Pharmacokinetics and pharmacodynamics of trimazosin in man

Pharmacokinetics and pharmacodynamics trimazosin in man

of

In this study the pharmacokinetics of trimatosin and its major metabolite, CP 23445, were determined, as were also the effects on blood pressure, heart rate, and peripheral vascular alpha-l receptors. Trimarosin was administered intravenously (100 mg) and orally (200 mg) to six healthy volunteer subjects (24 to 39 years of age). Blood samples were withdrawn from an indwelling intravenous cannula, and supine and erect blood pressure and heart rate recordings were made at frequent intervals over 12 hours. Concentrations of trimazosin and CP 23445 in whole blood were measured by a sensitive and specific high-pressure liquid chromatography-fluorescence assay developed in our laboratory. Both drug and metabolite could be measured after oral and intravenous administration. Pharmacokinetic parameters were obtained by computer-assisted, nonlinear, least-squares-fitting regression analysis. The pharmacokinetic profile of trimazosin was best described by a two-compartment model. The mean (*SD) terminal elimination half-life of trimarosin was 2.73 (20.90) hours. The metabolite concentrations were then incorporated into the model and a simultaneous fit of drug and metabolite carried out. This did not alter the pharmacokinetic profile of the parent drug but did allow the elimination half-life of the metabolite to be calculated as 1.47 + 0.65 hours. The bioavailability of oral trimazosin was 61 * 28%. Erect blood pressure was reduced for at least 6 hours after both oral and intravenous administration. The fall in pressure was associated with alpha-l adrenoceptor blockade as determined from pressor sensitivity to intravenous phenylephrine (pressor dose required to raise mean pressure 20 mm Hg [PD,,] was 1.96 ? 1.3 pg/kg/min after placebo and 3.82 + 2.1 Ag/kg/min after intravenous trimazosin). (AM HEART J 106:1222, 1983.)

John L. Reid, D.M., Peter A. Meredith, Glasgow, Scotland

Ph.D., and Henry L. Elliott,

Trimazosin (2-hydroxy-2-methylpropyl4-(4 amino6,7,8-trimethoxy-2-quinazolinyl)-1-piperazine carboxylic acid) is a quinazoline derivative structurally related to prazosin (Fig. 1). Like prazosin, trimazosin has been reported to lower blood pressure in animals’~ 2 and humans.3-6 Preliminary reports suggest that trimazosin has alpha-l-adrenoceptorblocking properties19 ’ like prazosin,7 but other mechanisms of hypotensive action may also be present in animals.’ Trimazosin is extensively metabolized in the liver, and its major metabolite (CP 23445) in humans is an alkyl-hydroxylated derivative of the parent drug (Fig. 1). The pharmacokinetics of oral and intravenous trimazosin were studied in healthy volunteer subjects. Whole blood concentrations of trimazosin and metabolite were measured simultaneously. In addi-

From the Clinical Medica, University Reprint Stobhill

1222

Pharmacology of Glasgow.

requests: J. L. General Hospital,

Research

Unit,

Department

Reid, D.M., Department of Glasgow, G21 3UW, Scotland.

Materia

of Materia Medica,

M.D.

tion, blood pressure, heart rate, and vascular alphaadrenoceptor responsiveness were determined to permit investigation into the relationship between concentrations of drug and metbolite and pharmacologic effect. METHODS

The study was performed in six healthy, normotensive men aged 24 to 39 years. The subjects gave informed consent to a protocol approved by the Ethical Review Committee of the Northern District of the Greater Glasgow Health Board. The definitive study followed doseranging ones in which intravenous trimazosin (10,25, and 50 mg) was studied. On three occasions at least 7 days apart, each subject received either a single dose of 100 mg trimazosin intravenously, 200 mg orally, or an intravenous injection of 0.9% saline vehicle. Blood samples (5 ml) were collected on 10 occasions after oral administration and on 16 occasions after intravenous administration for simultaneous measurement of whole blood trimazosin and CP 23445 by a sensitive specific assay based on high-pressure liquid chromatography (HPLC) with fluorescence detection, a method developed in our laboratory.* Blood pressure was measured with a Roche Arteriosonde semiautomatic sphygmoma-

