Systems pharmacology modeling of drug-induced modulation of thyroid hormones in dogs and translation to human.

PURPOSE: To develop a systems pharmacology model based on hormone physiology and pharmacokinetic-pharmacodynamic concepts describing the impact of thyroperoxidase (TPO) inhibition on thyroid hormone homeostasis in the dog and to predict drug-induced changes in thyroid hormones in humans. METHODS: A population model was developed based on a simultaneous analysis of concentration-time data of T4, T3 and TSH in dogs following once daily oral dosing for up to 6-months of a myeloperoxidase inhibitor (MPO-IN1) with TPO inhibiting properties. The model consisted of linked turnover compartments for T4, T3 and TSH including a negative feedback from T4 on TSH concentrations. RESULTS: The model could well describe the concentration-time profiles of thyroid hormones in dog. Successful model validation was performed by predicting the hormone concentrations during 1-month administration of MPO-IN2 based on its in vitro dog TPO inhibition potency. Using human thyroid hormone turnover rates and TPO inhibitory potency, the human T4 and TSH concentrations upon MPO-IN1 treatment were predicted well. CONCLUSIONS: The model provides a scientific framework for the prediction of drug induced effects on plasma thyroid hormones concentrations in humans via TPO inhibition based on results obtained in in vitro and animal studies.

Investigation of absolute and relative response for three different liquid chromatography/tandem mass spectrometry systems; the impact of ionization and detection saturation.

RATIONALE: The investigations in this article were triggered by two observations in the laboratory; for some liquid chromatography/tandem mass spectrometry (LC/MS/MS) systems it was possible to obtain linear calibration curves for extreme concentration ranges and for some systems seemingly linear calibration curves gave good accuracy at low concentrations only when using a quadratic regression function. METHODS: The absolute and relative responses were tested for three different LC/MS/MS systems by injecting solutions of a model compound and a stable isotope labeled internal standard. The analyte concentration range for the solutions was 0.00391 to 500 µM (128,000×), giving overload of the chromatographic column at the highest concentrations. The stable isotope labeled internal standard concentration was 0.667 µM in all samples. RESULTS: The absolute response per concentration unit decreased rapidly as higher concentrations were injected. The relative response, the ratio for the analyte peak area to the internal standard peak area, per concentration unit was calculated. For system 1, the ionization process was found to limit the response and the relative response per concentration unit was constant. For systems 2 and 3, the ion detection process was the limiting factor resulting in decreasing relative response at increasing concentrations. CONCLUSIONS: For systems behaving like system 1, simple linear regression can be used for any concentration range while, for systems behaving like systems 2 and 3, non-linear regression is recommended for all concentration ranges. Another consequence is that the ionization capacity limited systems will be insensitive to matrix ion suppression when an ideal internal standard is used while the detection capacity limited systems are at risk of giving erroneous results at high concentrations if the matrix ion suppression varies for different samples in a run.

Comparison of the QT interval response during sinus and paced rhythm in conscious and anesthetized beagle dogs.

INTRODUCTION: The aim of the present study was to compare sensitivity in detecting the drug-induced QT interval prolongation in three dog models: conscious telemetered at sinus rhythm and conscious and anesthetized dogs during atrial pacing. The test substances used represent different chemical classes with different pharmacological and pharmacokinetic profiles. METHOD: Dofetilide and moxifloxacin were tested in all models, whereas cisapride and terfenadine were tested in the conscious telemetered and paced models. All substances were given as two consecutive 1.5-h intravenous infusions (infusions 1 and 2). The individual concentration-time courses of dofetilide, moxifloxacin, and cisapride were linked to the drug-induced effects on the QT interval and described with a pharmacokinetic-pharmacodynamic model to obtain an estimate of the unbound plasma concentrations at steady state that give a 10- and 20-ms drug-induced QT interval prolongation (CE10ms and CE20ms). RESULTS: In the conscious telemetered, conscious paced, and anesthetized dog models, the mean CE10ms values were 1.4, 4.0, and 2.5 nM for dofetilide and 1300, 1800, and 12,200 nM for moxifloxacin. For cisapride, the CE10ms values were 8.0 and 4.4 nM in the conscious telemetered and conscious paced dog models. The drug-induced QT interval prolongation during the last 30 min of infusions 1 and 2 was comparable in the conscious models, but smaller in the anesthetized dog model. Terfenadine displayed a marked delay in onset of response, which could only be detected by the extended ECG recording. DISCUSSION: All dog models investigated detected QT interval prolongation after administration of the investigated test substances with similar sensitivity, except for a lower sensitivity in the anesthetized dogs following moxifloxacin administration. The conscious telemetered dog model was favorable, mainly due to the extended continuous ECG recording, which facilitated detection and quantification of delayed temporal differences between systemic exposure and drug-induced QT interval prolongation.

