[Show abstract][Hide abstract] ABSTRACT: Plasma ionized calcium (Ca(2+)) concentrations are tightly regulated in the body and maintained within a narrow range; thus it is challenging to quantify calcium absorption under normal physiologic conditions. This study aimed to develop a mechanistic model for the parathyroid hormone (PTH) response after calcium intake and indirectly compare the difference in oral calcium absorption from PTH responses. PTH and Ca(2+) concentrations were collected from 24 subjects from a clinical trial performed to evaluate the safety and calcium absorption of Geumjin Thermal Water in comparison with calcium carbonate tablets in healthy subjects. Indirect response models (NONMEM Ver. 7.2.0) were fitted to observed Ca(2+) and PTH data, respectively, in a manner that absorbed but unobserved Ca(2+) inhibits the secretion of PTH. Without notable changes in Ca(2+) levels, PTH responses were modeled and used as a marker for the extent of calcium absorption.
Korean Journal of Physiology and Pharmacology 06/2014; 18(3):217-23. · 1.26 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Moxifloxacin 400 mg is a widely used positive control in thorough QT (TQT) studies, but its QT-prolonging effects in Korean subjects have not been studied. The present study was conducted to collect pilot data in Korean subjects after moxifloxacin administration to evaluate the adequacy of moxifloxacin as a positive control.
Thirty-eight, healthy, Korean, male subjects were recruited for pharmacokinetic (PK) blood sampling and electrocardiography (ECG) recordings at three different study sites. On day 1, a baseline 12-lead ECG was recorded, and on day 2, ECG recordings were conducted after placebo, or moxifloxacin 400- or 800-mg administration. Baseline-corrected, placebo-adjusted, corrected QT (ΔΔQTc) values were calculated. Blood samples were collected after moxifloxacin administration and PK parameters were assessed.
A total of 33 subjects completed the study. The largest time-matched ΔΔQTc occurred approximately 4 h after dosing, with ΔΔQTcI (QT interval corrected by individual QT-RR regression model) values of 11.66 ms (moxifloxacin 400 mg) and 20.96 ms (800 mg). The mean and 90 % confidence intervals of ΔΔQTcI did not include zero at any of the measurement time points. There was a positive correlation between plasma moxifloxacin concentration and ΔΔQTcI (r = 0.422). Dose-proportional PK profiles were observed.
Moxifloxacin 400 mg is an adequate positive control in Korean TQT studies. Our results indicate that moxifloxacin 400 mg can be used to evaluate the cardiac safety of a drug in Korean subjects.
[Show abstract][Hide abstract] ABSTRACT: SKL10406, triple monoamine reuptake inhibitor, is a novel antidepressant candidate. A PET study was performed to investigate the occupancies of serotonin and dopamine transporters (SERT and DAT) in human brain, and the relationship between SKL10406 concentration and SERT occupancy was assessed using pharmacokinetic-pharmacodynamic (PK-PD) modeling methods. Fifteen healthy volunteers were given SKL10406 100 mg/day for 6 days or 150 mg/day for 6 days after 100 mg/day for 4 days. Each subject underwent full PK sampling for SKL10406 and PET scans at predose, 4 h and 16 h after dosing at a steady state to investigate the occupancies of SERT and DAT using 11C-DASB and 11C-PE2I, respectively. Naïve pooled method (NPM) and nonlinear mixed-effect methods (ME) including a direct ME (DME) and an effect compartmental ME (EME) were used (NONMEM Ver. 7.2). Six and five subjects completed the studies for SERT and DAT, respectively. The final estimates of Emax (53.4%) and EC50 (11.8 ng/mL) from DME were relatively lower than those from NPM (Emax, 74.1%; EC50, 36.8 ng/mL) and EME (Emax, 68.6%; EC50, 40.2 ng/mL). DAT occupancy results were not modeled because of lower occupancies. The results showed that the dosage regimens may be applied in patient studies. However, difference between estimation methods alerts that ME may not be a recommendable analysis tool for sparsely sampled PET scan data.
Translational and Clinical Pharmacology. 01/2014; 22(2):83.