Volume Number5,

106 Part2

nometer (Model 1225) and heart rate from precordial ECG leadsdisplayed on a Grasspolygraph. Blood pressure wasmeasuredafter subjectshad rested at least 10 minutes in the supine position and again 5 minutes after they had been standing erect. These measurementswere made before drug administration and at intervals up to 24 hours thereafter. If subjects developed orthostatic symptoms while standing or systolic blood pressurefell below 80 mm Hg, they were advised to lie down. Alpha-adrenoceptor responsivenesswas assessedafter intravenous trimazosin or placebo by constructing pressor dose-responsecurves to intravenous phenylephrine (0.5 to 10 pg/kg/min) and noradrenaline (0.01 to 1.0 pg/kg/min).g Each agonist wasgiven in the sameorder for each subject, at the sametime after administration, between 2.25 and 3.5 hours, by continuous infusions of increasingdosesby a Braun Perfusor pump for 5 minutes at each doselevel. At least four doselevels were studied for each agonist. Blood pressureand heart rate were measuredevery minute and the dosesindividually titrated to give a rise of at least 30 mm Hg but not more than 45 mm Hg in systolic blood pressure. Blood pressure (systolic, diastolic, and mean pressure) for each dose level was calculated and the dose-responsecurves analyzed by using all data points collected and a quadratic transformation of the results as recently described.I” From the quadratic function, the pressordose of agonist required to increase systolic, diastolic, or mean pressureby 20 mm Hg (PDQn)was calculated. Concentrations of trimazosin and its major metabolite, CP 23445,were determined in whole blood by HPLC with fluorescencedetection,” with the useof a simple and rapid method of sample preparation based on an ether and acid-back extraction. Trimazosin, CP 23445,and internal standard (doxazosin) were chromatographed by reverse phaseasion pairs with pentane sulfonic acid. The analytes are detected by native fluorescencewith excitation at 254 nm and with the emissionspectrum monitored at 370 nm. A typical chromatogram of drug metabolite and internal standard is shown in Fig. 2, along with the chromatogram of 1 ml control blood taken through the extraction procedure. The method is sensitive and reproducible with limits of detection of 1 rig/ml for trimazosin and 0.5 rig/ml for CP 23445;coefficients of variation are 5.2% and 6.2% , respectively. Statistical analysis. The pharmacokinetic data were analyzed by using a two-compartment model for the intravenous data (Cp(t) = Ae+’ + Be-‘j’t)), a one-compartment model for the oral data (Cp(t) = A(e-B’t-t~ag) - e-ka’t-tlag’), and a computer-assistedleast-squaresfitting. Blood pressure and heart rate are expressed as mean (+SD), and time effects are assessedby using repeatedmeasuresanalysisof variance. Pressoramine infusion data were analyzed after quadratic transformation by using all data points. loModeling of drug and metabolite concentration and hypotensive effect was undertaken as previously described.“, I? For many drugs the pharmacodynamic profile following

Clinical

pharmacology

of trimazosin

1223

CH30

CHqO

INH2

Trimazosin

CP 23.445

CHJO WH*

Prazosin

Fig. 1. Structure of trimazosin ancl its metabolite CP 23445in relation to prazosin.

intravenous administration is not in phase with the amount of drug in any of the predetermined kinetic compartments (central, peripheral, or metabolite). Thus the responseto the drug, as assessedby a fall in blood pressure,is not directly mediated by the amount of drug in one of the three compartments defined by the appropriate kinetic model. A novel approach that recognizes and models the fact that effect is not in phase with drug and metabolite concentration in any particular compartment hasbeen described.” This approach involves extending the pharmacokinetic models by an explicitly defined “effect” compartment in such a way that this modification does not influence the pharmacokinetic parameters defined by the original model. The effect is then described asa function of drug concentration in the “effect compartment.” Thus, with the pharmacokinetics defined, the effect model parameters are determined by fitting the concentration of drug in the effect compartment versus measuredeffect. RESULTS Whole

blood

concentration

of trimazosin

and

CP

23445. The pharmacokinetic profile of trimazosin following oral and intravenous administration is

shown in Figs. 3 and 4. The mean elimination half-life was 156 + 49 minutes after oral and 184 -+ 76 minutes after intravenous administration. There was no significant difference between these values. CP 23445 levels increased after oral and intravenous administration to amounts comparable to that of free drug. Expansion of the two-compartment model to include a discrete metabolite compartment

November,

1224

Reid, Meredith,

American

and Elliott

1983

Heart

Journal

i.s.

