CHIRALITY 4:265-267 (1992)

Stereospecific Gas Chromatographic Method for Determination of Methadone in Serum K. KRISTENSEN AND H.R. ANGEL0 The Royal Danish School of Phurnuuy, DK-2100 Copenhagen 0 and Department of clinical Chmistly,

Bispebjerg Hospital, DK-2400 Copenhagen W,Denmark

ABSTRACT rac-Methadone is used clinically for the chronic maintenance treatment of heroin addiction and for the relief of pain. As the pharmacological activity of methadone is due primarily to the ( - )-(R)-enantiomer,stereospecificmeasurements of methadone serum concentrations in methadone-treated patients are expected to be more relevant for clinical studies than earlier described total drug measurements. This study describes a stereospecificgas chromatographic (GC) method for the determination of methadone in serum. The extracted methadone was derivatizised with ( - )-menthy1 chloroformate. The diastereometric derivatives were analysed by GC on a capillary column and detected with a nitrogen-phosphorusdetector. The resolution factor obtained for the methadone enantiomers was 1.1 with a relatively short time of analysis (30 min). By analysing the pure (- )-@)-enantiomer,no racemization was seen during the analysis. The lower limit of quantitation was 75 nmol/l for each enantiomer. Measurements of the ratio between (-)-@)- and ( +)-(S)-methadoneconcentrations in serum from five methadone-treatedpatients showed interindividual differences (range 0.5-1.1). The patient results correlated well with those from another GC method measuring total methadone.

KEY WORDS: methadone, GC, ( - )-menthylchloroformate,derivatization, stereospecific INTRODUCTION Methadone is commercially available as a racemic mixture of its ( + )-(S)- and (- )-(R)-enantiomers.The analgesic and abstinence relieving effect is due primarily to the (-)-@)-enantiomer.' Dose adjustment based only on its therapeutic effect is difficult. Although the use of measurements of total methadone serum concentrations have shown some improvement in the treatment of heroin addicts in a methadone maintenance program,2 clinical studies based on measurements of separate determinations of the active (-)-@)- and the inactive (+)-(S)-enantiomerare expected to be a better alternative. To our knowledge no clinically applicable analytical method for stereospecific determination of methadone in serum has been described. The aim of this study was to develop a stereospecific chromatographic method for the determination of ( - )-(R)-and ( + )-(S)-methadonein serum from methadone-treatedpatients. The described gas chromatographic (GC) method is based on the formation of diastereomeric derivatives with ( - )-menthylchloroformate,separation on a capillary GC column and detection by a nitrogen-phosphorus detector.

GC Conditions

HP-5890 gas chromatograph with a nitrogen-phosphorusdetector, a 3393A integrator, 9114B disc drive, and a capillary split/ split-less injector were used. GC was performed on a 25 m x 0.25 mm (ID) fused silica column coated with TAP-CB (Chrompack), using helium as carrier gas (flow rate 1.0 ml/min), temperature programming from 100 "C to 245 "C, injection temperature 225 "C and a purge off time of 0 - 1.4 min. The rubidium bead power and detector gases were optimized according to the instructions in the HP manual.

Extaction and Deriuatization Procedure 1.0 ml of serum was added to 1.0 ml of 1M sodium carbonate buffer, pH 10, containing 30 nM internal standard (orphenadrine) and 5.0 ml n-hexane.The mixture was shaken for 15min, and after centrifugation for 5 min at 1300 g and cooling in dry ice-acetone bath for 1 min, the n-hexane phase was easily decanted into a Kimax reaction tube and evaporated to dryness at 50 "C under nitrogen. The residue was dissolved in 200 pl of ethyl acetate followed by addition of 10 mg of water-free sodium carbonate and 25 p1 of (-)-menthy1chloroformate. The MATERIALS AND METHODS reaction tubes were sealed with caps and the mixture was Chemicals heated at 110 "Con a heating block for 45 min. After cooling rac-Methadone hydrochloride (Ph. Eur., 2nd ed.) was supto room temperature, the reaction mixture was shaken with 0.5 plied from the hospital pharmacy and ( - )-menthylchloroforml alcoholic alkali for 5 min and 1.0 ml water for 5 min. This mate from Aldrich, Germany. ( - )-@)-methadonehydrochlomixture was shaken with 4.0 ml n-hexaneand after centrifugaride (Levomethadon-hydrochloridB)and orphenadrine were tion and cooling in a dry ice-acetonebath, the n-hexane phase gifts from Hoechst AG, Germany, and GEA A/S, Denmark, was decanted into a fresh glass vessel. The n-hexanephase was respectively. Alcoholic alkali was prepared by dissolving 2.8 g of potassium hydroxide in 95 ml of methanol and 25 ml of water. All chemicals were of analytical quality. Received for publication August 22,1991; accepted January 2, 1992. Presented at the Second International Symposium on Chiral Discrimination, May 27-31, 1991, Rome, Italy. @

1992 Wiley-Liss, Inc.

Address reprint requests to Helle R. Angelo, Dept. of Clinical Chemistry, Bispebjerg Hospital, DK-2400 copenhagen NV, Denmark. This work was performed at the Department of Clinical Chemistry, Bispebjerg Hospital.

