Liquid Chromatographic Assay and Pharmacokinetics of Halatepam and Its Metabolite in Humans

Liquid Chromatographic Assay and Pharmacokinetics of Halatepam and Its Metabolite in Humans SAMlR K. GUPTA' AND EVEREIT H. ELLINWOOD Received September 30, 1988, from the Department of Psychiatry, Duke University Medical Center, Box 3870, Durham, NC 27710. Accepted for publication November 10, 1989. Abstract 0A reversed-phase high-performance liquid chromatographic

method is described for simultaneous quantification of halazepam and its major active metabolite, nordiazepam, in plasma. The method uses a solid-phase extraction procedure to prepare plasma samples. After extraction, the methanolic extract is evaporated, and the residue is then reconstituted in a small volume of mobile phase (a 40:60,v/v, mixture of 0.02 M phosphate buffer, pH 4.0, and methanol) and chromatographed.The total chromatographytimeforasinglesampleis -10 min. Asensitivityof 1 ng/mL for halazepam and nordiazepam is attained when 1 mL of plasma is extracted.Analytical recovery of halazepam and nordiazepam added to the plasma ranged from 89 to 96%. The maximum within-day and day-to-day coefficientsof variation for each compound at the concentrationrange of 2 to 100 ng/mL were 8.7 and 10.3%, respectively. Suitability of the method was assessed in a preliminary pharmacokinetic study in which three subjects were given a single 20-mg oral dose of halazepam following an overnight fast. It appeared from our kinetic analysis that halazepam is adrug with a fairly rapid absorption phase that is followed by a slow elimination phase. Mean oral plasma clearance of halazepam was 24 Uh. The mean apparent elimination half-life of nordiazepam (45.22 h) is considerably longer than that of halazepam (21.15 h).

Halazepam (HZ) is a fluorinated 1,Cbenzodiazepine derivative. It has an N-1-trifluoroethyl group in place of the N-1-methyl group of diazepam. Halazepam was found to possess pronounced central nervous system activity coupled with a low order of toxicity.l.2 Presently, HZ is approved by the FDA for the treatment of anxiety and tension. Oral studies in humans have shown that HZ is rapidly and almost completely absorbed.3 Maximum plasma concentration occurs within 2-3 h after dosing.3 Halazepam has been considered a prodrug of nordiazepam (NDZ),4 as it undergoes rapid first-pass metabolism in the liver with the removal of trifluoroethyl side chain.3.4 In fact, NDZ is also the metabolic product of several other benzodiazepines, including diazepam, medazepam, prazepam, and clorazepate. In addition to NDZ, 3-hydroxyhalazepam (HOHZ), another metabolite of HZ, has also been identified in human plasma (see structure). However HOHZ is not present in plasma in measurable amounts because it undergoes rapid glucuronidation and is subsequently excreted in urine. At present, two gas chromatographic (GC) methods and no liquid chromatographic methods are available in the literature for determination of HZ and NDZ concentration in biological samples.5.6 Both GC methods are time consuming because of the long extraction procedures and a two-part assay procedure requiring two different internal standards536 and two different columns.6 We describe here a simple and time-efficient solid-phase extraction procedure for plasma sample preparation and a rapid and sensitive high-performance liquid chromatographic (HPLC) method for simultaneous determination of HZ and NDZ in plasma. This method was used to determine the pharmacokinetics (a preliminary study) of these compounds in healthy subjects following oral administration of HZ. 822 / Journal of Pharmaceutical Sciences Vol. 79, No. 9, September 1990

