KR101753561B1 - Model for prediction of QT prolongation by drugs - Google Patents

Abstract

The present invention relates to a pharmacokinetic / pharmacodynamic model for the evaluation of drug-induced QT prolongation syndrome in the development of a new drug, and has been used as a limitation of the pharmacokinetic / pharmacodynamic model for predicting the QT prolongation syndrome, Induced QT prolongation syndrome by eliminating what could not distinguish the QTc interval change caused by the drug-induced QT prolongation syndrome. Therefore, the pharmacokinetic / pharmacodynamic model for the prediction of QT syndrome according to the present invention is expected to more accurately predict the risk of drug-induced arrhythmia in the development of new drugs, thereby enhancing the stability of the drug.

Description

The QT prolongation syndrome (QT prolongation syndrome)

The present invention relates to a pharmacokinetic / pharmacodynamic model for the evaluation of drug-induced QT prolongation syndrome in the course of drug development.

The second-generation antihistamine was developed to overcome sedation, a side effect that can occur in first-generation antihistamines. However, in many patients severe cardiac anomalies have been found to be caused by noncardiovascular drugs, including second-generation antihistamines (Crumb W. et al., 1999).

Astemizole is a second-generation antihistamine and has a long duration and was developed in 1997 (Laduron P. M. et al., 1982). This drug is metabolized primarily by CYP2D6 as desmethyl astemizole. However, astemizole has disappeared from the market because astemizole and its active metabolite desmethylastimizole have been shown to cause arrhythmias due to QT prolongation. Astemizole has been used predominantly as a positive control in cardiovascular stability testing at preclinical stage (Valerio L. G., et al., 2013).

Reports of ventricular arrhythmias associated with sedating antihistamines were increased in 1991 and the sudden death of patients taking these drugs was associated with delayed ventricular repolarization and torsades de pointes, (Fermini B. et al., 2003), which is associated with QT prolongation syndrome.

The ICH Guidelines recommend in vitro and in vivo studies to assess the risk of arrhythmia. In vitro studies assess IC ₅₀ values obtained from hERG assays that measure the inhibitory activity of hERG (human ether-a-go-go-related gene), which indicates the activity of the potassium channel of the heart (Sanguinetti MC et al., 1995). Inhibition of the hERG channel by drugs inhibits cardiac repolarization by the transport of potassium ions and leads to an initial depolarization state with calcium ions (January CT et al., 1991 ). The in vivo study predicts the QT prolongation syndrome by measuring the QT interval (QT interval) of animals after drug injection and comparing and evaluating the results with the positive control group (mainly astemizole). The QT interval represents the depolarization and repolarization of the left ventricle and the ventricle, and is influenced by various physiological factors, especially heart rate.

The stirring for modeling the change in QTc (corrected QT) interval by the drug / pharmacodynamic model was found, these models have been described by the change in QTc interval with the S-shaped E _max model (sigmoidal E _max model). However, this model can explain drug-induced QT prolongation syndrome, but the change in QTc interval induced by circadian rhythm can not be distinguished from the change in QTc interval induced by the drug, There is a possibility to underestimate or overestimate the effect.

The circadian rhythm refers to the endogenous biological action repeated 24 hours a cycle, also called a biological clock or biorhythm. The activity cycle is affected by light and temperature (Cagnacci A. et al., 1992; Wever RA et al., 1983), and heart rate and electrocardiogram (ECG) et al., 2000; Nakagawa M. et al., 1998).

The harmonic model uses a fourier function and is developed to explain the change of activity cycle (Krzyzanski W. et al., 2000). The Fourier function uses cosine and sine functions to describe the periodic phenomena (Chakraborty A. et al., 1999). However, there is no case in which an active daily cycle model is applied to predict QT syndrome.

Therefore, the present inventors have developed a new pharmacokinetic / pharmacodynamic model which is constructed by combining a daily activity cycle using a harmonic model in a pharmacokinetic / pharmacodynamic model to increase the prediction accuracy of a drug-induced QT syndrome, Induced QT prolongation syndrome by separating changes in the QTc interval due to the effect of the present invention.

