Fundam CIin Pharmacol(l990) 4, 175-189 0 Elsevier, Paris
175
Pharmacokinetics of perindopril and its metabolites in healthy volunteers J P Devissaguet I * , N Ammoury ', M Devissaguet 2, L Perret
' UA CNRS I218, Laboratoire de Pharmacie Galhique et Biopharmacie,
Universiti Paris Sud, 92290 Chatenay-Malabry; zIRIS Cie & Dkveloppement, Neuilly-sur-Seine, France (Received 6 June 1989; accepted 31 July 1989)
Summary - Perindopril, an angiotensin converting enzyme (ACE) inhibitor, is converted in vivo to its active diacid metabolite, perindoprilat and to a perindoprilat glucuronide. The pharmacokinetic parameters of perindopril, perindoprilat and perindoprilat glucuronide were evaluated after single administration to healthy volunteers ( N = 12) of 8 mg of perindopril tert-butylamine salt by oral route (treatment A), by intravenous route (bolus in 5 min, treatment B) and of an equimolar dose of perindoprilat (6.1 mg) by intravenous route (infusion over 2 h, treatment C). The treatments were administered as a randomised 3-way cross-over design. Plasma samples were collected up to 96 h and urines up to 120 h. Perindopril is rapidly absorbed with an oral bioavailability of 95% and is mainly eliminated by metabolic processes. The formation of perindoprilat is slow and about 20% of the available parent drug is transformed into this metabolite. Elimination profile of perindoprilat is biphasic, with a rapid renal excretion of the free fraction and a long terminal half-life of the fraction bound to ACE. Perindoprilat glucuronide is mainly obtained from perindopril by a pre-systemic first pass metabolism. perindopril I ACE inhibitor I human pharmacokinetics
Introduction Angiotensin converting enzyme (ACE) inhibitors have now been demonstrated to be safe and efficacious in the clinical management of human hypertension and heart failure. Following the successful introduction of captopril, non-thiol compounds were developed. Perindopril** (Laubie et al, 1984) is a structurally new ACE inhibitor devoid of the sulfhydril moiety (fig I). It is converted in vivo to its active diacid form perindoprilat by hydrolysis of the ester function and to a perindoprilat glucu-
* Correspondence and reprints ** Coversyl Laboratoires Servier @,
176
JP Devissaguet et al
ronide (Drummer et al, 1987) (fig 1). Clinical studies have shown that perindopril is a useful ACE inhibitor (Lees and Reid, 1987a; Morgan et al, 1987). This study was designed to evaluate the pharmacokinetics of the parent drug and of its metabolites in healthy volunteers after ora1 and intravenous administration of perindopril and infusion of perindoprilat.
Patients and Methods Subjects
Twelve normotensive healthy volunteers (6 males and 6 females) were recruited after screening by history, physical examination, blood pressure (systolic and diastolic), electrocardiogram, urinalysis, and routine laboratory tests of haematology and blood biochemistry. Written informed consent was obtained from all subjects and the study design was approved by the local Ethical Committee. The mean age of the volunteers was 28 yr (range 23-36 yr); mean weight (W) was 67 kg (range 48-79 kg); and mean height was 171 cm (range 152-182 cm). Study design
The study was an open 3-way cross-over single treatment with 8 mg of perindopril tert-butylamine salt administered via the oral (treatment A : 2 x 4 mg tablets) and the intravenous route (treatment B: IV bolus in 5 min), and 6.1 mg of perindoprilat, the active metabolite of perindopril (treatment C : IV infusion in 2 h). The subjects were randomly allocated to the rows of a latin square to determine the order of treatment. Thus, each volunteer attended on 3 occasions at least 2 weeks apart and was given equimolar doses of drug or active metabolite on each occasion. No salt restriction was imposed and subjects were asked to take their usual diet throughout
R
R‘
Perlndopril
C2H5
H
Perindoprilat
H
H
Perindoprilat glucuronide
H
G
Perhdopril glucuronide
C2HS
G
Fig 1. Chemical structure of perindopril and its metabolites perindoprilat and perindoprilat glucuronide (perindopril glucuronide was demonstrated in vitro but not identified in vivo).
Perindopril pharmacokinetics
177
the study period. Subjects had fasted since 900 pm during the previous night and had avoided all drugs for the previous 2 weeks. No caffeine or alcohol ingestion was permitted and tobacco was limited to 5 cigarettes per day. Blood pressure and heart rate were measured after 10 min supine rest and 2 min standing at time 0 , 4 , 8 and 24 h after dosing. Blood samples were drawn for plasma concentration assay before dosing and at the following intervals after dosing: - treatment A (oral perindopril): 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 10, 12, 18, 24, 30, 36, 48, 72 and 96 h ; - treatment B (IV perindopril): 0.08, 0.25, 0.5, 1, 2, 3 h and then same as treatment A ; - treatment C (IV perindoprilat): 0.25, 0.5, 1, 1.5 and 2 h (during infusion period), then: 0.08,0.25,0.5, 1 , 2 , 3 , 4 , 5 , 6 , 8 , 10, 12, 16,22,28,34,46,70 and 94 h after the end of infusion. Urine was collected before dosing and during the following intervals after dosing : 0-2,2-4, 4-6,6-8,8-12, 12-24,24-48,48-72,72-96 and 96-120 h. Plasma was separated by centrifugation at + 4" c; all samples were kept frozen at -20" C until assay.
Laboratory methods Perindopril, perindoprilat and perindoprilat glucuronide were isolated from plasma and urine samples by chromatography on a Dowex AG 1 x 2 column. Perindoprilat and its glucuronide form were measured by a specific radioimmunoassay (Doucet et al, submitted). As the assay is insensitive to perindopril, the drug was measured after alkaline hydrolysis. The assay sensitivity was 0.5 n g m - ' .
Pharmacokinetic parameters Sets of individual plasma concentration data were analyzed by model independent methods. Peak concentrations (C,,,=) and time to reach the peak (t,,,=) were ascertained directly from the analytical data. Areas under the experimental curves (AUC,) were derived using the trapezoidal rule; total areas under the curves (AUCa) were obtained by adding AUC, to extrapolated areas (AUC,,,), derived from : C,/K,, where C, is the last experimental concentration and K, the elimination rate constant for the terminal log-linear phase. Total plasma clearances (Cl,) were derived from dose/AUCao after intravenous administration. The biotransformation ratio of perindopril to perindoprilat (fm) was estimated from AUCw ratios for perindoprilat after treatments C and B. Absolute bioavailability of perindopril (F) was derived from AUCao ratios for perindopril after treatments A and B. The availability of perindoprilat after oral administration of perindopril (F') was obtained from AUCw ratios for perindoprilat after treatments A and B. The absolute availability of perindoprilat after oral administration of perindopril (F") was derived from : fm x F'. Half-lives were estimated by log-linear regressions of experimental data and volumes were derived from Cl,/K,, where Kj is the rate constant corresponding to the half-life. The fraction of the dose excreted as perindopril (fe) was derived from cumulative urinary excretions after treatment B. F, F', F" and fm were also calculated from urine data for perindopril or perindoprilat by the ratio of cumulative excretions (Ae) after treatments A, B or C. Excretion rate-constants were derived from log-Iinear regressions of dAe/dt versus midpoint-time curves, or from loglinear regressions of ARE versus time curves, where ARE was the amount remaining to be excreted at time t. Renal clearances were obtained from Aeti/AUC,,, where Ae is the amount excreted and ti is the last experimental time common to urine and plasma data.
178
JP Devissaguet et a/
Results Figures 2, 3 and 4 illustrate the plasma concentration time profiles (log-linear scales) of perindopril, perindoprilat and perindoprilat glucuronide respectively (mean values of 12 subjects), after oral (treatment A) and intravenous (treatment B) administration of perindopril. Figure 5 illustrates the cumulative excretion of the three molecules after the same treatments A and B. The mean plasma concentration-profile for perindoprilat (log-linear scales) after 2 h infusion of perindoprilat (treatment C) is illustrated in figure 6. Excretion profiles for perindoprilat and its glucuronide after treatment C are not presented. Kinetic parameters of perindopril, perindoprilat and perindoprilat glucuronideafter treatments A and/or B and/or C are presented in tables I, I1 and I11 respectively. Perindopril kinetics Plasma concentration time profiles of perindopril are polyphasic after both oral and intravenous administration. Following the IV bolus, the initial decay is very rapid and characterized by a half-life (t vrj) of 14 k 3 min (mean f SD) and a volume of distribution (VJ of 7.3 +- 1.8 1. The second phase, common to both routes, can be characterized by a half-life (t nc)of 0.88 & 0.22 h after treatment A (oral) and 0.98 & 1.8 h after treatment B (IV), to which a volume of distribution corresponds
1000
PERINDOPRIL
A
E
--P \
c
9
Treatment B
100
0
c
2
C
c 0
2 8
10
2
in!