Experimental Section Reagents and Standards-Halazepam (HZ) and nordiazepam (NDZ)were provided by Schering Corporation (Bloomfield, NJ), and diazepam (DZ) was provided by Hoffmann-LaRoche (Nutley,NJ). The HPLC-grade methanol, phosphoric acid, and potassium monobasic phosphate were obtained from Fisher Scientific (Pittsburgh, PA). Glycine was obtained from Sigma Chemical (St. Louis, MO). Extraction Apparatus-The C, Bond-Elut columns and a Vac-Elut SPS24 apparatus were obtained from Analytichem International (Harbor City, CA). Instrumentation and Chromatographic Conditions-The HPLC system was equipped with a Waters Associates (Milford, MA) dual piston, positive displacement solvent delivery system (model 5011, an automatic injection module (model 712, WISP), a programmable multiwavelength, multichannel detector (model 7301, and an electronic integrator (model 745B). Chromatographic separations were made on a Alltech Associates (Deerfield, IL) Adsorbosphere C, column (10 cm x 4.6 mm i.d., 3-pm particle size). The mobile phase was 0.02 M phosphate buffer (pH 4.0kmethanol (40:60,v/v) filtered through a nylon 0.45-pm membrane (Schleicher and Schuell, Keene, NH). The chromatograph was operated at ambient temperature using a flow rate of 1 mL/min (1100 psi). Effluents were monitored at 240 nm. In order to determine the unknown plasma HZ and NDZ concentrations, standard curves were constructed from relative peak heights (HZ or NDZ to DZ) obtained from the integrator and drug concentrations. Standard Solutions-Working standard solutions were prepared by dissolving 10 mg of HZ, NDZ, and DZ (internal standard; see structure) in 100 mL of methanol. Sequential dilutions to 1 pg/mL were then made in 1M glycine buffer (pH 10.5).Calibrationstandards were prepared by adding HZ and NDZ to drug-free plasma to obtain concentrationsranging from 1 to 100 ng/mL. Preparation of Samples-Bond-Elut columns were placed on top of the Vac-Elut vacuum manifold. With the vacuum on, each column was washed with 2 mL of methanol and deionized distilled water. To prevent the column from drying out, the vacuum was shut off as soon as the water had run through each column. A 100-pL volume of DZ Ri

Halazepam Nordiazepam

-CH,CF,

-H

-H

-H

3-OH-halzepam -CH,CF,

-OH

Oxazepam

-H

-OH

Diazepam (I.S.)

-CH,

-H

0022-3549/90/0900-0822$0 1.0010 0 1990, American Pharmaceutical Association

solution containing an appropriate concentration of DZ was added (see Table I) to each column; this was followed by addition of 0.5 mL of standard (2-100 ng/mL) or 1 mL of (1 ng/mL) standard or sample plasma onto each column. Then, the vacuum was applied to draw the standards or samples through the column, the column matrix was washed with 2 mL of deionized distilled water followed by 50 pL of methanol, the vacuum was disconnected, and the eluant was discarded. A 10 x 75-mm silanized glass tube (appropriately labeled) was placed under each column to collect the eluant. A 200-pL volume of methanol was added to each column and the vacuum was then applied to draw the methanol into the collection tubes. The process was repeated with two 200-pL volumes of methanol. The combined methanol eluant was evaporated to dryness at 37 "C under a gentle stream of nitrogen, and the residue was reconstituted with 100 pL of mobile phase and then transferred to automatic sampling vials. Following this process, aliquots of 50-80 pL were injected into the chromatographic system by the automatic sampler. Analytical Recovery-Drug-free plasma spiked with 2, 5, 20, 50, and 100 ng/mL of HZ and NDZ was analyzed according to the above-described method without any added internal standard. Carefully measured aliquots of the reconstituted extract were injected and the peak heights of each compound were measured. Absolute recovery was calculated by comparing these peak heights with those obtained by direct injection of drug standards. Pharmacokinetic Study-Three healthy, nonsmoking men aged 24, 28, and 30 years and weighing 71, 78, and 74 kg, respectively, participated in this study. All subjects were normal by physical examination and laboratory screening profiles and did not have a history of recent chronic disease. The purpose and procedure were explained to the subjects and written informed consent was obtained. Alcohol and OTC drugs were excluded for 72 h prior to the study session. Prescription medication was excluded 2 weeks prior to the study day. Each subject received a single 20-mg oral dose of HZ (Paxipam Tablet 7NCH2; Schering, Kenilworth, NJ) with 200 mL of water following an overnight fast. A 7-mL venous blood sample was drawn into a heparinized vacutainer at zero (pre-dose), 0.25, 0.5, 1, 2, 4, 8, 10,24,30,48,56,72,80,96,120,144,192, and 240 h after the dosing from an obturated indwelling catheter in a subject's left forearm. The blood samples were centrifuged and plasma was harvested and frozen as soon as possible. Plasma was stored frozen until assayed. Data Analysis-The observed plasma HZ and NDZ concentrationtime data were used to determine the maximum plasma concentration (CmaX) and the time to reach the maximum plasma concentration (tma.J The apparent elimination rate constant was estimated from the terminal log-linear portion of the concentration-time curve by linear regression analysis.' The area under the plasma concentration-time curve to the last time point (AUCJ was determined by the trapezoidal rule. The total area under the plasma concentration-time curve (AUC) was calculated by adding the AUC, and the residual area calculated by dividing C, with the apparent elimination rate constant, where C, is the plasma concentration at the last time point a t which concentration is determined. The oral plasma clearance (assuming 100% drug absorption from GI tract)3 and the apparent elimination half-life were calculated in the usual manner.8

