Plasma concentrations of the enantiomers of methadone and therapeutic response in methadone maintenance treatment

Drug and Alcohol Dependence 61 (2000) 47 – 54 www.elsevier.com/locate/drugalcdep

Plasma concentrations of the enantiomers of methadone and therapeutic response in methadone maintenance treatment Chin B. Eap a,*, Michel Bourquin b, Jean-Louis Martin b, Jacques Spagnoli a, Santino Livoti b, Kerry Powell a, Pierre Baumann a, Jean-Jacques De´glon b a

Unite´ de Biochimie et Psychopharmacologie Clinique, De´partement Uni6ersitaire de Psychiatrie Adulte, Hoˆpital de Cery, CH-1008 Prilly-Lausanne, Switzerland b Fondation Phe´nix, Route de Cheˆne 100, 1224 Cheˆne-Bougeries, Gene6a, Switzerland Received 19 November 1999; received in revised form 1 February 2000; accepted 4 February 2000

Abstract Methadone is a 50:50 mixture of two enantiomers and (R)-methadone accounts for the majority of its opioid effect. The aim of this study was to determine whether a blood concentration of (R)-methadone can be associated with therapeutic response in addict patients in methadone maintenance treatment. Trough plasma concentrations of (R)-, (S)- and (R,S)-methadone were measured in 180 patients in maintenance treatment. Therapeutic response was defined by the absence of illicit opiate or cocaine in urine samples collected during a 2-month period prior to blood sampling. A large interindividual variability of (R)-methadone concentration-to-dose-to-weight ratios was found (mean, S.D., median, range: 112, 54, 100, 19 – 316 ng × kg/ml× mg). With regard to the consumption of illicit opiate (but not of cocaine), a therapeutic response was associated with (R)- (at 250 ng/ml) and (R,S)-methadone (at 400 ng/ml) but not with (S)-methadone concentrations. A higher specificity was calculated for (R)- than for (R,S)-methadone, as the number of non-responders above this threshold divided by the total number of non-responders was higher for (R,S)-methadone (19%) than for (R)-methadone (7%). The results support the use of therapeutic drug monitoring of (R)-methadone in cases of continued intake of illicit opiates. Due to the variability of methadone concentration-to-dose-to-weight ratios, theoretical doses of racemic methadone could be as small as 55 mg/day and as large as 921 mg/day to produce a plasma (R)-methadone concentration of 250 ng/ml in a 70-kg patient. This demonstrates the importance of individualizing methadone treatment. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Methadone; Enantiomer; Blood; Addiction; Cytochrome P-450

1. Introduction Methadone is an effective medication to treat opiate dependence if it is administered in adequate doses (Bertschy, 1995). A clear inverse correlation has been found between the dose and the risk of leaving treatment (Caplehorn and Bell, 1991). There is considerable inter-individual variation of methadone plasma concentrations following a treatment with the same dose (Holmstrand et al., 1978; Tennant, 1987). Several studies have tried to find a minimum methadone blood concentration that can support effective methadone* Corresponding author. Tel.: +41-21-6436438; fax: + 41-216436444. E-mail address: [email protected] (C.B. Eap).

maintenance therapy (MMT) reliably (Horns et al., 1975; Holmstrand et al., 1978; Tennant, 1987; Bell et al., 1988, 1990; Loimer and Schmid, 1992; Kell, 1995; Torrens et al., 1998; Dyer et al., 1999). Unfortunately most of these studies have been performed in only a small number of patients. In some studies, a threshold was not found (Horns et al., 1975; Bell et al., 1990; Torrens et al., 1998; Dyer et al., 1999), while various values, ranging from 50 to 600 ng/ml were found in other studies (Holmstrand et al., 1978; Bell et al., 1988; Dole, 1988; Loimer and Schmid, 1992; Wolff et al., 1992; Kell, 1995). A concentration of 400 ng/ml is often considered necessary to stabilize maintenance (Loimer et al., 1991; Payte and Khuri, 1993) and is used as a reference value when performing therapeutic drug monitoring of methadone. However, to our knowledge,

0376-8716/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 6 - 8 7 1 6 ( 0 0 ) 0 0 1 2 1 - 6