1
0
2
4
6
8
10
Tlme (h) Fig 2. Mean plasma concentration-time profiles of perindopril after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 mg) to healthy volunteers.
179
Perindopril pharmacokinetics
100
PERINDOPRILAT
=
+
E 0) e
\
Q
Y
TreatmentA TreatmentB
10
C
0
H
c C
8
i 1
E (I)
0
h ,1 0
12
6
18
24 Time (h)
30
42
36
48
Fig 3. Mean plasma concentration-time profiles of perindoprilat after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 mg) to healthy volunteers.
100 7
PERINDOPRILAT GLUCURONIDE
A
i
Y
5
10
5
E 0
1:
E j n $1 0
I
I
I
I
I
2
4
6
8
10
Time (h) Fig 4. Mean plasma concentration-time profiles of perindoprilat glucuronide after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 mg) to healthy volunteers.
180
JP Devissaguet el 4l
20
-
h
8
15-
0 '0
E. E
0 10-
+
Perlndoprll (Tr.A)
0
-0
Perlndopril (Tr.6)
4
Perlndoprllat (Tr.A)
4
Perlndoprllat (Tr.B)
z
:: -3e
5-
*
Perlndoprllat gluc. (Tr.A) Perindoprllat gluc. (Tr.6)
+ 0
I
I
I
I
I
I
Fig 5. Mean cumulative urine excretion of perindopril, perindoprilatand perindoprilat glucuronide after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 rng) to healthy volunteers.
1000
-2
1-
A
E
-5
PERINDOPRILAT
100
T!
z 0
=8
10
0 C
3 0
p'
1 0
1
I
6
12
1
18
24 Tlme (h)
'
1
30
.
1
36
42
48
Fig 6. Mean plasma concentration-time profiles of perindoprilat after intravenous infusion of 6.1 rng in 2 h (treatment C) to healthy volunteers.
181
Perindopril pharmacokinetics
(V,) of 31 f 1 1 1. Six h after dosing, the residual concentrations of the drug represent 0.35 k 0.02% of the initial IV concentration (0.08h) and 1.3 f 0.2% of oral peak concentrations. The terminal phase corresponds to very low concentrations of the drug and characterization by a half-life leads to poor fits: t nt : 44* 18 h and 523~18 h after oral and IV routes, respectively. Oral bioavailability of perindopril is large when calculated from both plasma (F = 94.4%) and urine (F= 95.1'7'0) data. Absorption from gastro-intestinal tract is rapid and leads to peak concentrations of 164f 64 n g m - ' in 0.9 f 0.5 h. Total clearance of perindopril is 362 f 62 ml-min- and seems to be related to body weight (5.4k0.5 mlamin- lnkg-') as indicated by a lower dispersion of values. Elimination of the drug occurs essentially by biotransformation and/or biliary excretion and renal excretion of the unchanged drug is only 15.6f 5.7% of the IV dose. The renal clearance is 683~27ml-min-' and 62k15 ml-min-' after IV and oral routes, respectively. The excretion profiles are biphasic and almost parallel to plasma profiles with an initial rapid decay characterized by a half-life (t nc)comprised between 0.8 f 0.3 h
Table 1. Mean ( f SD) parameters for perindopril after oral (treatment A) and intravenous (treatment B)
administration of 8 mg of perindopril (tert-butylamine salt) to 12 volunteers. * Geometric mean.
Parameter Plasma Cmaa tm,
AUCm F (9 CIT ClT,W t*,
Treatment
Treatment A
164f64 0.9f0.5 304 f 80 94.4
-
"3
vc
0.88 f 0.22 -
t nt
44f 18
tvr,
Urine Aet F (9 Clr fe-F
Treatment B
317f61 362 f 62 5.4f0.5 14f3 7.3 f 1.8 0.98*0.18 31f11 5 2 f 18
0.94 rt 0.17 95.1 6 2 f 15 14.0f2.6
1 .04f0.38
68 f 27 15.6f5.7 [F= I]
1.5 f 0 . 5 1.1*0.2
l.lk0.3 0.8f0.3
2 2 f 10 2 4 f 12
22f8 3 6 f 13
-
t*c
ARE method dAe/dt method t nt
ARE method dAe/dt method
182
JP Devissaguet et a/
and 1.5 f0.5 h, depending on the route of administration and the method of calculation (ARE or dAe/dt method). More than 99% of the total excretion is completed in 8 h after both IV and oral administration. The terminal phase is characterized by a half-life (t LZJ between 22 k 8 h and 36 f 13 h depending on the route of administration and method of calculation.
Perindoprilat kinetics Infusion of perindoprilat over a 2 h period results in plasma concentrations of 280 f77 ng-ml- I . Following the infusion period, a biphasic decay of plasma concentrations is observed. The initial rapid phase is characterized by a half-life (t mi) of 0.93 f0.25 h and a volume of distribution (Vi) of 9.3 k 3.0 1 . The second and slow phase is characterized by a half-life (t tat) of 29 -t 20 h. With perindopril, this phase corresponds to low concentrations, highly dispersed values for t tat and poor fits.
Table 11. Mean (fSD) parameters for perindoprilat after oral (treatment A) and intravenous (treatment B) administration of 8 mg of perindopril (tert-butylamine salt) and intravenous infusion in 2 H (treatment C) of an equimolar dose (6.1 mg) of perindoprilat to 12 volunteers. Parameter
Treatment
Treatment A
Treatment B
( n g - m l I)~
11.1 f 4 . 0
15.Ok 3.9
3 . 0 f 1.0 145 + 36 -
3.8k 1 . 1 207 f48 -
-
-
-
-
-
-
-
Treatment C
Plasma
c,,,
5.1 + 1.7 3 0 f 15
280 k I7 (end of infusion period)
PI 906 f226 120+34 l.Sk0.4 0.93 k0.25 9.3 3.0 -
+
72+ 16 16.8k4.8
5.4k1.4 25fll 23.4 4.4 [1W -
0.78 i 0.16 12.8+2.6
1.09f0.32 17.9f5.3
7.0+ 1.2 114+20
6.5k1.9 4.9 f 1.9
6.9 k 2.4 5.6 f2.6
2.2 + 0.6 1.5k0.2
22+2 30+ 1 1
20+3 26+7 15.7 k 3.6 [I001
20+3 23+4 [I001
-
[1001 141 k42
-
-
77 + 25 1 l . 4 + 2.8 102 k23
*
93 k 24
29 f 20 [IWI [I001
-
183
Perindopril pharmacokinetics
Total clearance for perindoprilat is 120f 34 ml.min-' or 1.8 f 0 . 4 ml.min-'.kg-'. Renal excretion of perindoprilat after infusion is higher than dose : 114 f 20'70, leading to a renal clearance of 141 f 4 2 ml-min-', higher than total plasma clearance. These results could be of analytical origin since the first highly concentrated urine samples were diluted in order to fall within the range of the calibration curve of assay; they could also be explained if the true administered dose is higher than the theoretical one. Excretion profiles are biphasic with an initial decay characterized by a half-life (t mi)of 1.5 0.2 h or 2.2 f 0.6 h depending on the calculation method; 96.6 f 0.4% of total perindoprilat excretion is completed in 12 h. The terminal phase of excretion is characterized by a half-life (t m,)of 20 f 3 h (ARE method) or 23 k 4 h (dAe/dt method). Intravenous and oral administration of perindopril leads to peak perindoprilat concentrations of 15.0k3.9 ng-ml-' and 1 1 . 1 k4.0 ng-ml-' after 3 . 8 + 1 . 1 h and 3.0 f 1 .O h respectively. The fraction (fm) of injected perindopril converted by hydrolysis to perindoprilat is 23.4 f 4.4% of the administered dose. The relative availability (F') of perindoprilat after oral perindopril is 72 f 16% of perindoprilat after IV injection of the parent drug. In this case, the fraction of orally administered perindopril dose present as perindoprilat (F") is 16.8 f 4.8%. As a result of the urinary excretion of perindoprilat exceeding the infused dose, fm and F" values calculated from urine data are less than plasma values: 15.7 f 3.6% and 11.4 f 2.8% respectively. Oral availability of perindoprilat (F') from urine data is 77 f25%, close to the plasma value. Following both routes of administration of perindopril, the decay
Table 111. Mean (kSD) parameters for perindoprilat glucuronide after oral (treatment A) and intravenous (treatment B) administration of 8 mg of perindopril (tert-butylamine salt) and intravenous infusion in 2 h (treatment C) of an equimolar dose (6.1 mg) of perindoprilat to 12 volunteers.
