- Email: [email protected]
The Pharmacokinetic-pharmacodynamic Assessment of the Hypotensive Effect after Coadministration of Losartan and Hydrochlorothiazide in Spontaneously Hypertensive Rats
Drug Metab. Pharmacokinet. 27 (2): 207215 (2012).
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
Regular Article The Pharmacokinetic-pharmacodynamic Assessment of the Hypotensive Effect after Coadministration of Losartan and Hydrochlorothiazide in Spontaneously Hypertensive Rats Ryosuke S HIMIZU 1,2, Makoto M IYAZAKI 2, Kazunori I WANAGA 2 and Masawo K AKEMI 2, * 1
Drug Metabolism & Pharmacokinetics, Drug Developmental Research Laboratories, Shionogi & Co., Ltd., Toyonaka, Japan 2 Division of Pharmaceutics, Osaka University of Pharmaceutical Sciences, Takatsuki, Japan
Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk Summary: The interactive hypotensive effect of the combination treatment of losartan (LOS) and hydrochlorothiazide (HCTZ) was assessed using a pharmacokinetic-pharmacodynamic (PK-PD) model in spontaneously hypertensive rats. Intravenous coadministration of these drugs showed a prolonged and enhanced time-course of the hypotensive effect. A population PK analysis revealed the delayed elimination of LOS after coadministration. The time-course of the plasma renin activity (PRA) was measured, and showed a more continuative time profile after coadministration compared with the administration of LOS alone. An indirect response model was applied to describe the relationship between the PK of LOS and the PRA profile, and the E max value for the increase of the PRA by LOS was increased with the dose of HCTZ. Blood pressure was linked to the PRA through an effect compartment. The model successfully described the relationship between the doses of LOS and HCTZ and their interactive hypotensive effect. These results indicate that the interaction for blood pressure in the combination treatment of LOS and HCTZ can be estimated using the doses of the drugs and the PRA-mediated PK-PD model. Keywords: losartan; hydrochlorothiazide; PK-PD model; plasma renin activity; drugdrug interaction
enhance each otherös hypotensive effect. ARBs also suppress such adverse effects as hypokalemia and hyperuricemia from thiazide diuretics.3¥ Activation of the renin-angiotensin system is responsible primarily to enhance the hypotensive effect after administration of ARBs with diuretics. As a result, the plasma renin activity ¤PRA¥ and the plasma concentrations of angiotensin II ¤Ang II¥ and angiotensin ¤1®7¥ ªAng ¤1®7¥« increase.4¥ Furthermore, plasma concentrations of nitric oxide and bradykinin, which are hypotension factors, also increase via the angiotensin type 2 receptor ¤AT2R¥ and the Mas receptor.5¥ However, little attention has been given to the quantitative relationship among the doses of drugs, the PRA, and the hypotensive effect after administration of ARBs with diuretics. A contribution of a pharmacokinetic ¤PK¥ interaction has not been revealed in the enhanced hypotensive effect either. The objectives of this study are to evaluate the PK and pharmacodynamic ¤PD¥ interactions for the hypotensive effect after the coadministration of losartan ¤LOS¥ and
Introduction Hypertension is a risk factor for several cardiovascular events. Avoiding the risk of hypertension by lowering blood pressure is important; in fact: úthe lower, the better.û1¥ The 2009 guidelines for treating hypertension by the Japanese Society of Hypertension ¤JSH2009¥ reported that middle age patients with hypertension, but without other risks, should maintain a systolic and diastolic pressure of less than 130/ 85 mmHg. However, in the Japan Home versus Office Blood Pressure Measurement Evaluation ¤J-HOME¥2¥ study, blood pressure management was reported as insufficient, and about half of patients treated with a single anti-hypertension drug could not maintain the target blood pressure. Thus, a combination therapy of anti-hypertensive drugs with different mechanism of actions is recommended to enhance the hypotensive effect. For combination therapies, angiotensin receptor blockers ¤ARBs¥ and thiazide diuretics are widely used, because they
Received June 16, 2011; Accepted October 26, 2011 J-STAGE Advance Published Date: November 10, 2011, doi:10.2133/dmpk.DMPK-11-RG-060 *To whom correspondence should be addressed: Masawo KAKEMI, Ph.D., Division of Pharmaceutics, Osaka University of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki 569-1094, Japan. Tel. +81-72-690-1048, Fax. +81-72-690-1048, E-mail: [email protected] 207
208
Ryosuke SHIMIZU, et al.
hydrochlorothiazide ¤HCTZ¥ in spontaneously hypertensive rats ¤SHRs¥, and to help estimation for the time course of blood pressure after coadministiration of both drugs. The responsible interactive factor will be revealed using a PK-PD compartmental analysis. Methods Chemicals: Losartan potassium ¤LOS¥ was purchased from Wako Pure Chemical Industries ¤Osaka, Japan¥ and hydrochlorothiazide ¤HCTZ¥ was bought from Sigma Chemical Co. ¤St. Louis, MO¥. Monoethanolamine was purchased from Nacalai Tesque ¤Kyoto, Japan¥. Saline was obtained from Otsuka Pharmaceuticals ¤Tokyo, Japan¥. All other reagents and solvents were commercial products of reagent grade. Animal surgery: Male SHRs ¤Japan SLC Inc., Shizuoka, Japan, 15®25 weeks old¥ were used. Rats were housed in constant environmental facilities ¤temperature: 24 + 1ôC, humidity: 55 + 10%¥, exposed to 12:12 h lightdark cycles ¤06:00 h/18:00 h¥ for more than 1 week, and allowed free access to the standard diet and tap water. On the day before the experiment, rats were lightly anesthetized with ethyl ether, and were implanted surgically with a SP10 catheter ¤Natume, Tokyo, Japan¥ connected to a PE50 ¤Clay Adams, Parsippany, NJ¥ catheter in the femoral artery for blood sampling. Rats were also implanted with a Phicon tube ¤Fuji Systems, Tokyo, Japan¥ connected to a PE50 catheter ¤Clay Adams¥ in the jugular vein for drug administration. Both catheters were externalized through the back in the neck region and secured. Unless otherwise specified, all animal experiments were carried out under non-restraining and non-anesthetic conditions and in a fasting state. These animal experiments were approved by the Animal Experimentation Committee of the Osaka University of Pharmaceutical Sciences. Animal experiments: LOS was dissolved in saline and HCTZ was dissolved in saline containing 1.7% monoethanolamine following the method of Kim et al.6¥ LOS ¤1 ® 10 mg/kg¥ and HCTZ ¤1 ® 10 mg/kg¥ were administrated through the jugular vein ¤n © 3®5¥. Blood samples were withdrawn from the femoral artery at designated postdose intervals. The blood samples were transferred into tubes containing heparin ¤1 IU¥, and then centrifuged ¤10,000 rpm for 3 min¥. The isolated plasma was stored frozen at %20ôC until analysis. Plasma LOS concentrations were determined by a HPLC-UV method7,8¥ and HCTZ concentrations were determined by a UPLCMS/MS method.9¥ The analytical methods for LOS and HCTZ are summarized in Table 1. Blood pressure was measured by the tail-cuff method ¤BP-98A-L, Softron, Tokyo, Japan¥ at predose ¤before surgery and drug administration¥ and at designated postdose intervals ¤n © 3®4¥. The average systolic blood pressure ¤SBP¥ value of 5 repeated measurements was defined as the observed data. To determine the PRA ¤n © 2®3¥, blood samples withdrawn
Table 1. The analytical methods for LOS and HCTZ HPLC-UV conditions for LOS column
Unison UK-C18 ¤3 µm¥ ¤75 mm ' 3 mm, Imtakt, Kyoto, Japan¥
UV wavelength
230 nm
mobile phase
25 mmol/L phosphoric acid:acetonirile © 80:20 ¤v/v¥
UPLC-MS/MS conditions for HCTZ column
ACQUITY UPLC BEH C18 ¤1.7 µm¥ ¤Waters Co. Ltd., Milford, NE, USA¥
mass
m/z 295.94 h m/z 269
mobile phase
10 mmol/L ammonium acetate:methanol © 70:30 ¤v/v¥
from the femoral artery were transferred into tubes containing EDTA-2Na, and then centrifuged ¤10,000 rpm for 3 min¥. The isolated plasma was frozen and stored at %20ôC until analysis. The PRA was determined by a radioimmunoassay method ¤Yamasa, Chiba, Japan¥. PK-PD modeling and data analysis: The concentration-time data of the drugs was analyzed by a nonlinear mixed effect model method using NONMEM software ¤version VI, level 2.0¥ to evaluate the PK interaction with the coadministration of LOS and HCTZ. The first-order method was employed throughout the analysis. Both 3-compartment and 2-compartment open models, as implemented in the NONMEM-PREDPP library subroutines ADVAN11/TRANS4 and ADVAN4/TRANS4, respectively, were investigated using the Akaike Information Criterion. Inter-individual variability for PK parameters and residual variability were estimated using an exponential error model. Model comparisons were based on the objective function values in NONMEM using the likelihood ratio test. The significance level was set at p g 0.05, which corresponds to a reduction of 3.84 in the objective function value to discriminate between the two nested structural models after inclusion of one additional parameter. Visual predictive checks were performed for the final model. The plasma drug concentration, PRA and blood pressure were evaluated in different animals. Inter-individual and intra-individual variability for the PD parameters was not characterized using a population analysis. The PK-PD model for the hypotensive effect of LOS was analyzed by a nonlinear regression program, FKDM,10¥ based on ordinary least squares. The differential equations were solved by the Runge-Kutta-Gill method.11¥ Fitting procedures were performed on a PC/AT compatible computer ¤NEC, Tokyo, Japan¥ running under Windows XP with the Fortran compiler Compaq Visual Fortran ¤version 6.1¥. The areas under the effect-time curve ¤AUE¥ were calculated by the linear trapezoidal method.