In the meantime, US Patent Application Publication No. 2009-0326401 discloses a method for diagnosing cardiac function disorder risk by using electrocardiogram data showing electrocardiographic changes due to treatment of a compound. In Korean Patent Laid-Open Publication No. 2013-0091574, Fragility prediction model is disclosed, but it differs in its composition from the QT prolongation syndrome prediction model through the pharmacokinetic / pharmacodynamic model and the activity period information of the present invention. Korean Patent Laid-Open No. 2009-0041817 discloses a technique for discriminating a patient's QT prolongation syndrome based on a patient's electrocardiogram data. However, the present invention is not limited to the diagnosis of QT prolongation syndrome according to the present invention, And the configuration is different.

US Patent Application Publication No. 2009-0326401, Method and system for dynamical systems modeling of electrocardiogram data, December 31, 2009. Korean Patent Laid-Open Publication No. 2013-0091574, an apparatus and method for generating atrial fibrillation prediction model, and an apparatus and method for predicting atrial fibrillation. Korean Patent Publication No. 2009-0041817, Cardiovascular Adverse Reaction Classification System, 2009. 04. 29. Disclosure.

Cagnacci A. et al., Melatonin: a major regulator of the circadian rhythm of core temperature in humans, J. Clin. Endocrinol. Metab., 75, 447-452,1992. Chakraborty A. et al., Mathematical modeling of circadian cortisol concentration using indirect response models: comparison of several methods, J. Pharmacokinet. Biopharm., 27, 23-43, 1999. Crumb W. et al., QT interval prolongation by non-cardiovascular drugs: issues and solutions for novel drug development, Pharm. Sci. Technol. Today, 2, 270-280, 1999. Fermini B. et al., The impact of drug-induced QT interval prolongation on drug discovery and development, Nat. Rev. Drug Discov., 2, 439-447, 2003. January C.T. et al., Triggered activity in the heart: cellular mechanisms of early after-depolarizations, Eur Heart J., 12, 4-9, 1991. Krzyzanski W. et al., Algorithm for application of Fourier analysis for biorhythmic baselines of pharmacodynamics indirect response models, Chronobiol. Int., 17, 77-93, 2000. Laduron P.M. et al., In vitro and in vivo binding characteristics of a novel long-acting histamine H1 antagonist, astemizole, Mol. Pharmacol., 21, 294-300, 1982. Massian M.M. et al., Circadian rhythm of heart rate and heart rate variability, Arch. Dis. Child, 888, 179-182, 2000. Nakagawa M. et al., Circadian rhythm of the signal averaged electrocardiogram and its relation to heart rate variability in healthy subjects, Heart, 79, 493-496, 1998. Valerio L.G. et al., Development of cardiac safety translational tools for QT prolongation and torsade de pointes, Expert Opin. Drug Metab. Toxicol., 9, 801-815, 2013. Sanguinetti M.C., et al. A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel, Cell, 81, 299-307, 1995. Wever R.A. et al., Bright light affects human circadian rhythms, Pflugers Arch., 396, 85-87, 1983.

It is an object of the present invention to provide a pharmacokinetic / pharmacodynamic model capable of evaluating drug-induced QT prolongation syndrome during drug development.

The present invention relates to a pharmacokinetic / pharmacodynamic model of a drug for predicting the QT prolongation syndrome expressed by the following formula.

The pharmacokinetic / pharmacodynamic model of the drug is a model in which an activity cycle is applied to a pharmacokinetic (pharmacodynamic) model. It is divided into QTc interval change by drug and QTc interval change by activity cycle, Extension syndrome can be detected.

The medicament may comprise a drug capable of causing QT prolongation syndrome.

The present invention also relates to a method for predicting QT prolongation syndrome using the pharmacokinetic / pharmacodynamic model of the above drug.

Hereinafter, the present invention will be described in detail.