Parameter
Treatment
Treatment A
Treatment B
Treatment C
64 k 20 1.4k0.4 178 50
20+6 0.7 k 0.3 59+ 17
1.1 i 0 . 4
1.3 k0.2
1.7ka.3
-
1.46k 0.28 15.6k3.0 142 k 33
0.58k0.18 6.3 k 2.0 169 + 57
0.07 + 0.02 0.1 k0.2
1.6k0.1 1.5 k0.2
2.0+0.3 1.6 t 0 . 4
2.8 k0.6 1.9+0.5
5.7 + 2.0 6.2 k 1.7
6.7 + 1.7 8.2 + 2.0
7.2k1.7 1.7 + 2.5
1.8k0.3 -
-
184
JP Devissaguet et al
of perindoprilat concentration after the peak is biphasic with a half-life (t nC)of 5.1 f 1.7 h and 5.4 f 1.4 h after oral and intravenous routes respectively, and a terminal half-life (t q)of 30 f 15 h and 25 f 11 h after oral and intravenous routes, respectively. The same comment regarding dispersion and fit applies to terminal half-lives of perindoprilat after perindopril administration. Excretion of perindoprilat is 12.8f2.6% and 17.9f5.3'70 of the dose after oral and intravenous administration of perindopril respectively. The renal clearance of perindoprilat following perindopril administration is 102f23 ml-min- and 93 f24 mlamin- after oral and intravenous routes respectively, these values being more realistic than that obtained after perindoprilat infusion. Excretion profiles of perindoprilat reflect plasma concentration profiles with half-lives between 4.9k 1.9 and 6.9f2.4 h for the first phase and between 2 0 f 3 h and 30+: 11 h for the terminal phase, depending on the calculation method and route of administration.
'
Perindoprilat glucuronide kinetics Very low plasma concentrations and urinary excretion of perindoprilat glucuronide are obtained after IV infusion of the active metabolite, leading to the conclusion that this conjugation process only has a minor role in the elimination pathway of perindoprilat. A . peak plasma concentration of 1.1 f0.4 ng-ml- was obtained in 1.8 0.3 h but area and half-life calculations were not possible. Urinary excretion constituted 0.7 f0.2% of the dose. Excretion was biphasic with half-lives of 2.8 f0.6 h or 1.9 f0.5 h for the first phase and 7.2 f 1.7 h or 7.7 f2.5 h for the terminal phase, depending on the calculation method, ARE and dAe/dt respectively. After IV administration of the parent drug perindopril, peak plasma concentrations of perindoprilat glucuronide are 2 0 f 6 ng-ml-' for a peak time at 0.7f0.3 h. Plasma concentrations rapidly fall to undetectable values with a half-life of 1.7 f0.3 h. Renal excretion represents 6.3 f2.9% of the dose and the renal clearance of perindoprilat glucuronide is 169k 57 mlsrnin-'. Excretion profiles clearly exhibit 2 phases with half-lives of 2.0 f0.3 h (ARE) or 1.6 f0.4 h (dAe/dt), and 6.7 f 1.7 h (ARE) or 8.2 f2.0 h (dAe/dt). After oral administration of perindopril, peak plasma concentrations for the perindoprilat glucuronide are higher (64f20 ng-ml- ') and are obtained after 1.4 f0.4 h. The area under the curve is 178f50 ng-ml-'ah after oraI administration of perindopril but only 5 9 f 17 ngm-'.h after IV injection of the drug. The half-life for plasma concentration decay is 1.3 f0.2 h. Renal excretion of perindoprilat glucuronide is 15.6 f3.0% of the perindopril dose and the renal clearance is 142f 33 ml-min-'. The biphasic profile of excretion is characterized by half-lives of 1.6 f0.1 h (ARE) or 1.5 f0.2 h (dAe/dt), and 5.7 f2.0 h (ARE) or 6.2 f 1.7 h (dAe/dt).
*
Discussion and Conclusion The present study associating intravenous infusion of the active metabolite perindoprilat with both oral and intravenous administration of the parent drug perindopril
Perindopril pharmacokinetics
185
was designed to allow precise estimations of bioavailability and conversion ratio of the drug to its diacid. Furthermore, the ability of the assay to measure plasma and urine levels of perindoprilat glucuronide provides an insight into the metabolic pathway of perindopril. Kinetic analysis
Polyexponential equations, assuming multicompartmental models, have been used to describe plasma concentration time profiles of non-thiol ACE inhibitors (Hockings et al, 1986). Recently perindoprilat kinetics after single bolus IV injection were described through an open 3-compartment model (Lees and Reid, 1987b). However, multicompartmental models were unable to simultaneously follow kinetics after single and multiple dosings, and accumulation ratios were much lower than predicted from the long terminal half-lives observed (Till et al, 1984). A onecompartment model with saturable binding was used to accurately relate kinetic and dynamic data, ie plasma concentration of ACE inhibitors and ACE inhibition in plasma (Francis et al, 1987). High-affinity, limited-capacity and specific binding of inhibitors to plasma ACE, in conjunction with low-affinity, high-capacity and nonspecific binding to other proteins, may explain their long terminal half-life and solve the paradox of the low accumulation ratio (Nussberger et al, 1987). It can then be assumed that high plasma concentrations of ACE inhibitors are mainly related to the free drug or drug bound to non-specific proteins and that low concentrationsare related to the drug bound to ACE, the slow dissociation rate of the complex acting as a limiting step for the elimination of the drug. In the present study, model-independent methods were used to characterize the kinetics of perindopril and its metabolites. Polyphasic profiles for some plasma and urine data have led to the calculation of 2 or 3 half-lives. These half-lives were directly derived from true experimental values during apparent log-linear decay but not from polyexponential analysis. From a theoretical point of view this method is debatable, but from a clinical point of view it allows a separate determination, with sufficient accuracy for therapeutic purposes, of the essential kinetic parameters of the free and bound drug. Perindopril kinetics
From both plasma and urine data, gastrointestinal absorption of perindopril is high and leads to an oral bioavailability of about 95% of the dose. The onset of absorption is rapid and peak concentrations are obtained in about 1 h. Disposition of the parent drug after both oral and IV routes is polyphasic. The initial decay after IV injection (t n: 14 min) can be attributed to dilution into blood and/or extravascular diffusion into well perfused tissues. The main phase after both oral and IV administration corresponds to a half-life of about 1 h and during this phase more than 99% of the renal excretion of unchanged perindopril
186
JP Devissaguet et a1
is completed, when residual plasma concentrations are about or less than 1070 of the peak values. Considering this half-life as the biological half-life of perindopril, the volume of distribution ( z 3 1 1) corresponds to body water, in good agreement with the hydrophilic character of the drug. The terminal phase corresponds to very low plasma concentrations and urine excretion rates of perindopril and poor fittings were obtained for half-life calculations. Repeated administration of pgrindopril has not led to accumulation (Bromet et al, 1988), indicating that this long terminal half-life does not correspond to a deep compartment. From both plasma and urine data, the terminal half-life of perindopril is closely related to the terminal half-life of perindoprilat and this result could suggest a saturable binding of the parent drug to ACE. In vitro data clearly indicate that binding of perindopril cannot be responsible for its long terminal half-life (Jackson et al, 1987). One possible explanation is that the terminal half-life is an apparent one and an artefact of analytical origin. This assumption has not been verified but the assay of perindopril includes a chromatographic separation from metabolites, followed by hydrolysis of the samples and determination of the active diacid. As the chromatographic samples corresponding to perindopril also contain proteins, some perindoprilat bound to plasma ACE could be present as a contaminant of the parent drug. Elimination of perindopril is mainly due to metabolic processes, and renal excretion of the unchanged drug represents only 15.6% of the IV dose. Nearly half of the IV dose elimination (45.5%) was obtained by adding to excretion the biotransformations of perindopril to the active diacid perindoprilat (23.4%) and to the perindoprilat glucuronide (6.3%). Total plasma clearance of perindopril is high (362 ml-min-l), compared to a renal clearance of about 65 ml-min-', a metabolic clearance of about 85 ml-min- for the formation of perindoprilat and a metabolic clearance of about 22 ml-min- * corresponding to perindoprilat glucuronide. Moreover, this glucuronide was 3-fold more abundant in plasma and urine after oral administration of perindopril, but present in very low amounts after IV infusion of perindoprilat . In agreement with Fournel-Gigleux et a1 (1988), those results strongly suggest that perindoprilat glucuronide is a hydrolysis residue of a parent drug glucuronide and that perindopril is subjected to a presystemic first-pass effect after oral administration. By substituting the urine excretions of perindoprilat glucuronide after oral and IV administration of perindopril, it can be concluded that this first-pass effect concerns = 9% of the dose, in agreement with the bioavailability of perindopril. This result also suggests that other metabolites, and especially the active diacid are formed after absorption, the first-pass effect acting only by glucuronidation of the parent drug.