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
209
PRA-mediated PK-PD Model for Antihypotensive Drugs
A
LOS 5 mg/kg+HCTZ 1 mg/kg LOS 5 mg/kg+HCTZ 5mg/kg
Lk12
Lk21
ALOS1 Lk31
ALOS3
X2
k10Ὁhh
Lk13 Xk in
Plasma LOS Conc. (µ g/mL)
i.v. bolus
LOS 5 mg/kg
100
Pharmacokinetics
ALOS2
Xk12 Xk21
X1
Xkout
S(t)
APRA
Population average for LOS (single)
10
Population average for LOS (coadministration)
1
Rkout
Pkin
CE
LOS 5 mg/kg+HCTZ 10 mg/kg
0.1 0
Hypotensive Effect
2
4
6
8
10
Time (hr)
Pkout Pharmacodynamics
B
HCTZ 5 mg/kg
10
HCTZ 5 mg/kg+LOS 5 mg/kg
Theoretical: Figure 1 represents the PK-PD model for the hypotensive effect after administration of LOS. The PK of LOS in plasma was described by a linear threecompartment open model, as follows: dALOS1 ¼ Lk21 ALOS2 þ Lk31 ALOS3 dt ðLk12 þ Lk13 þ Lk10h Þ ALOS1 ð1Þ dALOS2 ¼ Lk12 ALOS1 Lk21 ALOS2 ð2Þ dt dALOS3 ¼ Lk13 ALOS1 Lk31 ALOS3 ð3Þ dt ALOS1 CLOS1 ¼ ð4Þ LV where ALOS1 is the amount of LOS ¤µg¥ in the central compartment; ALOS2 and ALOS3 are the amount of LOS concentrations in the peripheral compartments; LV is the distribution volume of LOS in the central compartment; Lk12, Lk21, Lk13, Lk31, and Lk10 are the first-order rate constants ¤min%1¥ of LOS; CLOS1 is the plasma LOS concentration ¤µg/mL¥; and at t © 0, ALOS1 © Dose and ALOS2 © ALOS3 © 0. In the PK-PD analysis, the estimates of the population mean were used in the PK parameters, because animals were individually applied in the PK and pharmacological experiments. A hybrid model of an indirect response model and an indirect link model was used to describe a sequence of variations in the PRA and the hypotensive effect after LOS administration. In this study, we focused on the contribution of the PRA to the hypotensive effect of LOS. Then, we assumed the following: a¥ LOS increases the PRA via the accumulative delay compartments of X1 and X2. b¥ The influx rate into the X1/X2 compartment is negligible because LOS influx follows a first-order process but the received mass is negligible compared to the PK model. c¥ Increased PRA decreases blood pressure through the effect compartment CE. The differential equations of the model are as follows:
HCTZ 10 mg/kg
Plasma HCTZ Conc. ( µ g/mL)
Fig. 1. Diagrammatic representation of the PK-PD model for the hypotensive effect
HCTZ 10 mg/kg+LOS 5 mg/kg Population average for HCTZ 5 mg/kg
1
Population average for HCTZ 10 mg/kg
0.1
0.01 0
1
2
3 Time (hr)
4
5
6
Fig. 2. Pharmacokinetic interaction for coadministration of LOS and HCTZ A) The effect of HCTZ in plasma LOS concentration after administration of LOS with and without HCTZ. B) The effect of LOS in plasma HCTZ concentrations after administration of HCTZ with and without LOS. Each symbol indicates an individual animal (n ¦ 3–5).
dCX1 ¼ Xkin CLOS1 Xkout CX1 dt Xk12 CX1 þ Xk21 CX2 dCX2 ¼ Xk12 CX1 Xk21 CX2 dt dAPRA ¼ Rkin SðtÞ Rkout APRA dt REmax CX1 SðtÞ ¼ 1 þ REC50 þ CX1
ð5Þ ð6Þ ð7Þ ð8Þ
where CX1 and CX2 are the concentrations ¤µg/mL¥ in the compartments X1 and X2, respectively; Xk12, Xk21, Xkin, and Xkout are first-order constants ¤min%1¥; APRA is the plasma renin activity ¤ng&Ang I/mL/h¥ in the PRA compartment; REmax is the maximum effect of the PRA ¤ng&Ang I/mL/h¥; REC50 is the LOS concentration ¤µg/ mL¥ producing 50% of REmax; Rkin is the zero-order constant ¤ng&Ang I/mL/h2¥ concerned with the PRA production; and Rkout is the first-order constant ¤min%1¥ concerned with the PRA degradation. Before LOS is administered, the baseline level of APRA is maintained as APRA0 ¤ng&Ang I/mL/h¥, which is determined by Rkout © Rkin&APRA0. The PRA decreases blood pressure as follows:
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
210
Ryosuke SHIMIZU, et al.
dCE ¼ Pkin APRA dt
Pkout CE
ð9Þ
PEmax CE ð10Þ PEC50 þ CE where CE is the value of the PRA ¤ng&Ang I/mL/h¥ in the effect compartment, and Pkin and Pkout are first-order constants ¤min%1¥; PEmax is the maximum hypotensive effect of the PRA ¤mmHg¥; PEC50 is the PRA ¤ng&Ang I/mL/h¥ producing 50% of PEmax; and E0 is the basal level of blood pressure ¤mmHg¥. Blood pressure ¼ E0
Table 2. Population pharmacokinetic parameters of LOS and HCTZ Fixed effect
Random effect
Losartan ¤LOS¥ CLLOS ¤mL/min¥ © 2.54 ' 0.759HT CLLOS-ETA © 0.367 Q3LOS ¤mL/min¥ © 7.38 V1LOS ¤mL¥ © 44.2
SIGMALOS © 0.0847
CV ¤%¥ © 29.1
Hydrochlorothiazide ¤HCTZ¥ CLHCTZ ¤mL/min¥ © 14.7
CLHCTZ-ETA © 0.0210 CV ¤%¥ © 39.3
Q2HCTZ ¤mL/min¥ © 10.3 V1HCTZ ¤mL¥ © 39.3 V2HCTZ ¤mL¥ © 802 SIGMAHCTZ © 0.0404 CV ¤%¥ © 20.1 HT: administration with HCTZ © 1, administration without HCTZ © 0. CLLOS, CLHCTZ: the total clearances for LOS and HCTZ; Q2LOS, Q3LOS, Q2HCTZ: the clearances for LOS and HCTZ for the peripheral compartments; V1LOS, V1HCTZ: the distribution volume of the central compartment for LOS and HCTZ; V2LOS, V3LOS, V2HCTZ: the distribution volume of the peripheral compartment for LOS and HCTZ; CLLOS-ETA, CLHCTZ-ETA: variance of the random effect for CLLOS and CLHCTZ; V1LOS-ETA: variance of the random effect for V1LOS; CV: coefficient of variance; SIGMALOS, SIGMAHCTZ: residual variability for LOS and HCTZ.
200
Individual predicted LOS Conc. (µg/mL)
Population predicted LOS Conc. (µg/mL)
CV ¤%¥ © 69.9
V3LOS ¤mL¥ © 75
200
150
100
50
150
100
50
0
0 0
50
100
150
200
250
0
Observed Losartan Conc. (µg/mL)
50
100
150
200
250
Observed Losartan Conc. (µg/mL)
5 4 3
4
2 1 0 -1 -2
1
3 2 WRES
WRES
V1LOS-ETA © 0.487
V2LOS ¤mL¥ © 210
Results Pharmacokinetic and pharmacological interaction in the combination treatment: After i.v. bolus coadministration of LOS and HCTZ in each individual animal, the plotting points in Figure 2A show the effect of coadministered HCTZ on the time-course of plasma LOS concentration. A two- or three-exponential decline in plasma was observed after administration of LOS alone. Since linearity was shown in these dose-normalized timecourses for LOS ¤5, 10, 20 mg/kg, data not shown¥, a linear three-compartment model was selected to describe the PK of LOS. As shown by covariate adjustment, coadministered HCTZ, as a covariate, significantly and dose-independently decreased the elimination clearance of LOS by about 25%, but did not affect the distribution volume ¤Table 2, solid lines in Fig. 2A¥. The goodness-of-fit of the final model is shown in Figure 3. The plotting points in Figure 2B
0 -1 -2 -3
-3 -4 -5
-4 -5 0
50
100
150
200
Observed Losartan Conc. (µg/mL)
250
CV ¤%¥ © 60.6
Q2LOS ¤mL/min¥ © 0.938
0
200
400
600
Time (min)
Fig. 3. Goodness-of-fit for the final population pharmacokinetic model for LOS Plots are individual data. Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
800
211
PRA-mediated PK-PD Model for Antihypotensive Drugs
represent the effect of LOS in the plasma HCTZ concentration®time course after i.v. bolus coadministration of LOS and HCTZ in each individual animal. The plasma concentration of HCTZ was fitted to a two-compartment model ¤Table 3, solid line in Fig. 2B¥. LOS did not affect
any PK parameters of HCTZ. The time-courses of SBP after i.v. bolus administration of LOS and HCTZ are shown in Figures 4A and 4B, and the dose-dependency of the hypotensive effect in the AUE is shown in Figures 4C and 4D. After i.v. bolus administration of LOS ¤5 mg/kg¥ or HCTZ ¤5 mg/kg¥, SBP decreased slightly and continued for about 72 h ¤Figs. 4A and 4B¥. After coadministration of LOS and HCTZ, SBP is remarkably decreased up to 24 h after dosing and a lower SBP profile is maintained for about 144 h. This interactive enhancement of the hypotensive effect depended on the doses of LOS and HCTZ ¤Figs. 4C and 4D¥. This hypotensive interaction was also synergistic; e.g. AUE was 2,144.4 mmHg&h with a combination of LOS 5 mg/kg and HCTZ 5 mg/kg compared to 662.6 mmHg&h and 521.1 mmHg&h, respectively, for each drug alone. The effect of the combination treatment on the plasma renin activity: The time-course of the PRA after i.v. bolus coadministration of LOS and HCTZ is represented in Figure 5. The PRA temporally increased and then gradually recovered to baseline up to 72 h after administration, although the eliminations of LOS and HCTZ from plasma were rapid. Thus, there was a typical counterclockwise hysteresis between plasma LOS concentrations and the PRAs ¤data not shown¥. Dose-dependent increases in the PRA were observed in LOS and HCTZ coadministration studies ¤Fig. 5¥. PK-PD analysis for PRA enhancement: To describe the time-course of the PRA after administration of LOS with HCTZ, an indirect response model was applied to the PK-PD model for LOS as shown in Figure 1 and Eqs. ¤1¥®¤7¥. However, the compartments X1 and X2 were
Table 3. Pharmacodynamic parameter estimates of the hypotensive effect after coadministration of LOS and HCTZ in rats, and for E-3174 after i.v. bolus administration of LOS in rats Parameters
Estimates
Xk12
¤min%1¥
0.00732 + 0.0168
Xk21
¤min%1¥
0.00123 + 0.000720
Xkout
¤min%1¥
0.0162 + 0.0264
REC50
¤µg/mL¥
0.585 + 0.0717
Rkout
¤h%1¥
3.14 + 26.5
REmax ¤HCTZ 1 mg/kg¥
14.0 + 2.52
REmax ¤HCTZ 5 mg/kg¥
15.7 + 2.36
REmax ¤HCTZ 10 mg/kg¥
21.5 + 3.61
Pkout
¤min%1¥
PEC50
¤ng Ang I/mL/h¥
85.5 + 118
PEmax
mmHg
180 + 208
¤min%1¥
0.0228 + 0.0016
0.00244 + 0.000440
for E-3174 aa
%1
a
¤min ¥
0.0110 + 0.00285
Mk21a
¤min%1¥
0.00590 + 0.00211
Mkouta
¤min%1¥
0.0219 + 0.00205
Mk12
a
Parameters of the model for E-3174 ¤see Fig. 11¥. Data represent the computer-fitted value + SD.