48

C.B. Eap et al. / Drug and Alcohol Dependence 61 (2000) 47–54

there is no study that clearly demonstrates the existence of such a threshold. Methadone is extensively metabolised in the body, and bioavailabilities were found to range from 41 to 99% (Meresaar et al., 1981). The main metabolite of methadone (2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine) is inactive: it is formed by N-demethylation and subsequent spontaneous cyclisation (Sullivan and Due, 1973). In most countries (except for Germany), methadone is marketed as a 50:50 mixture of two enantiomers called (R)- or (L)- and (S)- or (D)-methadone. (R)-methadone accounts for the majority, if not the totality, of the opioid effect of racemic methadone (Kristensen et al., 1995). We have previously shown a large interindividual variability in the (R)/(S) ratios of methadone (range 0.63 – 2.4) as measured in the plasma samples of 22 addict patients under racemic methadone maintenance treatment (Eap et al., 1996). This suggests that measuring methadone enantiomers could be more reliable than determining the (R,S)-concentrations when correlating concentrations of methadone in blood with the therapeutic response. Furthermore, in another study, we confirmed the large interindividual variability of methadone blood concentrations: in 50 opiate-dependent patients under oral MMT, (R)-methadone concentrations were found to vary from over a sevenfold range at the same daily dose corrected for body weight (Eap et al., 1998). In the present study, we measured the (R)-methadone plasma trough concentrations in 211 patients undergoing methadone treatment. We first wanted to test, with a larger number of subjects, the reproducibility of our previous results on the inter-individual variability of (R)-methadone concentrations. However, the main goal of this study was to determine whether a minimum concentration of this active enantiomer can be associated with a good treatment response. Response was assessed by the presence of illicit drugs (opiate and cocaine) in the urine samples collected during a 2month period prior to the blood sampling in a large group of patients.

2. Methods

2.1. Subjects Blood samples were collected from outpatients under methadone treatment in the Phe´nix foundation (Geneva) and were sent to our laboratory for routine therapeutic monitoring of (R,S)-methadone. The patients fulfilled DSM-IV criteria for heroin dependence (304.02) and had not been included in our previous studies (Eap et al., 1996, 1997, 1998). For the present retrospective study, we re-analyzed these plasma samples (stored at −20°C) with a stereoselective analytical

method which allows for the separate measurement of (R)- and (S)-methadone (Eap et al., 1996). Patients for whom a plasma sample was available were contacted and the aim of the study was explained to them (no additional blood sampling was necessary. Written informed consent was required for inclusion in the study. Out of the 229 potential subjects, a written informed consent was obtained for 216 patients (failures to obtain consent were mainly due to patients who had left the center or who were on holiday during that period: their plasma samples were discarded). The results of two patients, free of comedication, in particular of drug metabolism inducer substance, were discarded from subsequent analysis because no methadone was detected in their plasma samples (limit of quantitation: 10 ng/ml for each enantiomer). Finally, three other patients were also discarded because they had been included into the present MMT less than 2 months before the blood sampling. Of the remaining 211 patients, 206 took methadone in a single dose (usually in the morning) and the other five split their daily dose into two portions. One hundred and eighty patients were in methadone maintenance treatment (MMT) and 31 patients were in the process of reducing their dose in an attempt to stop using methadone. Blood samples had been collected at least 5 days after any change of methadone dose and before the intake of the first methadone daily dose; this allowed determinations of the steady-state trough plasma concentrations.

2.2. Analysis (R)- and (S)-methadone were measured by high performance liquid chromatography with the use of a chiral column as previously described (Eap et al., 1996). Data from routine monitoring of (R,S)-methadone by gas chromatography –nitrogen/phosphorus detector (Bertschy et al., 1994) were used as a control for high performance liquid chromatography values: a good correlation and concordance was found between the two methods (r= 0.98, y (HPLC)= 0.89(GC)+ 15, n= 211). The presence of opiate and cocaine metabolites in urine samples was measured by Emit on an ETS Plus Syva autoanalyser (Behring Diagnostic, Zurich, Switzerland). The low concentration calibrators (300 ng/ml for morphine and 150 ng/ml for benzoylecgonine) were defined as the cut-off value for the limit of detection. Determination of opiates in urine samples was performed on a twice-a-week schedule, except for stabilized patients for whom analysis was performed on a once-a-week schedule. Determination of cocaine was performed on a once-a-month schedule and, if positive, analysis was performed every week (however, results for cocaine analysis are missing for two patients of the 180 patients in maintenance treatments). For each substance, the success criteria were defined as absence of

