmacokinetic Modeling and Human Placenta Perfusion Experiments

Background and Objective

Doravirine is currently not recommended for pregnant women living with human immunodeficiency virus because efficacy and safety data are lacking. This study aimed to predict maternal and fetal doravirine exposure by integrating human placenta perfusion experiments with pregnancy physiologically based pharmacokinetic (PBPK) modeling.

Methods

Ex vivo placenta perfusions were performed in a closed–closed configuration, in both maternal-to-fetal and fetal-to-maternal directions (n = 8). To derive intrinsic placental transfer parameters from perfusion data, we developed a mechanistic placenta model. Next, we developed a maternal and fetal full-body pregnancy PBPK model for doravirine in Simcyp, which was parameterized with the derived intrinsic placental transfer parameters to predict in vivo maternal and fetal doravirine exposure at 26, 32, and 40 weeks of pregnancy. The predicted total geometric mean (GM) trough plasma concentration (C_trough) values were compared with the target (0.23 mg/L) derived from in vivo exposure–response analysis.

Results

A decrease of 55% in maternal doravirine area under the plasma concentration–time curve (AUC)_0–24h was predicted in pregnant women at 40 weeks of pregnancy compared with nonpregnant women. At 26, 32, and 40 weeks of pregnancy, predicted maternal total doravirine GM C_trough values were below the predefined efficacy target of 0.23 mg/L. Perfusion experiments showed that doravirine extensively crossed the placenta, and PBPK modeling predicted considerable fetal doravirine exposure.

Conclusion

Substantially reduced maternal doravirine exposure was predicted during pregnancy, possibly resulting in impaired efficacy. Therapeutic drug and viral load monitoring are advised for pregnant women treated with doravirine. Considerable fetal doravirine exposure was predicted, highlighting the need for clinical fetal safety data.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40262-022-01127-0.

Population pharmacokinetics and exposure-response relationships of astegolimab in patients with severe asthma

Astegolimab is a fully human immunoglobulin G2 monoclonal antibody that binds to the ST2 receptor and blocks the interleukin‐33 signaling. It was evaluated in patients with uncontrolled severe asthma in the phase 2b study (Zenyatta) at doses of 70, 210, and 490 mg subcutaneously every 4 weeks for 52 weeks. This work aimed to characterize astegolimab pharmacokinetics, identify influential covariates contributing to its interindividual variability, and make a descriptive assessment of the exposure‐response relationships. A population pharmacokinetic model was developed using data from 368 patients in the Zenyatta study. Predicted average steady‐state concentration was used in the subsequent exposure‐response analyses, which evaluated efficacy (asthma exacerbation rate) and biomarker end points including forced expiratory volume in 1 second, fraction exhaled nitric oxide, blood eosinophils, and soluble ST2. A 2‐compartment disposition model with first‐order elimination and first‐order absorption best described the astegolimab pharmacokinetics. The relative bioavailability for the 70‐mg dose was 15.3% lower. Baseline body weight, estimated glomerular filtration rate, and eosinophils were statistically correlated with pharmacokinetic parameters, but only body weight had a clinically meaningful influence on the steady‐state exposure (ratios exceeding 0.8‐1.25). The exposure‐response of efficacy and biomarkers were generally flat with a weak trend in favor of the highest dose/exposure. This study characterized astegolimab pharmacokinetics in patients with asthma and showed typical pharmacokinetic behavior as a monoclonal antibody–based drug. The exposure‐response analyses suggested the highest dose tested in the Zenyatta study (490 mg every 4 weeks) performed close to the maximum effect, and no additional response may be expected above it.

Highly Variable Paracetamol Pharmacokinetics After Multiple Oral Dosing in Frail Older People: A Population Pharmacokinetic Analysis

INTRODUCTION

Paracetamol pharmacokinetics (PK) is highly variable in older fit adults after intravenous administration. Frailty and oral administration likely result in additional variability. The aim was to determine oral paracetamol PK and variability in geriatric inpatients.

METHODS

A population PK analysis, using NONMEM 7.2, was performed on 245 paracetamol samples in 40 geriatric inpatients (median age 87 [range 80-95] years, bodyweight 66.4 [49.3-110] kg, 92.5% frail [Edmonton Frail Scale]). All subjects received paracetamol 1000 mg as tablet (72.5%) or granulate (27.5%) three times daily. Simulations of dosing regimens (1000 mg every 6 hours [q6h] or q8h) were performed to determine target attainment, using mean steady-state concentration (C_ss-mean) of 10 mg/L as target.

RESULTS

A one-compartment model with first order absorption and lag time best described the data. The inter-individual variability was high, with absorption rate constant containing the highest variability. The inter-individual variability could not be explained by covariates. Simulations of 1000 mg q6h and q8h resulted in a C_ss-mean of 10.8 [25-75th percentiles 8.2-12.7] and 8.13 [6.3-9.6] mg/L, respectively, for the average geriatric inpatient. The majority of the population remained off-target (22.2% [q6h] and 52.2% [q8h] <8 mg/L; 31.3 [q6h] and 7.6% [q8h] >12 mg/L).

CONCLUSION

A population of average geriatric inpatients achieved target C_ss-mean with paracetamol 1000 mg q6h, while q8h resulted in underexposure for the majority of them. Due to high unexplained variability, a relevant proportion remained either above or below the target concentration of 10 mg/L. Research focusing on PK, efficacy and safety is needed to recommend dosing regimens.

Taking Kinetic Evaluations of Degradation Data to the Next Level with Nonlinear Mixed-Effects Models

When data on the degradation of a chemical substance have been collected in a number of environmental media (e.g., in different soils), two strategies can be followed for data evaluation. Currently, each individual dataset is evaluated separately, and representative degradation parameters are obtained by calculating averages of the kinetic parameters. However, such averages often take on unrealistic values if certain degradation parameters are ill-defined in some of the datasets. Moreover, the most appropriate degradation model is selected for each individual dataset, which is time consuming and then requires workarounds for averaging parameters from different models. Therefore, a simultaneous evaluation of all available data is desirable. If the environmental media are viewed as random samples from a population, an advanced strategy based on assumptions about the statistical distribution of the kinetic parameters across the population can be used. Here, we show the advantages of such simultaneous evaluations based on nonlinear mixed-effects models that incorporate such assumptions in the evaluation process. The advantages of this approach are demonstrated using synthetically generated data with known statistical properties and using publicly available experimental degradation data on two pesticidal active substances.

Relevance of primary lesion location, tumour heterogeneity and genetic mutation demonstrated through tumour growth inhibition and overall survival modelling in metastatic colorectal cancer

AIMS

The aims of this work were to build a semi-mechanistic tumour growth inhibition (TGI) model for metastatic colorectal cancer (mCRC) patients receiving either cetuximab + chemotherapy or chemotherapy alone and to identify early predictors of overall survival (OS).

METHODS

A total of 1716 patients from 4 mCRC clinical studies were included in the analysis. The TGI model was built with 8973 tumour size measurements where the probability of drop-out was also included and modelled as a time-to-event variable using parametric survival models, as it was the case in the OS analysis. The effects of patient- and tumour-related covariates on model parameters were explored.