ternal standard; see structurej gave symmetric, well-resolved peaks (Figure 1A) with retention times of 6.52,7.76,and 9.31 min for NDZ, DZ, and HZ, respectively. Extracts of pooled human plasma yielded no interference from endogenous plasma components, as shown in Figure 1B. Linearity of the detector response was evaluated by injecting known volumes of various methanolic standard solutions containing HZ, NDZ, and DZ in amounts ranging from 1 to 100 ng. The detector response (peak height) was linear over this range for each compound. Relative peak height ratios of NDZ and HZ to DZ from extracted plasma samples were also linearly related to concentration over the range of 1 to 100 ng/mL. The calibration curves obtained for both the low (1 to 5 ng/mL) and the high (10 to 100ng/mL) concentration ranges

I

(A)

v, LL

3

U

In

0 0. 0 c

W v)

z

0

a

v, W

QI QI

E W

t-

W 0

0

5 10 TIME, min

(6)

Results and Discussion Resolution, Linearity, and Sensitivity-Under the described chromatographic conditions, HZ, NDZ, and DZ (inTable I-Regression Analysis of the Calibration Curves of Nordiazepam (NDZ) and Haiazepam (HZ) with Diazepam (DZ) Drug DZ Drug Concentration, Concentration, ng/mL ng/mL NDZ

HZ

1-5 2-1 0 10-1 00 1-5 2-1 0 10-1 00

5 10 50 51 10 50

Equationa y = 0.27x - 0.07 y = 0.1ox + 0.02 y = 0.04~ - 0.01

r2

nb

0.990 5 0.991 5 0.997 6

0.988 5 y = 0.093~+ 0.04 y = 0.087x - 0.03 0.993 5 y = 0.027~+ 0.02 0.995 6

of drug to diazepam (internal standard). Number of points on the curve. = drug concentration, y = relative peak height ratio

1

5 10 TIME, min

Figure 1-(A) Chromatogram of extracted plasma containing 20 ng/mL of nordiazepam ( l ) ,diazepam (2) and halazepam (3). (B) Chromatogram of extracted drug-free plasma. Journal of Pharmaceutical Sciences I 823 Vol. 79, No. 9, September 7990

€or NDZ and HZ were straight lines. The constants of the respective linear regression are listed in Table I. The limits of the detection, allowing a signal-to-noise ratio of 3, are 1 ng of HZ and NDZ. The sensitivity of the method allows for quantitation of at least 2 ngimL of each compound extracted from 0.5 mL of plasma; however, the sensitivity is improved to 1 ng/mL by using 1-mL plasma samples. Recovery and Precision-The extraction of HZ, NDZ, and DZ from plasma by the solid-phase extraction (C, Bond-Elut column) method was good. Halazepam, NDZ, and DZ were added to drug-free pooled plasma to achieve concentrations of 2, 5, 20, 50, and 100 ngImL. Average recovery ranged from 92.6 to 95.7%for NDZ, 89.5 to 93.1%for DZ, and 88.9 to 93.7% for HZ (Table 11). There was no perceivable dependence on drug concentration over the range studied. "he precision of the method was assessed by repeated analyses of spiked plasma samples containing five known concentrations of NDZ and HZ (Table 111). The coefficient of variation (CV%)ranged from 1.1 to 8.7% for within-day determinations and from 1.1to 10.3%for day-today determinations. Interference Studies-A series of benzodiazepine derivatives, including 3-hydroxyhalazepam, were tested for potential interference with our procedure by comparing the retention times of these compounds with those for NDZ, DZ, and HZ. The absolute and relative retention times for those compounds that were detected by our system are listed in Table IV. Of those tested, only triazolam represents a risk of interference with the analysis of NDZ. However, triazolam would not, in clinical practice, be administered with halazepam or vice versa. Since DZ is used in this method as the internal standard, HZ and NDZ cannot be quantified correctly if DZ is present in the patient's plasma prior to HZ dosing. Perhaps because we used healthy subjects in our study, we may not have encountered this problem; however, it is also unlikely that the patient would receive DZ and HZ simultaneously. In the event that DZ is suspected from an examination of the chromatogram obtained from subject's plasma prior to HZ dosing, an alternate internal