C.B. Eap et al. / Drug and Alcohol Dependence 61 (2000) 47–54

positive urine determination during the 2-month period prior to blood sampling

2.3. Statistical analysis The demographic and clinical characteristics of responders (no positive urine samples during the 2-month period prior to blood sampling) and non-responders (one or more positive urine samples) were compared by the unpaired t-test. Pearson product-moment correlations were used for correlations. Unless otherwise stated, all tests were two-tailed, and an a level of 0.05 was considered significant. The optimum plasma (R)-, (S)-, or (R,S)-methadone concentrations for response were determined by classifying patients into responders and non-responders and by calculating the sensitivity and the specificity at various thresholds for methadone, as described previously (Kronig et al., 1995). The sensitivity (percentage of true positives) was defined by the number of responders above threshold divided by the total number of responders. The specificity (percentage of true negatives) was defined by the number of non-responders at/or below threshold divided by the total number of non-responders. A x 2-test was applied to test for significant differences at the various thresholds (Kronig et al., 1995).

3. Results The group of patients consisted of 164 male and 47 female subjects. The mean age, mean methadone dose and mean duration of the present methadone treatment of the whole group (n =211, mean9 S.D., range) were 32 97 years, 19–50 years; 100 9 58 mg/day, 5–350 mg/day and 43950 months, 2 – 236 months, respectively. For patients in maintenance treatment (n =180),

49

these values were 339 7 years, 19–50 years; 109957 mg/day, 10–350 mg/day and 459 53 months, 2–236 months, respectively, and for those reducing their doses, they were 299 6 years, 21–47 years, 49927 mg/day, 4.5–100 mg/day, and 29 9 29 months, 3–125 months, respectively. The mean age and mean dose, but not the mean duration of treatment, were significantly lower in the patients reducing their dose than in patients in maintenance treatment (unpaired t=2.63, 5.74, 1.71; df= 209, 209, 209, PB 0.01, PB0.0001, P= 0.09, respectively). Within the whole group, 112 patients received comedications (mainly antidepressants and benzodiazepines) and 99 were without comedication. The mean (R)-, (S)-, (R,S)-methadone concentrations (in ng/ml) were (n= 211, mean9 S.D., median, range) 144987, 133, 10–491; 1369 88, 123, 10–485; 2819 169, 251, 16–976, respectively. Table 1 shows the mean (R)-, (S)-, (R,S)-concentrations per mg of administered methadone dose (corrected for body weight) as determined in the whole group; the table also shows the data for the patients with and without comedications, for the patients in maintenance therapy, and for those reducing their dose. There were no differences between patients without and with comedication with regard to the (R)-, (S)- and (R,S)-methadone concentration-to-dose-toweight ratios (unpaired t= 0.23, 1.22, 0.80; df=209, 209, 209; P= 0.82, 0.22, 0.42, respectively). However, the range of values was wider in the group receiving comedication and resulted in a 41-fold difference between the smallest and the largest ratio; a 17-fold difference could be found in the group without comedication. Fig. 1 shows the histogram of (R)-methadone concentrations per mg of administered methadone dose corrected for body weight in the group of patients without comedication. It is noteworthy that the smallest (R)- and (S)-methadone concentration-to-

Table 1 (R)-, (S)- and (R,S)-methadone concentrations per mg of administered methadone dose corrected for body weight (in ng×kg/ml×mg) as determined in different groups of patientsa

All patients (n =211) Patients without comedications (n= 99) Patients with comedications (n = 112) Patients in maintenance treatment (n =180) Patients reducing their dose (n=31)