[Show abstract][Hide abstract] ABSTRACT: This study aimed to evaluate the change in the pharmacokinetics (PK) of cyclosporine in the non-steady-state period in the first week after renal transplantation; the factors influencing this change, including genetic variability; and the time point concentration that correlated best with drug exposure. Data were obtained from 69 patients, and PK studies were conducted on postoperative days (PODs) 2, 3, and 7. Samples were taken pre-dose and at 1, 2, 3, 4, 6, 8, and 12 hours after drug administration. MDR1, CYP3A4, and CYP3A5 were genotyped. A population PK analysis and correlational analysis between the concentration at each time point and the area under the time-concentration curve were performed. A two-compartment model with first-order absorption was chosen. The rate and extent of drug absorption showed a significant increase on POD3, followed by a slight decrease on POD7. Until POD3, 8 hours post-dose was the single time point concentration that correlated best with drug exposure and 3 hours was the best time point on POD7. In both analyses, the MDR1 genotype showed potential as a factor influencing PK change. We conclude that oral administration of cyclosporine and dose adjustment based on a single concentration measurement might result in unexpected drug exposure during this early posttransplantation period.
Drug Design, Development and Therapy 01/2014; 8:2241-9. · 3.03 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In this study, we developed a pharmacokinetic (PK)- pharmacodynamic (PD) model of a new sustained release formulation of interferon-alpha-2a (SR-IFN-alpha) using the blood concentration of IFN-alpha and neopterin in order to quantify the magnitude and saturation of neopterin production over time in healthy volunteers. The SR-IFN-alpha in this study is a solid microparticular formulation manufactured by spray drying of a feeding solution containing IFN-alpha, a biocompatible polymer (polyethylene glycol) and sodium hyaluronate.
The full PK and PD (neopterin concentration) datasets from 24 healthy subjects obtained after single doses of 9, 18, 27 and 36 MIU of subcutaneous SR-IFN-alpha were used to build the mixed-effect model using NONMEM (version 7.2) with the GFORTRAN compiler.
A one-compartment model with first-order elimination and a mixture of zero- and first-order absorption was chosen to describe the PK of SR-IFN-alpha. The time-concentration profile of neopterin, the PD marker, was described by a turnover model combined with a single transit compartment. The saturable pattern of the neopterin response blurring the dose--response relationship of SR-IFN-alpha was addressed by introducing the concept of the EC50 increasing over time.
The PK-PD model of SR-IFN-alpha developed in this study has presented a quantitative tool to assess the time-course of a saturable neopterin response in humans.
Journal of Translational Medicine 10/2013; 11(1):240. · 3.99 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Background The aim of this study was to assess the efficacy and safety of combination regimen of capecitabine plus everolimus in patients with refractory gastric cancer who have failed to at least two cytotoxic regimens. Methods Patients received capecitabine 650 mg/m(2) twice daily (D1-14) and everolimus 5 mg twice daily (D1-21) every 3 weeks until disease progression or unacceptable toxicity. The primary endpoint of the study was overall response (partial or complete response) and the secondary endpoints were progression-free survival (time between registration and disease progression or death) and overall survival. Pharmacokinetic analysis was also performed. Patients who have failed to at least two cytotoxic regimens were enrolled. Results Between March 2010 and June 2012, 47 patients were enrolled. 33 patients (70.2 %) had received more than three previous regimens prior to enrolment. Among 43 evaluable patients for treatment response, 5 patients achieved confirmed partial response and 18 patients showed stable disease, resulting in an overall response rate (ORR) of 10.6 % (95 % C.I.: 1.8-19.4 %) and disease control rate of 48.9 % (95 % C.I.:34.6-63.2 %). At a median follow-up of 106 weeks (range, 21-141 weeks), the median progression-free survival and overall survival were 11.0 weeks (95 % C.I.: 5.7-16.3 weeks) and 21.0 weeks (95 % C.I.: 14.3-27.7 weeks), respectively. Grade 3 nausea, diarrhea and stomatitis occurred in two, three and three patients, respectively. Elevated liver enzyme was observed in 21 patients and no patient had pulmonary fibrosis. Conclusions The combination of capecitabine 650 mg/m(2) twice daily and everolimus 5 mg twice daily was found to be effective in a small subset of GC patients who were heavily pre-treated.