Table IV-Absolute and Relative Retention Times, Capacity, and Separation Factors of Various Benzodiazepines

Absolute Relative" Capacity Separation" Retention Retention Factor Factor (K) (4 Time, min Time

Compound Alprazolam Chlordiazepoxide Clonazepam Desmethyldiazepam Desmethylflunitrazepam Diazepam Halazepam 3-hydroxyhalazepam Lorazepam Methylclonazepamb N-desalkyl2-oxoquazepamb~" Oxazepam Prazepam Quazepamd Temazepam Triazolam

7.06 10.27 6.16 6.52 5.82 7.76 9.31 8.56 5.60 5.80 5.11

0.91 1.32 0.79 0.84 0.75 1.00 1.20 1.10 0.72 0.75 0.66

6.20 9.47 5.29 5.65 4.94 6.92 8.50 7.73 4.71 4.92 4.21

0.90 1.37 0.75 0.82 0.71 1 .oo 1.23 1.12 0.68 0.71 0.61

5.71 10.40 9.86 6.16 6.79

0.74 1.34 1.27 0.79 0.86

4.83 9.61 9.06 5.29 5.93

0.70 1.39 1.31 0.76 0.86

a Relative to diazepam used as the internal standard. Can be used as an alternate internal standard. Same as N-desalkylflurazepam. Plasma containing quazepam at high therapeutic concentration was extracted and analyzed.

standard must be used; for examples, methylclonazepam, with a retention time of 5.8 min, or N-desalkyl-2-oxoquapam, with a retention time of 5.11 min (see Table IV). Plasma samples supplemented with quazepam, another trifluoroethyl1,Cbenzodiazepinederivative used clinically as a sedative and hypnotic agent, were also assayed. However, this method does not provide the desired sensitivity for the pharmacokinetic study of quazepam and its metabolites because the chromatographic conditions for quazepam are

Table iCAnaiytical Recovery

Percentage Recovery (? SD)a

Spiked Concentration, ng/mL

NDZ

DZ

HZ

2 5 25 50 100

94.4 (4.9) 92.6 (4.3) 94.5 (3.9) 92.9 (3.1) 95.7 (2.7)

93.1 (5.1) 92.4 (3.7) 90.7 (3.5) 89.5 (4.8) 92.9 (2.9)

91.3 (7.8) 89.7 (6.2) 91.5 (6.4) 88.9 (5.6) 93.7 (3.2)

*n

=

5.

Table Ill-Preclslon of Assay for Nordiazepam (NDZ) and Halazepam (HZ)

Compound

Within-Day Concenn tration,

CV, %

ng/mL

NDZ

HZ

Day-to-Day Concentration, n CV, % ng1mL

2.3 5.1 24.7 58.3 105.1

12 12 12 12 12

6.2 7.9 5.6 2.1 1.3

1.8 3.6 23.5 52.8 103.8

12 15 12 10 12

9.4 9.2 7.6 2.9 1.1

2.1 4.6 22.3 54.6 107.2

12 12 12 12 12

8.2 8.7 6.8 2.9 1.1

1.7 3.2 19.3 48.7 102.9

12 15 15 12 12

10.3 10.1 9.2 4.7 1.4

824 I Journal of Pharmaceutical Sciences Vol. 79, No. 9, September 1990

Q 0

24

48 72

6

120 lh 168 192 216 240

TIME, HOURS Figure 2-Mean plasma halazepam (0)and nordiazepam (A) concentration-time profiles following oral administration of halazepam. The points represent the observed concentrations( 5 SD)and the solid lines represent the theoretic lines obtained by the curve-fitting procedure.