The present invention relates to a pharmacokinetic / pharmacodynamic model of a drug for predicting QT prolongation syndrome expressed by the following equation,

In the equation T ₁ and T ₃ denotes the QTc amplification caused by the action-day cycle _{and, (time-T 2) /} 24 , and (time-T ₄₎ / 24 denotes a cycle period of one activity, QT ₀ is The baseline QTc value is the QTc value in the steady state, X _j (j = 4, 5) is the drug concentration of each living body compartment, and EC ₅₀ is the drug concentration when the drug effect is halved.

In addition, the pharmacokinetic / pharmacodynamic model of the drug can be used to underestimate or overestimate the drug-induced QT prolongation syndrome induced effect by separating changes in the QTc interval by the drug and the activity cycle by applying an activity cycle to the drug power model Can be prevented.

Circadian rhythm refers to the endogenous biological action that is repeated every 24 hours. It is also referred to as a biological clock or biorhythm, and affects heart rate and electrocardiogram.

A basic harmonic model can be modified and applied to apply the activity period to the model. The harmonic model is developed to explain the change of the activity period by using the fourier function. It uses the cosine and sine functions.

As the above-mentioned drugs capable of causing the QT extension syndrome, antihistamines astemizole, terfenadine, azelastine, cinnarizine, dimenhydrinate, fexofenadine, For example, diphenhydramine, cyproheptadine, hydroxyzine, ebastine, loratadine and doxepine. In the present invention,

In addition, the method for predicting the QT extension syndrome may include: measuring a pulse rate and an electrocardiogram of a patient before and after administration of the drug; Measuring the blood drug concentration with time of the patient to which the drug is administered; Inputting the measured electrocardiogram and administration information of the administered drug; Performing modeling through measurements of the blood drug concentration; And determining a QTc interval change by applying the inputted ECG, blood drug concentration, and drug administration information to a pharmacokinetic / pharmacodynamic model constructed.

The present invention relates to a pharmacokinetic / pharmacodynamic model for the evaluation of drug-induced QT prolongation syndrome in the development of a new drug, and has been used as a limitation of the pharmacokinetic / pharmacodynamic model for predicting the QT prolongation syndrome, Induced QT prolongation syndrome by eliminating what could not distinguish the QTc interval change caused by the drug-induced QT prolongation syndrome. Therefore, the pharmacokinetic / pharmacodynamic model for the prediction of QT syndrome according to the present invention is expected to more accurately predict the risk of drug-induced arrhythmia in the development of new drugs, thereby enhancing the stability of the drug.

Figure 1 shows changes in the QTc interval and the change in QTc interval due to drug administration during a daily cycle in a normal state without drug administration.
Fig. 2 shows the changes in plasma concentrations of astemizole (2A) and desmethylastimizole (2B) after administration of astemizole.
Figure 3 shows a paddock / pharmacodynamic model for prediction of the QT extension syndrome of the present invention.
FIG. 4 shows the accuracy of the result of predicting pharmacokinetics with the model constructed in the present invention through visual prediction test results. 4A is the concentration of astemx in plasma after administration of 10 mg / kg of astemizole, 4B is the concentration of desmethylasteminol in plasma after administration of 10 mg / kg of astemizole, 4C is astemizole of 30 mg / 4D is the concentration of desmethylasteminol in the plasma after administration of 30 mg / kg of astemizole.
FIG. 5 shows a visual prediction test result of a model for explaining the change of the QTc interval according to the activity period in the normal state without administration of the drug.
FIG. 6 shows the result of the visual prediction test as a result of confirming the accuracy of the pharmacokinetic / pharmacodynamic model for predicting the QT extension syndrome constructed in the present invention. 6A shows the change in QTc interval after administration of 10 mg / kg of astemizole, and 6B shows the accuracy of the results of estimating changes in QTc interval after administration of 30 mg / kg of astemizole.

Hereinafter, preferred embodiments of the present invention will be described in detail. However, the present invention is not limited to the embodiments described herein but may be embodied in other forms. Rather, the intention is to provide an exhaustive, complete, and complete disclosure of the principles of the invention to those skilled in the art.

