Pharmacokinetics of Methyprylon Following a Single Oral Dose

Pharmacokinetics of Methyprylon Following a Single Oral Dose PETERR. GwILT**', MARVINC. PANKASKIE*', JOHN E. THORNBURG§, ROBERTZUSTIAK*, AND DONALDR. SHOENTHAL' Received April 24, 1984, from the 'School of Pharmay ,Ferris,State College, Big Rapids, MI 49307, the §Departmentsof Famil Medicine, Pharmacology, and Toxicology, School of Human Me i n e , Michigan State University, Lansing, MI 48824, and the TSchoolof harmacy, West Virginia University Medical Center, Morganfown, WV 26506. Accepted for publication May 31, 1985. Present Addresses: *School of Pharmacy, West Virginia University Medical Center, Morgantown, WV 26506, and the 'Department of Biomedicinal Chemistry, College of Pharmacy, University of Nebraska, Omaha, NB 68105.

Abstract 0 Single oral doses of 300 mg of methyprylon were administered to 10 healthy volunteers. Plasma concentrations of methyprylon

and its dehydro metabolite were measured using a recently developed HPLC assay. Plasma concentration-time data were fitted to a twocompartment model with either first-order absorption, zero-order absorption, or two consecutive, discontinuous, first-order absorption rate constants. Based on the criteria of visual inspection, the correlation coefficient, standard deviations of the parameter estimates, and the residual sum of squares, it was concluded that the zero-order absorption model fit the data best. Mean (5SD) values for the half-life (9.2 f 2.2 h), apparent clearance, (11.91 2 4.42 mUh/kg) and apparent steady-state volume of distribution, (0.97 f 0.33 Ukg) were found.

Methyprylon, a nonbarbiturate hypnotic, has been in clinical use since 1956. However, studies on the pharmacokinetics of this drug have not yet appeared in the literature. The only available information on the disposition of methyprylon in humans comes from a n early study by Randall et a1.l in which only three blood samples were drawn from each subject and also from reports on patients being treated for methyprylon intoxication.24 Data from these latter studies yield conflicting estimates of the half-life of the drug. Plasma concentrations published by Burnstein and Strauss2 indicate a half-life of 5 h whereas two more recent report half-lives of 50 h in overdosed patients. One reason for these inconsistencies is the lack of a suitable assay for methyprylon. Early studies relied on a colorimetric assay which is selective for phenolic compounds but is not specific for methyprylon. More recent assays for methyprylon use GC.3-5 While these methods are more sensitive, they demonstrate poor resolution between the drug and its major metabolite. Recently, a new HPLC assay has been developed which has equal or greater sensitivity compared to GC procedures and clearly distinguishes between methyprylon and its known metabolites.6 The assay has been used to measure plasma concentrations of methyprylon and its dehydro metabolite in the dog.7 Using this new assay the pharmacokinetics of the parent drug and the plasma concentrations of the metabolite were determined in human subjects following an oral therapeutic dose of methyprylon.

Experimental Section Subjects-Ten healthy, nonobese (four male, six female) subjects all between 19 and 28 years old and weighing between 54 and 88 kg participated in the study after giving informed consent. All subjects had normal prestudy laboratory values. Following an overnight fast, each subject ingested a 300-mg capsule of methyprylon (Noludar, Hoffmann-LaRoche, Nutley, NJ) with 6 ounces (180 mL) of water. Blood samples were drawn from the antecubital vein using a 21-

0022-3549/85/0900-1001$0 1 .OO/O 0 1985, American Pharmaceutical Association

gauge butterfly cannula with a heparin lock. Samples were drawn into heparinized tubes just prior to dosing and a t 0.33,0.67,1.0,1.33, 1.67,2,3,4,6, 10, 12, and 24 h after the drug was administered. For subject I (Fig. 1) the last sampling time was 26 h after drug administration. Plasma was immediately frozen until assayed. Subjects refrained from eating or drinking (except water) for 4 h after dosing and remained supine during this period. Assay-Methyprylon (3,3-diethyl-5-methyl-2,4-piperidinedione) and its major metabolite (3,3-diethyl-5-methyl-1,2,3,4-tetrahydro2,4-piperidinedione) were assayed by HPLC. To 1.0 mL of plasma was added 100 pL of 0.1 M hydrochloric acid, 50 pL of internal standard (12.0 pL of pyrithyldione per mL of distilled water), and 10 mL of diethyl ether. The mixture was vortexed for 20 s and the layers separated by centrifugation at 1500xg for 5 min. The ether layer was transferred to a 15-mL conical centrifuge tube and evaporated under a stream of nitrogen a t 40°C. The residue was then dissolved in 50 pL of methanol and 25-40 pL aliquots were injected onto the column, Analyses were performed on a Waters HPLC system consisting of a model 6000A pumping module, a model U6K injector, and a model 450 variable-wavelength UV detector. Separations were obtained on a pporasil column (10 pm, 3.9 mm i.d. x 30 cm) using a mobile phase composed of hexane:tetrahydrofuran:methanol(72:6:2) at a flow rate of 2.0 mL/min. Chromatographic peaks were detected at 214 nm. The injection of 2.0 pg methyprylon resulted in peaks of near 50% full scale response at a sensitivity of 0.04 AUFs. Retention times for the metabolite, internal standard, and methyprylon were 5.0, 7.0, and 8.5 min, respectively. Intra-assay linearity and precision were determined over the plasma concentration ranges from 0.1 to 10 pg/mL for methyprylon and from 0.01 to 1.0 pg/mL for the metabolite. Triplicate samples of each concentration were analyzed. For methyprylon a correlation coefficient of 0.998 was obtained with an average coefficient of variation of 9.3% over the concentration range studied. Recovery averaged 85% with a lower limit of sensitivity of 0.1 pg/mL using 1.0-mL samples. For the metabolite, a correlation coefficient of 0.999 was obtained with an average coefficient of variation of 8.4%. Recovery averaged 89% and a lower limit of sensitivity of 0.01 pg/mL was obtained. Pharmacokinetic Calculations-Individual plasma methyprylon concentration-time data were fit to a two-compartment model with either first-order or zero-order absorption using NLINSAS.8 Data were weighted inversely to the concentration. Preliminary attempts to fit the data showed that better fits were obtained using data weighted inversely to the concentration compared to nonweighted data. The program required to fit the data to a zero-order absorption model included a procedure for estimating the time at which absorption ceased. Preliminary estimates of the absorption lag time were obtained from ESTRIP using the first-order absorption model.9