A novel approach to data processing of the QT interval response in the conscious telemetered beagle dog.

INTRODUCTION: Drug-induced QT interval prolongation may lead to ventricular arrhythmias. The aim of the study was to optimize QT interval data processing to quantify drug-induced QT interval prolongation in the telemetry instrumented conscious dog model. METHODS: The test substances cisapride, dofetilide, haloperidol, and terfenadine and corresponding vehicles were given to male and female beagle dogs during two consecutive 90-min intravenous infusions. Cardiovascular parameters were recorded for 24 h and exposure to the drugs was measured. The delayed response in the QT interval after an abrupt change in heart rate was investigated. Eight mathematical models to describe the QT interval-heart rate relationship were compared and different sets of covariates were used to quantify the drug-induced effect on the QT interval. RESULTS: After an abrupt decrease in heart rate, a 75% adaptation of the QT interval was reached after 54+/-9 s. A linear model was preferred to correct the drug-induced effect on the QT interval for heart rate, vehicle effect, serial correlation, plasma concentration and time of day. All test substances significantly prolonged the QT interval. DISCUSSION: To optimize the processing of QT interval data, the delay in QT interval response after an abrupt change in heart rate should be considered. The QT interval-heart rate relationship and vehicle response were individual-specific and corrections were therefore made individually. When estimating the drug-induced effect on the QT interval it is considered advantageous to use plasma concentration as a covariate, as well as adjusting for vehicle effect and serial correlation in measurements. The conscious dog model detected significant increases in the QT interval for all test substances investigated.

Direct quantification in bioanalytical LC-MS/MS using internal calibration via analyte/stable isotope ratio.

The possibility to rationalize and simplify bioanalysis, without compromising the analytical quality, by omitting the calibration curves was studied. Using mass spectrometry (MS) and a stable isotope labeled internal standard it was possible to get equally good results by calculating the results directly from the analyte/internal standard area ratio and a predetermined response factor as by the traditional way, using a calibration curve run at the same occasion. To be able to use this simplified quantification method, that we call internal calibration, in its most simple form there are some prerequisites that must be considered: (1) The relative response should not be concentration dependent. (2) The relative response should be constant between batches/days. (3) The level of analyte in the internal standard should not be detectable. (4) There should be no influence from naturally occurring isotopes of the analyte on the internal standard peak area. A bioanalytical LC-MS/MS method for a research compound was validated both with and without calibration curves and no significant differences were found regarding precision and accuracy. It was shown that all four prerequisites above were fulfilled. Validation data were very good for the whole concentration range, 0.010-30 micromol/L. Long-term data for QC samples showed excellent precision and accuracy.

Pharmacokinetic-pharmacodynamic modeling of drug-induced effect on the QT interval in conscious telemetered dogs.

INTRODUCTION: To assure drug safety, the investigation of the relationship between plasma concentration and drug-induced prolongation of the QT interval of the ECG is a challenge in drug discovery. For this purpose, dofetilide was utilized to demonstrate the benefits of characterizing the complete time course of concentrations and effect in conscious beagle dogs in the assessment of drug safety. METHOD: On two separate occasions, four male and two female beagle dogs were given vehicle or the test substance, dofetilide (0.25 mumol/kg), over a 3-h intravenous infusion. Cardiovascular parameters, including QT intervals, were recorded for 24-h using radiotelemetry. The QT interval was corrected individually for heart rate, vehicle treatment, and serial correlation (QT(c)). Exposure (plasma concentration) to dofetilide was measured and described by a two-compartment model. The individual concentration-time course of dofetilide was linked to the QT(c) interval via an effect compartment and a pharmacodynamic E(max) model, to account for the observed hysteresis. RESULTS: Dofetilide induced a concentration-dependent increase in the QT(c) interval, with an EC(50) of 9 nM (3-30 nM, 95% C.I.) and an E(max) of 59+/-9 ms. A hysteresis loop was observed by plotting plasma concentrations vs. QT interval in time order, indicating a delay in onset of effect. It was found to have an equilibrium half-life of 11+/-8 min. Based on the parameters potency and E(max), a representation was made of the drug-induced changes to the QT interval. DISCUSSION: An effect compartment model was found to accurately mimic the QT interval prolongation following administration of the test substance, dofetilide. The assessment of the individual concentration-effect relationship and confounding factors such as hysteresis might provide a better prediction of the safety profiles of new drug candidates.

Studies of drug binding to plasma proteins using a variant of equilibrium dialysis.