≪ Example 1: Animal experiment &

Example 1-1. Animal management and drug administration

In the animal experiment, 8 beagle dogs (male, 8 ~ 9 kg) were used. The beagle dogs were adapted to individual stainless steel weights under conditions of 20 ° C to 29 ° C temperature, humidity 40% to 70%, 12 hours night / day cycle, 150 lux to 300 lux illumination, 10 to 20 ventilation per hour. Water and irradiated water were supplied randomly.

Before drug administration, the animals were fasted for one day and fed from 4 hours after administration of the drug. As a drug, astramizole dissolved in a 0.5% methylcellulose solution was used. On the first day, animals were orally administered a 0.5% methylcellulose solution as a vehicle. The following day, the group was divided into two groups: 10 mg / kg of astemizole in one group and 30 mg / kg of astemizole in the other group.

Examples 1-2. Electrocardiogram and heart rate measurement

The arterial blood pressure, heart rate and electrocardiogram were measured after attaching a radio telemetry transmitter (TL11M2-D70-PCT, Data Sciences International, St Paul, MN) to the beagle dog. Electrocardiograms were measured with M-X and R-L using a Holter electrocardiograph (QR1300, Fukuda Me.E., Beijing, Japan).

The ECG waveform morphology assessment during the entire monitoring period of the ECG was performed by a pharmacologist or cardiologist. Standard electrocardiographic intervals (PR, RR, QRS, and QT) were measured automatically and recorded every 10 minutes. Signals were collected with Data Sciences International System hardware. After collecting data for 10 minutes, three waveforms were selected and the average of three waveforms at each time point was recorded. QTc intervals were calibrated using the Fridericia formulation.

Electrocardiogram and heart rate were measured at 0, 0.5, 1, 1.5, 2, 2.5, 4, 5, 6, 7, 8 and 24 hours after the administration of the control group, The QTc interval results obtained by electrocardiogram measurement are shown in FIG. 1, after 1.5, 2, 2.5, 3, 3.5, 4, 5, 6, 7, 8, 12, 16 and 24 hours.

As shown in FIG. 1, the change in QTc interval occurred at the beginning of the administration of the control group, but it did not change after a certain time. On the other hand, when astemizole was administered, the QTc interval change was larger when compared with the control group.

Examples 1-3. Blood drug concentration measurement

The plasma samples were analyzed by HPLC-MS / MS (high performance liquid chromatography-mass spectrometry) to determine the concentrations of astemizole and desmethyl astemizole in the blood after administration of astemizole in animals. Respectively.

20 μl of diphenhydramine dissolved in methanol was mixed with 40 μl of plasma, followed by addition of an extraction solution (ethyl acetate: methanol = 9: 1), followed by centrifugation at 13,200 rpm for 5 minutes. After centrifugation, 50 μl of the supernatant was mixed with 200 μl of 0.04% trifluoroacetic acid dissolved in acetonitrile, and then injected into HPLC-MS / MS using 5 μl of the supernatant. HILIC column (silica , 3 占 퐉, 2.1 占 50 mm). At this time, the mobile phase was separated at a flow rate of 0.25 ml / min using an isotonic solvent consisting of 86% acetonitrile containing 0.02% trifluoroacetic acid and 14% 10 mM ammonium acetate.

Mass spectrometry shows that the precursor-product ion transition ([M + H] ⁺ ) used in the analysis of astemizole, desmethyl astemizole and diphenhydramine is 459.2? 135.2, 115.2 → 204.3 and 256.2 → 167.1. The concentrations of astemizole and desmethylastimizole in the blood were measured, and non-compartmental analysis was performed using Phoenix. The results are shown in FIG. 2 and Table 1.