Investigational New Drugs 09/2013; 31(6). · 3.50 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Background We performed a phase I study to determine the dose and safety of everolimus as a combination chemotherapy in peripheral T-cell lymphoma (PTCL). Methods Four dose levels (2.5 to 10 mg) of everolimus from days 1 to 14 with CHOP (750 mg/m2 cyclophosphamide, 50 mg/m(2) doxorubicin, and 1.4 mg/m(2) (maximum 2 mg) vincristine on day 1, and 100 mg/day prednisone on days 1 to 5) every 21 days were planned. Results Fifteen patients newly diagnosed with stage III/IV PTCL were enrolled. One of 6 patients at dose level 2 (5 mg everolimus) had grade 3 hepatotoxicity and 3 of 6 patients at level 3 (7.5 mg everolimus) had grade 4 hematologic toxicities (two grade 4 thrombocytopenia and one grade 4 neutropenia with fever lasting more than 3 days). The recommended dose of everolimus for combination was 5 mg. There were no differences in steady state trough concentrations of everolimus between cycles 1 and 2 for all three dose levels. All evaluable patients achieved response (8 complete and 6 partial). Conclusions Everolimus (5 mg) can be safely combined with CHOP leading to a feasible and effective regimen for PTCL. The subsequent phase II is now in progress.
Investigational New Drugs 08/2013; · 3.50 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Colistin is increasingly used as a salvage therapy of nosocomial infections caused by multidrug-resistant gram-negative bacteria such as Pseudomonas aeruginosa and Acinetobacter baumannii. However, the available pharmacokinetic (PK) data of colistin are limited to guide dosing. The aim of this study was to develop a population PK model of colistin and to identify the optimal dosage regimens for burn patients.Fifty patients with burns ranging from 4% to 85% of total body surface area treated with colistimethate sodium (CMS) were studied. CMS which is hydrolyzed in vivo to an active metabolite was intravenously administered every 12 h. Blood samples were collected at 0, 1, 2, 4, 6, and 8 h after more than five infusions to measure the colistin concentration using a LC/MS/MS system. The population PK model was developed using nonlinear mixed effect modeling (NONMEM, ver. 6.2).A one-compartment linear PK model for colistin best described the data. The covariates included in the final model were creatinine clearance on the relative fraction of CMS converted into colistin and the presence of edema on the turnover rate constant of CMS converted into colistin. Steady-state 24 h the area under the concentration-time curve was simulated from 1,000 virtual patients receiving 150 mg colistin base activity every 12 h using the final model. Unlike previous reports in critically ill patients, the elimination half-life of colistin (6.6 h) was much shorter and continuous renal replacement therapy was not a significant covariate for any PK parameters.
Antimicrobial Agents and Chemotherapy 02/2013; · 4.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The pharmacokinetic (PK) property of fluconazole might be significantly altered in major burn patients by medical interventions and physiologic changes. In this study, we were to investigate fluconazole PK in burn patients using population approach and to recommend the optimal fluconazole regimen based upon the predicted therapeutic outcome. At steady-state, blood samples for PK analysis were obtained from 60 burn patients receiving daily 100 ∼ 400 mg fluconazole. A mixed-effect modeling was performed and the therapeutic outcome of antifungal therapy was predicted from 10,000 virtual patients using NONMEM (Ver.7.2). Minimum inhibitory concentration (MIC) values were sampled from the MIC distribution at the study site. fAUC/MIC > 25 hr was used as the criterion for therapeutic success. When the same dose was given, plasma concentration of fluconazole was predicted to be lower in burn patients compared to non-burn population because of the large PK parameter (Clearance, Volume of Distribution) estimates and continuous renal replacement therapy (CRRT). This tendency was particularly predominant when the patients were within 30 post-burn days. Thus, conventional fluconazole regimen might not guarantee a successful treatment against Candida spp. in burn patients even when the strain was susceptible. Based upon our findings, 400 mg/day fluconazole is recommended to obtain therapeutic successes in major burn patients.