Results and Discussion Initial attempts to fit the individual serum concentrationtime profiles to a two-compartment model with first-order absorption consistently failed to fit the peak concentrations. However, when the data were fit to a pharmacokinetic model Journal of Pharmaceutical Sciences / 1001 Vol. 74, No. 9, September 1985

with zero-order absorption, a considerably better fit was obtained (Fig. 1).McNamara et a1.lO and Colburn and Gibaldill have advocated three criteria by which a comparison can be made of how well two different models fit a set of data. They suggest comparing (a)goodness of fit visually, (b) the correlation coefficients, and ( c ) the computer estimated asymptotic SDs of the parameters common to both models expressed as coefficientsof variation. Visual inspection of the plots of individual data sets clearly revealed a better fit by the zero order model (Fig. 1)with the exception of subject E. In this case the models seem to fit equally well. The other two criteria plus an additional parameter, the residual or unexplained sum of squares, support the suitability of the zeroorder absorption model. The disposition of methyprylon is best described by a twocompartment model. Parameter values are summarized in Tables I and 11. The average half-life (harmonic mean) of the drug was 9.2 h. The steady-state volume of distribution V d , was calculated, assuming complete absorption, by the following equation:12

Vd,,

=

Dose * AUMC AUC2

-

Dose Absorption Time 2 AUC

-

served for several other drugs including ethano1,13 sulfisoxazole,l*griseofulvin,16erythromycin,11.16and hydroflumethiazide.lO It is difficult to suggest a single mechanism, such as dissolution rate limited absorption, which would explain this

(1)

which ave a mean value of 0.97 L/kg., A value of 11.9 mL * h-q kg-’ was found for the mean total body clearance of the drug, again assuming complete absorption. Metabolite plasma concentrations were much lower than those of the parent drug and declined more slowly. Apparent zero-order absorption kinetics have been ob-

I

I

I

I

I

I

I

4

8

12

16

20

24

28

TIME, h

-

Figure 1-Plasma concentrationof (0)methyprylon and (0) its dehydro metabolite following a 300-mg oral dose of methyprylon to subject 1. The zero-order absorption lines represent regression curves for the (-) model and (- -) the first-order absorption model.

-

Table 1-Methypryion Pharmacokinetlc Parameters Obtained by Compartmental Analysis with Zero-Order Absorption Subject

Lag Time, h

Absorption Time, h a

A

0.34 0.34 0.48 0.24 0.00 0.27 0.00 0.27 0.31 0.42

1.23 1.28 2.26 2.17 0.53 1.17 1.60 1.78 2.38 2.05

27.25 14.08 3.92 8.29 21.29 32.52 52.82 19.35 8.62 7.06

7.748 11.737 1.388 4.950 1 1.957 17.322 15.437 9.670 2.061 1.478

0.078 0.1 1 1 0.093 0.071 0.075 0.052 0.077 0.079 1.069 0.051

2.376 2.956 0.634 1.347 5.008 2.939 0.944 1.525 0.464 0.579

0.998 0.966 0.989 0.961 0.993 0.989 0.970 0.995 0.979 0.979

0.27 0.16

1.64 0.59

19.52 14.96

8.375 5.813

0.076 0.018

1.877 1.447

0.981 0.012

B

C D E F

Q H I J

x SD

a,

T2

h-’

aTime at which absorption ceases. b K = apparent zero order rate constant; F = fraction of dose absorbed; V, = volume of the central compartment. is the correlation coefficientassociated with curvilinear regression and is calculated by ( S S ~ O T A L - SSERROR)/SSTOTAL) (see ref. 21). Table ii-NonCompartmentai Characteristics of Methypryion Absorption and Disposition Subject A B