Astemizole Desmethylasteminol density 10 mg / kg 30 mg / kg 10 mg / kg 30 mg / kg C _max (ng / ml) 27.00 ± 30.51 228.35 ± 262.67 49.13 + - 41.28 240.75 ± 161.30 t _1/2 (hr) 35.2 ± 5.06 20.10 ± 2.88 32.75 + - 5.61 28.38 + - 10.75 AUC _inf
(ng · hr / ml) 739.04 ±
627.22 4892.53 ±
4694.51 2325.18 ±
2173.52 11454.38 ±
7688.73

FIG Referring to Figure 2, astemizole (Fig. 2A) and the desmethyl astemizole was to determine the blood change in concentration with time (Fig. 2B), As shown in Table 1, maximum plasma concentration (C _max of the drug ), Half-life (t _1/2 ), area under the curve (AUC _inf ), and so on.

< Example  2. Pharmacokinetics and drug power modelling >

Pharmacokinetics and pharmacokinetic modeling were performed using NONMEM version 7.3. The first-order conditional estimation-with-interaction (FOCE + I) method was used for variable estimation. An exponential model was used to explain inter-individual variability (IIV). A constant coefficient of variation (CCV) and an additive model were used to explain the residual variability of pharmacokinetic and pharmacokinetic models, respectively.

Example 2-1. Pharmacokinetic modeling

The pharmacokinetic model is shown in Equation (1).

[Equation 1]

Here, k _a is a parameter that accounts for the primary absorption rate after oral administration of the drug, and k _m is a parameter that accounts for the primary metabolic rate at which the drug is metabolized in the central compartment. V _c and V _m are the apparent distribution volumes of astemizole and desmethylasteminol, and k _el and k _el-m are the primary elimination rates of astemizole and desmethyl astemizole. The group representative estimates of each variable are denoted by θ and the inter-individual variance is expressed as η.

Based on the above equation (1), the parameters of the pharmacokinetic model after administration of astemizole are shown in Table 2.

parameter Estimated value (% RES) IIV (CV%) (% RES) Bootstrap results of estimated mean (2.5 ~ 97.5% percentiles) k _a (h ^-1 ) 0.49 ^a - 0.49 ^a V _c (L) 4950.00 (46.7%) 141.6% (19.7%) 4952.00 (2451.02 to 12271.54) K _el (h ^-1 ) 0.0127 (33.5%) 33.4 (34.2%) 0.011 (0.0068 to 0.024) k _m (h ^-1 ) 0.0095 (39.1%) - 0.0095 (0.0023-0.017) V _m (L) 20 ^a 141.6 (19.7%) 20 ^a K _el-m (h ^-1 ) 0.95 (50.3%) 33.4% (34.2%) 0.99 (0.20 to 2.49) Residual variability CV% (RSE%) Proportional error 81.2% (16.7%) ^a fixed variable

As shown in Table 2, the appropriate pharmacokinetic and individual differences of astemizole and desmethylastimizole could be calculated, and the median value of the parameters calculated from the bootstrap results was estimated by the model The stability of the parameters could be confirmed through similar.

Example 2-2. Pharmacokinetic modeling

A modified harmonic model was used as the pharmacokinetic model to explain the change of the QTc interval under the normal condition of the activity cycle.

The equation of the basic harmonic model is shown in Equation (2).

&Quot; (2) &quot;

Here, CR ₁ and CR ₂ describe the activity period using sine and cosine functions. T ₁ and T ₃ are parameters expressing the QTc amplification induced by the activity period. (time-T ₂ ) / 24 and (time-T ₄ ) / 24 are parameters representing the activity period period. QT ₀ was the baseline QTc value and was measured after administration of the control group at steady state.

To establish a model to explain the change in QTc interval under steady state, a model using the QTc interval measured after oral administration of the control group was made, and the model was set as a basic model. The QTc interval induced by the drug effect We established a pharmacokinetic model to explain the change. Estimated parameters of the final pharmacokinetic model are shown in Table 3.

parameter Estimated value (% RES) IIV (CV%) (% RES) QT ₀ (ms) 230 (1.9%) 2.4% (30.6%) T ₁ -2.15 (53.0%) - T ₂ -4.08 (19.6%) - T ₃ 5.09 (75.4%) - T ₄ 11.3 (8.9%) - Residual variation (RSE%)
Additional error
7.71 (12.2%)
10.8 (7.2%)

Referring to Table 3, we could confirm the appropriate pharmacokinetic parameters and the difference between individuals to explain the change of QTc interval by activity period in normal state.