Antimicrobial Agents and Chemotherapy 12/2012; · 4.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: We analyzed the pharmacokinetics of C3G on data from twelve subjects, after 2-week multiple dosing of black bean (Phaseolus vulgaris, Cheongjakong-3-ho) seed coat extract, using the mixed effect analysis method (NONMEM, Ver. 6.2), as well as the conventional non-compartmental method. We also examined the safety and tolerability. The PK analysis used plasma concentrations of the C3G on day 1 and 14. There was no observed accumulation of C3G after 2-week multiple dosing of black bean seed coat extract. The typical point estimates of PK were CL (clearance)=3,420 l/h, V (volume)=7,280 L, Ka (absorption constant)=9.94 h(-1), ALAG (lag time)=0.217 h. The black bean seed coat extract was well tolerated and there were no serious adverse events. In this study, we confirmed that a significant amount of C3G was absorbed in human after given the black bean seed coat extract.
Korean Journal of Physiology and Pharmacology 08/2012; 16(4):249-53. · 1.26 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: PURPOSE: Fimasartan is a non-peptide angiotensin II receptor antagonist which selectively blocks the AT(1) receptor. The aim of our study was to perform a population pharmacokinetic-pharmacodynamic (PK-PD) analysis of fimasartan to evaluate the effect of food on the mechanistic PK-PD relationship. METHODS: This was a food-drug interaction single-center study involving 24 healthy subjects that was designed as a randomized, open-label, single-dosing, two-way crossover trial. Extensive PK data was obtained on blood samples collected at 0, 0.5, 1, 2, 2.5, 3, 4, 6, 8, 12, and 24 h post-dosing and five systolic/diastolic blood pressure (BP) measurements made at 0, 4, 8, 12 and 24 h post-dosing and used to construct a mixed effect model (NONMEM, ver. 6.2). RESULTS: A two-compartment linear PK model with zero-order (fasted) or Weibull (fed with high-fat diet) absorption best described the PK of fimasartan. Relative bioavailability decreased by 37 % when the subjects were given a high-fat diet. CONCLUSIONS: The turnover PK-PD model combined with pre-defined cosine function for circadian rhythm described the BP changes measured within 24 h after dosing better than the effect compartment or transduction models. To predict the influence of a high-fat diet on the blood pressure-lowering effect of fimasartan in healthy subjects, we simulated changes in BP when fimasartan was given daily for 30 days. The overlapping pattern of simulated BP curves in the fasted versus fed group demonstrated that a high-fat diet would not cause a clinically significant reduction in the BP-lowering effect of fimasartan, despite a significant reduction in bioavailability.
European Journal of Clinical Pharmacology 06/2012; 69(1). · 2.70 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The aim of this study was to develop a non-linear mixed effect circadian rhythm model of acetylcholinesterase (AChE) activity variation and to evaluate the inhibitory effect of acorn extract (2 g) and galantamine (16 mg), used as positive control, on human AChE in red blood cells (RBC).
This was an open-label, randomized, three-way crossover study involving 12 healthy subjects who received one of the treatments in each study period: no treatment, acorn extract, and galantamine. RBC AChE activity was measured in peripheral blood samples collected at 0 (pre-dose), 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 16 and 24 h post-dose administration. Non-linear mixed effect modeling was performed using NONMEM (ver. 7.0).
The circadian variation of AChE activity was best described using two mixed effect cosine functions, with periods of 24 and 12 h, respectively. When the inhibitory effect terms were added, the model was significantly improved for both acorn extract and galantamine. In terms of the effect, a 2-g single dose of acorn extract showed AChE inhibition (about 5%) similar to that of a 16-mg single dose of galantamine, in the first 24 h after administration.
Based on the very pronounced inter- and intra-day variation in AChE activity in RBC, we conclude that the model-based approach is essential for the proof of concept and quantitation of AChE inhibition in human subjects.
European Journal of Clinical Pharmacology 12/2011; 68(5):599-605. · 2.70 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: To develop a limited sampling strategy (LSS) to predict area under the concentration-time curve (AUC) ratios of omeprazole (AUC(OPZ)) to its metabolites 5-hydroxyomeprazole (AUC(5OH)) and omeprazole sulfone (AUC(SUL)) as phenotyping parameters for cytochrome P450 (CYP) 2C19 and 3A.