210
A
200 190 180 170
LOS 5 mg/kg
160
LOS 5 mg/kg+HCTZ 1 mg/kg
150
LOS 5 mg/kg+HCTZ 5 mg/kg LOS 5 mg/kg+HCTZ 10 mg/kg
140 0
24
48
72
96
120
Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
210
190 180 170 HCTZ 5 mg/kg
160
HCTZ 5 mg/kg+LOS 1 mg/kg
150
HCTZ 5 mg/kg+LOS 5 mg/kg HCTZ 5 mg/kg+LOS 10 mg/kg
140 0
144
Time (hr) 4500
C
4000
4000
3500
3500
AUE (mmHg hr)
AUE (mmHg
hr)
4500
3000 2500 2000 1500
B
200
24
D
48
72
96
120
144
Time (hr)
3000 2500 2000 1500
1000
1000
500
500 0
0 LOS 5 mg/kg
LOS 5 mg/kg +HCTZ 1 mg/kg
Dose
LOS 5 mg/kg +HCTZ 5 mg/kg
LOS 5 mg/kg +HCTZ 10 mg/kg
HCTZ 5 mg/kg
LOS 1 mg/kg +HCTZ 5 mg/kg
LOS 5 mg/kg +HCTZ 5 mg/kg
LOS 10 mg/kg +HCTZ 5 mg/kg
Dose
Fig. 4. Changes in blood pressure (A, B) and AUE (C, D) after coadministration of LOS and HCTZ Values represent the mean + S.E.M. (n ¦ 3–4). Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
212
Ryosuke SHIMIZU, et al.
required to explain the delayed profiles of the PRA against plasma LOS concentration profiles. The solid lines, shown in Figure 6, represent the results of fitting the observed data after coadministration of LOS and HCTZ to the model described in the theoretical section. The regression curves fit the observed data well. The estimated PD parameters are
listed in Table 3. REmax values increased with the dose of HCTZ ¤Fig. 7¥. PK-PD analysis for the PRA-mediated hypotensive effect: The PK-PD analysis for the PRA-mediated hypotensive effect after i.v. bolus administration of LOS was carried out according to Eqs. ¤1¥®¤10¥. Since there was a typical counter-clockwise hysteresis between the PRA level and the hypotensive effect, an additional effect compartment was introduced between the PRA and the hypotensive effect model. The solid lines shown in Figure 8 represent the theoretical hypotensive effect and fit the observed data well. The estimated PD parameters are listed in Table 2. To evaluate the validity of the model, we simulated the timecourses of the PRA and SBP after i.v. bolus administration of 5 mg/kg LOS and 8 mg/kg HCTZ. In this simulation, we
Plasma Renin Activity (ng Ang 㻌㻌 mL/hr)
70 60 LOS 5 mg/kg
50
LOS 5 mg/kg+HCTZ 1 mg/kg 40
LOS 5 mg/kg+HCTZ 5 mg/kg LOS 5 mg/kg+HCTZ 10 mg/kg
30 20 10 0 24
0
48
72
96
120
25
144
Time (hr)
20 REmax value
Plasma Renin Activity (ng Ang 㻌 mL/hr)
70 60 50
HCTZ 5 mg/kg HCTZ 5 mg/kg+LOS 1 mg/kg
40
HCTZ 5 mg/kg+LOS 5 mg/kg HCTZ 5 mg/kg+LOS 10 mg/kg
15 10
y = 0.8493x + 12.53 R2 = 0.9461
5
30 20
0 0
10
4
8
12
Hydrochlorothiazide Dose (mg/kg) 0 0
24
48
72
96
120
144
Fig. 7. Relationship between the REmax value and the dose of HCTZ for the hypotensive effect after coadministration of LOS with HCTZ The solid line represents a regression line. Plots represent the model-calculated value.
Time (hr)
Fig. 5. Time-course of plasma renin activity after coadministration of LOS and HCTZ Plots represent the mean + S.E.M. (n ¦ 2–3)
LOS 5 mg/kg+HCTZ 5 mg/kg
60
60
60
50 40 30 20 10
Plasma Renin Activity (ng Ang 䊠/mL/hr)
70
0
50 40 30 20 10 0
-24
0
24
48
72
96
0
24
48
72
96
120 144
Time (hr)
HCTZ 5 mg/kg+LOS 1 mg/kg
60
Plasma Renin Activity (ng Ang mL/hr)
70
60
40 30 20 10 0
40 30 20 10 -24
0
24
48
72
96
120
Time (hr)
HCTZ 5 mg/kg+LOS 10 mg/kg
70
50
50
0 -24
120 144
Time (hr)
Plasma Renin Activity (ng Ang䊠 mL/hr)
LOS 5 mg/kg+HCTZ 10 mg/kg
70 Plasma Renin Activity (ng Ang mL/hr)
Plasma Renin Activity (ng Ang /mL/hr)
LOS 5 mg/kg+HCTZ 1 mg/kg 70
50 40 30 20 10 0
-24
0
24
48
72
Time (hr)
96
120 144
-24
0
24
48
72
96
120 144
Time (hr)
Fig. 6. Comparison of the model-fitted value with the observed data for the PRA after coadministration of LOS and HCTZ Solid lines are the model-fitted profiles. Plots represent the mean + S.E.M. (n ¦ 2–3). Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
213
PRA-mediated PK-PD Model for Antihypotensive Drugs LOS 5 mg/kg+HCTZ 1 mg/kg
LOS 5 mg/kg+HCTZ 5 mg/kg
200 190 180 170 160 150 140 0
24
48
72
96
120
200 190 180 170 160 150 140
144
0
24
48
Time (hr)
200 190 180 170 160 150 140 48
72
96
120
144
96
120
210 200 190 180 170 160 150 140 0
24
48
72
96
120
144
Time (hr)
HCTZ 5 mg/kg+LOS 10 mg/kg Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
HCTZ 5 mg/kg+LOS 1 mg/kg
24
72 Time (hr)
210
0
LOS 5 mg/kg+HCTZ 10 mg/kg
210
Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
210
144
210 200 190 180 170 160 150 140 0
24
Time (hr)
48
72
96
120
144
Time (hr)
Systolic Blood Pressure (mmHg)
Fig. 8. Comparison of the model-fitted value with the observed data for the hypotensive effect after coadministration of LOS and HCTZ Solid lines represent the model-fitted values. Plots are the observed data and the mean + S.E.M. (n ¦ 3–4).
200 190 180 170 160 150 140 0
24
48
72
96
120
144
168
Time (hr)
Fig. 9. Comparison of the model-simulated SBP with the observed data after i.v. bolus coadministration of 5 mg/kg LOS with 8 mg/kg HCTZ The solid line represents the simulated values and the plots are the mean of observed data + S.E.M. (n ¦ 3).
calculated the value of REmax based on the correlation between REmax and the dose of HCTZ ¤Fig. 7¥ and used 19.09 mmHg. The other parameters were used from Table 2. The solid line shown in Figure 9 represents the model-predicted profiles of SBP and explain the observed data well. Discussion Pharmacokinetic and Pharmacodynamic interaction: When HCTZ is coadministered with LOS, no effect is observed for the elimination of HCTZ alone, but the total elimination clearance of LOS decreased by about 25%. ¤Table 2, Fig. 2¥. LOS is primarily metabolized by the hepatic cytochrome P450 ¤CYP2C9¥ in rats, and then more than 94% of the dose is excreted in the feces7¥; however, most of the administered HCTZ was excreted in the urine as
an intact drug without any metabolites.12¥ These reports indicate that coadministered LOS and HCTZ do not compete with each other for their metabolism or excretion. The plasma protein binding ratio of LOS is reported to be more than 99%,5¥ and that of HCTZ is about 22%.13¥ Thus, HCTZ would not affect the protein binding of LOS. On the other hand, 10 mg/kg HCTZ significantly increased the urine volume up to 4 h after oral treatment for 5 days in SHRs.14¥ Change in the urine volume was not investigated before and after the drug administration in the present study. However, decrease of the body fluid by the diuretic effect of HCTZ ¤1®10 mg/kg¥ might concentrate the level of LOS in plasma, and then the apparent clearance of LOS would decrease. This speculation also supports the invariable PK of HCTZ after coadministration with LOS. The interactive hypotensive effect was observed after administration of LOS and HCTZ. It was greater than the aggregate of their respective effects ¤Fig. 4¥, even after considering the PK interaction by HCTZ. The hypotensive interaction depended on the doses of LOS and HCTZ, which means there is a synergistic effect. As shown in Figure 10, LOS blocks the angiotensin type 1 receptor ¤AT1R¥. This interaction of LOS with the AT1R suppresses vasoconstriction and activation of the AT2R with the elevated Ang II, and consequently yields a decreased blood pressure.15¥ As a result of blocking AT1R, plasma renin is also activated through suppressing the negative feedback mechanism for the renin-angiotensin system, and subsequently elevates Ang I, Ang II, and Ang ¤1®7¥ levels. Continuous increases of the PRA and Ang II level in plasma were observed after administration of valsartan.16¥ Ang ¤1®7¥ releases nitric oxide and decreases blood pressure after binding to the Mas receptor.17¥ Collister
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
214
Ryosuke SHIMIZU, et al.