CP tr I

nject

Jn.ied

I

16

0

8

I

24

16

8

0

time IminI 2. HPLC separation and fluorescencedetection of trimazosin (tr), CP 23445 (cp), and the internal standard (i.s.), doxazosin. Fig.

w

Trimarosin

O-O

Trimozosin

12--c\ CP 23445

I

I loo

I 200

I 300 Time

I 400 (min

1 500

I 000

r 700

I

3. Mean blood concentration of trimazosin and CP 23445after oral administration of 200 mg trimazosin in six normotensive subjects. Fig.

allows simultaneous fitting of the drug and the metabolite. The elimination half-life of CP 23445 was estimated to be 93 + 29 minutes and 71 rf: 20 minutes after intravenous and oral dosing, respectively. The other derived pharmacokinetic parameters are summarized in Table I. Clearance of trimazosin calculated from intravenous data was 66.1 k 28.3 ml/min, and bioavailability was 61 +- 28%. Bldod pressure and heart rate. There were no signif-

loo

I

I

I

I

I

I

200

300

400

500

600

700

Time

(min

)

Fig. 4. Mean blood concentration of trimazosin and CP

23445after intravenous injection of 100 mg trimazosin in six normotensive subjects.

icant changes in supine systolic or diastolic blood

pressure or heart rate after oral or intravenous trimazosin in normotensive volunteers. However, both routes of administration

led to significant falls

0, < 0.05) in standing blood pressure at 2 and 5 minutes, and heart rate increased (p < 0.05). Erect systolic blood pressure and heart rate after 5 minutes’ standing are shown in Fig. 5. The effects on blood pressure and heart rate were prominent at

Volume

106

Number

5. Part 2

Clinical

pharmacology

of trimazosin

1225

Svs tal ic ‘201a~ o.rr ^

c--a prazosin 0 oplacebo Heart Rate [beats/ mtnl

-‘s-d 0

2

4

6

8

--?4

Time I hl 5. Effect of intravenous trimazosin, 100mg, or placebo on systolic blood pressureand heart rate in six normotensive subjectsafter standing erect for 5 minutes. For comparison, the effects of intravenous prazosin, 1 mg in a similar study under similar conditions are shown. (Prazosin data from Elliott HL, Sumner DJ, McLean K, Reid JL: Effect of age on the responsivenessof vascular alpha-adrenoceptorsin mqn. J Cardiovasc Pharmacol 4:388, 1982.) Fig.

Table

1. Summary of pharmacokinetic parameters from oral and intravenous trimazosin studies Trimazosin

Pt 4’2 *

(min) Subject

Hioavailability f9cl.J

Clearance (mllmin)

CP 23445 I&t (mini

‘/‘if

Oral

IV

Oral

IV

1 2 3 4 6 6

90 27 18 88 32 51

99.6 77.9 90.3 56.8 24.7 47.3

136 103 102 178 210 204

178 136 112 122 266 288

45 61 59 91 71 95

43 a3 103 86 124 117

Mean ? SD

61 A 28

66.1 + 28.3

156 + 49

184 _t 76

71 * 20

93 + 29

*fit% = drug terminal elimination tK,,t’% = metabolite elimination

half-life. half-life.

less than 1 hour and between 4 and 6 hours after administration. The figure also shows blood pressure on a control day and, for comparison, changes in blood pressure and heart rate after 1 mg intravenous prazosin from a similar study.s Pressor responses to alpha-adrenoceptor antagonists. Log dose-response curves were constructed for

systolic, diastolic, and mean arterial pressure and were analyzed after quadratic transformation to determine the PD,, for amine. Trimazosin caused a parallel shift to the right of the pressor response to phenylephrine but not to noradrenaline, when responses 2.5 to 3.5 hours after trimazosin were compared with a placebo treatment day. The PD,, for phenylephrine was significantly greater (p < 0.05) after trimazosin than after placebo, whether systolic, diastolic, or mean pressure was measured (Fig. 6). The mean dose ratio (PD,, after