Perindoprilat kinetics Disposition behaviour of the active diacid metabolite after its IV infusion and after perindopril administration by both oral and IV routes is quite different. IV infusion of perindoprilat leads to a rapid elimination of the free drug (tm: 0.93 h) by renal
Perindopril pharmacokinetics
187
excretion. Extravascular diffusion of perindoprilat is very limited, with a volume of distribution (9 1) that is smaller than the volume of the parent drug, in good agreement with the more hydrophilic character of the active metabolite. After IV or oral dosing of perindopril, formation of perindoprilat is quite slow and the peak time for the plasma concentrations is between 3-4 h. The decay of plasma concentration after peak (15 ng-ml-I and 11 ng-ml-' after IV and oral dosing respectively) is also quite slow compared to the decay after IV infusion of perindoprilat (tvr: z 5.5 h from plasma and urine data). This feature, which is in agreement with the results described by Lees and Reid (1987a, b) could reflect the rate of formation of perindoprilat from the parent drug and as this process is only a small part of the total elimination process for perindopril, it should have no apparent effect on the half-life of the parent drug. The terminal long half-life of perindoprilat (tvz: 20-30 h) is present after any drug dose or route of administration and can be characterized from both plasma and urine data. It is assumed that this slow elimination corresponds to perindoprilat bound to ACE and reflects the rate of dissociation of the complex. After perindopril dosing, formation of perindoprilat represents about 20% of the available perindopril (23% from plasma data and 18% from urine data). This result is in good agreement with other experiments (Lees and Reid, 1987b) and corresponds to z 1.2 mg of active metabolite available after 8 mg oral dosing of perindopril as tert-butylamine salt. A 1-mg dose of perindoprilat has been shown to totally inhibit plasma ACE after IV injection, more than 50% of inhibition being obtained over ::20 h with an IC,, of 1.8 ngem1-I (Lees and Reid, 1987a). The mean plasma concentrations of perindoprilat are 3.16 ng-ml- 12 h after oral administration of 8 mg of perindopril and 1.15 ng-ml-' after 24 h. A prolonged pharmacological response is therefore expected after oral dosing with 8 mg and has been reported as being at 60% of ACE inhibition 24 h after treatment (Lees and Reid, 1987a). As for other non-thiol ACE inhibitors, perindoprilat elimination is essentially an excretion process with a renal clearance of z 100 mlamin-I, close to the glomerular filtration rate. As previously stated, the formation of perindoprilat glucuronide from perindoprilat is negligible.
Perindoprilat glucuronide kinetics Perindoprilat glucuronide, mainly resulting from a first-pass effect after oral administration of perindopril, reaches higher plasma concentrations (:: 64 ng-ml- I), more rapidly (tmax: 1.4 h) than perindoprilat. These results confirm that this glucuronide does not come from the active diacid. Formation of the glucuronide is rapid even after IV administration of perindopril (t,,, : 0.7 h), but peak concentrations remain quite low (20 n g m - '). The elimination of perindoprilat glucuronide from plasma is rapid and monophasic (t vz= 1.5 h) and concentrations reach undetectable levels before any slow phase appears, in contrast with urine where a second half-life (:: 7 h) can be calculated with good accuracy. Repeated oral administration of perindopril
188
JP Devissaguet et
a1
does not lead to accumulation of the perindoprilat glucuronide (Bromet et al, 1988) and it has been shown that this metabolite can bind to ACE, but with a lower affinity (IC50 = 64 nM) than the diacid metabolite of perindopril (IC50 = 2.4 nM). Thus, as for ACE inhibitors, the slow elimination phase of the perindoprilat glucuronide could reflect the dissociation of the complex. The renal clearance of perindoprilat glucuronide ( z 155 ml-min-’ from both oral and IV routes for perindopril) is the highest of the 3 renal clearances calculated in this study, but there is no experimental evidence to indicate that renal excretion is the only process involved in the glucuronide elimination.
Conclusion The protocol design of this study allowed the estimation of essential kinetic parameters for the parent drug perindopril, for its active diacid metabolite perindoprilat and for perindoprilat glucuronide. Furthermore, this study allowed some useful insights into the metabolic pathway of perindopril. All treatments were well tolerated during the study. Perindopril is intensively and rapidly available, its elimination, mainly by metabolic process, is rapid and leads to z 20% of the available dose being converted to the active metabolite. Perindoprilat exhibits well-known characteristics of non-thiol ACE inhibitors, such as rapid renal excretion of free drug and long terminal half-life of bound drug. Perindoprilat glucuronide, mainly present after oral administration of perindopril, seems to be formed from the parent drug itself and not by conjugation of the active diacid.
References Bromet N, Lelievre E, Vidal D, Grislain L, Devissaguet M, Jochemsen R, Ammoury N, Devissaguet J P (1988) Etude pharmacodynamique et pharmacocinttique chez I’homme d’un nouvel inhibiteur de l’enzyme de conversion de I’angiotensine I, le perindopril, apres administration unique et repttte. Journtes de I’HTA, Paris, 15-16 December Drummer OH, Rowley K, Johnson H, Worland P, Workman B, Harris QLG, Conway EL, Louis WJ (1987)Metabolism and pharmacodynamics of angiotensin converting enzyme with special reference to perindopril. Excerpta Med Znt Congr Ser 750, 545-550
Fournel-Gigleux S, Magdalou J, Lafaurie C, Grislain L, Garnier NH. Dabe JF, Luijten W, Bromet N, Devissaguet M (1988) Glucuronidation of perindopril by hepatic microsomes : interspecies comparison. Cellular and Molecular Aspects of Glucuronidation, Montpellier, 27-29 April Francis RJ, Brown AN, Kler L, Fasanella d’Amora T, Nussberger J, Waeber B, Brunner BR (1987) Pharmacokinetics of the converting enzyme inhibitor cilazapril in normal volunteers and the relationship to enzyme inhibition: development of a mathematical model. J Cardiovasc Pharmacol9, 32-38 Hockings NB, Ajayi AA, Reid JL (1986)Age and the pharmacokinetics of angiotensin
Perindopril pharmacokinetics
converting enzyme inhibitors enalapril and enalaprilat. Br J Clin Pharmacol21, 341-348
Jackson B, Cubela RB, Johnston CI (1987) Angiotensin converting enzyme inhibitors : measurement of relative inhibitory potency and serum drug levels by radio-inhibitor binding displacement assay. J Cardiovasc Pharmacol 9, 699-704 Laubie M, Schiavi P, Vincent M, Schmitt H (1984) Inhibition of angiotensin I converting enzyme with S 9490: biochemical effects, interspecies differencies and role of sodium diet in hemodynamic effects J Cardiovasc Pharmacol 6, 1076- 1082
Lees KR, Reid JL (1987a) Effects of intravenous S 9780, an angiotensin converting enzyme inhibitor, in normotensive subjects. J Cardiovasc Pharmacol 10, 129-135
Lees KR, Reid JL (1987b) The haemo-
189
dynamics and humoral effects of treatment for one month with the angiotensin converting enzyme inhibitor perindopril in salt replete hypertensive patients. Eur J Clin Pharmacol 3 1, 5 19-524 Morgan T, Anderson A, Wilson D, Murphy J, Nowson C (1987) The effect of perindopril on blood pressure in humans on different sodium intakes. J Cardiovasc Pharmacol 10, S 116, S 118 Nussberger J, Fasanella d’Amora T, Porchet M (1987) Repeated administration of the converting enzyme inhibitor cilazapril to normal volunteers. J Cardiovasc Pharrnacol9, 39-44 Till AE, Gomez HJ, Hichens M, Bolognese JA, Mc Nabb WA, Brooks BA, Noormohamed F, Lant AF (1984) Pharmacokinetics of repeated single oraI doses of enalapril maleate (MK 421) in normal volunteers. Biopharm & Drug Dispos 5 , 273-280
175
Pharmacokinetics of perindopril and its metabolites in healthy volunteers J P Devissaguet I * , N Ammoury ', M Devissaguet 2, L Perret
' UA CNRS I218, Laboratoire de Pharmacie Galhique et Biopharmacie,
Universiti Paris Sud, 92290 Chatenay-Malabry; zIRIS Cie & Dkveloppement, Neuilly-sur-Seine, France (Received 6 June 1989; accepted 31 July 1989)
Summary - Perindopril, an angiotensin converting enzyme (ACE) inhibitor, is converted in vivo to its active diacid metabolite, perindoprilat and to a perindoprilat glucuronide. The pharmacokinetic parameters of perindopril, perindoprilat and perindoprilat glucuronide were evaluated after single administration to healthy volunteers ( N = 12) of 8 mg of perindopril tert-butylamine salt by oral route (treatment A), by intravenous route (bolus in 5 min, treatment B) and of an equimolar dose of perindoprilat (6.1 mg) by intravenous route (infusion over 2 h, treatment C). The treatments were administered as a randomised 3-way cross-over design. Plasma samples were collected up to 96 h and urines up to 120 h. Perindopril is rapidly absorbed with an oral bioavailability of 95% and is mainly eliminated by metabolic processes. The formation of perindoprilat is slow and about 20% of the available parent drug is transformed into this metabolite. Elimination profile of perindoprilat is biphasic, with a rapid renal excretion of the free fraction and a long terminal half-life of the fraction bound to ACE. Perindoprilat glucuronide is mainly obtained from perindopril by a pre-systemic first pass metabolism. perindopril I ACE inhibitor I human pharmacokinetics
Introduction Angiotensin converting enzyme (ACE) inhibitors have now been demonstrated to be safe and efficacious in the clinical management of human hypertension and heart failure. Following the successful introduction of captopril, non-thiol compounds were developed. Perindopril** (Laubie et al, 1984) is a structurally new ACE inhibitor devoid of the sulfhydril moiety (fig I). It is converted in vivo to its active diacid form perindoprilat by hydrolysis of the ester function and to a perindoprilat glucu-
* Correspondence and reprints ** Coversyl Laboratoires Servier @,
176
JP Devissaguet et al
ronide (Drummer et al, 1987) (fig 1). Clinical studies have shown that perindopril is a useful ACE inhibitor (Lees and Reid, 1987a; Morgan et al, 1987). This study was designed to evaluate the pharmacokinetics of the parent drug and of its metabolites in healthy volunteers after ora1 and intravenous administration of perindopril and infusion of perindoprilat.