E3174
Hydrochlorothiazide
ALOS2
Angiotensinogen
M2
Plasma Renin activity
Lk12
Angiotensin
Lk21
ACE
Negative Feedback
a
i.v. bolus Angiotensin ACE2
Hypertensive effect etc.
AT 2 receptor
K10
Mkout
ALOS3
Hypotensive effect etc.
Plasma E3174 Conc. (mg/mL)
Fig. 10. Effect of LOS and HCTZ in the renin-angiotensin system cascade
and Hendel18¥ reported that the increase of Ang ¤1®7¥ in plasma develops the hypotensive effect. Furthermore, the PRA increased after administration of HCTZ.14¥ In our study, the PRA continuously increased for 72 h after LOS administration ¤Fig. 5¥. HCTZ administered with LOS further raised the PRA synergistically. These results suggest that PRA is a pharmacological mediator for PK and the hypotensive effect. PRA-mediated PK-PD model: The hypotensive effect after i.v. coadministration of LOS and HCTZ was described well using the PK-PD model with PRA enhancement ¤Fig. 6¥. In this model, a hypothetical maximum effect for enhancement of the PRA ¤REmax¥ increased depending on the dose of HCTZ ¤Fig. 7¥. The delay compartment X1/X2 was necessary to describe the change in the PRA, because of the delay for activation of plasma renin against the plasma LOS concentration profile. LOS has an active metabolite, E-3174, which shows a hypotensive effect. E-3174 also has a long half-life and a strong affinity compared to LOS. We evaluated the possibility that the X1/X2 compartment represented the PK of E-3174 using the reported PK data.19¥ Figure 11 represents the PK model for E-3174 that was modified based on the previous model ¤Fig. 1¥. The PK of E3174 in plasma is described by Eqs ¤11¥®¤14¥, as follows: dALOS1 ¼ Lk21 ALOS2 þ Lk31 ALOS3 dt ðLk12 þ Lk13 þ Lk10 Þ ALOS1 ð11Þ dM1 ¼ a CLOS1 ðMk12 þ Mkout Þ M1 þ Mk21 M2 dt ð12Þ dM2 ¼ Mk12 M1 Mk21 M2 ð13Þ dt where M1 and M2 are the concentration of E-3174 ¤µg/mL¥; and Mk12, Mk21, Mkout, and a are the first-order rate constants ¤min%1¥ of E-3174. These parameter values were estimated using the values shown in Table 2 and the reported plasma concentrations of E-3174 after i.v. bolus administration of LOS without HCTZ. The regression
Lk13
Lk31
Mas receptor
Mk21
M1
ALOS1
Angiotensin(1-7)
Losartan AT 1 receptor
Mk12
10
1
0.1 0
2
4
6
8
10
12
Time (hr)
Fig. 11. Pharmacokinetic model for E-3174 after i.v. bolus administration of LOS (upper) and the time-course profile of plasma E-3174 concentration (lower) M1, M2: the central and peripheral compartment for E-3174; Mk12, Mk21, a, Mkout: the first-order constants.
curves fit the literature data well ¤Fig. 11, lower¥, and the estimated PK parameters are listed in Table 3. However, as compared with the parameters Xk12, Xk21 and Xkout in Table 3, these parameters indicated a faster elimination of the E-3174 profile in plasma. That is, the delay compartments X1/X2 do not represent the E-3174 profile only. These compartments also include other mediators, such as unknown metabolites and endogenous factors concerned with the PRA production. In addition, this result indicates the validity of the present model based on plasma LOS concentration, and the pharmacokinetic data of E-3174 is not necessary for the estimation of the interactive hypotensive effect. Furthermore, an additive effect compartment ¤CE¥ was required to link the PRA and the hypotensive effect. PRA raises the plasma Ang II and Ang ¤1®7¥ concentrations. Then these Angs bind to the AT2R and the MasR ¤Fig. 10¥. As a result, a decrease in blood pressure is observed. Therefore, the delays described by the effect compartment may represent the transit time required for the sequence of these pharmacological events of the renin-angiotensin system. The HCTZ dose-dependency of the maximum hypotensive effect also reasonably predicted the observed data. In conclusion, the interactive hypotensive effect is shown after i.v. bolus coadministration of LOS and HCTZ, and the
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
PRA-mediated PK-PD Model for Antihypotensive Drugs
relationship between plasma LOS concentration and blood pressure is described using the PK-PD model. The model quantitatively reveals that HCTZ principally enhances the effect of LOS dose-dependently through an increase of the PRA. This relationship can estimate the hypotensive effect for combination treatment using different doses and kinds of ARBs and thiazide diuretics.
9¥
10¥
References 1¥ Staessen, J. A., Gasowski, J., Wang, J. G. et al.: Risks of untreated and treated isolated systolic hypertension in the elderly: metaanalysis of outcome trials. Lancet, 355¤9207¥: 865®872 ¤2000¥. 2¥ Ohkubo, T., Obara, T., Funahashi, J. et al.: J-HOME Study Group., Control of blood pressure as measured at home and office, and comparison with physiciansö assessment of control among treated hypertensive patients in Japan: First Report of the Japan Home versus Office Blood Pressure Measurement Evaluation ¤J-HOME¥ study. Hypertens. Res., 27¤10¥: 755®763 ¤2004¥. 3¥ Nakashima, M., Uematsu, T., Kosuge, K. and Kanamaru, M.: Pilot study of the uricosuric effect of DuP-753, a new angiotensin II receptor antagonist, in healthy subjects. Eur. J. Clin. Pharmacol., 42: 333®335 ¤1992¥. 4¥ Villamil, A., Chrysant, S. G., Calhoun, D., Schober, B., Hsu, H., Matrisciano-Dimichino, L. and Zhang, J.: Renin inhibition with aliskiren provides additive antihypertensive efficacy when used in combination with hydrochlorothiazide. J. Hypertens., 25¤1¥: 217® 226 ¤2007¥. 5¥ Walters, P. E., Gaspari, T. A. and Widdop, R. E.: Angiotensin¤1-7¥ acts as a vasodepressor agent via angiotensin II type 2 receptors in conscious rats. Hypertension, 45: 960®966 ¤2005¥. 6¥ Kim, G. H., Na, K. Y., Kim, S. Y., Joo, K. W., Oh, Y. K., Chae, S. W., Endou, H. and Han, J. S.: Up-regulation of organic anion transporter 1 protein is induced by chronic furosemide or hydrochlorothiazide infusion in rat kidney. Nephrol. Dial. Transplant., 18¤8¥: 1505®1511 ¤2003¥. 7¥ Takayama, F., Saito, K., Yoshinaga, T., Morita, M., Hata, S., Esumi, Y., Jin, Y. and Okamura, Y.: Metabolic fate of losartan, a new angiotensin II receptor antagonist ¤1¥: absorption, distribution, metabolism and excretion after single administration in rats. Xenobiotic Metabolism and Disposition, 10: 223®243 ¤1995¥. 8¥ Farthing, D., Sica, D., Fakhry, I., Pedro, A. and Gehr, T. W.: Simple high-performance liquid chromatographic method for determination of losartan and E-3174 metabolite in human plasma,
11¥ 12¥ 13¥ 14¥
15¥ 16¥
17¥
18¥ 19¥
215
urine and dialysate. J. Chromatogr. B Biomed. Sci. Appl., 704¤1-2¥: 374®378 ¤1997¥. Li, H., Wang, Y., Tang, Y., Wang, J., Zhao, L. and Gu, J.: A liquid chromatography/tandem mass spectrometry method for the simultaneous quantification of valsartan and hydrochlorothiazide in human plasma. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 852¤1-2¥: 436®442 ¤2007¥. Lu, W., Endoh, M., Katayama, K., Kakemi, M. and Koizumi, T.: Pharmacokinetic and pharmacodynamic studies of piretanide in rabbits. I. Effect of different hydrated conditions. J. Pharmacobiodyn., 10: 356®363 ¤1987¥. Nelson, W., Tong, Y. L., Lee, J.-K. and Halberg, F.: Methods for cosinor-rhythmometry. Chronobiologia, 6: 305®323 ¤1979¥. Beermann, B., Groschinsky-Grinf, M. and Rosen, A.: Absorption, metabolism, and excretion of hydrochlorothiazide. Clin. Pharmacol. Ther., 19¤5 Pt 1¥: 531®537 ¤1976¥. Package insert of Preminent¬ tablet, ver. 9, Tokyo, MSD K.K., 2011. Wienen, W. and Schierok, H. J.: Effects of telmisartan, hydrochlorothiazide and their combination on blood pressure and renal excretory parameters in spontaneously hypertensive rats. J. Renin Angiotensin Aldosterone Syst., 2¤2¥: 123®128 ¤2001¥. Walters, P. E., Gaspari, T. A. and Widdop, R. E.: Angiotensin¤1-7¥ Acts as a Vasodepressor Agent Via Angiotensin II Type 2 Receptors in Conscious Rats. Hypertension, 45: 960®966 ¤2005¥. Azizi, M., Ménard, J., Bissery, A., Guyenne, T. T., Bura-Rivière, A., Vaidyanathan, S. and Camisasca, R. P.: Pharmacologic demonstration of the synergistic effects of a combination of the renin inhibitor aliskiren and the AT1 receptor antagonist valsartan on the angiotensin II-renin feedback interruption. J. Am. Soc. Nephrol., 15¤12¥: 3126®3133 ¤2004¥. Sampaio, W. O., Souza dos Santos, R. A., Faria-Silva, R., da Mata Machado, L. T., Schiffrin, E. and Touyz, R. M.: Angiotensin-¤1-7¥ through receptor Mas mediates endothelial nitric oxide synthase activation via Akt-dependent pathways. Hypertension, 49¤1¥: 185®192 ¤2007¥. Collister, J. P. and Hendel, M. D.: The role of Ang¤1-7¥ in mediating the chronic hypotensive effects of losartan in normal rats. J. Renin Angiotensin Aldosterone Syst., 4: 176®179 ¤2003¥. Moon, C. H., Lee, H. J., Jung, Y. S., Lee, S. H. and Baik, E. J.: Pharmacokinetics of losartan and its metabolite, EXP3174, after intravenous and oral administration of losartan to rats with streptozotocin-induced diabetes mellitus. Res. Commun. Mol. Pathol. Pharmacol., 101¤2¥: 147®158 ¤1998¥.