(R)-methadone/dose/ weight

P-value (S)-methadone/dose/ weight

P-value (R,S)-methadone/dose/ weight

111960, 97, 10–407 1129 54, 100, 19–316

112 978, 93, 6–531 0.817b 119 969, 101, 12–357

224 9134, 191, 16–938 0.225b 232 9 119, 207, 19–673

111 9 64, 96, 10–407

106 985, 83, 6–531

2179146, 181, 16–938

1089 56, 97, 10–326

0.046c 106 9 70, 90, 6–447

0.004c 214 9122, 187, 16–706

131 9 77, 113, 32–407

Values are means 9 S.D., median and range. As compared to the group of patients with comedications. c As compared to the group of patients reducing their dose. a

b

150 9109, 135, 16–531

281 9 183, 255, 48–938

P-value

0.423b

0.009c

C.B. Eap et al. / Drug and Alcohol Dependence 61 (2000) 47–54

50

Fig. 1. Distribution of (R)-methadone concentrations per mg of administered methadone dose corrected for body weight (ng × kg/ ml× mg) in a group of patients without comedications (n =99). Brackets are round or square: (25, 50] means superior to 25 and inferior or equal to 50.

dose-to-weight ratios (10 and 6 ng× kg/ml × mg, respectively) were in a patient receiving tetrabamate (300 mg/day), a precursor of febarbamate, difebarbamate and phenobarbital; it is well known that barbiturates induce the metabolism of methadone (Liu and Wang, 1984). Interestingly, the mean (R)-, (S)-, and (R,S)methadone concentration-to-dose-to-weight ratios were significantly higher in the group of patients reducing their dose than in the group in maintenance treatment (unpaired t= −2.01, − 2.94, − 2.60; df= 209, 209, 209; P B 0.05, B 0.005, B0.01, respectively). Table 2 shows the correlations between drug dose corrected for body weight and (R)-, (S)-, and (R,S)-methadone concentrations in the whole group and in the groups of patients with and without comedications. As expected, the highest correlations were found in the latter group. The ratios of (R)-, (S)-, and (R,S)-concentrations to dose to weights were not significantly different between male and female subjects (unpaired t= 1.29, 1.43, 1.42; df= 209, 209, 209; P = 0.20, 0.16, 0.16, respectively) and were not correlated with age (r = 0.04, 0.00, 0.02, P = 0.52, 1.00, 0.77, respectively). On the other hand, (R)-, (S)-, and (R,S)-concentration-to-dose-to-weight ratios were significantly and inversely correlated with methadone doses (r= − 0.22, − 0.32, − 0.29; PB 0.005, B 0.0001, B 0.0001, respectively). Accordingly,

the mean (R)-, (S)-, and (R,S)-concentration-to-doseto-weight ratios measured in patients with doses higher than or equal to 100 mg/day (a dose considered as an upper limit in many centers) were significantly lower than those measured in patients with doses lower than 100 mg/day (mean value 99, 88, 187 ng × kg/ml ×mg vs. 124, 135, 258 ng × kg/ml ×mg; unpaired t =3.11, 4.56, 4.02, df= 209, 209, 209, PB 0.005, B0.0001, B0.0001, respectively). Finally, it should be mentioned that the mean (R)/(S)-methadone ratio was (n =211; mean9 S.D., range): 1.149 0.37, 0.55–2.55. This ratio was significantly correlated with the dose (r=0.41, PB 0.0001). In order to determine a threshold of (R)-, (S)- or (R,S)-methadone concentrations associated with therapeutic response, the data from the 31 patients reducing their doses of methadone were discarded from subsequent analyses as they represent a very different subset of patients as compared with those in MMT. Patients were then classified into responders and non-responders (response was defined as absence of opiate positive urine determination during the 2-month period prior to blood sampling). Thresholds from 100 to 500 ng/ml for (S)- and (R)-methadone, and from 100 to 1000 ng/ml for (R,S)-methadone, were examined in increments of 50 and 100 ng/ml, respectively. For each threshold, the sensitivity and the specificity were calculated as previously described (Kronig et al., 1995) (see Section 2). The results, with the x 2-values for each concentration, are shown in Table 3 (for (R)-methadone), Table 4 (for (S)-methadone) and Table 5 (for (R,S)-methadone). For (R)-methadone, x 2-values reached statistical significance at the thresholds of 200, 250 and 300 ng/ml but the lowest P-value was found for 250 ng/ml ( B0.002). For (S)-methadone, x 2-values did not reach statistical significance at any thresholds examined. For (R,S)methadone, a statistically significant x 2-value was found for one concentration (400 ng/ml). Finally, patients were also classified into responders and non-responders by the absence (in 125 patients) or presence (in 53 patients) of cocaine metabolite in urine samples collected during the 2-month period prior to blood sampling. No association was found between therapeutic response and (R)-, (S)-, or (R,S)-methadone concentrations (data not shown).