The plasma protein binding of three model compounds was investigated using a variant of equilibrium dialysis, denoted comparative equilibrium dialysis (CED), and the results were compared with those obtained with ultrafiltration (UF). In CED, the buffer that the plasma is dialysed against in traditional equilibrium dialysis is replaced by, for example, plasma from other species. The CED method has the advantage that the unbound concentration (C(u)) does not need to be measured, which can be difficult for drugs with extremely small unbound fractions. Instead, the ratio of the total drug concentration (C(tot)) on either side of the dialysis membrane at equilibrium is a direct measure of the relative binding properties of the two plasma types. For the first model compound, having an unbound fraction (f(u)) of about 0.05% in human plasma, the time to reach equilibrium was too long (> or =40 h) to make the CED technique feasible in practice. For the second model compound, the more weakly bound drug NAD-299 (with an unbound fraction of about 2% in human plasma), the CED equilibration times were considerably shortened (< or =16 h), and the technique was applied to plasma from three different species. Large discrepancies between the CED and UF results were seen, CED always giving rise to much lower C(tot) differences than expected from the UF results. It is suspected that this discrepancy was due to equilibration between the dialysis chambers of all plasma components with a molecular weight less than the cut-off of the membrane. This equilibration causes altered binding properties compared to the initial plasma. When performing ultrafiltration on plasma where drug was added to untreated plasma or added to blank plasma that was equilibrated against plasma from the same or from another species, the change of binding properties was confirmed. To ensure that the results were not specific for NAD-299, a third model compound, tolterodine, was also included. The same trends as for NAD-299 were seen. Because of the long equilibration times for compounds with high protein binding and, in particular, the suspected partial mixture of low molecular weight compounds from the two plasma types and the subsequent change of binding properties, we cannot recommend the CED method as a tool for studying relative protein binding.

The bioanalytical challenge of determining unbound concentration and protein binding for drugs.

Knowledge regarding unbound concentrations is of vital importance when exploring the PK and PD of a drug. The accurate and reproducible determination of plasma protein binding and unbound concentrations for a compound/drug is a serious challenge for the bioanalytical laboratory. When the drug is in equilibrium with the binding protein(s), this equilibrium will shift when physiological conditions are not met. Furthermore, the true unbound fraction/concentration is unknown, and there are numerous publications in the scientific literature reporting and discussing data that have been produced without sufficient control of the parameters influencing the equilibrium. In this Review, different parameters affecting the equilibrium and analysis are discussed, together with suggestions on how to control these parameters in order to produce as trustworthy results for unbound concentrations/fractions as possible.

Capillary microsampling in the regulatory environment: validation and use of bioanalytical capillary microsampling methods.

Capillary microsampling (CMS) has recently been introduced as a response to the demands for more ethical use of laboratory animals according to the 3R principles. In CMS, an exact volume of the blood, plasma or other biofluid is collected in a capillary from which it is washed out, resulting in a diluted sample that can be handled using the existing equipment in the bioanalytical laboratory. CMS differs from traditional large volume sampling as the microsample is diluted before further handling and analysis, and reanalysis is performed using the diluted sample. This has some implications for the validation and this report is an attempt to clarify how to validate and use CMS methods in a regulatory environment. CMS also shows some distinct new opportunities: labile analytes can be immediately stabilized at sample collection and the addition of the internal standard to the whole sample can improve analytical performance. The experiences from 5 years use of CMS of plasma and blood for determination of drug exposure in animal studies are reviewed.

Validation of a bioanalytical method using capillary microsampling of 8 µl plasma samples: application to a toxicokinetic study in mice.

BACKGROUND: Capillary microsampling was recently introduced as a new technique for simplified collection and handling of small, exact volumes of liquid matrices. In this article, a bioanalytical method was developed and fully validated for 8 µl plasma samples and applied to a toxicology study in mice. RESULTS: The method was validated in the concentration range 0.06-30 µM. A procedure where 32 µl of blood was collected for preparation of 8 µl plasma was successfully implemented at the animal facility. All the results for the method and study validation met the requirements of a regulated assay. CONCLUSION: It is shown that 8 µl plasma microsamples can be sampled and analyzed with consistently excellent accuracy and precision. Capillary microsampling of plasma offers a possibility to combine ethical, scientific and economic benefits while still maintaining the advantage of having drug exposure data in plasma.

Capillary microsampling of 25 µl blood for the determination of toxicokinetic parameters in regulatory studies in animals.

BACKGROUND: Capillary microsampling (CMS) is a new technique for simplified collection, handling and analysis of small, exact volumes of liquid matrices. CMS was compared with conventional large volume sampling, in toxicology studies in rat and dog. RESULTS: Bioanalytical validation data were well within acceptance limits. Toxicokinetic (TK) parameters from microsampling were in agreement with data from conventional volume sampling. Clinical pathology parameters in rats measured 2 days after repeated microsampling were not affected when compared with rats not sampled. CONCLUSION: The fast collection and simple handling of small, exact volumes of liquid blood makes the CMS technique generic and flexible, as well as easily implemented and automated. Presented data support that TK measurements can be performed in main study rats, instead of dosing additional satellite animals only for TK sampling, giving both a higher scientific value and a substantial reduction of animal numbers in preclinical development.