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
Regular Article The Pharmacokinetic-pharmacodynamic Assessment of the Hypotensive Effect after Coadministration of Losartan and Hydrochlorothiazide in Spontaneously Hypertensive Rats Ryosuke S HIMIZU 1,2, Makoto M IYAZAKI 2, Kazunori I WANAGA 2 and Masawo K AKEMI 2, * 1
Drug Metabolism & Pharmacokinetics, Drug Developmental Research Laboratories, Shionogi & Co., Ltd., Toyonaka, Japan 2 Division of Pharmaceutics, Osaka University of Pharmaceutical Sciences, Takatsuki, Japan
Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk Summary: The interactive hypotensive effect of the combination treatment of losartan (LOS) and hydrochlorothiazide (HCTZ) was assessed using a pharmacokinetic-pharmacodynamic (PK-PD) model in spontaneously hypertensive rats. Intravenous coadministration of these drugs showed a prolonged and enhanced time-course of the hypotensive effect. A population PK analysis revealed the delayed elimination of LOS after coadministration. The time-course of the plasma renin activity (PRA) was measured, and showed a more continuative time profile after coadministration compared with the administration of LOS alone. An indirect response model was applied to describe the relationship between the PK of LOS and the PRA profile, and the E max value for the increase of the PRA by LOS was increased with the dose of HCTZ. Blood pressure was linked to the PRA through an effect compartment. The model successfully described the relationship between the doses of LOS and HCTZ and their interactive hypotensive effect. These results indicate that the interaction for blood pressure in the combination treatment of LOS and HCTZ can be estimated using the doses of the drugs and the PRA-mediated PK-PD model. Keywords: losartan; hydrochlorothiazide; PK-PD model; plasma renin activity; drugdrug interaction
enhance each otherös hypotensive effect. ARBs also suppress such adverse effects as hypokalemia and hyperuricemia from thiazide diuretics.3¥ Activation of the renin-angiotensin system is responsible primarily to enhance the hypotensive effect after administration of ARBs with diuretics. As a result, the plasma renin activity ¤PRA¥ and the plasma concentrations of angiotensin II ¤Ang II¥ and angiotensin ¤1®7¥ ªAng ¤1®7¥« increase.4¥ Furthermore, plasma concentrations of nitric oxide and bradykinin, which are hypotension factors, also increase via the angiotensin type 2 receptor ¤AT2R¥ and the Mas receptor.5¥ However, little attention has been given to the quantitative relationship among the doses of drugs, the PRA, and the hypotensive effect after administration of ARBs with diuretics. A contribution of a pharmacokinetic ¤PK¥ interaction has not been revealed in the enhanced hypotensive effect either. The objectives of this study are to evaluate the PK and pharmacodynamic ¤PD¥ interactions for the hypotensive effect after the coadministration of losartan ¤LOS¥ and
Introduction Hypertension is a risk factor for several cardiovascular events. Avoiding the risk of hypertension by lowering blood pressure is important; in fact: úthe lower, the better.û1¥ The 2009 guidelines for treating hypertension by the Japanese Society of Hypertension ¤JSH2009¥ reported that middle age patients with hypertension, but without other risks, should maintain a systolic and diastolic pressure of less than 130/ 85 mmHg. However, in the Japan Home versus Office Blood Pressure Measurement Evaluation ¤J-HOME¥2¥ study, blood pressure management was reported as insufficient, and about half of patients treated with a single anti-hypertension drug could not maintain the target blood pressure. Thus, a combination therapy of anti-hypertensive drugs with different mechanism of actions is recommended to enhance the hypotensive effect. For combination therapies, angiotensin receptor blockers ¤ARBs¥ and thiazide diuretics are widely used, because they
Received June 16, 2011; Accepted October 26, 2011 J-STAGE Advance Published Date: November 10, 2011, doi:10.2133/dmpk.DMPK-11-RG-060 *To whom correspondence should be addressed: Masawo KAKEMI, Ph.D., Division of Pharmaceutics, Osaka University of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki 569-1094, Japan. Tel. +81-72-690-1048, Fax. +81-72-690-1048, E-mail: [email protected] 207
208
Ryosuke SHIMIZU, et al.
hydrochlorothiazide ¤HCTZ¥ in spontaneously hypertensive rats ¤SHRs¥, and to help estimation for the time course of blood pressure after coadministiration of both drugs. The responsible interactive factor will be revealed using a PK-PD compartmental analysis. Methods Chemicals: Losartan potassium ¤LOS¥ was purchased from Wako Pure Chemical Industries ¤Osaka, Japan¥ and hydrochlorothiazide ¤HCTZ¥ was bought from Sigma Chemical Co. ¤St. Louis, MO¥. Monoethanolamine was purchased from Nacalai Tesque ¤Kyoto, Japan¥. Saline was obtained from Otsuka Pharmaceuticals ¤Tokyo, Japan¥. All other reagents and solvents were commercial products of reagent grade. Animal surgery: Male SHRs ¤Japan SLC Inc., Shizuoka, Japan, 15®25 weeks old¥ were used. Rats were housed in constant environmental facilities ¤temperature: 24 + 1ôC, humidity: 55 + 10%¥, exposed to 12:12 h lightdark cycles ¤06:00 h/18:00 h¥ for more than 1 week, and allowed free access to the standard diet and tap water. On the day before the experiment, rats were lightly anesthetized with ethyl ether, and were implanted surgically with a SP10 catheter ¤Natume, Tokyo, Japan¥ connected to a PE50 ¤Clay Adams, Parsippany, NJ¥ catheter in the femoral artery for blood sampling. Rats were also implanted with a Phicon tube ¤Fuji Systems, Tokyo, Japan¥ connected to a PE50 catheter ¤Clay Adams¥ in the jugular vein for drug administration. Both catheters were externalized through the back in the neck region and secured. Unless otherwise specified, all animal experiments were carried out under non-restraining and non-anesthetic conditions and in a fasting state. These animal experiments were approved by the Animal Experimentation Committee of the Osaka University of Pharmaceutical Sciences. Animal experiments: LOS was dissolved in saline and HCTZ was dissolved in saline containing 1.7% monoethanolamine following the method of Kim et al.6¥ LOS ¤1 ® 10 mg/kg¥ and HCTZ ¤1 ® 10 mg/kg¥ were administrated through the jugular vein ¤n © 3®5¥. Blood samples were withdrawn from the femoral artery at designated postdose intervals. The blood samples were transferred into tubes containing heparin ¤1 IU¥, and then centrifuged ¤10,000 rpm for 3 min¥. The isolated plasma was stored frozen at %20ôC until analysis. Plasma LOS concentrations were determined by a HPLC-UV method7,8¥ and HCTZ concentrations were determined by a UPLCMS/MS method.9¥ The analytical methods for LOS and HCTZ are summarized in Table 1. Blood pressure was measured by the tail-cuff method ¤BP-98A-L, Softron, Tokyo, Japan¥ at predose ¤before surgery and drug administration¥ and at designated postdose intervals ¤n © 3®4¥. The average systolic blood pressure ¤SBP¥ value of 5 repeated measurements was defined as the observed data. To determine the PRA ¤n © 2®3¥, blood samples withdrawn
Table 1. The analytical methods for LOS and HCTZ HPLC-UV conditions for LOS column
Unison UK-C18 ¤3 µm¥ ¤75 mm ' 3 mm, Imtakt, Kyoto, Japan¥
UV wavelength
230 nm
mobile phase
25 mmol/L phosphoric acid:acetonirile © 80:20 ¤v/v¥
UPLC-MS/MS conditions for HCTZ column
ACQUITY UPLC BEH C18 ¤1.7 µm¥ ¤Waters Co. Ltd., Milford, NE, USA¥
mass
m/z 295.94 h m/z 269
mobile phase
10 mmol/L ammonium acetate:methanol © 70:30 ¤v/v¥
from the femoral artery were transferred into tubes containing EDTA-2Na, and then centrifuged ¤10,000 rpm for 3 min¥. The isolated plasma was frozen and stored at %20ôC until analysis. The PRA was determined by a radioimmunoassay method ¤Yamasa, Chiba, Japan¥. PK-PD modeling and data analysis: The concentration-time data of the drugs was analyzed by a nonlinear mixed effect model method using NONMEM software ¤version VI, level 2.0¥ to evaluate the PK interaction with the coadministration of LOS and HCTZ. The first-order method was employed throughout the analysis. Both 3-compartment and 2-compartment open models, as implemented in the NONMEM-PREDPP library subroutines ADVAN11/TRANS4 and ADVAN4/TRANS4, respectively, were investigated using the Akaike Information Criterion. Inter-individual variability for PK parameters and residual variability were estimated using an exponential error model. Model comparisons were based on the objective function values in NONMEM using the likelihood ratio test. The significance level was set at p g 0.05, which corresponds to a reduction of 3.84 in the objective function value to discriminate between the two nested structural models after inclusion of one additional parameter. Visual predictive checks were performed for the final model. The plasma drug concentration, PRA and blood pressure were evaluated in different animals. Inter-individual and intra-individual variability for the PD parameters was not characterized using a population analysis. The PK-PD model for the hypotensive effect of LOS was analyzed by a nonlinear regression program, FKDM,10¥ based on ordinary least squares. The differential equations were solved by the Runge-Kutta-Gill method.11¥ Fitting procedures were performed on a PC/AT compatible computer ¤NEC, Tokyo, Japan¥ running under Windows XP with the Fortran compiler Compaq Visual Fortran ¤version 6.1¥. The areas under the effect-time curve ¤AUE¥ were calculated by the linear trapezoidal method.