Data were obtained from 37 (4 women) Caucasian, Chinese, and Korean healthy adults from three published studies. The AUC(OPZ), AUC(5OH), and AUC(SUL) were calculated via noncompartmental analysis. Observed AUC(OPZ, OBS)/AUC(5OH, OBS) and AUC(OPZ, OBS)/AUC(SUL, OBS) were determined. Plasma concentrations of omeprazole, 5-hydroxyomeprazole, and omeprazole sulfone at 1, 1.5, 2, 3, 4, 6, and 8 h post-dose were used to generate limited sampling strategy (LSS) models to predict AUC(OPZ,PRE)/AUC(5OH,PRE) and AUC(OPZ,PRE/)AUC(SUL,PRE). Bias and precision were assessed via percentage mean prediction error (%MPE) and percentage mean absolute error (%MAE), with acceptable limits being <15%.
For CYP2C19, the AUC(OPZ,OBS)/AUC(5OH,OBS) was [mean ± standard deviation (SD)] 2.10 ± 1.63. Five LSS models of AUC(OPZ,PRE)/AUC(5OH,PRE) were generated, but none met the bias or precision criteria. Upon stratification by CYP2C19 genotype and ethnicity, a three-timepoint (at 1, 2, and 4 h) LSS model accurately predicted AUC(OPZ)/AUC(5OH) in Caucasian CYP2C19*1/*1 subjects. For CYP3A, AUC(OPZ,OBS)/AUC(SUL,OBS) (mean ± SD) was 1.79 ± 0.67. All LSS models had unacceptable %MAE, even when stratified by CYP2C19 genotype and ethnicity.
A LSS model to predict AUC(OPZ)/AUC(5OH), and thus CYP2C19 activity, was generated for Caucasian CYP2C19*1/*1 subjects. However, additional model validation is needed prior to general use. LSS models to predict AUC(OPZ)/AUC(SUL), and thus CYP3A activity, were not possible, even upon stratification by CYP2C19 genotype and ethnicity.
European Journal of Clinical Pharmacology 10/2011; 68(4):407-13. · 2.70 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Voriconazole is a triazole agent with excellent antifungal activity against Aspergillus species. However, despite its potential advantages, the occurrence of unpredictable toxicities might be critical in immunocompromised patients. The aim of this study was to analyze risk factors for voriconazole-related severe adverse events (SAEs).
This prospective observational study was conducted in Korean patients with hematological malignancies and invasive aspergillosis on intravenous voriconazole therapy between June 2008 and April 2009.
Of the 25 patients enrolled, eight (32%) showed voriconazole-related SAEs, which included hepatotoxicities (n=5), cardiac tachyarrhythmias (n=2), and neurotoxicity (n=1). Sex, age, underlying hematological malignancies, voriconazole dose, the co-administration of a proton pump inhibitor, and CYP2C19 genotype were not found to be related to the occurrence of SAEs. However, trough plasma concentrations of voriconazole were found to be significantly higher in the patients with an SAE: median 6.32 mg/l (interquartile range (IQR) 2.86-9.71 mg/l) vs. median 2.15 mg/l (IQR 0.92-4.00 mg/l); p=0.011. Receiver operating characteristic curve analysis identified a cut-off trough concentration for SAEs of 5.83 mg/l (sensitivity 62.5% and specificity 94.1%). Furthermore, multivariate analysis showed that a trough concentration of ≥ 5.83mg/l was the only significant independent risk factor of an SAE.
This study shows that therapeutic drug monitoring is indicated in patients with a voriconazole-related SAE and that dose adjustment is required if the trough concentration of voriconazole exceeds 5.83 mg/l.
International journal of infectious diseases: IJID: official publication of the International Society for Infectious Diseases 08/2011; 15(11):e753-8. · 2.17 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The present study was conducted to determine and compare the target attainment rate (TAR) between microorganism-nonspecific (C(trough)) and microorganism- specific (AUC24/MIC) targets over two weeks of teicoplanin administration according to several dose regimens for the treatment of Staphylococcus aureus in Korean patients with neutropenic fever.