Table 2 Coefficients of correlation (r) and coefficients of determination (r 2) between methadone drug dosage corrected for body weight and (R)-, (S)- and (R,S)-methadone concentrations in various groups of patientsa

All patients (n = 211) Patients without comedications (n= 99) Patients with comedications (n = 112) a

PB0.0001 in all cases.

(R)-methadone

(S)-methadone

(R,S)-methadone

r= 0.58, r 2 =0.33 r= 0.65, r 2 =0.42 r =0.46, r 2 =0.21

r =0.34, r 2 =0.12 r =0.48, r 2 =0.23 r= 0.20, r 2 =0.04

r= 0.48, r 2 =0.23 r =0.57, r 2 =0.33 r =0.34, r 2 =0.12

C.B. Eap et al. / Drug and Alcohol Dependence 61 (2000) 47–54

51

Table 3 Sensitivity and specificity of various threshold (R)-methadone plasma levels for illicit opiate intakes Plasma levels (ng/ml)

Sensitivity (%)a

100 150 200 250 300 350 400 450 500

72.9 48.2 32.9 24.7 11.8 4.7 3.5 1.2 0

a b

Specificity (%)b 26.3 58.9 84.2 92.6 97.9 98.9 100 100 100

x 2-value (df =1)

P-value

0.01 0.94 7.26 10.27 6.73 2.22 3.41 1.12

0.910 0.333 B0.008 B0.002 B0.01 0.136 0.065 0.289

Number of responders above threshold divided by total number of responders (n = 85). Number of non-responders at or below threshold divided by total number of non-responders (n =95).

Table 4 Sensitivity and specificity of various threshold (S)-methadone plasma levels for illicit opiate intakes Plasma levels (ng/ml)

Sensitivity (%)a

100 150 200 250 300 350 400 450 500

63.5 38.8 23.5 12.9 8.2 4.7 3.5 1.2 0

a b

Specificity (%)b 38.9 62.1 76.8 89.5 95.8 98.9 98.9 98.9 100

x 2-value (df =1)

P-value

0.12 0.02 0.00 0.25 1.27 2.22 1.27 0.01

0.732 0.898 0.953 0.614 0.260 0.136 0.260 0.937

Number of responders above threshold divided by total number of responders (n = 85). Number of non-responders at or below threshold divided by total number of non-responders (n =95).

Table 5 Sensitivity and specificity of various threshold (R,S)-methadone plasma levels for illicit opiate intakes Plasma levels (ng/ml)

Sensitivity (%)a

100 200 300 400 500 600 700 800 900 1000

94.1 68.2 44.7 31.8 16.5 8.2 4.7 2.4 1.2 0

a b

Specificity (%)b 9.5 30.5 63.2 81.1 92.6 97.9 97.9 100 100 100

x 2-value (df =1)

P-value

0.81 0.03 1.15 3.93 3.61 3.55 0.94 2.26 1.12

0.369 0.858 0.283 B0.05 0.058 0.060 0.332 0.133 0.289

Number of responders above threshold divided by total number of responders (n = 85). Number of non-responders at or below threshold divided by total number of non-responders (n =95).

