Summary. A metabolite of prazosin was detected in serum from hypertensive patients treated with pra- zosin. Its structure as 2-(1-piperazinyl)-4-amino-.
European Journal of Clinical Pharmacology
Eur J Clin Pharmacol (1984) 27:275-280
© Springer-Verlag 1984
Identification of a Prazosin Metabolite and Some Preliminary Data on its Kinetics in Hypertensive Patients V. K. Piotrovskii, N. N. Veiko, O. S. Ryabokon, S. F. Postolnikov, and V. I. Metelitsa Institute of Preventive Cardiology, USSR Cardiology Research Centre, Academy of Medical Sciences, Moscow, USSR
Summary. A metabolite of prazosin was detected in serum from hypertensive patients treated with prazosin. Its structure as 2-(1-piperazinyl)-4-amino6,7-dimethoxyquinazoline was established by UV, IR, and mass-spectrometry. An assay method for simultaneous determination of prazosin and its metabolite in serum, urine and saliva is described. Preliminary data about the kinetics of prazosin and the metabolite after a single oral dose of prazosin 1 mg, and after multiple doses of 1 to 5 mg t.i.d, for 6-82 days in 7 patients with hypertension, are presented. After the single dose the metabolite level was much lower than that of intact drug, even though the former was eliminated much more slowly than the latter. The slow elimination of the metabolite led to its eventual accumulation in serum during multiple administration. The mean accumulation ratio of the metabolite was estimated to be at least 5.5 (from 3.0 to 7.9). Prazosin itself had a low accumulation ratio, so the mean steady-state level of the intact drug on multiple administration was several times lower than that of metabolite. As this metabolite has some hypotensive effect in animals, it may account for part of the therapeutic activity of parzosin in patients. The mean steady-state concentration of intact prazosin during the course of treatment were found to be significantly lower than that predicted from a single dose study.
patients [6-8], but many aspects of its fate in man, especially in patients, remain unclear; for example, prazosin metabolism in man has not been well studied, and little is known about its kinetics during chronic treatment. The present report describes certain results obtained during an extensive continued study of the disposition and effects of prazosin in hypertensive patients. For drug assay the authors' own method based on ion-exchange high performance liquid chromatography was used [9]. A pilot study had shown that chromatograms of extracts of serum from patients obtained after oral administration of prazosin contained a peak corresponding to unchanged drug and a new peak which was absent from blank serum. The latter peak was attributed to a possible metabolite of prazosin, so attention was concentrated on its isolation, identification and quantitation. It proved necessary to modify the assay method because the initial version [9] proved unsatisfactory for the metabolite. The results of this newer work are presented here together with preliminary data about the kinetics of the metabolite and the parent drug.
Key words: prazosin, hypertensive patients; prazosin metabolite, HPLC assay, pharmacokinetics, hypotensive effect
To 1 ml aliquot of a biofluid (serum, saliva or urine) was added 0.1 ml internal standard solution (p 359, Orion Pharmaceutica, Helsinki, Finland) and 1 N potassium hydroxide 0.2 ml. The mixture was vortexed with 5 ml ethyl acetate. After centrifugation for 5 min at 2000 g, the organic layer was transferred to a conical tube and was extracted with 0.1 N H2SO 4 0.1 ml. After brief centrifugation, 10-50 ~tl of the acidic layer was chromatographed.
Prazosin is an antihypertensive agent with alphaladrenergic blocking activity [1]. Its pharmacokinetics has been studied in healthy volunteers [2-5] and in
Materials and Methods
Assay Method
276 The chromatography was performed with a Model B-100-S-2 Eldex high pressure pump, Model 7125 Rheodine injection valve and Model FS-970 Schoeffel fluorescence detector (Schoeffel Instruments, Westwood, N J, USA). The chromatograms were processed using a Model C-E1B Chromatopac processor (Shimadzu Corp, Kyoto, Japan). The column used was 4.6 m m x 25 cm packed with Partisil 10-SCX cation-exchanger (Alltech Ass., Deerfild, IL, USA). The mobile phase consisted of (by volume) 14.7% acetonitrile, 83.4% deionized water, 0.5% diethylamine and 1.3% orthophosphoric acid. The flow rate was 2 ml/min. The detector excitation wavelength was maintained at 246 nm. An emission filter with a 370 nm cut off was used. Quantitation was done by means of the internal standard method.