One thousand virtual concentrations were obtained for each dose using the population pharmacokinetic parameters of teicoplanin adopted from a published study. Simulation of 1,000 virtual MICs was performed using the MICs of 78 clinical isolates of S. aureus collected from a hospital in Korea. Thereafter, these simulated MICs were randomly allocated to 1,000 virtual patients in whom the TARs for AUC24/MIC>125 [or 345] and C(trough)>10 [or 20] mg/L were determined. The relationship of the maintenance dose with the steady-state TAR was predicted with respect to the AUC24/MIC>125 [or 345] using logistic analysis.
The standard dose regimen of teicoplanin showed TARs of about 70% [or 33%] and 70% [or 20%] at steady-state in cases with AUC24/MIC>125 [or 345] and C(trough)>10 [or 20] mg/L, respectively.
The current standard dose regimen was predicted to be insufficient to adequately treat S. aureus in Korean patients with neutropenic fever. To assure at least an 80% TAR in this population, dose adjustment of teicoplanin should be considered.
Yonsei medical journal 07/2011; 52(4):616-23. · 0.77 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Sorafenib is a multi-kinase inhibitor, which was approved as first-line treatment for patients with advanced hepatocellular carcinoma (HCC). We conducted a phase 1 study of sorafenib plus S-1 in patients with advanced HCC.
We designed to escalate S-1 at 4 different dose levels with fixed dose of sorafenib. Four dose levels were as follows: level 1, D1-14 S-1 50 mg/m(2)/day + D1-21 sorafenib 400 mg bid; level 2, D1-14 S-1 60 mg/m(2)/day + D1-21 sorafenib 400 mg bid; level 3,, D1-14 S-1 70 mg/m(2)/day + D1-21 sorafenib 400 mg bid; level 4, D1-14 S-1 80 mg/m(2)/day + D1-21 sorafenib 400 mg bid. The treatment was repeated every 3 weeks.
From August 2009 to July 2010, 20 patients with advanced HCC were enrolled. The median age was 48 years (range, 29-74). Eighteen (90%) patients had hepatitis B viral infection and 19 (95%) patients were rated as Child-Pugh class A. The dose-limiting toxicities were grade 4 infection and thrombocytopenia. After a median follow-up duration of 8.6 months (range, 3.7-14.2 months), median PFS was 3.9 months (95% CI, 0.8-7.0 months) and median OS was 10.4 months (95% CI, 0-22.4 months). In pharmacokinetic analysis, there was no statistically significant drug interaction between sorafenib and S-1.
The combination of sorafenib and S-1 showed tolerable toxicity profile and modest clinical efficacy in patients with advanced HCC. The recommended dose of sorafenib and S-1 was 400 mg twice daily and 40 mg/m(2) twice daily, respectively.
Investigational New Drugs 06/2011; 30(4):1540-7. · 3.50 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Population pharmacokinetic (PK) and pharmacodynamic (PD) modeling of clopidogrel was developed from pooled data from healthy volunteers (n = 44) and stroke patients (n = 35). The PK modeling used plasma concentrations of the clopidogrel metabolite (SR26334), and the PD modeling used platelet aggregation. The models were developed using NONMEM and evaluated via visual predictive check (VPC). Data were analyzed by 2-compartment modeling with Erlang's absorption and first-order elimination. There was no statistically significant covariate for each model parameter. The typical point estimates of PK were k(tr) (identical transfer rate constant) = 5.97 h(-1), k(e) (elimination rate constant) = 0.126 h(-1), k(d) (distribution rate constant) = 0.212 h(-1), V(2) (volume of central compartment) = 21.0 L, and V(3) (volume of peripheral compartment) = 38.8 L. The typical point estimates of PD were k(in) (input rate) = 27.9 h(-1), E(max) (maximum effect on input rate) = 0.292 h(-1), EC(5) (0) (median effective concentration) = 0.00629 ng/mL, and BASE (predose aggregation) = 66.7%. The final model was used to estimate individual parameters using patient data and showed good predictions using VPC.
The Journal of Clinical Pharmacology 05/2011; 52(7):985-95. · 2.47 Impact Factor