RESULTS

Chemotherapy and cetuximab effects were included in an additive form in the TGI model. Development of resistance was found to be faster for chemotherapy (drug effect halved at wk 8) compared to cetuximab (drug effect halved at wk 12). KRAS wild-type status and presenting a right-sided primary lesion were related to a 3.5-fold increase in cetuximab drug effect and a 4.7× larger cetuximab resistance, respectively. The early appearance of a new lesion (HR = 4.14), a large tumour size at baseline (HR = 1.62) and tumour heterogeneity (HR = 1.36) were the main predictors of OS.

CONCLUSIONS

Semi-mechanistic TGI and OS models have been developed in a large population of mCRC patients receiving chemotherapy in combination or not with cetuximab. Tumour-related predictors, including a machine learning derived-index of tumour heterogeneity, were linked to changes in drug effect, resistance to treatment or OS, contributing to the understanding of the variability in clinical response.

A population K-PD model analysis of long-term testosterone inhibition in prostate cancer patients undergoing intermittent androgen deprivation therapy

Intermittent androgen deprivation therapy with gonadotropin-releasing-hormone (GnRH) agonists can prevent or delay disease progression and development of castration resistant prostate cancer for subpopulations of prostate cancer patients. It may also reduce risk and severity of side effects associated with chemical castration in prostate cancer (PCa) patients. One of the earliest comprehensively documented clinical trials on this was reported in a Canadian patient population treated with leuprorelin preceded by a lead-in with cyproterone acetate. A systems-based mixed effect analysis of testosterone response in active and recovery phases allows inference of new information from this patient population. Efficacy of androgen deprivation therapy is presumed to depend on a treshold value for testosterone at the nadir, below which no additional beneficial effects on PSA reponse can be expected, and occurance of testosterone breakthroughs during active therapy. The present analysis results in a mixed effect model, incorporating GnRH receptor activation, testosterone turnover and feedback mechanisms, describing and predicting testosterone inhibition under intermittent androgen deprivation therapy on the individual and population level, during multiple years of therapy. Testosterone levels in these patients decline over time with an estimated first order rate constant of 0.083 year^-1(T_1/2 = 8.4 y), with a substantial distribution among this patient population, compared to the general population. PCa patients leaving the trial due to unmanageble PSA relapse appear to have slightly higher testosterone levels at the nadir than sustained responders. These findings are expected to contribute to an increased understanding of the role of testosterone in long term disease progression of prostate cancer.

A population pharmacokinetics analysis assessing the exposure of raltegravir once-daily 1200 mg in pregnant women living with HIV

Once-daily two 600 mg tablets (1200 mg q.d.) raltegravir offers an easier treatment option compared to the twice-daily regimen of one 400 mg tablet. No pharmacokinetic, efficacy, or safety data of the 1200 mg q.d. regimen have been reported in pregnant women to date as it is challenging to collect these clinical data. This study aimed to develop a population pharmacokinetic (PopPK) model to predict the pharmacokinetic profile of raltegravir 1200 mg q.d. in pregnant women and to discuss the expected pharmacodynamic properties of raltegravir 1200 mg q.d. during pregnancy based on previously reported concentration-effect relationships. Data from 11 pharmacokinetic studies were pooled (n = 221). A two-compartment model with first-order elimination and absorption through three sequential transit compartments best described the data. We assessed that the bio-availability of the 600 mg tablets was 21% higher as the 400 mg tablets, and the bio-availability in pregnant women was 49% lower. Monte-Carlo simulations were performed to predict the pharmacokinetic profile of 1200 mg q.d. in pregnant and nonpregnant women. The primary criteria for efficacy were that the lower bound of the 90% confidence interval (CI) of the concentration before next dose administration (C_trough ) geometric mean ratio (GMR) of simulated pregnant/nonpregnant women had to be greater than 0.75. The simulated raltegravir C_trough GMR (90% CI) was 0.51 (0.41-0.63), hence not meeting the primary target for efficacy. Clinical data from two pregnant women using 1200 mg q.d. raltegravir showed a similar C_trough ratio pregnant/nonpregnant. Our pharmacokinetic results support the current recommendation of not using the raltegravir 1200 mg q.d. regimen during pregnancy until more data on the exposure-response relationship becomes available.

Comparing Circulating Tumor Cell Counts with Dynamic Tumor Size Changes as Predictor of Overall Survival: A Quantitative Modeling Framework

PURPOSE

Quantitative relationships between treatment-induced changes in tumor size and circulating tumor cell (CTC) counts, and their links to overall survival (OS), are lacking. We present a population modeling framework identifying and quantifying such relationships, based on longitudinal data collected in patients with metastatic colorectal cancer (mCRC) to evaluate the value of tumor size and CTC counts as predictors of OS.

EXPERIMENTAL DESIGN

A pharmacometric approach (i.e., population pharmacodynamic modeling) was used to characterize the changes in tumor size and CTC count and evaluate them as predictors of OS in 451 patients with mCRC treated with chemotherapy and targeted therapy in a prospectively randomized phase III study (CAIRO2).

RESULTS

A tumor size model of tumor quiescence and drug resistance was used to characterize the tumor size time-course, and was, in addition to the total normalized dose (i.e., of all administered drugs) in a given cycle, related to the CTC counts through a negative binomial model (CTC model). Tumor size changes did not contribute additional predictive value when the mean CTC count was a predictor of OS. Treatment reduced the typical mean count from 1.43 to 0.477 (HR = 3.94). The modeling framework was applied to explore whether dose modifications (increased and reduced) would result in a CTC count below 1/7.5 mL after 1 to 2 weeks of treatment.

CONCLUSIONS

Time-varying CTC counts can be useful for early predicting OS in patients with mCRC, and may therefore have potential for model-based treatment individualization. Although tumor size was connected to CTC, its link to OS was weaker.

Variability in the Log Domain and Limitations to Its Approximation by the Normal Distribution

Pharmacometric models using lognormal distributions have become commonplace in pharmacokinetic-pharmacodynamic investigations. The extent to which it can be interpreted by traditional description of variability through the normal distribution remains elusive. In this tutorial, the comparison is made using formal approximation methods. The quality of the resulting approximation was assessed by the similarity of prediction intervals (PIs) to true values, illustrated using 80% PIs. Approximated PIs were close to true values when lognormal standard deviation (omega) was smaller than about 0.25, depending mostly on the desired precision. With increasing omega values, the precision of approximation worsens and starts to deteriorate at omega values of about 1. With such high omega values, there is no resemblance between the lognormal and normal distribution anymore. To support dissemination and interpretation of these nonlinear properties, some additional statistics are discussed in the context of the three regions of behavior of the lognormal distribution.

The effect of food and formulation on the population pharmacokinetics of cholesteryl ester transferase protein inhibitor DRL-17822 in healthy male volunteers

We aimed to characterise the population pharmacokinetics of cholesteryl ester transferase protein inhibitor DRL-17822 in healthy males and explore the effect of food and formulation on the oral absorption of DRL-17822 in 4 phase I studies. DRL-17822 was dosed orally (2-1000 mg) in 2 different drug formulations (nanocrystal formulation and amorphous solid dispersion formulation) after either an overnight fast, or a low-fat, continental or high-fat breakfast. A 2-compartment model with 6 transit absorption compartments best characterised the data. Additionally, a strong interaction of food and formulation on bioavailability was observed and parsimoniously characterised in the model by binning combinations of food state and formulation with similar bio-availabilities. The final model adequately characterised the pharmacokinetic data of DRL-17822 in healthy males including the complex interaction of food and drug formulation. The amorphous solid dispersion formulation has a lower food effect on bioavailability compared with the nanocrystal formulation.