Metabolite Isolation, Synthesis and Identification After alkalinization the prazosin metabolite was extracted with ethyl acetate from combined urine samples obtained from several patients treated with prazosin. The extract was then shaken with 0.1 N H2SO4 and the metabolite was isolated by preparative liquid chromatography on a 10 m m x 25 cm Partisil 10-SCX column (Altex Sci., Berkeley, CA, USA). The mobile phase was prepared from acetonitrile (20% by volume), water (77%), diethylamine (0.8%) and glacial acetic acid (1.6%). The flow rate was 4.2 ml/min. Fractions of the effluent containing the metabolite were evaporated to dryness, the residue was redissolved in 0.1 N HC1 and chromatographed on paper in isopropanol - ammonia - water (7:1 : 2). The metabolite was eluted from the paper with an acetonitrile - water mixture (1 : 4) and lyophilised. The metabolite standard was prepared from prazosin hydrochloride (Orion Pharmaceutica). 10rag prazosin HC1 was heated with 6 N HCI 15 ml in a boiling water bath for 3 h. After cooling and neutralization, the metabolite was isolated as above. The structure of the metabolite was established by mass spectrometry, IR and UV spectrophotometry. Mass spectra were obtained on a MAT-112 VarJan GC-MS system. Methanolic solutions of samples were directly injected into the ion source operating at 200 °(3. Ionizing energy was 50 eV. IR spectra of samples in Nujol were recorded with a UR-20 spectrophotometer (Carl Zeiss, Jena, GDR). UV spectra in water were recorded with a Specord UV-VIS spectrophotometer (Carl Zeiss). The identity of the metabolite isolated from urine and that prepared from prazosin was confirmed by HPLC on two different columns, cation-exchange and reversed phase (4.6 mm x 25 cm Ultrasphere
V.K.Piotrovskiiet at.: PrazosinMetabolitesin HypertensivePatients ODS, Altex Sci.). For the latter the mobile phase consisted of 12.5% acetonitrile, 87.5% water, 0.1% diethylamine and 0.1% orthophosphoric acid.
Pharmacokinetic Study Seven men with hypertension (WHO Stage II) took part in the parmacokinetic study of prazosin (Pratsiol, Orion Pharmaceutica). Their mean age was 46 years (35-56 years), and mean body weight 87 kg (77-106 kg). Both liver and renal functions were normal, as assessed by history, physical examination and appropriate laboratory tests. Three were smokers (Nos 1, 2 and 4; see Tables 1, 2). The patients did not take any drugs for at least 2 weeks prior to the study. Each patient took part in two separate pharmacokinetic studies, first with a single oral dose of 1 mg prazosin, and second after the last dose of a prolonged course of increasing (in all but one patient) doses given three times daily. The overall duration of the course, the size of the last dose and the period for which the latter was administered are presented in Table 2. On both occasions the drug was given 2 h after a standard, light breakfast with 100 ml water. No food was permitted for 4 h after drug administration. Venous blood samples were collected before and 15, 30 and 45 min, and 1, 1.5, 2, 3, 4, 6, 8, 10 and 24h after dosing. In some cases blood was taken also 30 h after a single dose and 30, 36, 48 and 72 h after the last dose of the prolonged course. Blood samples were allowed to clot and serum was separated and frozen at - 1 8 °C until analysed.