trimazosin/PD,, after placebo) was 2.38 -t 1.8 for phenylephrine. In contrast, although there was some interindividual variation, the mean PD, for noradrenaline was not altered by this dose of trimazosin under the conditions of this study (Fig. 7). Further, there was no change in systolic or diastolic pressure PD, after noradrenaline. Concentration-effect relationships. The parameters derived from fitting data to the effect model described previously are as follows: (1) slope, which is the slope of the equation for effect versus drug concentration and therefore a measure of the drug sensitivity, and (2) K,, a first-order rate constant, which characterizes the discrepancy between the blood concentration and effect time profiles and thus is a measure of the drug’s duration of action. In the six subjects receiving intravenous trimazosin, the effect profile (represented by the placebo-

1226

Reid, Meredith, and Elliott Control

American

Trimazosin

6. Pressor responsivenessto phenylephrine 2.5 to 3.5 hours after intravenous trimazosin, 100 mg compared to the sameperiod after vehicle saline control. Individual results for PD, (pg/kg/min) are shown, together with the mean -t SD (p < 0.05) when the two groups were compared with paired Student’s t test. Fig.

fall in systolic blood pressure after 5 minutes’ standing) was fitted, as assessed by the general linear test, by the effect model combining both drug and metabolite (Fig. 8). corrected

DISCUSSION

This investigation of normotensive subjects confirms that trimazosin is well absorbed after oral administration and that metabolism to the alkylhydroxylated metabolite CP 23445 is an important route of elimination. The pharmacokinetic parameters reveal a similar elimination half-life of parent drug after oral and intravenous administration. A similar half-life (3.4 hours) after intravenous administration has recently been reported in another study.13 The metabolite is found in equivalent molar amounts to the unchanged drug in plasma, but it has a shorter elimination half-life, making it unlikely that the metabolite will accumulate during longterm administration. Trimazosin had little effect on supine blood pressure in normotensive volunteers but did have a modest action on standing blood pressure associated with a tachycardia. The latter presumably reflects a baroreceptor reflex-mediated response to the fall in pressure. However, no subject experienced marked hypotension and none had orthostatic symptoms. After intravenous administration, the fall in blood pressure lasted for 6 hours. This is in contrast to

Control

November, 1983 Heart Journal

Trimazosin

Fig. 7. Pressorresponsiveness to noradrenaline 2.5 to 3.5 hours after intravenous trimazosin, 100 mg, compared to the same period after vehicle saline control. Individual results for PD,, (pg/kg/min) are shown, together with the mean?SD. There was no significant difference between the two groups when compared by paired Student’s t test.

experience with prazosin in similar studies under the same conditions?, l4 in which the fall in blood pressure lasts only 2 to 3 hours and symptomatic hypotension may occur. Trimazosin increased the dose of phenylephrine but not of noradrenaline required to raise mean arterial pressure by 20 mm Hg. This suggests that trimazosin has selective alpha-1-adrenoceptor antagonist properties. There is no evidence of alpha2-receptor antagonist activity, Recently, several groups, including our ~wr-r,l~-~~have demonstrated in animals, in vitro and in vivo, that alpha-2 receptors situated at postjunctional sites on vascular smooth muscle may contribute to the regulation of peripheral vascular resistance. The results of pressor amine responses in the present study are consistent with the existence of postjunctional vascular alpha-2 adrenoceptors in man. These receptors would be activated by noradrenaline or adrenaline and do not appear to be blocked by trimazosin. Preliminary experiments in conscious rabbits indicate that trimazosin, like prazosin, does not alter the acute peripheral alpha-2-mediated pressor response to intravenous guanabenz (Vincent J, Reid JL: unpublished observations). However, the present experiments do not establish that trimazosin exerts its hemodynamic actions solely by alpha-l-receptor blockade. Alpha-l blockade was relatively modest (dose ratio 2.33) and noted at a time when there had

Volume Number

106 5. Part 2

Clinical

2 z 2 x a”

pharmacology

of

trimazosin

1227

-15 -

-2o. -25

.