Population Pharmacokinetics of Upadacitinib in Healthy Subjects and Subjects with Rheumatoid Arthritis: Analyses of Phase I and II Clinical Trials

Background and Objectives

Upadacitinib is a janus kinase (JAK) 1 inhibitor being developed for the treatment of rheumatoid arthritis (RA) and other inflammatory diseases. This work characterized upadacitinib population pharmacokinetics in healthy subjects and RA patients and the effects of covariates on upadacitinib exposure.

Methods

Upadacitinib plasma concentrations (n = 6399) from 107 healthy subjects and 466 RA patients from three phase I and two 12-week RA phase IIb trials (1–48 mg immediate-release doses across studies) were analyzed using non-linear mixed-effects modeling. The models were qualified using bootstrap and stochastic simulations.

Results

A two-compartment model with first-order absorption and elimination described upadacitinib pharmacokinetics. Estimates (95% bootstrap confidence interval) for upadacitinib oral clearance, steady-state volume of distribution, absorption lag time, and mean absorption time were 39.7 (37.8–41.5) L/h, 210 (196–231) L, 0.48 (0.47–0.49) h, and 0.08 (0.04–0.12) h, respectively, for a typical healthy male. Matching on other covariates, a 16 and 32% higher upadacitinib area under the concentration–time curve (AUC) was estimated for females relative to males, and for subjects with RA relative to healthy volunteers, respectively. Subjects with RA with mild or moderate renal impairment were estimated to have 16 and 32% higher upadacitinib AUC, respectively, compared with subjects with RA with normal renal function. Upadacitinib clearance was not correlated with body weight.

Conclusions

Upadacitinib pharmacokinetics follow dose-proportional, bi-exponential disposition. A slightly lower upadacitinib clearance is estimated in subjects with RA than in healthy volunteers, consistent with observations for other JAK inhibitors. Other covariates (weight, sex, mild or moderate renal impairment) are not associated with clinically relevant effects on upadacitinib exposure.

Trial Registration

ClinicalTrials.gov (https://clinicaltrials.gov/) identifiers: NCT01741493, NCT02066389, and NCT01960855.

Electronic supplementary material

The online version of this article (doi:10.1007/s40262-017-0605-6) contains supplementary material, which is available to authorized users.

Extrapolation of a Brivaracetam Exposure-Response Model from Adults to Children with Focal Seizures

Introduction

Prediction of brivaracetam effects in children was obtained by scaling an existing adult pharmacokinetic/pharmacodynamic (PK/PD) model for brivaracetam to children, using an existing population PK model for brivaracetam in children. The scaling was supported by estimating the change from adults to children in the concentration–effect relationship parameters for levetiracetam, a compound interacting with the same target protein (synaptic vesicle protein SV2A).

Methods

The existing adult PK/PD model for brivaracetam was applied to a combined adult–pediatric dataset of levetiracetam. This model was then used to predict the effective oral twice-daily dose of brivaracetam in children aged ≥4 to <16 years as adjunctive treatment for focal (partial onset) seizures. The existing model described daily seizure counts using a negative binomial distribution, taking previous-day seizure frequencies into account, and using a mixture model to separate ‘placebo-like’ and ‘responder’ subpopulations. The model was adapted to describe aggregated monthly seizure counts for adult patients in the levetiracetam studies: daily seizure counts were only available for children in the levetiracetam studies.

Results

The levetiracetam PK/PD model successfully described both the adult and pediatric data using the same drug effect parameters, and using a model structure similar to the existing adult brivaracetam PK/PD model.

Conclusion

Simulation with the adult brivaracetam PK/PD model in combination with an existing pediatric brivaracetam population PK model allowed characterization of the dose–response curve, suggesting maximum response at brivaracetam 4 mg/kg/day dosing (capped at 200 mg/day, the maximum adult dose) in children aged ≥4 years.

Electronic supplementary material

The online version of this article (doi:10.1007/s40262-017-0597-2) contains supplementary material, which is available to authorized users.

Population Pharmacokinetics of the Antituberculosis Agent Pretomanid

A population pharmacokinetic (PopPK) model for pretomanid was developed using data from 14 studies in the pretomanid development program: six phase 1 studies, six phase 2 studies, and two phase 3 studies. The final analysis data set contained 17,725 observations from 1,054 subjects, including healthy subjects and subjects with drug-sensitive, multidrug-resistant, or extensively drug-resistant pulmonary tuberculosis dosed pretomanid in monotherapy or combination therapy for up to 6 months. Pretomanid pharmacokinetic behavior was described by a one-compartment model that at a given dose was linear in its absorption and clearance processes but where the rate of absorption and extent of bioavailability changed with dose. Clearance and volume of distribution scaled allometrically with weight. Apparent clearance in females was 18% less than in males. Among HIV-positive subjects, absent the effect of CYP3A4-inducing antiretrovirals, apparent clearance was 6% higher. Some effects of total bilirubin and albumin were found, but the impacts on exposure were small. Bioavailability in the fasted condition was about half that in the fed condition. Relative bioavailability decreased with increasing dose in the fasted condition, but not for doses of ≤200 mg in the fed condition. HIV-positive subjects taking efavirenz and lopinavir/ritonavir had exposures that were reduced by 46 and 17%, respectively. There was little evidence for noteworthy effects of regimen partners on pretomanid. Standard diagnostics indicated that the model described the voluminous, diverse data well, so that the model could be used to generate exposure metrics for exposure/response analyses to be reported elsewhere.

Population Pharmacokinetic and Exposure-Response Analyses of Prasugrel in Pediatric Patients with Sickle Cell Anemia

BACKGROUND AND OBJECTIVE

Prasugrel, a P2Y₁₂ adenosine diphosphate (ADP) receptor antagonist, inhibits ADP-mediated platelet activation and aggregation in patients with sickle cell anemia (SCA). We developed a population pharmacokinetic (popPK) model in pediatric patients from 2 to <18 years of age with SCA, and performed exposure-response evaluations to characterize the effects of prasugrel in a subset of these patients who weighed 19 kg or more and experienced at least two episodes of vaso-occlusive crises (VOC) in the past year.

METHODS

A three-compartment popPK model adapted from that used in adults with acute coronary syndrome was used to describe the relationship between plasma concentrations of prasugrel's active metabolite (Pras-AM) and time using data from phase II and III clinical studies in children. A VOC event rate model was developed from the phase III study to explore the exposure-response relationship between Pras-AM exposure and VOC, and included evaluation of covariates.