4. Discussion Methadone is extensively metabolized in the organism by cytochrome P450 enzymes. Cytochrome P4501A2 (CYP1A2) (Yue et al., 1995), cytochrome P450IID6 (CYP2D6) (Yue et al., 1995; Eap et al., 1997), cytochrome P450IIIA4 (CYP3A4) (Iribarne et al., 1996; Moody et al., 1997; Foster et al., 1999), cytochrome P450IIC9 (CYP2C9) (Moody et al., 1997;

Foster et al., 1999), and cytochrome P450IIC19 (CYP2C19) (Foster et al., 1999) were reported to be involved, but the respective contribution of each isozyme has not been defined clearly. The CYP2D6 enzyme preferentially metabolizes (R)-methadone (Yue et al., 1995; Eap et al., 1997), while CYP3A4 probably metabolizes both enantiomers (Foster et al., 1999). In a previous study with 50 patients, we demonstrated a large interindividual variability of methadone concen-

52

C.B. Eap et al. / Drug and Alcohol Dependence 61 (2000) 47–54

trations: even when corrected for dose and weight, the (R)-methadone concentrations varied sevenfold (Eap et al., 1998). The present study raised the number of patients studied and a 17-fold difference was found between the smallest and the largest values of (R)-methadone concentration-to-dose ratios in the subset of patients who did not receive any comedication. This is not surprising, considering the interindividual variability of the activities of CYP enzymes, activities that are both genetically and environmentally controlled (Brøsen and Gram, 1989; Wilkinson et al., 1989; Bertilsson et al., 1993; Schweikl et al., 1993; Ketter et al., 1995; Yasar et al., 1999); it is also relevant that increasing the number of patients increases the probability of including patients with several CYP enzymes that have simultaneously low or high activities, as well as patients with variable renal clearance (Nilsson et al., 1982). In a study of (R,S)-methadone concentrations performed in a closed metabolic ward (where compliance could be controlled), it was found that the steady-state concentrations showed more than a 10-fold difference (with 60 mg/day) in a group of 17 patients (Holmstrand et al., 1978). In the present study, in the subset of patients receiving comedications, there was a 41-fold difference between the smallest and the largest ratios for (R)methadone: this is probably due to some inhibition and induction of metabolism in the patients with large and small ratios, respectively. Thus, the smallest ratio was found in a patient receiving barbiturates, a well-known CYP3A4 inducer (Liu and Wang, 1984). In the subset of patients without comedication, the correlation calculated for methadone dosage (corrected for body weight) with (R)-methadone concentration showed that less than half of the variability (r 2 = 42%) of the concentrations of (R)-methadone can be explained by drug dosage. This confirms our previous study (Eap et al., 1998). The significant (although weak) inverse correlation between concentration-todose-to-weight ratios and methadone dose may be explained by some induction of drug metabolism at large doses and/or by the selection of patients with high metabolic activity in the large-dose group (dose increases were based on criteria such as continued use of illicit drugs and/or complaints of being underdosed). This inverse correlation probably explains the significantly higher concentration-to-dose-to-weight ratios in the group of patients reducing their doses as compared with those in MMT because smaller doses of methadone were given to the former group. Finally, the significant correlation between (R)/(S) methadone ratios and methadone doses may be tentatively explained by a quantitatively more important contribution of CYP3A4, an enzyme of low affinity but high capacity, to high-dose methadone metabolism. Thus, the (S)enantiomer would be more metabolized by this non-