---

drug

-

drug

+metabolite

-

0

1 200

t 400 Time

t 600

I 600

(mini

Fig. 8. Concentration-effect relationship in a representative subject by using systolic blood pressure after 5 minutes’ standing, corrected for changes on placebo day. The full-effect model, including metabolite, gave a signifi&ntly better fit. -

been little evidence of circulatory effects of trimazosin (2.5 to 3.5 hours). The effects on blood pressure and heart rate were observed for more than 6 hours, when not only had plasma levels of trimazosin fallen to low levels but also alpha-receptor blockade might be expected to be falling off. These observations raiser the possibility either that trimazosin has important actions on blood pressure control in addition to alpha blockade or that it forms an active metabolite, which contributes to the hemodynamic actions either by augmenting alpha blockade or by another mechanism as yet undefined. CP 23445 is a possible candidate for the role of active metabolite. It has been reported to show some hypotensive activity in animals (Pfizer Central Research: personal communication) and is present in relatively large amounts after administration of trimazosin in humans. With the use of modeling techniques to investigate the relationship between drug concentration and effect when the two are not in phase,“, l2 we have investigated the relationship between drug level and hypotensive effect for trimazosinzO and for prazosin and related quinazolines.21 When a contribution from the metabolite, as well as from trimazosin, was included in the model, there was a significantly better “fit” as assessed by the general linear test than when the parent drug levels alone were examined. This analysis suggests that CP 23445 is indeed an “active metabolite” in humans and that its effects contribute to the action of trimazosin in humans.

fall the

The mechanism of the postulated hypotensive effect of CP 23445 remains to be established. CP 23445 does not compete for specific binding sites on rabbit brain membranes with either selective alpha-l radioligands (3H-prazosin) or alpha-2 ligands (3H-clonidine), suggesting that the metabolite has a low affinity for alpha adrenoceptors (Vincent J, Hamilton CA, Reid JL: unpublished observations). The metabolite may have peripheral vasodilator effects by nonspecific mechanisms that are as yet undefined. Conclusions. Trimazosin is an antihypertensive drug with a relatively long duration of action. It is rapidly cleared and extensively metabolized in the liver. CP 23445, its principal metabolite, can be measured in plasma and may contribute to the cardiovascular effects of trimazosin in humans. We are grateful to Dr. W. Singleton and colleagues, of Pfizer Central Research, Sandwich, England, for supplies of trimazosin. Dr. Andrew Kelman advised in data analysis, Marty Hughes skillfully assayed the blood samples, and Christine Lawrie, R.G.N., and Dr. John Vincent assisted in the studies. REFERENCES

1. Buyniski JP, Pircio AW, Schurig JE, Campbell JA: Effects of tiodazosin, prazosin, trimazosin, and phentolamine on blood pressure, heart rate, and on pre- and postsynaptic alphaadrenergic receptors in the rat. Clin Exp Hypertens 2:1039, 1980. 2. Constantine JW, Hess HJ: The cardiovascular effects of trimazosin. Eur J Pharmacol 74:227, 1981. 3. DeGuia D, Mendlowitz M, Russo C, Vlachakis ND, Antram S: The effect of trimazosin in essential hypertension. Curr Ther Res 15:339, 1973. 4. Vlachakis ND, Mendlowitz M, De Guzman D: Treatment of

Reid, Meredith,

5.

6.

7.

8.

9.

10.

11.

12. 13.

and Elliott

essential hypertension with trimazosin, a new vasodilator agent. Curr Ther Res 17564, 1975. Aronow WS, Tobias J, Hughes D, Siegel J, Easthope J: Comparison of trimazosin and methyldopa in hypertension. Clin Pharmacol Ther 22:425, 1977. Chrysant SG, Miller RF, Brown JL, Danisa K: Long-term hemodynamic and metabolic effects of trimazosin in essential hypertension. Clin Pharmacol Ther 30:600, 1981. Cambridge D, Davey MJ, Massingham R: Prazosin: A selective antagonist of postsynaptic alpha adrenoceptors. Br J Pharmacol 59:514, 1977. Hughes MA, Meredith PA, Elliott HL: The determination of trimazosin and its metabolite (CP 23445) in whole blood by high-pressure liquid chromatography using fluorescence detection. d Chromatogr (in press). Elliott HL, Sumner DJ, McLean K, Reid JL: Effect of age on the responsiveness of vascular alpha-adrenoceptors in man. J Cardiovasc Pharmacol 4:388, 1982. Sumner DJ, Elliott HL, Reid JL: The analysis of pressor dose response curves in human subjects. Br J Clin Pharmacol 14:128, 1982. Kelman AW, Whiting B: Modeling of drug response in individual subjects. J Pharmacokinet Biopharm 8:115, 1980. Whiting B, Kelman AW: The modeling of drug response. Clin Sci 59:311, 1981. Weidler DJ, Kessler KM, Anzola E, Taylor CR, Garg DC, Jallad NS: Hemodynamic effects and pharmacokinetics of

Pharmacokinetics of a su trimazosin tablet formuktion

14.