RESULTS

The final popPK model for children with SCA provided a reasonable fit to Pras-AM plasma concentrations over time, with estimates of apparent clearance (CL/F) (172 L/h) and apparent volume of distribution (Vd/F) (51.7 L) that were comparable to previous studies in adults. The final model included weight as a covariate on both CL/F and Vd/F, and age as a covariate on CL/F. Analyses of safety (bleeding events requiring medical intervention) and efficacy (VOC event rate) variables showed no apparent relationship to model-predicted Pras-AM exposure quartiles, and no statistically significant effects of intrinsic or extrinsic factors on the VOC event rate were identified in the VOC event rate model. The effect of post hoc exposures on the VOC event rate did not reach statistical significance.

CONCLUSIONS

A popPK model was developed that provided reasonable parameter estimates, goodness-of-fit diagnostics, and visual predictive checks when applied to Pras-AM plasma concentrations in pediatric patients with SCA. Post hoc exposures obtained from this model did not correlate with measures of VOC or bleeding events in this population.

Population Pharmacokinetics of Artemether, Dihydroartemisinin, and Lumefantrine in Rwandese Pregnant Women Treated for Uncomplicated Plasmodium falciparum Malaria

The artemisinin-based combination therapy artemether-lumefantrine is commonly used in pregnant malaria patients. However, the effect of pregnancy-related changes on exposure is unclear, and pregnancy has been associated with decreased efficacy in previous studies. This study aimed to characterize the population pharmacokinetics of artemether, its active metabolite dihydroartemisinin, and lumefantrine in 22 Rwandese pregnant women in their second (n = 11) or third (n = 11) trimester with uncomplicated Plasmodium falciparum malaria. These patients were enrolled from Rwamagana district hospital and received the standard fixed oral dose combination of 80 mg of artemether and 480 mg of lumefantrine twice daily for 3 days. Venous plasma concentrations were quantified for all three analytes using liquid chromatography coupled with tandem mass spectroscopy, and data were analyzed using nonlinear mixed-effects modeling. Lumefantrine pharmacokinetics was described by a flexible but highly variable absorption, with a mean absorption time of 4.04 h, followed by a biphasic disposition model. The median area under the concentration-time curve from 0 h to infinity (AUC_0-∞) for lumefantrine was 641 h · mg/liter. Model-based simulations indicated that 11.7% of the study population did not attain the target day 7 plasma concentration (280 ng/ml), a threshold associated with increased risk of recrudescence. The pharmacokinetics of artemether was time dependent, and the autoinduction of its clearance was described using an enzyme turnover model. The turnover half-life was predicted to be 30.4 h. The typical oral clearance, which started at 467 liters/h, increased 1.43-fold at the end of treatment. Simulations suggested that lumefantrine pharmacokinetic target attainment appeared to be reassuring in Rwandese pregnant women, particularly compared to target attainment in Southeast Asia. Larger cohorts will be required to confirm this finding.

Model-Based Prediction of Plasma Concentration and Enterohepatic Circulation of Total Bile Acids in Humans

Bile acids released postprandially can modify the rate and extent of lipophilic compounds' absorption. This study aimed to predict the enterohepatic circulation (EHC) of total bile acids (TBAs) in response to caloric intake from their spillover in plasma. A model for TBA EHC was combined with a previously developed gastric emptying (GE) model. Longitudinal gallbladder volumes and TBA plasma concentration data from 30 subjects studied after ingestion of four different test drinks were supplemented with literature data. Postprandial gallbladder refilling periods were implemented to improve model predictions. The TBA hepatic extraction was reduced with the high-fat drink. Basal and nutrient-induced gallbladder emptying rates were altered by type 2 diabetes (T2D). The model was predictive of the central trend and the variability of gallbladder volume and TBA plasma concentration for all test drinks. Integration of this model within physiological pharmacokinetic modeling frameworks could improve the predictions for lipophilic compounds' absorption considerably.

Population pharmacokinetics-pharmacodynamics of oral everolimus in patients with seizures associated with tuberous sclerosis complex

Everolimus is approved in Europe and in the USA for the adjunctive treatment of patients aged 2 years and older whose refractory partial-onset seizures, with or without secondary generalization, are associated with tuberous sclerosis complex. The objective of this analysis was to establish a population pharmacokinetic (PK)/pharmacodynamic model describing the relationship between seizure frequency and everolimus exposure to confirm the recommended target concentration range of 5-15 ng/mL. The PK model was a two-compartment model with first order absorption and clearance. CYP3A and P-gp inducers and body-surface area were shown to impact everolimus exposure, justifying dose adjustments. A Poisson distribution was found to adequately describe the random nature of daily seizure counts during the screening phase. A placebo effect on the Poisson seizure mean was implemented as an asymptotic exponential function of time leading to a new steady-state seizure mean. The everolimus effect was implemented as an inhibitory E_max function of C_min on the seizure mean, where E_max exhibited an asymptotic exponential increase over time to a higher steady-state value. Increasing age was found to decrease the baseline seizure mean and to prolong the half-life of the increase in E_max. The dependence of seizure frequencies on C_min was explored by simulation. The responder rate increased with increasing C_min. As C_min decreased below 5 ng/mL, variability in response became larger and responder rates decreased more rapidly. The results supported the recommended target concentration range for everolimus of 5-15 ng/mL to ensure treatment efficacy.

Modelling of mycobacterial load reveals bedaquiline's exposure-response relationship in patients with drug-resistant TB

Background

Bedaquiline has been shown to reduce time to sputum culture conversion (SCC) and increase cure rates in patients with drug-resistant TB, but the influence of drug exposure remains uncharacterized.

Objectives

To investigate whether an exposure–response relationship could be characterized by making better use of the existing information on pharmacokinetics and longitudinal measurements of mycobacterial load.

Methods

Quantitative culture data in the form of time to positivity (TTP) in mycobacterial growth indicator tubes obtained from a randomized placebo-controlled Phase IIb registration trial were examined using non-linear mixed-effects methodology. The link to individual bedaquiline exposures and other patient characteristics was evaluated.

Results

The developed model included three simultaneously fitted components: a longitudinal representation of mycobacterial load in patients, a probabilistic component for bacterial presence in sputum samples, and a time-to-event model for TTP. Data were described adequately, and time to SCC was well predicted. Individual bedaquiline exposure was found to significantly affect the decline in mycobacterial load. Consequently, the proportion of patients without SCC at week 20 is expected to decrease from 25% (95% CI 20%–31%) without bedaquiline to 17% (95% CI 13%–21%), 12% (95% CI 8%–16%) and 7% (95% CI 4%–11%), respectively, with half the median, median and double the median bedaquiline exposure observed in patients with standard dosing. Baseline bacterial load and level of drug resistance were other important predictors.

Conclusions

To our knowledge, this is the first successful description of bedaquiline’s exposure–response relationship and may be used when considering dose optimization. Characterization of this relationship was possible by integrating quantitative information in existing clinical data using novel models.