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
209
PRA-mediated PK-PD Model for Antihypotensive Drugs
A
LOS 5 mg/kg+HCTZ 1 mg/kg LOS 5 mg/kg+HCTZ 5mg/kg
Lk12
Lk21
ALOS1 Lk31
ALOS3
X2
k10Ὁhh
Lk13 Xk in
Plasma LOS Conc. (µ g/mL)
i.v. bolus
LOS 5 mg/kg
100
Pharmacokinetics
ALOS2
Xk12 Xk21
X1
Xkout
S(t)
APRA
Population average for LOS (single)
10
Population average for LOS (coadministration)
1
Rkout
Pkin
CE
LOS 5 mg/kg+HCTZ 10 mg/kg
0.1 0
Hypotensive Effect
2
4
6
8
10
Time (hr)
Pkout Pharmacodynamics
B
HCTZ 5 mg/kg
10
HCTZ 5 mg/kg+LOS 5 mg/kg
Theoretical: Figure 1 represents the PK-PD model for the hypotensive effect after administration of LOS. The PK of LOS in plasma was described by a linear threecompartment open model, as follows: dALOS1 ¼ Lk21 ALOS2 þ Lk31 ALOS3 dt ðLk12 þ Lk13 þ Lk10h Þ ALOS1 ð1Þ dALOS2 ¼ Lk12 ALOS1 Lk21 ALOS2 ð2Þ dt dALOS3 ¼ Lk13 ALOS1 Lk31 ALOS3 ð3Þ dt ALOS1 CLOS1 ¼ ð4Þ LV where ALOS1 is the amount of LOS ¤µg¥ in the central compartment; ALOS2 and ALOS3 are the amount of LOS concentrations in the peripheral compartments; LV is the distribution volume of LOS in the central compartment; Lk12, Lk21, Lk13, Lk31, and Lk10 are the first-order rate constants ¤min%1¥ of LOS; CLOS1 is the plasma LOS concentration ¤µg/mL¥; and at t © 0, ALOS1 © Dose and ALOS2 © ALOS3 © 0. In the PK-PD analysis, the estimates of the population mean were used in the PK parameters, because animals were individually applied in the PK and pharmacological experiments. A hybrid model of an indirect response model and an indirect link model was used to describe a sequence of variations in the PRA and the hypotensive effect after LOS administration. In this study, we focused on the contribution of the PRA to the hypotensive effect of LOS. Then, we assumed the following: a¥ LOS increases the PRA via the accumulative delay compartments of X1 and X2. b¥ The influx rate into the X1/X2 compartment is negligible because LOS influx follows a first-order process but the received mass is negligible compared to the PK model. c¥ Increased PRA decreases blood pressure through the effect compartment CE. The differential equations of the model are as follows:
HCTZ 10 mg/kg
Plasma HCTZ Conc. ( µ g/mL)
Fig. 1. Diagrammatic representation of the PK-PD model for the hypotensive effect
HCTZ 10 mg/kg+LOS 5 mg/kg Population average for HCTZ 5 mg/kg
1
Population average for HCTZ 10 mg/kg
0.1
0.01 0
1
2
3 Time (hr)
4
5
6
Fig. 2. Pharmacokinetic interaction for coadministration of LOS and HCTZ A) The effect of HCTZ in plasma LOS concentration after administration of LOS with and without HCTZ. B) The effect of LOS in plasma HCTZ concentrations after administration of HCTZ with and without LOS. Each symbol indicates an individual animal (n ¦ 3–5).
dCX1 ¼ Xkin CLOS1 Xkout CX1 dt Xk12 CX1 þ Xk21 CX2 dCX2 ¼ Xk12 CX1 Xk21 CX2 dt dAPRA ¼ Rkin SðtÞ Rkout APRA dt REmax CX1 SðtÞ ¼ 1 þ REC50 þ CX1
ð5Þ ð6Þ ð7Þ ð8Þ
where CX1 and CX2 are the concentrations ¤µg/mL¥ in the compartments X1 and X2, respectively; Xk12, Xk21, Xkin, and Xkout are first-order constants ¤min%1¥; APRA is the plasma renin activity ¤ng&Ang I/mL/h¥ in the PRA compartment; REmax is the maximum effect of the PRA ¤ng&Ang I/mL/h¥; REC50 is the LOS concentration ¤µg/ mL¥ producing 50% of REmax; Rkin is the zero-order constant ¤ng&Ang I/mL/h2¥ concerned with the PRA production; and Rkout is the first-order constant ¤min%1¥ concerned with the PRA degradation. Before LOS is administered, the baseline level of APRA is maintained as APRA0 ¤ng&Ang I/mL/h¥, which is determined by Rkout © Rkin&APRA0. The PRA decreases blood pressure as follows:
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
210
Ryosuke SHIMIZU, et al.
dCE ¼ Pkin APRA dt
Pkout CE
ð9Þ
PEmax CE ð10Þ PEC50 þ CE where CE is the value of the PRA ¤ng&Ang I/mL/h¥ in the effect compartment, and Pkin and Pkout are first-order constants ¤min%1¥; PEmax is the maximum hypotensive effect of the PRA ¤mmHg¥; PEC50 is the PRA ¤ng&Ang I/mL/h¥ producing 50% of PEmax; and E0 is the basal level of blood pressure ¤mmHg¥. Blood pressure ¼ E0
Table 2. Population pharmacokinetic parameters of LOS and HCTZ Fixed effect
Random effect
Losartan ¤LOS¥ CLLOS ¤mL/min¥ © 2.54 ' 0.759HT CLLOS-ETA © 0.367 Q3LOS ¤mL/min¥ © 7.38 V1LOS ¤mL¥ © 44.2
SIGMALOS © 0.0847
CV ¤%¥ © 29.1
Hydrochlorothiazide ¤HCTZ¥ CLHCTZ ¤mL/min¥ © 14.7
CLHCTZ-ETA © 0.0210 CV ¤%¥ © 39.3
Q2HCTZ ¤mL/min¥ © 10.3 V1HCTZ ¤mL¥ © 39.3 V2HCTZ ¤mL¥ © 802 SIGMAHCTZ © 0.0404 CV ¤%¥ © 20.1 HT: administration with HCTZ © 1, administration without HCTZ © 0. CLLOS, CLHCTZ: the total clearances for LOS and HCTZ; Q2LOS, Q3LOS, Q2HCTZ: the clearances for LOS and HCTZ for the peripheral compartments; V1LOS, V1HCTZ: the distribution volume of the central compartment for LOS and HCTZ; V2LOS, V3LOS, V2HCTZ: the distribution volume of the peripheral compartment for LOS and HCTZ; CLLOS-ETA, CLHCTZ-ETA: variance of the random effect for CLLOS and CLHCTZ; V1LOS-ETA: variance of the random effect for V1LOS; CV: coefficient of variance; SIGMALOS, SIGMAHCTZ: residual variability for LOS and HCTZ.
200
Individual predicted LOS Conc. (µg/mL)
Population predicted LOS Conc. (µg/mL)
CV ¤%¥ © 69.9
V3LOS ¤mL¥ © 75
200
150
100
50
150
100
50
0
0 0
50
100
150
200
250
0
Observed Losartan Conc. (µg/mL)
50
100
150
200
250
Observed Losartan Conc. (µg/mL)
5 4 3
4
2 1 0 -1 -2
1
3 2 WRES
WRES
V1LOS-ETA © 0.487
V2LOS ¤mL¥ © 210
Results Pharmacokinetic and pharmacological interaction in the combination treatment: After i.v. bolus coadministration of LOS and HCTZ in each individual animal, the plotting points in Figure 2A show the effect of coadministered HCTZ on the time-course of plasma LOS concentration. A two- or three-exponential decline in plasma was observed after administration of LOS alone. Since linearity was shown in these dose-normalized timecourses for LOS ¤5, 10, 20 mg/kg, data not shown¥, a linear three-compartment model was selected to describe the PK of LOS. As shown by covariate adjustment, coadministered HCTZ, as a covariate, significantly and dose-independently decreased the elimination clearance of LOS by about 25%, but did not affect the distribution volume ¤Table 2, solid lines in Fig. 2A¥. The goodness-of-fit of the final model is shown in Figure 3. The plotting points in Figure 2B
0 -1 -2 -3
-3 -4 -5
-4 -5 0
50
100
150
200
Observed Losartan Conc. (µg/mL)
250
CV ¤%¥ © 60.6
Q2LOS ¤mL/min¥ © 0.938
0
200
400
600
Time (min)
Fig. 3. Goodness-of-fit for the final population pharmacokinetic model for LOS Plots are individual data. Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
800
211
PRA-mediated PK-PD Model for Antihypotensive Drugs
represent the effect of LOS in the plasma HCTZ concentration®time course after i.v. bolus coadministration of LOS and HCTZ in each individual animal. The plasma concentration of HCTZ was fitted to a two-compartment model ¤Table 3, solid line in Fig. 2B¥. LOS did not affect
any PK parameters of HCTZ. The time-courses of SBP after i.v. bolus administration of LOS and HCTZ are shown in Figures 4A and 4B, and the dose-dependency of the hypotensive effect in the AUE is shown in Figures 4C and 4D. After i.v. bolus administration of LOS ¤5 mg/kg¥ or HCTZ ¤5 mg/kg¥, SBP decreased slightly and continued for about 72 h ¤Figs. 4A and 4B¥. After coadministration of LOS and HCTZ, SBP is remarkably decreased up to 24 h after dosing and a lower SBP profile is maintained for about 144 h. This interactive enhancement of the hypotensive effect depended on the doses of LOS and HCTZ ¤Figs. 4C and 4D¥. This hypotensive interaction was also synergistic; e.g. AUE was 2,144.4 mmHg&h with a combination of LOS 5 mg/kg and HCTZ 5 mg/kg compared to 662.6 mmHg&h and 521.1 mmHg&h, respectively, for each drug alone. The effect of the combination treatment on the plasma renin activity: The time-course of the PRA after i.v. bolus coadministration of LOS and HCTZ is represented in Figure 5. The PRA temporally increased and then gradually recovered to baseline up to 72 h after administration, although the eliminations of LOS and HCTZ from plasma were rapid. Thus, there was a typical counterclockwise hysteresis between plasma LOS concentrations and the PRAs ¤data not shown¥. Dose-dependent increases in the PRA were observed in LOS and HCTZ coadministration studies ¤Fig. 5¥. PK-PD analysis for PRA enhancement: To describe the time-course of the PRA after administration of LOS with HCTZ, an indirect response model was applied to the PK-PD model for LOS as shown in Figure 1 and Eqs. ¤1¥®¤7¥. However, the compartments X1 and X2 were
Table 3. Pharmacodynamic parameter estimates of the hypotensive effect after coadministration of LOS and HCTZ in rats, and for E-3174 after i.v. bolus administration of LOS in rats Parameters
Estimates
Xk12
¤min%1¥
0.00732 + 0.0168
Xk21
¤min%1¥
0.00123 + 0.000720
Xkout
¤min%1¥
0.0162 + 0.0264
REC50
¤µg/mL¥
0.585 + 0.0717
Rkout
¤h%1¥
3.14 + 26.5
REmax ¤HCTZ 1 mg/kg¥
14.0 + 2.52
REmax ¤HCTZ 5 mg/kg¥
15.7 + 2.36
REmax ¤HCTZ 10 mg/kg¥
21.5 + 3.61
Pkout
¤min%1¥
PEC50
¤ng Ang I/mL/h¥
85.5 + 118
PEmax
mmHg
180 + 208
¤min%1¥
0.0228 + 0.0016
0.00244 + 0.000440
for E-3174 aa
%1
a
¤min ¥
0.0110 + 0.00285
Mk21a
¤min%1¥
0.00590 + 0.00211
Mkouta
¤min%1¥
0.0219 + 0.00205
Mk12
a
Parameters of the model for E-3174 ¤see Fig. 11¥. Data represent the computer-fitted value + SD.