Patients and Methods Subjects
Twelve normotensive healthy volunteers (6 males and 6 females) were recruited after screening by history, physical examination, blood pressure (systolic and diastolic), electrocardiogram, urinalysis, and routine laboratory tests of haematology and blood biochemistry. Written informed consent was obtained from all subjects and the study design was approved by the local Ethical Committee. The mean age of the volunteers was 28 yr (range 23-36 yr); mean weight (W) was 67 kg (range 48-79 kg); and mean height was 171 cm (range 152-182 cm). Study design
The study was an open 3-way cross-over single treatment with 8 mg of perindopril tert-butylamine salt administered via the oral (treatment A : 2 x 4 mg tablets) and the intravenous route (treatment B: IV bolus in 5 min), and 6.1 mg of perindoprilat, the active metabolite of perindopril (treatment C : IV infusion in 2 h). The subjects were randomly allocated to the rows of a latin square to determine the order of treatment. Thus, each volunteer attended on 3 occasions at least 2 weeks apart and was given equimolar doses of drug or active metabolite on each occasion. No salt restriction was imposed and subjects were asked to take their usual diet throughout
R
R‘
Perlndopril
C2H5
H
Perindoprilat
H
H
Perindoprilat glucuronide
H
G
Perhdopril glucuronide
C2HS
G
Fig 1. Chemical structure of perindopril and its metabolites perindoprilat and perindoprilat glucuronide (perindopril glucuronide was demonstrated in vitro but not identified in vivo).
Perindopril pharmacokinetics
177
the study period. Subjects had fasted since 900 pm during the previous night and had avoided all drugs for the previous 2 weeks. No caffeine or alcohol ingestion was permitted and tobacco was limited to 5 cigarettes per day. Blood pressure and heart rate were measured after 10 min supine rest and 2 min standing at time 0 , 4 , 8 and 24 h after dosing. Blood samples were drawn for plasma concentration assay before dosing and at the following intervals after dosing: - treatment A (oral perindopril): 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 10, 12, 18, 24, 30, 36, 48, 72 and 96 h ; - treatment B (IV perindopril): 0.08, 0.25, 0.5, 1, 2, 3 h and then same as treatment A ; - treatment C (IV perindoprilat): 0.25, 0.5, 1, 1.5 and 2 h (during infusion period), then: 0.08,0.25,0.5, 1 , 2 , 3 , 4 , 5 , 6 , 8 , 10, 12, 16,22,28,34,46,70 and 94 h after the end of infusion. Urine was collected before dosing and during the following intervals after dosing : 0-2,2-4, 4-6,6-8,8-12, 12-24,24-48,48-72,72-96 and 96-120 h. Plasma was separated by centrifugation at + 4" c; all samples were kept frozen at -20" C until assay.
Laboratory methods Perindopril, perindoprilat and perindoprilat glucuronide were isolated from plasma and urine samples by chromatography on a Dowex AG 1 x 2 column. Perindoprilat and its glucuronide form were measured by a specific radioimmunoassay (Doucet et al, submitted). As the assay is insensitive to perindopril, the drug was measured after alkaline hydrolysis. The assay sensitivity was 0.5 n g m - ' .
Pharmacokinetic parameters Sets of individual plasma concentration data were analyzed by model independent methods. Peak concentrations (C,,,=) and time to reach the peak (t,,,=) were ascertained directly from the analytical data. Areas under the experimental curves (AUC,) were derived using the trapezoidal rule; total areas under the curves (AUCa) were obtained by adding AUC, to extrapolated areas (AUC,,,), derived from : C,/K,, where C, is the last experimental concentration and K, the elimination rate constant for the terminal log-linear phase. Total plasma clearances (Cl,) were derived from dose/AUCao after intravenous administration. The biotransformation ratio of perindopril to perindoprilat (fm) was estimated from AUCw ratios for perindoprilat after treatments C and B. Absolute bioavailability of perindopril (F) was derived from AUCao ratios for perindopril after treatments A and B. The availability of perindoprilat after oral administration of perindopril (F') was obtained from AUCw ratios for perindoprilat after treatments A and B. The absolute availability of perindoprilat after oral administration of perindopril (F") was derived from : fm x F'. Half-lives were estimated by log-linear regressions of experimental data and volumes were derived from Cl,/K,, where Kj is the rate constant corresponding to the half-life. The fraction of the dose excreted as perindopril (fe) was derived from cumulative urinary excretions after treatment B. F, F', F" and fm were also calculated from urine data for perindopril or perindoprilat by the ratio of cumulative excretions (Ae) after treatments A, B or C. Excretion rate-constants were derived from log-Iinear regressions of dAe/dt versus midpoint-time curves, or from loglinear regressions of ARE versus time curves, where ARE was the amount remaining to be excreted at time t. Renal clearances were obtained from Aeti/AUC,,, where Ae is the amount excreted and ti is the last experimental time common to urine and plasma data.
178
JP Devissaguet et a/
Results Figures 2, 3 and 4 illustrate the plasma concentration time profiles (log-linear scales) of perindopril, perindoprilat and perindoprilat glucuronide respectively (mean values of 12 subjects), after oral (treatment A) and intravenous (treatment B) administration of perindopril. Figure 5 illustrates the cumulative excretion of the three molecules after the same treatments A and B. The mean plasma concentration-profile for perindoprilat (log-linear scales) after 2 h infusion of perindoprilat (treatment C) is illustrated in figure 6. Excretion profiles for perindoprilat and its glucuronide after treatment C are not presented. Kinetic parameters of perindopril, perindoprilat and perindoprilat glucuronideafter treatments A and/or B and/or C are presented in tables I, I1 and I11 respectively. Perindopril kinetics Plasma concentration time profiles of perindopril are polyphasic after both oral and intravenous administration. Following the IV bolus, the initial decay is very rapid and characterized by a half-life (t vrj) of 14 k 3 min (mean f SD) and a volume of distribution (VJ of 7.3 +- 1.8 1. The second phase, common to both routes, can be characterized by a half-life (t nc)of 0.88 & 0.22 h after treatment A (oral) and 0.98 & 1.8 h after treatment B (IV), to which a volume of distribution corresponds
1000
PERINDOPRIL
A
E
--P \
c
9
Treatment B
100
0
c
2
C
c 0
2 8
10
2
in!