C D E F

G H I J

z

SD a

8.38 3.51 3.80 5.48 4.96 5.41 6.91 5.03 6.17 6.54

1.33 1.33 1.87 2.17 1.37 1.03 1.37 1.33 2.00 2.08

100.68 33.23 34.59 64.54 65.11 94.00 91.35 61.38 64.70 137.92

11.80 21.48 14.24 10.11 9.85 6.40

8.73

0.59 1.43 1.11 0.81 0.76 0.63 0.72 1.21 0.94 1.50

5.62 1.45

1.59 0.40

74.75 31.84

11.91 4.42

0.97 0.33

CL = total body clearance assuming complete absorption. Vd, = steady-state volume of distribution assuming complete absorption.

1002 / Journal of Pharmaceutical Sciences Vol. 74, No. 9, September 1985

phenomenon, since the drugs represent a spectrum of aqueous and lipid solubilities from highly soluble (ethanol and methyprylon) t o almost insoluble (griseofulvin). An alternative absorption model for drugs with these types of absorption profiles has been proposed by Suverkrup.17The absorption profile is described by one or more sequential, discontinuous first-order rate constants. Excellent fits of the sulfisoxazole and griseofulvin data were exhibited in his paper. Zimmerman18has recently described the use of NONLINl9 to fit the absorption models reported by Suverkrup. Using this method, the methyprylon date. were fit to a model with a lag time and a single discontinuous first-order absorption rate constant and also to a model with a lag time and two consecutive rate constants. Neither model fit the data as well as a single zero-order input. The mean half-life of 9.2 h is notably shorter than that reported in overdosed patients. Both Bridges and Peat4 and Pancorbo et aL3 reported a half-life of 50 h before and 20 h after dialysis in overdosed patients with initial plasma concentrations of 168 pg/mL and 94 pg/mL, respectively. Since in both cases hemodialysis significantly lowered plasma concentrations, the elimination of methyprylon, like many other drugs studied in overdosed patients,20 appears to be concentration dependent at high plasma concentrations.

References and Notes 1. Randall, L. 0.; Iliev, V.; Brandman, 0.Arch. Znt. Phurmacodyn. 1956,106, 388. 2. Burnstein, N.; Strauss, H. K. J.Am. Med. Assoc. 1965,194,237. 3. Pancorbo, A. S.;Palagi, P. A.; Piecoro, J. J.; Wilson, H. D. J.Am. Med. Assoc. 1977,237, 470.

4. 5. 6. 7. 8.

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Brid es, R. R.; Peat, M. A. J.Anal. Toxicol. 1979,3, 21. Vanioven, M.; Sunshine, I. J.Anal. Toxicol. 1979,3, 174. Pankaskie, M. C.; Brooks, M. A. J. Chromato r. 1983,278,458. Gwilt, P. R.; Pankaskie, M. C.; Mitala, J. J. &;an.Anaesth. SOC. J. 1982,29, 381. “SAS User’s Guide”; Helwig, J. T.; Council, J. A., Eds.; SAS Institute, Inc.: Raleigh, NC, 1979. Brown, R. D.; Manno, J. E. J. Phnrm. Sci. 1978,67, 1687. McNamara, P. J.; Colburn, W. A.; Gibaldi, M. J. Clin.PhnrmaCOZ. 1978, 18; 190. Colburn, W. A,; Gibaldi, M. Can. J. Phnrm. Sci. 1977,12, 90. Gibaldi, Milo; Perrier, Donald G. “Pharmacokinetics,” 2nd ed.; Dekker: New York, 1982, 414. Welling, P. G.; Lyons, L. J!; Elliott, R.; Amidon, G. L. J. Clin. Pharmacol. 1977, 17, 199. Kaplan, S.A.; Weinfeld, R. E.; Abruzzo, C. W.; Lewis, M. J. Phnrm. Sci. 1972,61, 773. Bates, T. R.; Carri an, P. J. J.Pharm. Sci. 1975, 64, 1475. Colburn, W. A.; Di%anto, A. R.; Gibaldi, M. J. Clin. Phnrmacol. 1977, 17, 592. Suverkrup, R. J.Pharm. Sci. 1979,68, 1395. Zimmerman, J. J. J.Pharm. Sci. 1983, 72, 138. Metzler, Carl M.; Elfring, Gary L.; McEwen, Amber J. “A Users Manual For NONLIN and Associated Programs”; The Upjohn Co.: Kalamazoo, MI, 1974. Rosenberg, J.; Benowitz, N. C.; Pond, S. Clin. Pharmokinet. 1981.6. 161. Howell ’ David C. “Statistical Methods for Psychology”; PWS Publishers: Boston, MA, 1982; p 262.

Acknowledgments The authors acknowledge Dr. James Zimmerman, Stuart Pharmaceuticals for supplyin copies of DFUNC for the discontinuous/consecutive fits. This worf was supported by a grant from HoffmannLaRoche, Inc.

Journal of Pharmaceutical Sciences / 1003 Vol. 74, No. 9, September 1985