15.

16.

17.

18.

19.

20.

21.

intravenous trimazosin in mildly hypertensive man (abstr). Clin Pharmacol Ther 31:279, 1982. Elliott HL, McLean K, Sumner DJ, Meredith PA, Reid JL: Immediate cardiovascular responses to oral prazosin: Effects of concurrent beta blockers. Clin Pharmacol Ther 29:303, 1981. Drew GM, Whiting SB: Evidence for two distinct types of postsynaptic alpha adrenoceptors in vascular smooth muscle in vivo. Br J Pharmacol 67:207, 1979. Timmermans PBMWM, Kwa HY, van Zwieten PA: Possible subdivision of postsynaptic alpha receptors mediating pressor responses in the pithed rat. Naunyn Schmiedebergs Arch Pharmacol 310:189, 1979. Reid JL, Hamilton CA: Catecholamines and blood pressure regulation: The role of alpha receptors. J Cardiovasc Pharmaco1 S(supp1 3):325, 1980. Starke K, Dochery JR: Recent developments in alpha adrenoceptor research. d Cardiovasc Pharmacol S(supp1 3):269, 1980. McGrath JC: Evidence for more than one type of postjunctional alpha adrenoceptor. Biochem Pharmacol 31:467, 1982. Meredith PA, Elliott HL, Kelman AW, Reid JL: Concentration effect modeling of trimazosin and its major metabolite. Br J Clin Pharmacol 14:135, 1982. Meredith PA, Elliott HL, Kelman AW, Reid JL: Pharmacokinetic and pharmacodynamic modeling of doxazosin and trimazosin. Eur J Clin Invest 12:156, 1982.

-rekaee

Pharmacokinetics and bioavailability of a trimazosin sustained-release tablet (WIT) formulation (300 mg) were studied in healthy volunteers. In the first study of 12 subjects, the bioequivalence of trimazosin, 100 mg, in a standard tablet (ST) and in a capsule was demonstrated. After that, the bioavailability of one SRT (300 mg) was compared with that of the STs, (100 mg three times a day), in 19 subjects. Maximum plasma concentration (C,.) after SRT (8.1 + 3.0 mg/L) was significantly lower than that observed after ST (13.5 & 2.3 mg/L), time to peak was strongly delayed by a factor 7, and the time when plasma concentrations were higher than half of C,,. (t C,.,/2) was longer (10.4 f 3.2 vs 2.3 + 0.6 hours, p < 0.001). Bloavailabiiity of the SRT (300 mg) as measured by the area under the curve (AUC,) was about 65% of the ST. At the seventh dose, after single daily doses of the SRT in 12 subjects, the mean C,.. values were not significantly higher than after the first dose (8.6 + 3.2 mg/L vs 7.7 * 2.2 mg/L), t &,.,I2 values was comparable to AUC, calculated after the first were the same (10.4 hours), and the AU&,,, dose. Steady-state plasma concentration of trimazosin was obtained rapidly. No accumulation of trimazosin or its metabolite occurred. (AM HEART J 106:1228, 1983.)

B. Flouvat, Pharm.D., F. Fodor, Pharm.D., A. Roux, Pharm.D., J. R. Bianchine, M.D. Boulogne, France, and Columbus, Ohio

From the Departement de Pharmacologic cologie, HBpital Ambroise Park, and Ohio Reprint Clinique, Boulogne,

1228

requests: B. Flouvat, Pharm.D., Laboratoire de Toxicologic, France.

Clinique, Laboratoire State University.

de Toxi-

Departement de Pharmacologic H6pital Ambroise Park, 92100,

and

In the treatment of hypertension, a single daily dose regimen would be expected to improve patient compliance in drug taking. With a drug that has a short elimination half-life, it is necessary either to repeat