1
0
2
4
6
8
10
Tlme (h) Fig 2. Mean plasma concentration-time profiles of perindopril after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 mg) to healthy volunteers.
179
Perindopril pharmacokinetics
100
PERINDOPRILAT
=
+
E 0) e
\
Q
Y
TreatmentA TreatmentB
10
C
0
H
c C
8
i 1
E (I)
0
h ,1 0
12
6
18
24 Time (h)
30
42
36
48
Fig 3. Mean plasma concentration-time profiles of perindoprilat after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 mg) to healthy volunteers.
100 7
PERINDOPRILAT GLUCURONIDE
A
i
Y
5
10
5
E 0
1:
E j n $1 0
I
I
I
I
I
2
4
6
8
10
Time (h) Fig 4. Mean plasma concentration-time profiles of perindoprilat glucuronide after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 mg) to healthy volunteers.
180
JP Devissaguet el 4l
20
-
h
8
15-
0 '0
E. E
0 10-
+
Perlndoprll (Tr.A)
0
-0
Perlndopril (Tr.6)
4
Perlndoprllat (Tr.A)
4
Perlndoprllat (Tr.B)
z
:: -3e
5-
*
Perlndoprllat gluc. (Tr.A) Perindoprllat gluc. (Tr.6)
+ 0
I
I
I
I
I
I
Fig 5. Mean cumulative urine excretion of perindopril, perindoprilatand perindoprilat glucuronide after oral (treatment A) and intravenous (treatment B) administration of perindopril tert-butylamine (8 rng) to healthy volunteers.
1000
-2
1-
A
E
-5
PERINDOPRILAT
100
T!
z 0
=8
10
0 C
3 0
p'
1 0
1
I
6
12
1
18
24 Tlme (h)
'
1
30
.
1
36
42
48
Fig 6. Mean plasma concentration-time profiles of perindoprilat after intravenous infusion of 6.1 rng in 2 h (treatment C) to healthy volunteers.
181
Perindopril pharmacokinetics
(V,) of 31 f 1 1 1. Six h after dosing, the residual concentrations of the drug represent 0.35 k 0.02% of the initial IV concentration (0.08h) and 1.3 f 0.2% of oral peak concentrations. The terminal phase corresponds to very low concentrations of the drug and characterization by a half-life leads to poor fits: t nt : 44* 18 h and 523~18 h after oral and IV routes, respectively. Oral bioavailability of perindopril is large when calculated from both plasma (F = 94.4%) and urine (F= 95.1'7'0) data. Absorption from gastro-intestinal tract is rapid and leads to peak concentrations of 164f 64 n g m - ' in 0.9 f 0.5 h. Total clearance of perindopril is 362 f 62 ml-min- and seems to be related to body weight (5.4k0.5 mlamin- lnkg-') as indicated by a lower dispersion of values. Elimination of the drug occurs essentially by biotransformation and/or biliary excretion and renal excretion of the unchanged drug is only 15.6f 5.7% of the IV dose. The renal clearance is 683~27ml-min-' and 62k15 ml-min-' after IV and oral routes, respectively. The excretion profiles are biphasic and almost parallel to plasma profiles with an initial rapid decay characterized by a half-life (t nc)comprised between 0.8 f 0.3 h
Table 1. Mean ( f SD) parameters for perindopril after oral (treatment A) and intravenous (treatment B)
administration of 8 mg of perindopril (tert-butylamine salt) to 12 volunteers. * Geometric mean.
Parameter Plasma Cmaa tm,
AUCm F (9 CIT ClT,W t*,
Treatment
Treatment A
164f64 0.9f0.5 304 f 80 94.4
-
"3
vc
0.88 f 0.22 -
t nt
44f 18
tvr,
Urine Aet F (9 Clr fe-F
Treatment B
317f61 362 f 62 5.4f0.5 14f3 7.3 f 1.8 0.98*0.18 31f11 5 2 f 18
0.94 rt 0.17 95.1 6 2 f 15 14.0f2.6
1 .04f0.38
68 f 27 15.6f5.7 [F= I]
1.5 f 0 . 5 1.1*0.2
l.lk0.3 0.8f0.3
2 2 f 10 2 4 f 12
22f8 3 6 f 13
-
t*c
ARE method dAe/dt method t nt
ARE method dAe/dt method
182
JP Devissaguet et a/
and 1.5 f0.5 h, depending on the route of administration and the method of calculation (ARE or dAe/dt method). More than 99% of the total excretion is completed in 8 h after both IV and oral administration. The terminal phase is characterized by a half-life (t LZJ between 22 k 8 h and 36 f 13 h depending on the route of administration and method of calculation.
Perindoprilat kinetics Infusion of perindoprilat over a 2 h period results in plasma concentrations of 280 f77 ng-ml- I . Following the infusion period, a biphasic decay of plasma concentrations is observed. The initial rapid phase is characterized by a half-life (t mi) of 0.93 f0.25 h and a volume of distribution (Vi) of 9.3 k 3.0 1 . The second and slow phase is characterized by a half-life (t tat) of 29 -t 20 h. With perindopril, this phase corresponds to low concentrations, highly dispersed values for t tat and poor fits.
Table 11. Mean (fSD) parameters for perindoprilat after oral (treatment A) and intravenous (treatment B) administration of 8 mg of perindopril (tert-butylamine salt) and intravenous infusion in 2 H (treatment C) of an equimolar dose (6.1 mg) of perindoprilat to 12 volunteers. Parameter
Treatment
Treatment A
Treatment B
( n g - m l I)~
11.1 f 4 . 0
15.Ok 3.9
3 . 0 f 1.0 145 + 36 -
3.8k 1 . 1 207 f48 -
-
-
-
-
-
-
-
Treatment C
Plasma
c,,,
5.1 + 1.7 3 0 f 15
280 k I7 (end of infusion period)
PI 906 f226 120+34 l.Sk0.4 0.93 k0.25 9.3 3.0 -
+
72+ 16 16.8k4.8
5.4k1.4 25fll 23.4 4.4 [1W -
0.78 i 0.16 12.8+2.6
1.09f0.32 17.9f5.3
7.0+ 1.2 114+20
6.5k1.9 4.9 f 1.9
6.9 k 2.4 5.6 f2.6
2.2 + 0.6 1.5k0.2
22+2 30+ 1 1
20+3 26+7 15.7 k 3.6 [I001
20+3 23+4 [I001
-
[1001 141 k42
-
-
77 + 25 1 l . 4 + 2.8 102 k23
*
93 k 24
29 f 20 [IWI [I001
-
183
Perindopril pharmacokinetics
Total clearance for perindoprilat is 120f 34 ml.min-' or 1.8 f 0 . 4 ml.min-'.kg-'. Renal excretion of perindoprilat after infusion is higher than dose : 114 f 20'70, leading to a renal clearance of 141 f 4 2 ml-min-', higher than total plasma clearance. These results could be of analytical origin since the first highly concentrated urine samples were diluted in order to fall within the range of the calibration curve of assay; they could also be explained if the true administered dose is higher than the theoretical one. Excretion profiles are biphasic with an initial decay characterized by a half-life (t mi)of 1.5 0.2 h or 2.2 f 0.6 h depending on the calculation method; 96.6 f 0.4% of total perindoprilat excretion is completed in 12 h. The terminal phase of excretion is characterized by a half-life (t m,)of 20 f 3 h (ARE method) or 23 k 4 h (dAe/dt method). Intravenous and oral administration of perindopril leads to peak perindoprilat concentrations of 15.0k3.9 ng-ml-' and 1 1 . 1 k4.0 ng-ml-' after 3 . 8 + 1 . 1 h and 3.0 f 1 .O h respectively. The fraction (fm) of injected perindopril converted by hydrolysis to perindoprilat is 23.4 f 4.4% of the administered dose. The relative availability (F') of perindoprilat after oral perindopril is 72 f 16% of perindoprilat after IV injection of the parent drug. In this case, the fraction of orally administered perindopril dose present as perindoprilat (F") is 16.8 f 4.8%. As a result of the urinary excretion of perindoprilat exceeding the infused dose, fm and F" values calculated from urine data are less than plasma values: 15.7 f 3.6% and 11.4 f 2.8% respectively. Oral availability of perindoprilat (F') from urine data is 77 f25%, close to the plasma value. Following both routes of administration of perindopril, the decay
Table 111. Mean (kSD) parameters for perindoprilat glucuronide after oral (treatment A) and intravenous (treatment B) administration of 8 mg of perindopril (tert-butylamine salt) and intravenous infusion in 2 h (treatment C) of an equimolar dose (6.1 mg) of perindoprilat to 12 volunteers.