210
A
200 190 180 170
LOS 5 mg/kg
160
LOS 5 mg/kg+HCTZ 1 mg/kg
150
LOS 5 mg/kg+HCTZ 5 mg/kg LOS 5 mg/kg+HCTZ 10 mg/kg
140 0
24
48
72
96
120
Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
210
190 180 170 HCTZ 5 mg/kg
160
HCTZ 5 mg/kg+LOS 1 mg/kg
150
HCTZ 5 mg/kg+LOS 5 mg/kg HCTZ 5 mg/kg+LOS 10 mg/kg
140 0
144
Time (hr) 4500
C
4000
4000
3500
3500
AUE (mmHg hr)
AUE (mmHg
hr)
4500
3000 2500 2000 1500
B
200
24
D
48
72
96
120
144
Time (hr)
3000 2500 2000 1500
1000
1000
500
500 0
0 LOS 5 mg/kg
LOS 5 mg/kg +HCTZ 1 mg/kg
Dose
LOS 5 mg/kg +HCTZ 5 mg/kg
LOS 5 mg/kg +HCTZ 10 mg/kg
HCTZ 5 mg/kg
LOS 1 mg/kg +HCTZ 5 mg/kg
LOS 5 mg/kg +HCTZ 5 mg/kg
LOS 10 mg/kg +HCTZ 5 mg/kg
Dose
Fig. 4. Changes in blood pressure (A, B) and AUE (C, D) after coadministration of LOS and HCTZ Values represent the mean + S.E.M. (n ¦ 3–4). Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
212
Ryosuke SHIMIZU, et al.
required to explain the delayed profiles of the PRA against plasma LOS concentration profiles. The solid lines, shown in Figure 6, represent the results of fitting the observed data after coadministration of LOS and HCTZ to the model described in the theoretical section. The regression curves fit the observed data well. The estimated PD parameters are
listed in Table 3. REmax values increased with the dose of HCTZ ¤Fig. 7¥. PK-PD analysis for the PRA-mediated hypotensive effect: The PK-PD analysis for the PRA-mediated hypotensive effect after i.v. bolus administration of LOS was carried out according to Eqs. ¤1¥®¤10¥. Since there was a typical counter-clockwise hysteresis between the PRA level and the hypotensive effect, an additional effect compartment was introduced between the PRA and the hypotensive effect model. The solid lines shown in Figure 8 represent the theoretical hypotensive effect and fit the observed data well. The estimated PD parameters are listed in Table 2. To evaluate the validity of the model, we simulated the timecourses of the PRA and SBP after i.v. bolus administration of 5 mg/kg LOS and 8 mg/kg HCTZ. In this simulation, we
Plasma Renin Activity (ng Ang 㻌㻌 mL/hr)
70 60 LOS 5 mg/kg
50
LOS 5 mg/kg+HCTZ 1 mg/kg 40
LOS 5 mg/kg+HCTZ 5 mg/kg LOS 5 mg/kg+HCTZ 10 mg/kg
30 20 10 0 24
0
48
72
96
120
25
144
Time (hr)
20 REmax value
Plasma Renin Activity (ng Ang 㻌 mL/hr)
70 60 50
HCTZ 5 mg/kg HCTZ 5 mg/kg+LOS 1 mg/kg
40
HCTZ 5 mg/kg+LOS 5 mg/kg HCTZ 5 mg/kg+LOS 10 mg/kg
15 10
y = 0.8493x + 12.53 R2 = 0.9461
5
30 20
0 0
10
4
8
12
Hydrochlorothiazide Dose (mg/kg) 0 0
24
48
72
96
120
144
Fig. 7. Relationship between the REmax value and the dose of HCTZ for the hypotensive effect after coadministration of LOS with HCTZ The solid line represents a regression line. Plots represent the model-calculated value.
Time (hr)
Fig. 5. Time-course of plasma renin activity after coadministration of LOS and HCTZ Plots represent the mean + S.E.M. (n ¦ 2–3)
LOS 5 mg/kg+HCTZ 5 mg/kg
60
60
60
50 40 30 20 10
Plasma Renin Activity (ng Ang 䊠/mL/hr)
70
0
50 40 30 20 10 0
-24
0
24
48
72
96
0
24
48
72
96
120 144
Time (hr)
HCTZ 5 mg/kg+LOS 1 mg/kg
60
Plasma Renin Activity (ng Ang mL/hr)
70
60
40 30 20 10 0
40 30 20 10 -24
0
24
48
72
96
120
Time (hr)
HCTZ 5 mg/kg+LOS 10 mg/kg
70
50
50
0 -24
120 144
Time (hr)
Plasma Renin Activity (ng Ang䊠 mL/hr)
LOS 5 mg/kg+HCTZ 10 mg/kg
70 Plasma Renin Activity (ng Ang mL/hr)
Plasma Renin Activity (ng Ang /mL/hr)
LOS 5 mg/kg+HCTZ 1 mg/kg 70
50 40 30 20 10 0
-24
0
24
48
72
Time (hr)
96
120 144
-24
0
24
48
72
96
120 144
Time (hr)
Fig. 6. Comparison of the model-fitted value with the observed data for the PRA after coadministration of LOS and HCTZ Solid lines are the model-fitted profiles. Plots represent the mean + S.E.M. (n ¦ 2–3). Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
213
PRA-mediated PK-PD Model for Antihypotensive Drugs LOS 5 mg/kg+HCTZ 1 mg/kg
LOS 5 mg/kg+HCTZ 5 mg/kg
200 190 180 170 160 150 140 0
24
48
72
96
120
200 190 180 170 160 150 140
144
0
24
48
Time (hr)
200 190 180 170 160 150 140 48
72
96
120
144
96
120
210 200 190 180 170 160 150 140 0
24
48
72
96
120
144
Time (hr)
HCTZ 5 mg/kg+LOS 10 mg/kg Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
HCTZ 5 mg/kg+LOS 1 mg/kg
24
72 Time (hr)
210
0
LOS 5 mg/kg+HCTZ 10 mg/kg
210
Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
Systolic Blood Pressure (mmHg)
210
144
210 200 190 180 170 160 150 140 0
24
Time (hr)
48
72
96
120
144
Time (hr)
Systolic Blood Pressure (mmHg)
Fig. 8. Comparison of the model-fitted value with the observed data for the hypotensive effect after coadministration of LOS and HCTZ Solid lines represent the model-fitted values. Plots are the observed data and the mean + S.E.M. (n ¦ 3–4).
200 190 180 170 160 150 140 0
24
48
72
96
120
144
168
Time (hr)
Fig. 9. Comparison of the model-simulated SBP with the observed data after i.v. bolus coadministration of 5 mg/kg LOS with 8 mg/kg HCTZ The solid line represents the simulated values and the plots are the mean of observed data + S.E.M. (n ¦ 3).
calculated the value of REmax based on the correlation between REmax and the dose of HCTZ ¤Fig. 7¥ and used 19.09 mmHg. The other parameters were used from Table 2. The solid line shown in Figure 9 represents the model-predicted profiles of SBP and explain the observed data well. Discussion Pharmacokinetic and Pharmacodynamic interaction: When HCTZ is coadministered with LOS, no effect is observed for the elimination of HCTZ alone, but the total elimination clearance of LOS decreased by about 25%. ¤Table 2, Fig. 2¥. LOS is primarily metabolized by the hepatic cytochrome P450 ¤CYP2C9¥ in rats, and then more than 94% of the dose is excreted in the feces7¥; however, most of the administered HCTZ was excreted in the urine as
an intact drug without any metabolites.12¥ These reports indicate that coadministered LOS and HCTZ do not compete with each other for their metabolism or excretion. The plasma protein binding ratio of LOS is reported to be more than 99%,5¥ and that of HCTZ is about 22%.13¥ Thus, HCTZ would not affect the protein binding of LOS. On the other hand, 10 mg/kg HCTZ significantly increased the urine volume up to 4 h after oral treatment for 5 days in SHRs.14¥ Change in the urine volume was not investigated before and after the drug administration in the present study. However, decrease of the body fluid by the diuretic effect of HCTZ ¤1®10 mg/kg¥ might concentrate the level of LOS in plasma, and then the apparent clearance of LOS would decrease. This speculation also supports the invariable PK of HCTZ after coadministration with LOS. The interactive hypotensive effect was observed after administration of LOS and HCTZ. It was greater than the aggregate of their respective effects ¤Fig. 4¥, even after considering the PK interaction by HCTZ. The hypotensive interaction depended on the doses of LOS and HCTZ, which means there is a synergistic effect. As shown in Figure 10, LOS blocks the angiotensin type 1 receptor ¤AT1R¥. This interaction of LOS with the AT1R suppresses vasoconstriction and activation of the AT2R with the elevated Ang II, and consequently yields a decreased blood pressure.15¥ As a result of blocking AT1R, plasma renin is also activated through suppressing the negative feedback mechanism for the renin-angiotensin system, and subsequently elevates Ang I, Ang II, and Ang ¤1®7¥ levels. Continuous increases of the PRA and Ang II level in plasma were observed after administration of valsartan.16¥ Ang ¤1®7¥ releases nitric oxide and decreases blood pressure after binding to the Mas receptor.17¥ Collister
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
214
Ryosuke SHIMIZU, et al.