Dose/Exposure-Response Modeling to Support Dosing Recommendation for Phase III Development of Baricitinib in Patients with Rheumatoid Arthritis

Baricitinib is an oral inhibitor of Janus kinases (JAKs), selective for JAK1 and 2. It demonstrated dose‐dependent efficacy in patients with moderate‐to‐severe rheumatoid arthritis (RA) in a phase IIb study up to 24 weeks. Population pharmacokinetic/pharmacodynamic (PopPK/PD) models were developed to characterize concentration‐time profiles and dose/exposure‐response (D/E‐R) relationships for the key efficacy (proportion of patients achieving American College of Rheumatology 20%, 50%, or 70% response rate) and safety endpoints (incidence of anemia) for the phase IIb study. The modeling suggested that 4 mg q.d. was likely to offer the optimum risk/benefit balance, whereas 2 mg q.d. had the potential for adequate efficacy. In addition, at the same total daily dose, a twice‐daily regimen is not expected to provide an advantage over q.d. dosing for the efficacy or safety endpoints. The model‐based simulations formed the rationale for key aspects of dosing, such as dose levels and dosing frequency for phase III development.

Dose-Exposure-Response Relationship of the Investigational Anti-Interleukin-13 Monoclonal Antibody Tralokinumab in Patients With Severe, Uncontrolled Asthma

Interleukin (IL)-13 is involved in the pathogenesis of some types of asthma. Tralokinumab is a human immunoglobulin G₄ monoclonal antibody that specifically binds to IL-13. Two placebo-controlled phase II studies (phase IIa, NCT00873860 and phase IIb, NCT01402986) have been conducted in which tralokinumab was administered subcutaneously. This investigation aimed to characterize tralokinumab's dose-exposure-response (forced expiratory volume in 1 s (FEV₁ )) relationship in patients with asthma and to predict the most appropriate dose for phase III. An integrated population pharmacokinetic-pharmacodynamic (PK/PD) modeling analysis was required for phase III dose selection, due to differing phase II patient populations, designs, and regimens. Analysis of combined datasets enabled the identification of tralokinumab's dose-exposure-FEV₁ response relationship in patients with asthma. Near-maximal FEV₁ increase was predicted at a dose of 300 mg SC once every 2 weeks (Q2W). This dose was chosen for tralokinumab in the phase III clinical development program for treatment of severe, uncontrolled asthma.

Model-Based Approach to Predict Adherence to Protocol During Antiobesity Trials

Development of antiobesity drugs is continuously challenged by high dropout rates during clinical trials. The objective was to develop a population pharmacodynamic model that describes the temporal changes in body weight, considering disease progression, lifestyle intervention, and drug effects. Markov modeling (MM) was applied for quantification and characterization of responder and nonresponder as key drivers of dropout rates, to ultimately support the clinical trial simulations and the outcome in terms of trial adherence. Subjects (n = 4591) from 6 Contrave^® trials were included in this analysis. An indirect‐response model developed by van Wart et al was used as a starting point. Inclusion of drug effect was dose driven using a population dose‐ and time‐dependent pharmacodynamic (DTPD) model. Additionally, a population‐pharmacokinetic parameter‐ and data (PPPD)‐driven model was developed using the final DTPD model structure and final parameter estimates from a previously developed population pharmacokinetic model based on available Contrave^® pharmacokinetic concentrations. Last, MM was developed to predict transition rate probabilities among responder, nonresponder, and dropout states driven by the pharmacodynamic effect resulting from the DTPD or PPPD model. Covariates included in the models and parameters were diabetes mellitus and race. The linked DTPD‐MM and PPPD‐MM was able to predict transition rates among responder, nonresponder, and dropout states well. The analysis concluded that body‐weight change is an important factor influencing dropout rates, and the MM depicted that overall a DTPD model‐driven approach provides a reasonable prediction of clinical trial outcome probabilities similar to a pharmacokinetic‐driven approach.

Item Response Theory to Quantify Longitudinal Placebo and Paliperidone Effects on PANSS Scores in Schizophrenia

As biomarkers are lacking, multi-item questionnaire-based tools like the Positive and Negative Syndrome Scale (PANSS) are used to quantify disease severity in schizophrenia. Analyzing composite PANSS scores as continuous data discards information and violates the numerical nature of the scale. Here a longitudinal analysis based on Item Response Theory is presented using PANSS data from phase III clinical trials. Latent disease severity variables were derived from item-level data on the positive, negative, and general PANSS subscales each. On all subscales, the time course of placebo responses were best described with Weibull models, and dose-independent functions with exponential models to describe the onset of the full effect were used to describe paliperidone's effect. Placebo and drug effect were most pronounced on the positive subscale. The final model successfully describes the time course of treatment effects on the individual PANSS item-levels, on all PANSS subscale levels, and on the total score level.

A Pharmacometric Framework for Axitinib Exposure, Efficacy, and Safety in Metastatic Renal Cell Carcinoma Patients

The relationships between exposure, biomarkers (vascular endothelial growth factor (VEGF), soluble VEGF receptors (sVEGFR)-1, -2, -3, and soluble stem cell factor receptor (sKIT)), tumor sum of longest diameters (SLD), diastolic blood pressure (dBP), and overall survival (OS) were investigated in a modeling framework. The dataset included 64 metastatic renal cell carcinoma patients (mRCC) treated with oral axitinib. Biomarker timecourses were described by indirect response (IDR) models where axitinib inhibits sVEGFR-1, -2, and -3 production, and VEGF degradation. No effect was identified on sKIT. A tumor model using sVEGFR-3 dynamics as driver predicted SLD data well. An IDR model, with axitinib exposure stimulating the response, characterized dBP increase. In a time-to-event model the SLD timecourse predicted OS better than exposure, biomarker- or dBP-related metrics. This type of framework can be used to relate pharmacokinetics, efficacy, and safety to long-term clinical outcome in mRCC patients treated with VEGFR inhibitors. (ClinicalTrial.gov identifier NCT00569946.).

Pharmacometric Modeling of Liver Metastases' Diameter, Volume, and Density and Their Relation to Clinical Outcome in Imatinib-Treated Patients With Gastrointestinal Stromal Tumors

Three‐dimensional and density‐based tumor metrics have been suggested to better discriminate tumor response to treatment than unidimensional metrics, particularly for tumors exhibiting nonuniform size changes. In the developed pharmacometric modeling framework based on data from 77 imatinib‐treated gastrointestinal patients, the time‐courses of liver metastases' maximum transaxial diameters, software‐calculated actual volumes (V_actual) and calculated ellipsoidal volumes were characterized by logistic growth models, in which imatinib induced a linear dose‐dependent size reduction. An indirect response model best described the reduction in density. Substantial interindividual variability in the drug effect of all response assessments and additional interlesion variability in the drug effect on density were identified. The predictive ability of longitudinal tumor unidimensional and three‐dimensional size and density on overall survival (OS) and progression‐free survival (PFS) were compared using parametric time‐to‐event models. Death hazard increased with increasing V_actual. This framework may guide early clinical interventions based on three‐dimensional tumor responses to enhance benefits for patients with gastrointestinal stromal tumors (GIST).