Pharrnacokinetic Calculations The half-life (tO of prazosin was calculated from the slope constant (Ice) of the terminal linear part of the semilogarithmic plot of concentration versus time. The linear trapezoidal rule was used to estimate areas under serum concentration - time curves (AUC) within the sampling period, and for AUC from zero to infinity, extrapolated areas determined as Cz/ke (where Cz is the last measuremem of the drug concentration) were added. The AUCs of prazosin and the metabolite from zero to 8 h (AUC8p and AUC8m, respectively), both after single (s) and multiple (mlt) administrations, and the AUC of prazosin from zero to infinity after the single dose (AUC~,s) were calculated. The mean actual (Css,a) and predicted (Css, pd) steady-state levels of prazosin were assessed, dividing by 8 AUC~, mlt and AUC p, s, respectively. The accumulation ratios of prazosin and metabolite were estimated as AUC~, mlt/AUC~, s and AUC~,mlt/AUC~,s, respectively (AUCs after chronic treatment were adjusted for the size of the
V.K. Piotrovskii et al.: Prazosin Metabolites in Hypertensive Patients
277
last dose). All the calculations were made using an HP-85 personal computer (Hewlett-Packard, Avondale, USA). Statistical significance of differences was assessed by the paired t-test.
Results 2
Assay Method and Metabolite Identification c o
3 c0 [23
0 3 6 9 Time {rain) Fig. 1. Chromatograms of extract of blank serum a and of the extract of serum from a patient during prazosin treatment b. Peak identification: 1 - intact prazosin, 2 - prazosin metabolite, 3 - internal standard (p 359). For chromatography conditions see text
OCH3 H3C O . . . ~ ] ~ N
Prazosin
II
O OCH3 H3CO~N
H~N~N XN~N
Metabolite
H
Fig.2. Formulae ofprazosin and its metabolite
The method described provided good spearation of the substances analysed. Chromatograms of extracts of blank serum (a) and of serum from the patient on chronic treatment with prazosin (b) are shown in Fig. 1 ; Peaks 1, 2 and 3 correspond to intact prazosin, metabolite and the internal standard, respectively. Calibration graphs obtained for prazosin in the range 1-60 ng/ml and for the metabolite in the range 1-120 ng/ml were found to be linear. The determination limit was 0.1-0.2ng/ml for both the compounds, and the interassay variability was about 10% at 0.2 ng/ml and less than 7% at 1 ng/ml. In urine from patients receiving prazosin the same metabolite was found as in serum. Its level in urine was much higher than in serum. In the mass spectrum of the metabolite isolated from urine the m / e of the major peak was 233, as in the case of prazosin [10], which means that the dimethoxyquinazoline structure remains intact in the metabolite. The molecular ion of the metabolite had an m / e of 289 (53%). The same peak was present in the prazosin mass spectrum and corresponded to the loss of the 2-furoyl group. Other peaks of m / e 259 (22%), 245 (42%), 220 (70%) and 205 (31%) coincide in the mass spectra of the metabolite and intact prazosin. In the IR spectrum of the metabolite the characteristic bands for prazosin at 3050-3340cm -1, corresponding to stretching of the C-H bond of the furan ring, were absent. There were intense bands at 2360-2540cm -1, corresponding to N-H bond stretching in secondary amine salts. The UV spectrum of the metabolite contains an intense band of 2max= 246 nm. Absorption in this region is due to the quinazoline aromatic system. In prazosin this band is widened and has a shoulder at 270-280 nm, corresponding to the characteristic absorption of the furoyl group with ";max= 270 nm. This widening and shoulder were absent from the spectrum of the metabolite. The data obtained show that the metabolite has the structure of 2-(1-piperazinyl)-4-amino-6,7-dimethoxyquinazoline; its formula is presented in Fig. 2. The metabolite can be formed from prazosin by acid hydrolysis. By heating prazosin with 6 N HC1 it
278
E
V.K. Piotrovskii et al.: Prazosin Metabolites in Hypertensive Patients
20 ~ " ' ~ F'.-,
orient 3
8
20 10
50
1
]k