Table V-Pharmacoklnetic Parameters of Halazepam

Parameter Half-life,ha C, ng/mL Law

h

AUC, ng h/mL C L , Uh a

studies are needed to evaluate individual variability in the pharmacokinetic properties of HZ.

Subject 1

2

3

18.11 26.28 1.68 638.57 31.32

24.75 35.92 2.14 1029.07 19.42

21.65 29.60 2.57 826.10 24.21

Mean

2

21.50 t- 3.32 30.60 C 6.87 2.13 f 0.45 831.51 t- 195.71 24.98 2 5.99

Apparent elimination half-life.

Table VI-Pharmacokinetlc Parameters of Nordiazepamfollowing Oral Admlnlstratlon of Halazepam

Parameter Half-life,ha C, ng/mL ~rnax, h AUC, ng. h/mL

a

Subject 1

2

3

46.20 68.57 3.75 4950.70

48.92 58.38 4.54 4653.29

41.25 62.97 3.85 4115.90

Mean

References and Notes

SD

f SD

45.46 4 3.89 63.31 & 5.10 4.05 4 0.43 4573.30 ? 423.11

Apparent elimination half-life.

different from those for HZ (e.g., ultraviolet absorption maximum, column, and buffer strength of mobile phase1.9 Pharmacokinetic Study-Mean (+- SD) plasma HZ and NDZ concentration-time profiles of three healthy subjects following an oral dose of HZ are shown in Figure 2, and mean pharmacokinetic parameter values are presented in Tables V and VI. The concentration of HZ peaked at 2.13 h to reach a , C value of 30.6 nglmL, and HZ was eliminated with a mean apparent elimination half-life of 21.50 h. Our t,,, C,,, and the elimination half-life values of HZ are very similar to values reported previously.3.5~s.~o.l~ The interindividual variation in estimation of clearance (CL,) was about twofold. The NDZ appeared in the plasma fairly rapidly after HZ dosing and the C,,, was reached within 4 h. The mean C,,, value of NDZ (63.31 ng/mL) is twofold greater than t h a t of HZ (30.6 ng/mL). However, the AUC value of NDZ is more than fivefold greater than that of HZ. , and t,,, values of NDZ are very similar t o Our mean C previously reported values.6.10 The mean elimination halflife of NDZ (45.46h) is considerably larger t h a n t h a t of HZ (21.50 h). The apparent elimination half-life of NDZ calculated from our study is of the same order as those reported for NDZ in subjects receiving halazepam (50-100 h),3,5,"3-12 diazepam (50-120 h),13-16 medazepam (36-200 h),17 clorazepate (40-70 h),'s-ZO and prazepam (29-193 h).21,22 Nordiazepam (NDZ) is known to have considerable pharmacological activity.23-26 Perhaps it is by virtue of this active metabolite that prazepam and clorazepate exert their main pharmacological effects, in view of the considerably higher levels of NDZ than parent drugs in plasma. Although the metabolism of HZ is very similar to that of prazepam and clorazepate, HZ itself is active.27 Further