< Example  3. QT Extension syndrome  Predictive drug pumping / Pharmacodynamic modelling >

The correlation between pharmacokinetics and pharmacokinetics was expressed as a counterclock wise hysteresis and a biophase model was used to explain the relationship between pharmacokinetics and pharmacokinetics. The differential equation used to express the pharmacokinetic / pharmacodynamic model of the final drug is shown in Equation 3 below.

&Quot; (3) &quot;

In Equation 3, X _i (i = 1, 2, 3) is the amount of drug in each compartment, X _j (j = 4, 5) is the drug concentration of each living body compartment. k _eo and k _{eo -m} are distribution rate constants from the central compartment of the drug and metabolite to the bio-compartment.

Based on the above Equation (3), the pharmacokinetic / pharmacodynamic model of the drug for predicting the QT extension syndrome was constructed by using Equations (1) and (2) above, and the equations are shown in Equations (4) and (3) The estimated parameters of the pharmacokinetic / pharmacodynamic model are shown in Table 4.

&Quot; (4) &quot;

parameter Estimated value (% RES) IIV (CV%) (% RES) Bootstrap results of estimated mean (2.5 ~ 97.5% percentiles) QT ₀ (ms) 233 (1.4%) 3.0% (31%) 231.63 (222.91-238.10) T ₁ 3.31 (92.1%) - 3.50 (1.00 to 7.31) T ₂ -9.24 (30.7%) - -9.85 (-21.83 to -4.37) T ₃ 1.50 (81.3%) - 1.50 (0.20 to 7.81) T ₄ 1.85 (53.4%) - 1.89 (0.25 - 3.68) EC ₅₀ (ng / ml) 0.81 (69.9%) - 0.34 (0.14 to 9.60) k _eo (h ^-1 ) 0.04 ^a - 0.04 ^a k _eo-m (h- ¹ ) 0.03 ^a - 0.03 ^{a a} fixed variable

As shown in FIG. 3, the pandemic / pharmacodynamic model of the present invention shows that it is possible to predict the changes in serum concentration and QTc interval of astemizole, desmethylastestimazole due to the effects of the action period and drug.

In addition, Table 4 illustrates appropriate parameters and differences between individuals to explain the QTc change in the active day cycle and the effect of QTc change due to the administration of astemizole and desmethylasteminol in the normal state. The stability of the parameters can be confirmed through the fact that the median of the variables calculated as a result of the bootstrap is similar to the value estimated by the model.

<Example 4: Verification of model accuracy>

The accuracy of the final parameter estimates was assessed using bootstrapping. The median and standard error of the final pharmacokinetic / pharmacodynamic model obtained from the bootstrap analysis are shown in Tables 2 and 4.

The predictive power of the model was evaluated using a visual predictive check.

Example 4-1. Check the accuracy of pharmacokinetic model

FIG. 4 shows the visual prediction test result for confirming the accuracy of the pharmacokinetic model constructed in Example 2-1.

FIG. 4A shows the concentration of astemizole in the plasma after administration of 10 mg / kg of astemizole, FIG. 4B shows the concentration of desmethylastimizole in plasma after administration of 10 mg / kg astemizole, FIG. 4C shows the concentration of 30 mg / FIG. 4D shows the result of drug dynamics according to concentration of desmethylastimizole in plasma after administration of 30 mg / kg of astemizole. FIG. 4D shows the concentration of astemizole in plasma after administration of astemizole. As a result of comparing the median value and the 95% confidence interval of the 90th quintile of the simulation data obtained from the constructed pharmacokinetic model with the actual observation value, the confidence interval simulated by the model well included the observation value , Indicating that the predictive power of the pharmacokinetic model of astemizole and desmethylastestimol was sufficient.

Example  4-2. Check the accuracy of the pharmacokinetic model

FIG. 5 shows the visual prediction test results for confirming the accuracy of the pharmacokinetic model used in Example 2-2.