E3174
Hydrochlorothiazide
ALOS2
Angiotensinogen
M2
Plasma Renin activity
Lk12
Angiotensin
Lk21
ACE
Negative Feedback
a
i.v. bolus Angiotensin ACE2
Hypertensive effect etc.
AT 2 receptor
K10
Mkout
ALOS3
Hypotensive effect etc.
Plasma E3174 Conc. (mg/mL)
Fig. 10. Effect of LOS and HCTZ in the renin-angiotensin system cascade
and Hendel18¥ reported that the increase of Ang ¤1®7¥ in plasma develops the hypotensive effect. Furthermore, the PRA increased after administration of HCTZ.14¥ In our study, the PRA continuously increased for 72 h after LOS administration ¤Fig. 5¥. HCTZ administered with LOS further raised the PRA synergistically. These results suggest that PRA is a pharmacological mediator for PK and the hypotensive effect. PRA-mediated PK-PD model: The hypotensive effect after i.v. coadministration of LOS and HCTZ was described well using the PK-PD model with PRA enhancement ¤Fig. 6¥. In this model, a hypothetical maximum effect for enhancement of the PRA ¤REmax¥ increased depending on the dose of HCTZ ¤Fig. 7¥. The delay compartment X1/X2 was necessary to describe the change in the PRA, because of the delay for activation of plasma renin against the plasma LOS concentration profile. LOS has an active metabolite, E-3174, which shows a hypotensive effect. E-3174 also has a long half-life and a strong affinity compared to LOS. We evaluated the possibility that the X1/X2 compartment represented the PK of E-3174 using the reported PK data.19¥ Figure 11 represents the PK model for E-3174 that was modified based on the previous model ¤Fig. 1¥. The PK of E3174 in plasma is described by Eqs ¤11¥®¤14¥, as follows: dALOS1 ¼ Lk21 ALOS2 þ Lk31 ALOS3 dt ðLk12 þ Lk13 þ Lk10 Þ ALOS1 ð11Þ dM1 ¼ a CLOS1 ðMk12 þ Mkout Þ M1 þ Mk21 M2 dt ð12Þ dM2 ¼ Mk12 M1 Mk21 M2 ð13Þ dt where M1 and M2 are the concentration of E-3174 ¤µg/mL¥; and Mk12, Mk21, Mkout, and a are the first-order rate constants ¤min%1¥ of E-3174. These parameter values were estimated using the values shown in Table 2 and the reported plasma concentrations of E-3174 after i.v. bolus administration of LOS without HCTZ. The regression
Lk13
Lk31
Mas receptor
Mk21
M1
ALOS1
Angiotensin(1-7)
Losartan AT 1 receptor
Mk12
10
1
0.1 0
2
4
6
8
10
12
Time (hr)
Fig. 11. Pharmacokinetic model for E-3174 after i.v. bolus administration of LOS (upper) and the time-course profile of plasma E-3174 concentration (lower) M1, M2: the central and peripheral compartment for E-3174; Mk12, Mk21, a, Mkout: the first-order constants.
curves fit the literature data well ¤Fig. 11, lower¥, and the estimated PK parameters are listed in Table 3. However, as compared with the parameters Xk12, Xk21 and Xkout in Table 3, these parameters indicated a faster elimination of the E-3174 profile in plasma. That is, the delay compartments X1/X2 do not represent the E-3174 profile only. These compartments also include other mediators, such as unknown metabolites and endogenous factors concerned with the PRA production. In addition, this result indicates the validity of the present model based on plasma LOS concentration, and the pharmacokinetic data of E-3174 is not necessary for the estimation of the interactive hypotensive effect. Furthermore, an additive effect compartment ¤CE¥ was required to link the PRA and the hypotensive effect. PRA raises the plasma Ang II and Ang ¤1®7¥ concentrations. Then these Angs bind to the AT2R and the MasR ¤Fig. 10¥. As a result, a decrease in blood pressure is observed. Therefore, the delays described by the effect compartment may represent the transit time required for the sequence of these pharmacological events of the renin-angiotensin system. The HCTZ dose-dependency of the maximum hypotensive effect also reasonably predicted the observed data. In conclusion, the interactive hypotensive effect is shown after i.v. bolus coadministration of LOS and HCTZ, and the
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)
PRA-mediated PK-PD Model for Antihypotensive Drugs
relationship between plasma LOS concentration and blood pressure is described using the PK-PD model. The model quantitatively reveals that HCTZ principally enhances the effect of LOS dose-dependently through an increase of the PRA. This relationship can estimate the hypotensive effect for combination treatment using different doses and kinds of ARBs and thiazide diuretics.
9¥
10¥
References 1¥ Staessen, J. A., Gasowski, J., Wang, J. G. et al.: Risks of untreated and treated isolated systolic hypertension in the elderly: metaanalysis of outcome trials. Lancet, 355¤9207¥: 865®872 ¤2000¥. 2¥ Ohkubo, T., Obara, T., Funahashi, J. et al.: J-HOME Study Group., Control of blood pressure as measured at home and office, and comparison with physiciansö assessment of control among treated hypertensive patients in Japan: First Report of the Japan Home versus Office Blood Pressure Measurement Evaluation ¤J-HOME¥ study. Hypertens. Res., 27¤10¥: 755®763 ¤2004¥. 3¥ Nakashima, M., Uematsu, T., Kosuge, K. and Kanamaru, M.: Pilot study of the uricosuric effect of DuP-753, a new angiotensin II receptor antagonist, in healthy subjects. Eur. J. Clin. Pharmacol., 42: 333®335 ¤1992¥. 4¥ Villamil, A., Chrysant, S. G., Calhoun, D., Schober, B., Hsu, H., Matrisciano-Dimichino, L. and Zhang, J.: Renin inhibition with aliskiren provides additive antihypertensive efficacy when used in combination with hydrochlorothiazide. J. Hypertens., 25¤1¥: 217® 226 ¤2007¥. 5¥ Walters, P. E., Gaspari, T. A. and Widdop, R. E.: Angiotensin¤1-7¥ acts as a vasodepressor agent via angiotensin II type 2 receptors in conscious rats. Hypertension, 45: 960®966 ¤2005¥. 6¥ Kim, G. H., Na, K. Y., Kim, S. Y., Joo, K. W., Oh, Y. K., Chae, S. W., Endou, H. and Han, J. S.: Up-regulation of organic anion transporter 1 protein is induced by chronic furosemide or hydrochlorothiazide infusion in rat kidney. Nephrol. Dial. Transplant., 18¤8¥: 1505®1511 ¤2003¥. 7¥ Takayama, F., Saito, K., Yoshinaga, T., Morita, M., Hata, S., Esumi, Y., Jin, Y. and Okamura, Y.: Metabolic fate of losartan, a new angiotensin II receptor antagonist ¤1¥: absorption, distribution, metabolism and excretion after single administration in rats. Xenobiotic Metabolism and Disposition, 10: 223®243 ¤1995¥. 8¥ Farthing, D., Sica, D., Fakhry, I., Pedro, A. and Gehr, T. W.: Simple high-performance liquid chromatographic method for determination of losartan and E-3174 metabolite in human plasma,
11¥ 12¥ 13¥ 14¥
15¥ 16¥
17¥
18¥ 19¥
215
urine and dialysate. J. Chromatogr. B Biomed. Sci. Appl., 704¤1-2¥: 374®378 ¤1997¥. Li, H., Wang, Y., Tang, Y., Wang, J., Zhao, L. and Gu, J.: A liquid chromatography/tandem mass spectrometry method for the simultaneous quantification of valsartan and hydrochlorothiazide in human plasma. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 852¤1-2¥: 436®442 ¤2007¥. Lu, W., Endoh, M., Katayama, K., Kakemi, M. and Koizumi, T.: Pharmacokinetic and pharmacodynamic studies of piretanide in rabbits. I. Effect of different hydrated conditions. J. Pharmacobiodyn., 10: 356®363 ¤1987¥. Nelson, W., Tong, Y. L., Lee, J.-K. and Halberg, F.: Methods for cosinor-rhythmometry. Chronobiologia, 6: 305®323 ¤1979¥. Beermann, B., Groschinsky-Grinf, M. and Rosen, A.: Absorption, metabolism, and excretion of hydrochlorothiazide. Clin. Pharmacol. Ther., 19¤5 Pt 1¥: 531®537 ¤1976¥. Package insert of Preminent¬ tablet, ver. 9, Tokyo, MSD K.K., 2011. Wienen, W. and Schierok, H. J.: Effects of telmisartan, hydrochlorothiazide and their combination on blood pressure and renal excretory parameters in spontaneously hypertensive rats. J. Renin Angiotensin Aldosterone Syst., 2¤2¥: 123®128 ¤2001¥. Walters, P. E., Gaspari, T. A. and Widdop, R. E.: Angiotensin¤1-7¥ Acts as a Vasodepressor Agent Via Angiotensin II Type 2 Receptors in Conscious Rats. Hypertension, 45: 960®966 ¤2005¥. Azizi, M., Ménard, J., Bissery, A., Guyenne, T. T., Bura-Rivière, A., Vaidyanathan, S. and Camisasca, R. P.: Pharmacologic demonstration of the synergistic effects of a combination of the renin inhibitor aliskiren and the AT1 receptor antagonist valsartan on the angiotensin II-renin feedback interruption. J. Am. Soc. Nephrol., 15¤12¥: 3126®3133 ¤2004¥. Sampaio, W. O., Souza dos Santos, R. A., Faria-Silva, R., da Mata Machado, L. T., Schiffrin, E. and Touyz, R. M.: Angiotensin-¤1-7¥ through receptor Mas mediates endothelial nitric oxide synthase activation via Akt-dependent pathways. Hypertension, 49¤1¥: 185®192 ¤2007¥. Collister, J. P. and Hendel, M. D.: The role of Ang¤1-7¥ in mediating the chronic hypotensive effects of losartan in normal rats. J. Renin Angiotensin Aldosterone Syst., 4: 176®179 ¤2003¥. Moon, C. H., Lee, H. J., Jung, Y. S., Lee, S. H. and Baik, E. J.: Pharmacokinetics of losartan and its metabolite, EXP3174, after intravenous and oral administration of losartan to rats with streptozotocin-induced diabetes mellitus. Res. Commun. Mol. Pathol. Pharmacol., 101¤2¥: 147®158 ¤1998¥.
Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX)