Pharmacokinetic and pharmacodynamic modeling of romiplostim in animals.

PURPOSE: Romiplostim is a novel thrombopoiesis-stimulating peptibody that targets the thrombopoietin c-Mpl receptor, resulting in increased platelet production. The pharmacodynamic-mediated disposition (PDMDD) and its stimulatory effect on platelet production in Sprague-Dawley rats, rhesus monkeys, and cynomolgus monkeys following IV bolus and SC administration at various dose levels were determined. METHODS: The pharmacokinetic (PK) profile was described by a PDMDD model that accounts for romiplostim binding to the c-Mpl receptor. The PD model contained a series of aging compartments for precursor cells in bone marrow and platelets. The stimulatory function was described by an on-and-off function operating on the fractional receptor occupancy (RO). The threshold effect, RO(thr), and K(D) parameters were determinants of drug potency, whereas S(max) reflected drug efficacy. RESULTS: The model implicated that receptor-mediated clearance was negligible. RO(thr) estimated occupancies were 0.288, 0.385, 0.771 for rats, rhesus, and cynomolgus monkeys, respectively. The analogous estimated values of K(D) were 4.05, 2320, and 429 ng/mL, implying that romiplostim was much more potent in rats, which was confirmed by a dose-response (ratio of peak platelet count to baseline) relationship. CONCLUSIONS: The model adequately described romiplostim serum concentrations and platelet counts in rats, rhesus monkeys, and cynomolgus monkeys, and quantified linear clearance, PDMDD, and potency of romiplostim.

Population pharmacokinetic analysis of denosumab in patients with bone metastases from solid tumours.

BACKGROUND AND OBJECTIVE: Denosumab (XGEVA®; AMG 162) is a fully human IgG2 monoclonal antibody, which binds to the receptor activator of nuclear factor κ-B ligand (RANKL) and prevents terminal differentiation, activation and survival of osteoclasts. We aimed to characterize the population pharmacokinetics of denosumab in patients with advanced solid tumours and bone metastases. METHODS: A total of 14 228 free serum concentrations of denosumab from 1076 subjects (495 healthy subjects and 581 advanced cancer patients with solid tumours and bone metastases) included in 14 clinical studies were pooled. Denosumab was administered as either single intravenous (n = 36), single subcutaneous (n = 490) or multiple subcutaneous doses (n = 550) ranging from 30 to 180 mg (or from 0.01 to 3 mg/kg) and was given every 4 or 12 weeks for up to 3 years. An open two-compartment pharmacokinetic model with first-order absorption, linear distribution to a peripheral compartment, linear clearance and quasi-steady-state approximation of the target-mediated drug disposition was used to describe denosumab pharmacokinetics, using NONMEM Version 7.1.0 software. The influence of covariates (body weight, age, race, tumour type) was investigated using the full model approach. Model evaluation was performed through visual predictive checks. Model-based simulations were conducted to explore the role of covariates on denosumab serum concentrations and inferred RANKL occupancy. RESULTS: After subcutaneous administration, the dose-independent bioavailability and mean absorption half-life of denosumab were estimated to be 61% and 2.7 days, respectively. The central volume of distribution and linear clearance were 2.62 L/66 kg and 3.25 mL/h/66 kg, respectively. Clearance and volume parameters were proportional to body weight. Assuming 1 : 1 denosumab-RANKL binding, the baseline RANKL level, quasi-steady-state constant and RANKL degradation rate were inferred to be 4.46 nmol/L, 208 ng/mL and 0.00116 h-1, respectively. Between-subject variability in model parameters was moderate. Following 120 mg dosing every 4 weeks, the inferred RANKL occupancy at steady state exceeded 97% during the entire dosing interval in more than 95% of subjects, regardless of the patient covariates. CONCLUSIONS: The integration of pharmacokinetic data from 14 clinical studies demonstrated denosumab RANKL-mediated pharmacokinetics. Pharmacokinetics-based dosage adjustments on the basis of body weight, age, race and tumour type are not necessary in patients with bone metastases from solid tumours.

Denosumab dose selection for patients with bone metastases from solid tumors.

PURPOSE: To quantitatively characterize the longitudinal dose exposure-response [urinary N-telopeptide normalized to urinary creatinine (uNTx/Cr) suppression] relationship for denosumab in patients with bone metastases from solid tumors. EXPERIMENTAL DESIGN: Data from 373 patients who received denosumab as single or multiple subcutaneous doses ranging from 30 to 180 mg (or 0.01 to 3 mg/kg) administered every 4 or 12 weeks for up to 3 years were used in this analysis. An inhibitory sigmoid I(Max) model was used to characterize the time course of uNTx/Cr as a function of serum denosumab concentrations and the M3 method was used to analyze the 52% of uNTx/Cr values below the limit of quantification in the context of a mixed-effects model. Age, weight, sex, race, and cancer type were evaluated as potential covariates for model parameters. Model-based simulations were undertaken to explore and predict the role of denosumab dose and dosing intervals on uNTx/Cr suppression. RESULTS: The typical value (between-subject variability; %) for uNTx/Cr at baseline was 49.2 nmol/L/mmol/L (76.8%), denosumab maximal uNTx/Cr suppression (efficacy) was 93.7% (127%), and the denosumab concentration providing half-maximal uNTx/Cr suppression (potency) was 31.8 ng/mL (287%). No effect of covariates on denosumab efficacy and potency was identified. Simulations indicated that a s.c. denosumab dose of 120 mg administered every 4 weeks provides more than 90% suppression of uNTx/Cr in the maximum proportion of patients relative to other every 4- and 12-week doses evaluated. CONCLUSIONS: Over the wide range of dosing regimens examined, a s.c. denosumab dose of 120 mg administered every 4 weeks is the optimal dosing regimen to suppress uNTx/Cr in patients with bone metastases from solid tumors. Clin Cancer Res; 18(9); 2648-57. ©2012 AACR.

Population pharmacokinetic meta-analysis of denosumab in healthy subjects and postmenopausal women with osteopenia or osteoporosis.

BACKGROUND AND OBJECTIVE: Inhibition of the receptor activator of nuclear factor κ-B ligand (RANKL) is a therapeutic target for treatment of bone disorders associated with increased bone resorption, such as osteoporosis. The objective of this analysis was to characterize the population pharmacokinetics of denosumab (AMG 162; Prolia®), a fully human IgG2 monoclonal antibody that binds to RANKL, in healthy subjects and postmenopausal women with osteopenia or osteoporosis. METHODS: A total of 22944 serum free denosumab concentrations from 495 healthy subjects and 1069 postmenopausal women with osteopenia or osteoporosis were pooled. Denosumab was administered as either a single intravenous dose (n = 36), a single subcutaneous dose (n = 469) or multiple subcutaneous doses (n = 1059), ranging from 0.01 to 3 mg/kg (or 6-210 mg as fixed mass dosages), every 3 or 6 months for up to 48 months. An open, two-compartment pharmacokinetic model with a quasi-steady-state approximation of the target-mediated drug disposition model was used to describe denosumab pharmacokinetics, using NONMEM Version 7.1.0 software. Subcutaneous absorption was characterized by the first-order absorption rate constant (k(a)), with constant absolute bioavailability over the range of doses that were evaluated. Clearance and volume of distribution parameters were scaled by body weight, using a power model. Model evaluation was performed through visual predictive checks. RESULTS: The subcutaneous bioavailability of denosumab was 64%, and the k(a) was 0.00883 h-1. The central volume of distribution and linear clearance were 2.49 L/66 kg and 3.06 mL/h/66 kg, respectively. The baseline RANKL level, quasi-steady-state constant and RANKL degradation rate were 614 ng/mL, 138 ng/mL and 0.00148 h-1, respectively. Between-subject variability in model parameters was moderate. A fixed dose of 60 mg provided RANKL inhibition similar to that achieved by equivalent body weight-based dosing. The effects of age and race on the area under the serum concentration-time curve of denosumab were less than 15% over the range of covariate values that were evaluated. CONCLUSIONS: The non-linearity in denosumab pharmacokinetics is probably due to RANKL binding, and denosumab dose adjustment based on the patient demographics is not warranted.

Population pharmacokinetic/pharmacodynamic modeling for the time course of tumor shrinkage by motesanib in thyroid cancer patients.