Population Pharmacokinetics of Bedaquiline and Metabolite M2 in Patients With Drug-Resistant Tuberculosis: The Effect of Time-Varying Weight and Albumin

Albumin concentration and body weight are altered in patients with multidrug‐resistant tuberculosis (MDR‐TB) and change during the long treatment period, potentially affecting drug disposition. We here describe the pharmacokinetics (PKs) of the novel anti‐TB drug bedaquiline and its metabolite M2 in 335 patients with MDR‐TB receiving 24 weeks of bedaquiline on top of a longer individualized background regimen. Semiphysiological models were developed to characterize the changes in weight and albumin over time. Bedaquiline and M2 disposition were well described by three and one‐compartment models, respectively. Weight and albumin were correlated, typically increasing after the start of treatment, and significantly affected bedaquiline and M2 plasma disposition. Additionally, age and race were significant covariates, whereas concomitant human immunodeficiency virus (HIV) infection, sex, or having extensively drug‐resistant TB was not. This is the first population model simultaneously characterizing bedaquiline and M2 PKs in its intended use population. The developed model will be used for efficacy and safety exposure‐response analyses.

The variability in beta-cell function in placebo-treated subjects with type 2 diabetes: application of the weight-HbA1c-insulin-glucose (WHIG) model

AIM

The weight-glycosylated haemoglobin (HbA1C)-insulin-glucose (WHIG) model describes the effects of changes in weight on insulin sensitivity (IS) in newly diagnosed, obese subjects receiving placebo treatment. This model was applied to a wider population of placebo-treated subjects, to investigate factors influencing the variability in IS and β-cell function.

METHODS

The WHIG model was applied to the WHIG dataset (Study 1) and two other placebo datasets (Studies 2 and 3). Studies 2 and 3 consisted of nonobese subjects and subjects with advanced type 2 diabetes mellitus (T2DM). Body weight, fasting serum insulin (FSI), fasting plasma glucose (FPG) and HbA1c were used for nonlinear mixed-effects modelling (using NONMEM v7.2 software). Sources of interstudy variability (ISV) and potential covariates (age, gender, diabetes duration, ethnicity, compliance) were investigated.

RESULTS

An ISV for baseline parameters (body weight and β-cell function) was required. The baseline β-cell function was significantly lower in subjects with advanced T2DM (median difference: Study 2: 15.6%, P < 0.001; Study 3: 22.7%, P < 0.001) than in subjects with newly diagnosed T2DM (Study 1). A reduction in the estimated insulin secretory response in subjects with advanced T2DM was observed but diabetes duration was not a significant covariate.

CONCLUSION

The WHIG model can be used to describe the changes in weight, IS and β-cell function in the diabetic population. IS remained relatively stable between subjects but a large ISV in β-cell function was observed. There was a trend towards decreasing β-cell responsiveness with diabetes duration, and further studies, incorporating subjects with a longer history of diabetes, are required.

Brivaracetam Population Pharmacokinetics and Exposure-Response Modeling in Adult Subjects With Partial-Onset Seizures

Brivaracetam is a selective high-affinity ligand for synaptic vesicle protein 2A, recently approved as adjunctive therapy in the treatment of partial-onset (focal) seizures in patients 16 years of age and older with epilepsy. A population pharmacokinetic (PK) model and a population pharmacokinetic/pharmacodynamic (PKPD) model were developed describing brivaracetam plasma concentration and the relationship with daily seizure counts in adequate well-controlled efficacy trials. The effect of body weight on clearance and volume was implemented using allometric scaling, and a range of covariates were investigated for their influence on brivaracetam clearance. The PKPD model described daily seizure counts using a negative binomial distribution, taking previous day seizures into account, and using a mixture model to separate "placebo-like" and "response" subpopulations. The PK and PKPD models provided a good description of the data, documented using visual predictive checks. Coadministration with carbamazepine, phenytoin, and phenobarbital decreased brivaracetam exposure by 26%, 21%, and 19%, respectively, without significant effects on PD response. Covariate analysis indicated that levetiracetam coadministration reduced the fraction of subjects in the mixture model response population to 4% and identified baseline seizure frequency as a strong predictor for being assigned to the mixture model response population. Simulation allowed characterization of the dose-response curve, suggesting maximum response is obtained at brivaracetam 150-200 mg/day.

A strategy for residual error modeling incorporating scedasticity of variance and distribution shape

Nonlinear mixed effects models parameters are commonly estimated using maximum likelihood. The properties of these estimators depend on the assumption that residual errors are independent and normally distributed with mean zero and correctly defined variance. Violations of this assumption can cause bias in parameter estimates, invalidate the likelihood ratio test and preclude simulation of real-life like data. The choice of error model is mostly done on a case-by-case basis from a limited set of commonly used models. In this work, two strategies are proposed to extend and unify residual error modeling: a dynamic transform-both-sides approach combined with a power error model (dTBS) capable of handling skewed and/or heteroscedastic residuals, and a t-distributed residual error model allowing for symmetric heavy tails. Ten published pharmacokinetic and pharmacodynamic models as well as stochastic simulation and estimation were used to evaluate the two approaches. dTBS always led to significant improvements in objective function value, with most examples displaying some degree of right-skewness and variances proportional to predictions raised to powers between 0 and 1. The t-distribution led to significant improvement for 5 out of 10 models with degrees of freedom between 3 and 9. Six models were most improved by the t-distribution while four models benefited more from dTBS. Changes in other model parameter estimates were observed. In conclusion, the use of dTBS and/or t-distribution models provides a flexible and easy-to-use framework capable of characterizing all commonly encountered residual error distributions.

Modeling Testosterone Circadian Rhythm in Hypogonadal Males: Effect of Age and Circannual Variations

The objective of this study was to characterize the baseline circadian rhythm of testosterone levels in hypogonadal men. A total of 859 baseline profiles of testosterone from hypogonadal men were included in this analysis. The circadian rhythm of the testosterone was described by a stretched cosine function. Model parameters were estimated using NONMEM^® 7.3. The effect of different covariates on the testosterone levels was investigated. Model evaluation was performed using non-parametric bootstrap and predictive checks. A stretched cosine function deeply improved the data goodness of fit compared to the standard trigonometric function (p < 0.001; ΔOFV = −204). The effect of the age and the semester, defined as winter and spring versus summer and fall, were significantly associated with the baseline levels of testosterone (p < 0.001, ΔOFV = −15.6, and p < 0.001, ΔOFV = −47.0). Model evaluation procedures such as diagnostic plots, visual predictive check, and non-parametric bootstrap evidenced that the proposed stretched cosine function was able to model the time course of the diurnal testosterone levels in hypogonadal males with accuracy and precision. The circadian rhythm of the testosterone levels was better predicted by the proposed stretched cosine function than a standard cosine function. Testosterone levels decreased by 5.74 ng/dL (2.4%) every 10 years and were 19.3 ng/dL (8.1%) higher during winter and spring compared to summer and fall.

Population pharmacokinetic and pharmacodynamic modeling of epinephrine administered using a mobile inhaler

Inhaled epinephrine is a potential alternative to self-administered intramuscular epinephrine in imminent anaphylactic reactions. The objective was to develop a pharmacokinetic-pharmacodynamic model describing exposure and effects on heart rate of inhaled epinephrine. Data from a 4-phase cross-over clinical trial in 9 healthy volunteers including 0.3 mg intramuscular epinephrine, two doses of inhaled epinephrine (4 mg/mL solution administered during [mean] 18 and 25 min, respectively) using a mobile pocket inhaler, and an inhaled placebo were analyzed using mixed-effects modeling. Inhaled epinephrine was available almost immediately and more rapidly than via the intramuscular route (absorption half-live 29 min). Epinephrine plasma concentrations declined rapidly after terminating inhalation (elimination half-life 4.1 min) offering the option to stop exposure in case of adverse events. While the expected maximum concentration was higher for inhaled epinephrine, this was not associated with safety concerns due to only moderate additional hemodynamic effects compared to intramuscular administration. Bioavailability after inhalation (4.7%) was subject to high interindividual and interoccasional variability highlighting that training of inhalation would be essential for patients. The proposed model suggests that the use of a highly concentrated epinephrine solution via inhalation may offer an effective treatment option in anaphylaxis, while efficacy in patients remains to be shown.