As shown in FIG. 5, when the median value and the 95% confidence interval of the 90th quintile of the simulation data obtained using the pharmacokinetic model were compared with the actual observation values, the confidence interval simulated by the model was observed The results show that the predictive power of the drug power model to explain the change in the QTc interval over the period of activity in the normal state is sufficient.

Example  4-3. The medication / Pharmacodynamic  Check the accuracy of the model

FIG. 6 shows the visual prediction test results for confirming the accuracy of the pharmacokinetic / pharmacodynamic model of the drug for predicting the QT extension syndrome constructed in Example 3.

FIG. 6A shows changes in QTc intervals after administration of 10 mg / kg of astemizole, and FIG. 6B shows changes in QTc intervals after administration of 30 mg / kg of astemizole. As a result of comparing the median value and the 95% confidence interval of the 90th quintile of the simulated data obtained from the constructed pharmacokinetic / pharmacodynamic model with the actual observation values, the confidence interval simulated by the model was used as the observation value Well, it was found that the pharmacokinetic / pharmacodynamic model was predictive enough to predict changes in QTc interval due to daily dosing and drug administration.

<Experimental Example 1> Confirmation of the suitability of the model according to whether or not the activity days are applied>

The objective function value of the model was measured in order to determine whether the generalized model of the present invention was applied to the active phase / pharmacodynamic model and the active period without applying the activity period.

The objective function value was measured by the following equation.

n denotes the total number of data, obs _j denotes the observed value, pred _j denotes the predicted value by the model, and var _j denotes the variance of the residuals of the model. Therefore, the lower the objective function value, the less the residuals of the predicted values by the observed values and the model, and the statistically good model. If the objective function value is lowered by 18.47 when four variables are added in the model, (p < 0.001).

As a result, the objective function value of the pharmacokinetic / pharmacodynamic model according to the present invention was 73.68 lower than the objective function value of the general model without the activity period. In addition, since the present invention adds four variables as compared with the general model, but the objective function value is low at 73.68, the model according to the present invention is a much improved model.

Claims (7)

A method of predicting a QT syndrome using a pharmacokinetic / pharmacodynamic model of a drug,
Measuring the pulse rate and electrocardiogram of the patient before and after administration of the drug;
Measuring the blood drug concentration with time of the patient to which the drug is administered;
Inputting the measured electrocardiogram and administration information of the administered drug;
Performing modeling through measurements of the blood drug concentration; And
Applying the input ECG, blood drug concentration, and drug administration information to the established pharmacokinetic / pharmacodynamic model to determine the QTc interval change;
Wherein the QT prolongation syndrome prediction method comprises:
The pharmacokinetic / pharmacodynamic model of the drug,

Where T ₁ and T ₃ are the QTc amplifications induced by the activity cycle, (time-T ₂ ) / 24 and (time-T ₄ ) / 24 represent the activity cycle periods, QT ₀ is the base QTc X _j (j = 4, 5) means the drug concentration in each living body compartment, and EC ₅₀ means the concentration of the drug when the effect of the drug is halved. The method according to claim 1,
Wherein the pharmacokinetic / pharmacodynamic model of the drug is a combination of modified harmonic models to apply an activity period to a pharmacokinetic model. The method according to claim 1,
Wherein the pharmacokinetic / pharmacodynamic model of the drug is characterized by detecting drug-induced QT prolongation syndrome by classifying drug-induced QTc interval change and QTc interval change by activity cycle. The method of claim 3,
Wherein the medicament comprises a drug capable of causing a QT prolongation syndrome. 5. The method of claim 4,
Wherein the drug capable of causing the QT prolongation syndrome comprises an antihistamine agent. 6. The method of claim 5,
The antihistamine agent is selected from the group consisting of astemizole, terfenadine, azelastine, cinnarizine, dimenhydrinate, fexofenadine, diphenhydramine, A method for predicting QT prolongation syndrome characterized by comprising cyproheptadine, hydroxyzine, ebastine, loratadine, and doxepine.











delete

Images