OBJECTIVE: To develop a population pharmacokinetic/pharmacodynamic model describing the relationship between motesanib exposure and tumor response in a phase 2 study of motesanib in patients with advanced differentiated thyroid cancer or medullary thyroid cancer. METHODS: Data from patients (n = 184) who received motesanib 125 mg once daily were used for population pharmacokinetic/pharmacodynamic modeling. Motesanib concentrations were fitted to a 2-compartment population pharmacokinetic model. Observed change in tumor size was the drug response measure for the pharmacodynamic model. Exposure measures in the pharmacokinetic/pharmacodynamic model included dose, plasma concentration profile, or steady-state area under the concentration versus time curve (AUC( ss )). A longitudinal exposure-tumor response model of drug effect on tumor growth dynamics was used. RESULTS: Motesanib oral clearance in patients with medullary thyroid cancer was 67% higher than in patients with differentiated thyroid cancer patients (73.7 vs. 44 L/h). Patients' disease type (medullary thyroid cancer vs. differentiated thyroid cancer) was the most important covariate for explaining interpatient variability in clearance. The objective response rates were 14 versus 2% for differentiated thyroid cancer and medullary thyroid cancer, respectively. Motesanib exposure measures (AUC( ss ) or concentration profile) were better predictors of tumor response than motesanib dose. The estimated motesanib concentration yielding tumor stasis (1.9 ng/mL) was lower than the observed trough concentrations in differentiated thyroid cancer and medullary thyroid cancer patients. CONCLUSIONS: Differences in motesanib pharmacokinetics likely explain the difference in tumor response observed between differentiated thyroid cancer and medullary thyroid cancer patients. The population pharmacokinetic/pharmacodynamic model provides a tool for predicting tumor response to the drug to support the dosing regimen of motesanib in thyroid cancer patients.

Population pharmacokinetic analysis of panitumumab in patients with advanced solid tumors.

Panitumumab is a fully human monoclonal antibody targeted to the extracellular domain of human epidermal growth factor receptor (EGFR). A comprehensive population pharmacokinetic model of panitumumab was developed using nonlinear mixed-effects modeling of 1200 patients with advanced solid tumors in 14 clinical studies. The disposition of panitumumab was best described with a 2-compartment model with parallel linear and nonlinear (Michaelis-Menten) elimination pathways. For a typical male patient with colorectal cancer (80 kg, 60 years old), the estimates for the linear clearance (CL), the maximum nonlinear clearance (V(max)/K(m)), the central volume of distribution (V(1)), the peripheral volume of distribution (V(2)), and the Michaelis-Menten constant (K(m)) are 0.273 L/d, 28.4 L/d, 3.95 L, 2.59 L, and 0.426 mcg/mL, respectively. Baseline covariates such as body weight, cancer type, age, sex, and race were studied for their influence on panitumumab pharmacokinetics. Body weight was found to be the most influential covariate on panitumumab exposure, affecting CL, V(max), and V(1). The administration of concomitant chemotherapy (IFL, FOLFIRI, or paclitaxel/carboplatin) or intensity of baseline tumor EGFR expression did not alter the pharmacokinetics of panitumumab. The presence of antipanitumumab antibodies in patients (immunogenicity rate 3.4%) did not appear to affect panitumumab exposure substantially (AUC difference 8%). In support of a new drug application in Japan, the model was used to assess panitumumab pharmacokinetics in Japanese patients compared to other racial groups; there were no significant differences in model-predicted steady-state panitumumab AUC, C(max), or C(min) after accounting for the effect of body weight.

Population pharmacokinetics of darbepoetin alfa in healthy subjects.

AIM: To develop and evaluate a population pharmacokinetic (PK) model of the long-acting erythropoiesis-stimulating protein, darbepoetin alfa in healthy subjects. METHODS: PK profiles were obtained from 140 healthy subjects receiving single intravenous and/or single or multiple subcutaneous doses of darbepoetin alfa (0.75-8.0 microg kg(-1), or either 80 or 500 microg). Data were analysed by a nonlinear mixed-effects modelling approach using NONMEM software. Influential covariates were identified by covariate analysis emphasizing parameter estimates and their confidence intervals, rather than stepwise hypothesis testing. The model was evaluated by comparing simulated profiles (obtained using the covariate model) to the observed profiles in a test dataset. RESULTS: The population PK model, including first-order absorption, two-compartment disposition and first-order elimination, provided a good description of data. Modelling indicated that for a 70-kg human, the observed nearly twofold disproportionate dose-exposure relationship at the 8.0 microg kg(-1)-dose relative to the 0.75 microg kg(-1)-dose may reflect changing relative bioavailability, which increased from approximately 48% at 0.75 microg kg(-1) to 78% at 8.0 microg kg(-1). The covariate analysis showed that increasing body weight may be related to increasing clearance and central compartment volume, and that the absorption rate constant decreased with increasing age. The full covariate model performed adequately in a fixed-effects prediction test against an external dataset. CONCLUSION: The developed population PK model describes the inter- and intraindividual variability in darbepoetin alfa PK. The model is a suitable tool for predicting the PK response of darbepoetin alfa using clinically untested dosing regimens.