Lumefantrine and Desbutyl-Lumefantrine Population Pharmacokinetic-Pharmacodynamic Relationships in Pregnant Women with Uncomplicated Plasmodium falciparum Malaria on the Thailand-Myanmar Border

Artemether-lumefantrine is the most widely used antimalarial artemisinin-based combination treatment. Recent studies have suggested that day 7 plasma concentrations of the potent metabolite desbutyl-lumefantrine correlate better with treatment outcomes than those of lumefantrine. Low cure rates have been reported in pregnant women with uncomplicated falciparum malaria treated with artemether-lumefantrine in northwest Thailand. A simultaneous pharmacokinetic drug-metabolite model was developed based on dense venous and sparse capillary lumefantrine and desbutyl-lumefantrine plasma samples from 116 pregnant patients on the Thailand-Myanmar border. The best model was used to evaluate therapeutic outcomes with a time-to-event approach. Lumefantrine and desbutyl-lumefantrine concentrations, implemented in an Emax model, both predicted treatment outcomes, but lumefantrine provided better predictive power. A combined model including both lumefantrine and desbutyl-lumefantrine did not improve the model further. Simulations suggested that cure rates in pregnant women with falciparum malaria could be increased by prolonging the treatment course. (These trials were registered at controlled-trials.com [ISRCTN 86353884].).

Inter-study variability of preclinical in vivo safety studies and translational exposure-QTc relationships--a PKPD meta-analysis

BACKGROUND AND PURPOSE

Preclinical cardiovascular safety studies (CVS) have been compared between facilities with respect to their sensitivity to detect drug-induced QTc prolongation (ΔQTc). Little is known about the consistency of quantitative ΔQTc predictions that are relevant for translation to humans.

EXPERIMENTAL APPROACH

We derived typical ΔQTc predictions at therapeutic exposure (ΔQTcTHER ) with 95% confidence intervals (95%CI) for 3 Kv 11.1 (hERG) channel blockers (moxifloxacin, dofetilide and sotalol) from a total of 14 CVS with variable designs in the conscious dog. Population pharmacokinetic-pharmacodynamic (PKPD) analysis of each study was followed by a meta-analysis (pooling 2-6 studies including 10-32 dogs per compound) to derive meta-predictions of typical ΔQTcTHER . Meta-predictions were used as a reference to evaluate the consistency of study predictions and to relate results to those found in the clinical literature.

KEY RESULTS

The 95%CIs of study-predicted ΔQTcTHER comprised in 13 out of 14 cases the meta-prediction. Overall inter-study variability (mean deviation from meta-prediction at upper level of therapeutic exposure) was 30% (range: 1-69%). Meta-ΔQTcTHER predictions for moxifloxacin, dofetilide and sotalol overlapped with reported clinical QTc prolongation when expressed as %-prolongation from baseline.

CONCLUSIONS AND IMPLICATIONS

Consistent exposure-ΔQTc predictions were obtained from single preclinical dog studies of highly variable designs by systematic PKPD analysis, which is suitable for translational purposes. The good preclinical-clinical pharmacodynamic correlations obtained suggest that such an analysis should be more routinely applied to increase the informative and predictive value of results obtained from animal experiments.

Evaluation of Approaches to Deal with Low-Frequency Nuisance Covariates in Population Pharmacokinetic Analyses

Clinical studies include occurrences of rare variables, like genotypes, which due to their frequency and strength render their effects difficult to estimate from a dataset. Variables that influence the estimated value of a model-based parameter are termed covariates. It is often difficult to determine if such an effect is significant, since type I error can be inflated when the covariate is rare. Their presence may have either an insubstantial effect on the parameters of interest, hence are ignorable, or conversely they may be influential and therefore non-ignorable. In the case that these covariate effects cannot be estimated due to power and are non-ignorable, then these are considered nuisance, in that they have to be considered but due to type 1 error are of limited interest. This study assesses methods of handling nuisance covariate effects. The specific objectives include (1) calibrating the frequency of a covariate that is associated with type 1 error inflation, (2) calibrating its strength that renders it non-ignorable and (3) evaluating methods for handling these non-ignorable covariates in a nonlinear mixed effects model setting. Type 1 error was determined for the Wald test. Methods considered for handling the nuisance covariate effects were case deletion, Box-Cox transformation and inclusion of a specific fixed effects parameter. Non-ignorable nuisance covariates were found to be effectively handled through addition of a fixed effect parameter.

Incorporation of stochastic variability in mechanistic population pharmacokinetic models: handling the physiological constraints using normal transformations

The utilisation of physiologically-based pharmacokinetic models for the analysis of population data is an approach with progressively increasing impact. However, as we move from empirical to complex mechanistic model structures, incorporation of stochastic variability in model parameters can be challenging due to the physiological constraints that may arise. Here, we investigated the most common types of constraints faced in mechanistic pharmacokinetic modelling and explored techniques for handling them during a population data analysis. An efficient way to impose stochastic variability on the parameters of interest without neglecting the underlying physiological constraints is through the assumption that they follow a distribution with support and properties matching the underlying physiology. It was found that two distributions that arise through transformations of the normal, the logit-normal generalisation and the logistic-normal, are excellent for such an application as not only they can satisfy the physiological constraints but also offer high flexibility during characterisation of the parameters' distribution. The statistical properties and practical advantages/disadvantages of these distributions for such an application were clearly displayed in the context of different modelling examples. Finally, a simulation study clearly illustrated the practical gains of the utilisation of the described techniques, as omission of population variability in physiological systems parameters leads to a biased/misplaced stochastic model with mechanistically incorrect variance structure. The current methodological work aims to facilitate the use of mechanistic/physiologically-based models for the analysis of population pharmacokinetic clinical data.

Population pharmacokinetic analysis of voriconazole from a pharmacokinetic study with immunocompromised Japanese pediatric subjects

A population pharmacokinetic (PK) analysis was conducted to characterize the voriconazole pharmacokinetic profiles in immunocompromised Japanese pediatric subjects and to compare them to those in immunocompromised non-Japanese pediatric subjects. A previously developed two-compartment pharmacokinetic model with first-order absorption and mixed linear and nonlinear elimination adequately described the voriconazole intravenous and oral data from Japanese pediatric subjects with few modifications. Bayesian priors were applied to this analysis by using the NONMEM routine NWPRI, which allowed priors for the fixed-effect parameter vector and variance matrix of the random-effect parameters to be a normal distribution and an inverse Wishart distribution, respectively. Large intersubject variabilities in oral bioavailability and voriconazole exposure were observed in these pediatric subjects. The mean oral bioavailability estimated in Japanese pediatric subjects was 73% (range, 17% to 99%), which is consistent with the reported estimates of 64% in the previous model and less than what was originally estimated for healthy adults-96%. Voriconazole exposures in Japanese pediatric subjects were generally comparable to those in non-Japanese pediatric subjects receiving the same dosing regimens, given the large intersubject variability. Consistent with the previous findings, the CYP2C19 genotyping status did not have a clinically relevant effect on voriconazole exposure in Japanese pediatric subjects, although it was identified as a covariate in the model to help explain the intersubject variability in voriconazole exposure. The CYP2C19 genotyping status alone does not warrant dose adjustment of voriconazole. No other factors besides age and weight were identified to explain the PK variability of voriconazole.