264

KRISTENSEN AND ANGEL0

TABLE 1. Reproducibility of replicate analysis of R,S-methadone and R-methadone added to human serum

evaporated to dryness at 50 "C under nitrogen. The residue was then dissolved in 30 p1 of toluene. Samples of 2 p1 were analysed by GC.Quantitation was by peak height ratio with reference to graphs obtained by analysing serum standards simultaneously.

Serum coma (nmol/l) found

N=

R-methadone

228

232

3

11.5

R-methadone Smethadone

150 150

137 141

5 5

10.2 7.1

R-methadone S-methadone

225 225

226 221

5 5

11.3 6.2

R-methadone Smethadone

750 750

771 782

7 7

7.5 7.4

R-methadone S-methadone

900 900

962

944

6 6

9.6 11.1

Compound

RESULTS Figure 1 shows chromatograms obtained from analysis of serum samples. Using pure ( - )-@)-methadoneno racemization was seen during the derivatization and analytical procedures (Fig. 1C) and in no cases interfering peaks from endogenous compoundswere seen (Fig. 1A).Plots of the standard curves of the two enantiomersover the range 75-900 nmol/l were linear and passed through the origin (Fig. 2). The reproducibility was determined using spiked serum samples analysed at random on different days (Table 1).The lower limit of quantitation was 75 nmol/l for each enantiomer. Table 2 compares the results from analysis of serum from methadone-treatedpatients by the new stereospecific method with our earlier described method A

added

Coefficient of variation Yo

aMean values from N samples.

B

kC

inj

inj

1

1

20 22

l

l

l

l

24 26 28 30

1

1

1

1

l

1

20 22 24 26 28 30 TIME (min)

-

20 22 24 26 28 30

Fig. 1. Chromatograms of human serum samples, analysed as described above. (A) Blank serum;(B) mum to which ( - )@)- and ( + )-(S)-methadonewere added (900 and 900 nM,respectively); (C) serum to which ( - )@).methadone was added (225 nM). R-M = (-)@)-methadone derivative; SM = (+)-(S)-methadonederivative. I = internal standard (orphenadrine).

Peak height ratio

0

200

400

600

800

1000

Conc. (nmolll) +S-M

*R-M

Fig. 2. Calibration curves of serum standards. S M

= (

+ )-(Sbmethadone. R-M= ( - )-@)-methadone.

265

METHADONE IN SERUM

for assay of total methadone c~ncentration.~ The regression analysis of the comparison between the two methods indicates the acceptableaccuracy of the stereospecificmethod (y = 0.90~ + 45, r = 0.994). Measurements of the ratio between (-)-(R)- and ( + )-(S)-methadoneconcentrations in serum from five methadone-treated patients at steady state showed interindividual differences (Table 2). DISCUSSION The assay described in this work provides a stereospecific method that, unlike that of McGilliard et al.,4 has been validated for clinical studies. Earlier investigationsof the disposition and pharmacokinetic differences between the active and the inactive enantiomers of methadone were based on administration of a pseudo racemic mixture of differentisotope-labelled ( - )-(R)-and ( + )-@-methadone5,which is not applicable for clinical studies. A GC method was developed, because this technique offers the possibility of detectionwith the sensitivenitrogen-phospho-

rus detector. During method development, chiral GC on two commercially available capillary columns (Chirasil-L-Val and CP-cyclodextrin-P-2,3,6-m-19 from Chrompack) was tried, but since no separation of the two enantiomers of methadone was obtained (data not shown),an indirect method based on formation of diastereomeric derivatives was chosen instead. Derivatization of tertiary amines with chloroformate esters producing carbamateshas been used for both GCit and HPLC 7lo methods. Using the optically pure ( - )-menthylchloroformate, diastereomeric derivatives were formed (Fig.3), which could be separated on the TAP-CP column with a resolution factor of 1.1 (at 240 "C) for the methadone enantiomers. On capillary columns with less polarity (IB-1and HP-5),it was not possible to separate the diastereomeric derivatives; on HP-17, separation was obtained after approximately 1 hour (unpublished results). The lower limit of detection could not be less than 75 nmol/l for each enantiomer due to a relatively low yield of the derivative (about 20%), when compared with peak heights obtained from imipramine assuming a 100% yield for the derivative of this compound.6 This low yield is due to

TABLE 2. Comparison of results of analysis of serum samples from methadone-treatedpatients at steady state by the new stereoselective with our earlier described method measuring total methadone" Patient number 1. 2. 3. 3. 4. 5. 5.

Smethadone conc. (nmol/l)

R-methadone conc. (nmol/l)

R/S ratio

735 260 310 260b 188 167b 209

820 140 219 161 183 136b 190

1.1 0.5 0.7 0.6b 1.o 0.P 0.9

R + S-methadone conc. (nmol/l) 1555 400

529 421 371 303 399

"See Ref. 3. bThrough concentrations. All other values were peak concentrations (1-3 h after oral methadone administration).