Parameter
Treatment
Treatment A
Treatment B
Treatment C
64 k 20 1.4k0.4 178 50
20+6 0.7 k 0.3 59+ 17
1.1 i 0 . 4
1.3 k0.2
1.7ka.3
-
1.46k 0.28 15.6k3.0 142 k 33
0.58k0.18 6.3 k 2.0 169 + 57
0.07 + 0.02 0.1 k0.2
1.6k0.1 1.5 k0.2
2.0+0.3 1.6 t 0 . 4
2.8 k0.6 1.9+0.5
5.7 + 2.0 6.2 k 1.7
6.7 + 1.7 8.2 + 2.0
7.2k1.7 1.7 + 2.5
1.8k0.3 -
-
184
JP Devissaguet et al
of perindoprilat concentration after the peak is biphasic with a half-life (t nC)of 5.1 f 1.7 h and 5.4 f 1.4 h after oral and intravenous routes respectively, and a terminal half-life (t q)of 30 f 15 h and 25 f 11 h after oral and intravenous routes, respectively. The same comment regarding dispersion and fit applies to terminal half-lives of perindoprilat after perindopril administration. Excretion of perindoprilat is 12.8f2.6% and 17.9f5.3'70 of the dose after oral and intravenous administration of perindopril respectively. The renal clearance of perindoprilat following perindopril administration is 102f23 ml-min- and 93 f24 mlamin- after oral and intravenous routes respectively, these values being more realistic than that obtained after perindoprilat infusion. Excretion profiles of perindoprilat reflect plasma concentration profiles with half-lives between 4.9k 1.9 and 6.9f2.4 h for the first phase and between 2 0 f 3 h and 30+: 11 h for the terminal phase, depending on the calculation method and route of administration.
'
Perindoprilat glucuronide kinetics Very low plasma concentrations and urinary excretion of perindoprilat glucuronide are obtained after IV infusion of the active metabolite, leading to the conclusion that this conjugation process only has a minor role in the elimination pathway of perindoprilat. A . peak plasma concentration of 1.1 f0.4 ng-ml- was obtained in 1.8 0.3 h but area and half-life calculations were not possible. Urinary excretion constituted 0.7 f0.2% of the dose. Excretion was biphasic with half-lives of 2.8 f0.6 h or 1.9 f0.5 h for the first phase and 7.2 f 1.7 h or 7.7 f2.5 h for the terminal phase, depending on the calculation method, ARE and dAe/dt respectively. After IV administration of the parent drug perindopril, peak plasma concentrations of perindoprilat glucuronide are 2 0 f 6 ng-ml-' for a peak time at 0.7f0.3 h. Plasma concentrations rapidly fall to undetectable values with a half-life of 1.7 f0.3 h. Renal excretion represents 6.3 f2.9% of the dose and the renal clearance of perindoprilat glucuronide is 169k 57 mlsrnin-'. Excretion profiles clearly exhibit 2 phases with half-lives of 2.0 f0.3 h (ARE) or 1.6 f0.4 h (dAe/dt), and 6.7 f 1.7 h (ARE) or 8.2 f2.0 h (dAe/dt). After oral administration of perindopril, peak plasma concentrations for the perindoprilat glucuronide are higher (64f20 ng-ml- ') and are obtained after 1.4 f0.4 h. The area under the curve is 178f50 ng-ml-'ah after oraI administration of perindopril but only 5 9 f 17 ngm-'.h after IV injection of the drug. The half-life for plasma concentration decay is 1.3 f0.2 h. Renal excretion of perindoprilat glucuronide is 15.6 f3.0% of the perindopril dose and the renal clearance is 142f 33 ml-min-'. The biphasic profile of excretion is characterized by half-lives of 1.6 f0.1 h (ARE) or 1.5 f0.2 h (dAe/dt), and 5.7 f2.0 h (ARE) or 6.2 f 1.7 h (dAe/dt).
*
Discussion and Conclusion The present study associating intravenous infusion of the active metabolite perindoprilat with both oral and intravenous administration of the parent drug perindopril
Perindopril pharmacokinetics
185
was designed to allow precise estimations of bioavailability and conversion ratio of the drug to its diacid. Furthermore, the ability of the assay to measure plasma and urine levels of perindoprilat glucuronide provides an insight into the metabolic pathway of perindopril. Kinetic analysis
Polyexponential equations, assuming multicompartmental models, have been used to describe plasma concentration time profiles of non-thiol ACE inhibitors (Hockings et al, 1986). Recently perindoprilat kinetics after single bolus IV injection were described through an open 3-compartment model (Lees and Reid, 1987b). However, multicompartmental models were unable to simultaneously follow kinetics after single and multiple dosings, and accumulation ratios were much lower than predicted from the long terminal half-lives observed (Till et al, 1984). A onecompartment model with saturable binding was used to accurately relate kinetic and dynamic data, ie plasma concentration of ACE inhibitors and ACE inhibition in plasma (Francis et al, 1987). High-affinity, limited-capacity and specific binding of inhibitors to plasma ACE, in conjunction with low-affinity, high-capacity and nonspecific binding to other proteins, may explain their long terminal half-life and solve the paradox of the low accumulation ratio (Nussberger et al, 1987). It can then be assumed that high plasma concentrations of ACE inhibitors are mainly related to the free drug or drug bound to non-specific proteins and that low concentrationsare related to the drug bound to ACE, the slow dissociation rate of the complex acting as a limiting step for the elimination of the drug. In the present study, model-independent methods were used to characterize the kinetics of perindopril and its metabolites. Polyphasic profiles for some plasma and urine data have led to the calculation of 2 or 3 half-lives. These half-lives were directly derived from true experimental values during apparent log-linear decay but not from polyexponential analysis. From a theoretical point of view this method is debatable, but from a clinical point of view it allows a separate determination, with sufficient accuracy for therapeutic purposes, of the essential kinetic parameters of the free and bound drug. Perindopril kinetics
From both plasma and urine data, gastrointestinal absorption of perindopril is high and leads to an oral bioavailability of about 95% of the dose. The onset of absorption is rapid and peak concentrations are obtained in about 1 h. Disposition of the parent drug after both oral and IV routes is polyphasic. The initial decay after IV injection (t n: 14 min) can be attributed to dilution into blood and/or extravascular diffusion into well perfused tissues. The main phase after both oral and IV administration corresponds to a half-life of about 1 h and during this phase more than 99% of the renal excretion of unchanged perindopril
186
JP Devissaguet et a1
is completed, when residual plasma concentrations are about or less than 1070 of the peak values. Considering this half-life as the biological half-life of perindopril, the volume of distribution ( z 3 1 1) corresponds to body water, in good agreement with the hydrophilic character of the drug. The terminal phase corresponds to very low plasma concentrations and urine excretion rates of perindopril and poor fittings were obtained for half-life calculations. Repeated administration of pgrindopril has not led to accumulation (Bromet et al, 1988), indicating that this long terminal half-life does not correspond to a deep compartment. From both plasma and urine data, the terminal half-life of perindopril is closely related to the terminal half-life of perindoprilat and this result could suggest a saturable binding of the parent drug to ACE. In vitro data clearly indicate that binding of perindopril cannot be responsible for its long terminal half-life (Jackson et al, 1987). One possible explanation is that the terminal half-life is an apparent one and an artefact of analytical origin. This assumption has not been verified but the assay of perindopril includes a chromatographic separation from metabolites, followed by hydrolysis of the samples and determination of the active diacid. As the chromatographic samples corresponding to perindopril also contain proteins, some perindoprilat bound to plasma ACE could be present as a contaminant of the parent drug. Elimination of perindopril is mainly due to metabolic processes, and renal excretion of the unchanged drug represents only 15.6% of the IV dose. Nearly half of the IV dose elimination (45.5%) was obtained by adding to excretion the biotransformations of perindopril to the active diacid perindoprilat (23.4%) and to the perindoprilat glucuronide (6.3%). Total plasma clearance of perindopril is high (362 ml-min-l), compared to a renal clearance of about 65 ml-min-', a metabolic clearance of about 85 ml-min- for the formation of perindoprilat and a metabolic clearance of about 22 ml-min- * corresponding to perindoprilat glucuronide. Moreover, this glucuronide was 3-fold more abundant in plasma and urine after oral administration of perindopril, but present in very low amounts after IV infusion of perindoprilat . In agreement with Fournel-Gigleux et a1 (1988), those results strongly suggest that perindoprilat glucuronide is a hydrolysis residue of a parent drug glucuronide and that perindopril is subjected to a presystemic first-pass effect after oral administration. By substituting the urine excretions of perindoprilat glucuronide after oral and IV administration of perindopril, it can be concluded that this first-pass effect concerns = 9% of the dose, in agreement with the bioavailability of perindopril. This result also suggests that other metabolites, and especially the active diacid are formed after absorption, the first-pass effect acting only by glucuronidation of the parent drug.