1. Fann, W. E.; Sullivan, J. L.; Miller, R. D. Curr. Ther. Res. 1974, 16, 1281-1286. 2. Kellner, R.; Bruzzese, D.; Winslow, W. W.; Rada, R. T.; Wall, F. J. J . Clin. Pharmacol. 1978, 18, 203-209. 3. Fann, W. E.; Pitts, W. M.; Whaless, J. C. Pharmacotherapy 1982, 2. 72-79. 4. Greenblatt, D. J.; Divoll, M.; Abernethy, D. R.; Ochs, H. R.; Shader, R. I. Clin. Pharmacokinet. 1983, 8, 233-252. 5. Chung, M.; Hibert, J. M.; Gural, R. P.; Radwanski, E.; Symchowicz, S.; Zampaglione, N. Clin. Pharmacol. Ther. 1984, 35, 838-842. 6 . Greenblatt, D. J.;Lockiskar, A.; Shader,R. Psychophurmacology 1983,80, 178-180. 7. Metzler, C. M.; Elfring, G. L.: McEwen,A. J. Biometrics 1974,30, 562-563. 8. Gibaldi, M.; Perrier, D. In Pharmacokinetics; Marcel Dekker: New York, 1982; pp 81-88,336. 9. Gupta, S. K.; Ellinwood, E. H. Pharm. Res. 1988,5, 365-368. 10. S mchowicz, S.; Tabachnick, I.; Katchen, B.; Dixon, R.; Chung, Postgmd. Med. Custom. Comm. 1981, Oct., 3 3 3 7 . 11. Pecknold, J. C.; McClure, D. J.; Fleury, D.; Chang, H.; Elie, R. Curr. Ther. Res. 1982,32, 895-905. 12. Absorption, Metabolism, Excretion and Pharmacokinetics of 3H-Sch-12041 in Man; Schering Corporation: Kenilworth, NJ, 1977. 13. Klotz, U.; Antonin, K. H.; Brugel, H.; Bieck, P. R. Clin. Pharmacol. Ther. 1977,21, 430. 14. Mandelli, M.; Tognoni, G.; Garattini, S. Clin. Pharmacol. Ther. 1978.3. 72-91. 15. Greenblatt, D J.; Allen, M. D.; Harmatz, S.; Shader, R. I. Clin. Pharmacol. Ther. 1980,27, 301412. 16. Kaplan, S.,A.; Jack, M. L.; Alexander, K.; Weinfeld, R. E. J . Pharm. SCL.1973,62, 1789. 17. Viukari, M.: Linnolia. M. Anns. Clin. Res. 1979, 9, 284-286. 18. Carrigan, P.'J.; Chao,G. C.; Baker, W. M.; Hoffman; D. J.; Chun, A. H. C. J. Clin. Pharmacol. 1977,17, 18-22. 19. Chun, A. H. C.; Carrigan, P. J.; Hoffman, D. J.; Kershner, R. P.; Stuart, H. D. Clin. Pharmacol. Ther. 1977,22,329. 20. Wretlind, M.; Pilbrant, A.; Sundwall, A.; Vessman, J. Acta Pharmacol. Toxicol. 1977,40 (suppl. I), 28. 21. Allen, M. D.; Greenblatt, D. J.; Harmatz, J. S.; Shader, R. I. J. Clin. Pharmacol. 1979, 19, 445-450. 22. Smith, M. T.; Evans, L. E. J.; Eadie, M. J.; Tyrer, J. H. Eur. J. Clin. Pharmacol. 1979, 16, 141-147. 23. Tansella, M.; Burti, L.; Siciliani, 0.; Zimmerman, C.; Schiavon, M.; Gerna, M.; Tognoni, G.; Morselli, P. L. Psychopharmacologia 1975,41, 81-85. 24. Biscaldi, G. P.; Hattab, J.; Montanaro, N.; Scoz, R. Curr. Ther. Res. 1971. 13. 606-615. 25. Tognoni, G.; Gomeni, R.; De Maio, D.; Alberti, G. G.; Franciosi, P.: Scieghi. G. Br. J. Clin. Pharmacol. 1975.2, 227-232. 26. Greenblatt, D. J.; Shader, R. I. In Benzodkkepines in Clinical Practice; Raven: New York, 1974; p 17. 27. Rickels, K.; Pereira-Organ,J.;Csanolosi, I.; Morris, R. J.;Rosenfeld, H.; Sablosky, L.; Schless A,; Werblowsky, J. H. Psychopharmacology 1977,52, 129-136.

d

Acknowledgments This work was supported b a grant from the National Institute on Drug Abuse, #DA-01883. T te authors thank Joan Wall and Gary Hadden for their assistance in preparing this manuscript.

Journal of Pharmaceutical Sciences / 825 Vol. 79, No. 9, September 1990