+

METHADONE

(-)-MENTHYLCHLOROFORMATE

+

CH,C I

( - ) - M E N T H Y L C H L O R O F O R M A T E D E R I V A T I V E S OF M E T H A D O N E Fig. 3. Proposed reaction scheme for formation of methadone (- )menthylchloroformate derivatives.

Total methadone conc. (nmol/l) 1668 305 600 475b 330 329b 387

266

KRISTENSEN AND ANGEL0

0

+

C 1 3 C-C H2-0-C-C

II

Heat ___,

1

Tr i c h l o r o e thy1 chloroformate

Promethazine

iH3 +F-O-CH2-CC II

1

OH___)

H H

CH3

I

I YH3

K-C=C-H

+

+

"20

I FH3

HC 1

H-C-C-H

___)

I I

H C1

I1

End product Fig. 4. Suggested pathway for deaminating of promethazine.

deamination.l1 The same deamination reaction has been described by Hartvig et al. for other tertiary amines with similar structure: two carbons and a branching between the amine and a strongly electron-withdrawinggroup, like the phenothiazine nucleus of promethazine. The electron-withdrawing group placed near the amine results in a simple Hoffmann elimination of the amine, when carbarnation is tried. Further reaction results in the end product shown in Figure 4.The end product of this reaction has been proven by Wallace et al.,8 although they suggested another reaction pathway. Using acetonitrile as reaction medium and omitting the base destruction step, this deaminating process seems to be eliminated. lo We have found that the base destructionis a necessary step in the GC procedure probably because a quarternary ammonium ion, which can only be analysed by HPLC,is formed otherwise. Besides, in the base destruction step, deamination also seems to take place in the reaction medium. Hartvig et a1.l1 describe a complete deaminationfor methadone,which is faster in chloroform than in heptane. By experimentingwith different media and reaction conditions, we succeeded in obtaining a reasonable yield of the carbamate derivative, probably because the carbonyl groups, due to the distance, have only a small electron-withdrawing effect on the carbon in the 5th position (Fig. 3). It was not possible to improve the formation of the carbamate derivative further. In spite of the above-mentioned problems, we have been able to measure (-)-(R)- and (+)-(S)-methadonein serum samples

with an acceptable precision and agreement of results from analysis of serum from methadone-treatedpatients by the new stereospecificmethod with our earlier method measuring total methadone. ACKNOWLEDGMENTS

The study was supported by grant 12-9078from the Danish Medical Research Council. We are grateful to Jan Jorn Hansen, The Royal Danish School of Pharmacy, for his help in discussing the chemical reactions, and to Steen Gammeltoft, Department of Clinical Chemistry, Bispebjerg Hospital, for his help in reviewing this paper. REFERENCES 1. Judson, B.A., Horns, W.H., Goldstein, A. Side effects of levomethadone and racemic methadone in a maintenance program. Clin. Pharmacol. Ther. 20: 44.5449, 1976. 2. Tennant, F.S. Inadequate plasma concentrationsin some high-dose methadone maintenance patients. Am. J. Psychiatry 144:134%1350,1987. 3. Angelo, H.R., %Ian, H. A micro gas chromatographicmethod for determination of methadone in serum during methadone maintenance treatment. Acta Pharm. Toxicol. 57,suppl. 1:A-14, 1985. 4. McGilliard, K.L., Wilson,J.E., Olsen, G.D., Bartos,F. Stereospecificradioimmunoassay of d- and 1-methadone,Proc. West. Phamacol. SOC. 2463-466, 1979. 5. Nakamura, K., Hachey, D.L., Kreek, N.S., Irving, C.S., Klein, P.O. Quantitation of methadone enantiomm in humans using stable isotope-labelled [2H3]-, [2H5]and ['H,] methadone. J. Pharm. Sci. 714043,1982.

h4ETHADONE IN SERUM 6. Hartvig, P., Ahnfelt, N.O., Karlsson, K.E. Electron capture gas chromatogra-

phy of nanogram amounts of tertiary amines as trichlomthyl carbamates. A& Pharm. SUK. 13181-189, 1976. 7. Sintov, A,, Siden. R., Levy, R. Sensitive high-performance liquid matographic assay using 9-fluorenylmethylchloroformatefor monitoring controlled-release lidccaine in plasma. J. Chromatogr. 496:335-244, 1989. 8. Wallace, J.E., Shimek, Jr., EL., Harris,S.C., Stavchansky, S. Determination of promethazine in serum by liquid chromatography. Clin. Chem. 2725% 255, 1981.

267

9. F’rakash, C., Jajoo, H.K., Blair, LA., Mayol, R.F. Resolution of enantiomersof the antianhythmic drug encainide and its major metabolites by chiral derivatizationand high-performance liquid chromatography.J. Chromatogr. 493325335,1989. 10. Witte, D.T., de Zeeuw, R.A., Drenth, B.F.H. Chml derivatizationof prometw i n e with (- )-menthylchlorofonnate for enantiomeric separation by RPHPLC. J. High Resol. Chromatogr. 13569-571,1990. 11. W i g , P., Vessman, J. Electron capture gas chromatography of tertiary amines after a demethylation reaction to secondary amines. Acta Pharm. Suec. 11:115124,1974.