Perindoprilat kinetics Disposition behaviour of the active diacid metabolite after its IV infusion and after perindopril administration by both oral and IV routes is quite different. IV infusion of perindoprilat leads to a rapid elimination of the free drug (tm: 0.93 h) by renal
Perindopril pharmacokinetics
187
excretion. Extravascular diffusion of perindoprilat is very limited, with a volume of distribution (9 1) that is smaller than the volume of the parent drug, in good agreement with the more hydrophilic character of the active metabolite. After IV or oral dosing of perindopril, formation of perindoprilat is quite slow and the peak time for the plasma concentrations is between 3-4 h. The decay of plasma concentration after peak (15 ng-ml-I and 11 ng-ml-' after IV and oral dosing respectively) is also quite slow compared to the decay after IV infusion of perindoprilat (tvr: z 5.5 h from plasma and urine data). This feature, which is in agreement with the results described by Lees and Reid (1987a, b) could reflect the rate of formation of perindoprilat from the parent drug and as this process is only a small part of the total elimination process for perindopril, it should have no apparent effect on the half-life of the parent drug. The terminal long half-life of perindoprilat (tvz: 20-30 h) is present after any drug dose or route of administration and can be characterized from both plasma and urine data. It is assumed that this slow elimination corresponds to perindoprilat bound to ACE and reflects the rate of dissociation of the complex. After perindopril dosing, formation of perindoprilat represents about 20% of the available perindopril (23% from plasma data and 18% from urine data). This result is in good agreement with other experiments (Lees and Reid, 1987b) and corresponds to z 1.2 mg of active metabolite available after 8 mg oral dosing of perindopril as tert-butylamine salt. A 1-mg dose of perindoprilat has been shown to totally inhibit plasma ACE after IV injection, more than 50% of inhibition being obtained over ::20 h with an IC,, of 1.8 ngem1-I (Lees and Reid, 1987a). The mean plasma concentrations of perindoprilat are 3.16 ng-ml- 12 h after oral administration of 8 mg of perindopril and 1.15 ng-ml-' after 24 h. A prolonged pharmacological response is therefore expected after oral dosing with 8 mg and has been reported as being at 60% of ACE inhibition 24 h after treatment (Lees and Reid, 1987a). As for other non-thiol ACE inhibitors, perindoprilat elimination is essentially an excretion process with a renal clearance of z 100 mlamin-I, close to the glomerular filtration rate. As previously stated, the formation of perindoprilat glucuronide from perindoprilat is negligible.
Perindoprilat glucuronide kinetics Perindoprilat glucuronide, mainly resulting from a first-pass effect after oral administration of perindopril, reaches higher plasma concentrations (:: 64 ng-ml- I), more rapidly (tmax: 1.4 h) than perindoprilat. These results confirm that this glucuronide does not come from the active diacid. Formation of the glucuronide is rapid even after IV administration of perindopril (t,,, : 0.7 h), but peak concentrations remain quite low (20 n g m - '). The elimination of perindoprilat glucuronide from plasma is rapid and monophasic (t vz= 1.5 h) and concentrations reach undetectable levels before any slow phase appears, in contrast with urine where a second half-life (:: 7 h) can be calculated with good accuracy. Repeated oral administration of perindopril
188
JP Devissaguet et
a1
does not lead to accumulation of the perindoprilat glucuronide (Bromet et al, 1988) and it has been shown that this metabolite can bind to ACE, but with a lower affinity (IC50 = 64 nM) than the diacid metabolite of perindopril (IC50 = 2.4 nM). Thus, as for ACE inhibitors, the slow elimination phase of the perindoprilat glucuronide could reflect the dissociation of the complex. The renal clearance of perindoprilat glucuronide ( z 155 ml-min-’ from both oral and IV routes for perindopril) is the highest of the 3 renal clearances calculated in this study, but there is no experimental evidence to indicate that renal excretion is the only process involved in the glucuronide elimination.
Conclusion The protocol design of this study allowed the estimation of essential kinetic parameters for the parent drug perindopril, for its active diacid metabolite perindoprilat and for perindoprilat glucuronide. Furthermore, this study allowed some useful insights into the metabolic pathway of perindopril. All treatments were well tolerated during the study. Perindopril is intensively and rapidly available, its elimination, mainly by metabolic process, is rapid and leads to z 20% of the available dose being converted to the active metabolite. Perindoprilat exhibits well-known characteristics of non-thiol ACE inhibitors, such as rapid renal excretion of free drug and long terminal half-life of bound drug. Perindoprilat glucuronide, mainly present after oral administration of perindopril, seems to be formed from the parent drug itself and not by conjugation of the active diacid.
References Bromet N, Lelievre E, Vidal D, Grislain L, Devissaguet M, Jochemsen R, Ammoury N, Devissaguet J P (1988) Etude pharmacodynamique et pharmacocinttique chez I’homme d’un nouvel inhibiteur de l’enzyme de conversion de I’angiotensine I, le perindopril, apres administration unique et repttte. Journtes de I’HTA, Paris, 15-16 December Drummer OH, Rowley K, Johnson H, Worland P, Workman B, Harris QLG, Conway EL, Louis WJ (1987)Metabolism and pharmacodynamics of angiotensin converting enzyme with special reference to perindopril. Excerpta Med Znt Congr Ser 750, 545-550
Fournel-Gigleux S, Magdalou J, Lafaurie C, Grislain L, Garnier NH. Dabe JF, Luijten W, Bromet N, Devissaguet M (1988) Glucuronidation of perindopril by hepatic microsomes : interspecies comparison. Cellular and Molecular Aspects of Glucuronidation, Montpellier, 27-29 April Francis RJ, Brown AN, Kler L, Fasanella d’Amora T, Nussberger J, Waeber B, Brunner BR (1987) Pharmacokinetics of the converting enzyme inhibitor cilazapril in normal volunteers and the relationship to enzyme inhibition: development of a mathematical model. J Cardiovasc Pharmacol9, 32-38 Hockings NB, Ajayi AA, Reid JL (1986)Age and the pharmacokinetics of angiotensin
Perindopril pharmacokinetics
converting enzyme inhibitors enalapril and enalaprilat. Br J Clin Pharmacol21, 341-348
Jackson B, Cubela RB, Johnston CI (1987) Angiotensin converting enzyme inhibitors : measurement of relative inhibitory potency and serum drug levels by radio-inhibitor binding displacement assay. J Cardiovasc Pharmacol 9, 699-704 Laubie M, Schiavi P, Vincent M, Schmitt H (1984) Inhibition of angiotensin I converting enzyme with S 9490: biochemical effects, interspecies differencies and role of sodium diet in hemodynamic effects J Cardiovasc Pharmacol 6, 1076- 1082
Lees KR, Reid JL (1987a) Effects of intravenous S 9780, an angiotensin converting enzyme inhibitor, in normotensive subjects. J Cardiovasc Pharmacol 10, 129-135
Lees KR, Reid JL (1987b) The haemo-
189
dynamics and humoral effects of treatment for one month with the angiotensin converting enzyme inhibitor perindopril in salt replete hypertensive patients. Eur J Clin Pharmacol 3 1, 5 19-524 Morgan T, Anderson A, Wilson D, Murphy J, Nowson C (1987) The effect of perindopril on blood pressure in humans on different sodium intakes. J Cardiovasc Pharmacol 10, S 116, S 118 Nussberger J, Fasanella d’Amora T, Porchet M (1987) Repeated administration of the converting enzyme inhibitor cilazapril to normal volunteers. J Cardiovasc Pharrnacol9, 39-44 Till AE, Gomez HJ, Hichens M, Bolognese JA, Mc Nabb WA, Brooks BA, Noormohamed F, Lant AF (1984) Pharmacokinetics of repeated single oraI doses of enalapril maleate (MK 421) in normal volunteers. Biopharm & Drug Dispos 5 , 273-280