Evaluation of the long-term durability and glycemic control of fasting plasma glucose and glycosylated hemoglobin for pioglitazone in Japanese patients with type 2 diabetes

BACKGROUND

This study applied a pharmacodynamic model-based approach to evaluate the long-term durability and glycemic control of pioglitazone in comparison with other oral glucose-lowering drugs in Japanese type 2 diabetes mellitus (T2DM) patients.

SUBJECTS AND METHODS

Japanese T2DM patients were enrolled in a prospective, randomized, open-label, blinded-end point study and received pioglitazone with or without other oral glucose-lowering drugs (excluding another thiazolidinedione [TZD]) (n=293) or oral glucose-lowering drugs excluding TZD (n=294). Treatment was adjusted to achieve glycosylated hemoglobin (HbA1c) <6.9%, and samples for fasting plasma glucose (FPG) and HbA1c were collected over 2.5-4 years. A simultaneous cascading indirect response model structure was applied to describe the time course of FPG and HbA1c. HbA1c levels were described using both an FPG-dependent and an FPG-independent function. To account for titration, drug effects for both treatment groups were implemented using a time-dependent Emax model.

RESULTS

Pioglitazone was superior in both time to maximum effect and the magnitude of reduction achieved in FPG and HbA1c. A greater reduction in median FPG (-21 mg/dL vs. -9 mg/dL) was observed with pioglitazone (P<0.05). Maximum drug effect for FPG was predicted to occur earlier (11 months) for pioglitazone than for the control group (14 months). The simulated additional reduction in FPG and HbA1c achieved with pioglitazone was predicted to be maintained beyond the currently observed study duration.

CONCLUSIONS

Pioglitazone was found to result in improved glycemic control and durability compared with control treatment. This model-based approach enabled the quantification of differences in FPG and HbA1c for both treatment groups and simulation to evaluate longer-term durability on FPG and HbA1c.

Rilotumumab exposure-response relationship in patients with advanced or metastatic gastric cancer

PURPOSE

Rilotumumab is an investigational, fully human monoclonal antibody to hepatocyte growth factor. In a randomized phase II study, trends toward improved survival were observed with rilotumumab (7.5 or 15 mg/kg) plus epirubicin, cisplatin, and capecitabine (ECX) versus placebo plus ECX in gastric/gastroesophageal junction (GEJ) cancer patients, especially in MET-positive patients. Here, we quantitatively characterized the longitudinal exposure-response [tumor growth (TG) and overall survival (OS)] relationship for rilotumumab.

EXPERIMENTAL DESIGN

Rilotumumab concentrations, tumor sizes, and survival time from the phase II study were pooled to develop a longitudinal exposure versus TG model and parametric OS model that explored predictive/prognostic/treatment effects (MET expression, rilotumumab exposure, relative tumor size). Model evaluation included visual predictive checks, nonparametric bootstrap, and normalized prediction distribution errors. Simulations were undertaken to predict the relationship between rilotumumab dose and OS.

RESULTS

Rilotumumab exhibited linear time-independent pharmacokinetics not affected by MET expression. The TG model adequately described tumor size across arms. A Weibull distribution best described OS. Rilotumumab exposure and change in tumor size from baseline at week 24 were predictive of OS. MET-positive patients showed shorter survival and responded better to rilotumumab than MET-negative patients. Simulations predicted a median (95% confidence interval) HR of 0.38 (0.18-0.60) in MET-positive patients treated with 15 mg/kg rilotumumab Q3W.

CONCLUSIONS

Rilotumumab plus ECX demonstrated concentration-dependent effects on OS, influenced by MET expression, and tumor size in gastric/GEJ cancer patients. These findings support the phase II testing of rilotumumab 15 mg/kg every 3 weeks in MET-positive gastric/GEJ cancer (RILOMET-1; NCT01697072).

Mixed-effects modeling of clinical DCE-MRI data: application to colorectal liver metastases treated with bevacizumab

PURPOSE

Most dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data are evaluated for individual patients with cohorts analyzed to detect significant changes from baseline values, repeating the process at each posttreatment timepoint. Our study aimed to develop a statistically valid model for the complete time course of DCE-MRI data in a patient cohort.

MATERIALS AND METHODS

Data from 10 patients with colorectal cancer liver metastases were analyzed, including two baseline scans and four post-bevacizumab scans. Apparent changes in tumor median K(trans) were adjusted for changes in observed enhancing tumor fraction (EnF) by multiplying K(trans) by EnF (KEnF). A mixed-effects model (MEM) was defined to describe the KEnF time course for all patients simultaneously by assuming a three-parameter indirect response model with model parameters lognormally distributed across patients.

RESULTS

The typical cohort time course showed a KEnF reduction to 59% of baseline at 24 hours, returning to 65% of baseline values by day 12. Interpatient variability of model parameters ranged from 11% to 307%.

CONCLUSION

The MEM approach has potential for comparing responses at a group level in clinical trials with different doses, schedules, or combination regimens. Furthermore, the KEnF biomarker successfully resolved confounds in interpreting K(trans) arising from therapy induced changes in the volume of enhancing tumor.

Pharmacokinetic-pharmacodynamic modeling of antibacterial drugs

Pharmacokinetic-pharmacodynamic (PKPD) modeling and simulation has evolved as an important tool for rational drug development and drug use, where developed models characterize both the typical trends in the data and quantify the variability in relationships between dose, concentration, and desired effects and side effects. In parallel, rapid emergence of antibiotic-resistant bacteria imposes new challenges on modern health care. Models that can characterize bacterial growth, bacterial killing by antibiotics and immune system, and selection of resistance can provide valuable information on the interactions between antibiotics, bacteria, and host. Simulations from developed models allow for outcome predictions of untested scenarios, improved study designs, and optimized dosing regimens. Today, much quantitative information on antibiotic PKPD is thrown away by summarizing data into variables with limited possibilities for extrapolation to different dosing regimens and study populations. In vitro studies allow for flexible study designs and valuable information on time courses of antibiotic drug action. Such experiments have formed the basis for development of a variety of PKPD models that primarily differ in how antibiotic drug exposure induces amplification of resistant bacteria. The models have shown promise for efficacy predictions in patients, but few PKPD models describe time courses of antibiotic drug effects in animals and patients. We promote more extensive use of modeling and simulation to speed up development of new antibiotics and promising antibiotic drug combinations. This review summarizes the value of PKPD modeling and provides an overview of the characteristics of available PKPD models of antibiotics based on in vitro, animal, and patient data.