stereoselective enzyme at large than at small doses. On the other hand, CYP2D6, an enzyme with high affinity and small capacity and with a stereoselectivity towards the (R)-enantiomer (Yue et al., 1995; Eap et al., 1997), would be more involved at low doses. The results of our study indicate an absence of association between (R)-, (S)-, and (R,S)-methadone plasma concentrations and therapeutic response when examining the intake of cocaine. This is not surprising considering the known limited efficacy of methadone for reducing cocaine abuse in opioid-dependent patients (Schottenfeld et al., 1997). The most relevant result of this study shows that (R)- and (R,S)-methadone but not (S)-methadone plasma concentrations are associated with therapeutic response to methadone, when examining the intake of illicit opiates. As (R)-methadone contributes to the majority of the opioid effect of racemic methadone (Kristensen et al., 1995), the fact that no association was found between response and (S)-methadone concentrations was expected. Our data support the existence of a threshold of 400 ng/ml for (R,S)-methadone. Interestingly, this value is often considered necessary to provide stabilized maintenance (Loimer et al., 1991; Payte and Khuri, 1993). However, to our knowledge, a study which clearly demonstrates the existence of such a threshold has been lacking until now. Our data also support the existence of thresholds of 200, 250 and 300 ng/ml for (R)-methadone, but the best association was found at 250 ng/ml. At this value, the specificity of the threshold was found to be higher for (R)-methadone (93%) than for (R,S)-methadone (81%). In other words, at the threshold of 250 ng/ml of (R)-methadone, only 7% of the non-responders have plasma concentrations of (R)-methadone above this concentration, while this proportion was 19% at the threshold of 400 ng/ml for (R,S)-methadone. One can see that for both (R)- (at 250 ng/ml) and (R,S)-methadone (at 400 ng/ml), the calculated sensitivities are low: 25 and 32%, respectively. In other words, 75 and 68% of responders have plasma concentrations of (R)- and (R,S)-methadone, respectively, below this concentration. This reflects the well-known fact that patients in MMT may respond to their treatment at small doses and/or small concentrations of methadone. Interindividual response to methadone-maintenance treatment may depend on pharmacokinetic parameters (i.e. metabolism variations), and also on pharmacodynamic parameters (i.e. receptors’ variations). Moreover, other factors, such as psychological or social factors, are also extremely important. The present results demonstrate a low sensitivity but a high specificity at the thresholds mentioned above, associated with a wide interindividual variability of methadone concentrations; this suggests that therapeutic monitoring of methadone (preferably of the (R)-enantiomer) is not necessary in every patient

C.B. Eap et al. / Drug and Alcohol Dependence 61 (2000) 47–54

or for patients who respond to their treatment, but could be very useful in cases of non-response (i.e. continued use of illicit opiates). It should be mentioned that, at the threshold of 400 ng/ml of (R)-methadone, all non-responders had (R)-methadone plasma values below this concentration (specificity: 100%). Thus, for those few patients who still use illicit opiates, even if their (R)-methadone concentrations have reached 250 ng/ml, it might be worthwhile to increase further the dose in order to obtain a (R)-methadone concentration of 400 ng/ml. In a very recent study, it was found that very large doses of methadone (up to 780 mg/day) were necessary to prevent illicit opiate use in some patients (Maxwell and Shinderman, 1999). Methadone dose policy may vary between countries, states and programs. Patients in ‘low dose programs’ may receive doses that are too small to be effective (D’Aunno and Vaughn, 1992). To obtain a concentration of 250 ng/ml of (R)-methadone with the range of (R)-methadone concentration-to-dose-to-weight ratios that we found in patients without comedications (19– 316 ng× kg/ml ×mg, see Table 1), the theoretical doses of racemic methadone could be as small as 55 mg/day, and the maximum value would be as large as 921 mg/day in a 70-kg patient. This highlights the importance of individualizing methadone treatment and, in our opinion, it is a strong argument against the policy of arbitrarily fixed limits of methadone doses.

Acknowledgements The authors thank A.-C. Aubert for collecting and computerizing the urine analyses data. They also thank the editorial assistance of C. Bertschi and the bibliographical help of J. Rosselet, M. Gobin and T. Bocquet. This work has been supported in part by grants from the Swiss Federal Office of Public Health (No. 316 97 0629).

References Bell, J., Seres, V., Bowron, P., Lewis, J., Batey, R., 1988. The use of serum methadone levels in patients receiving methadone maintenance. Clin. Pharmacol. Ther. 43, 623–629. Bell, J., Bowron, P., Lewis, J., Batey, R., 1990. Serum levels of methadone in maintenance clients who persist in illicit drug use. Br. J. Addict. 85, 1599–1602. Bertilsson, L., Dahl, M.-L., Sjo¨qvist, F., Aberg-Wistedt, A., Humble, M., Johansson, I., Lundqvist, E., Ingelman-Sundberg, M., 1993. Molecular basis for rational megaprescribing in ultrarapid hydroxylators of debrisoquine. Lancet 341, 63. Bertschy, G., 1995. Methadone maintenance treatment: an update. Eur. Arch. Psychiatry Clin. Neurosci. 245, 114–124. Bertschy, G., Baumann, P., Eap, C.B., Baettig, D., 1994. Probable metabolic interaction between methadone and fluvoxamine in addict patients. Ther. Drug Monit. 16, 42–45.

53

Brøsen, K., Gram, L.F., 1989. Clinical significance of the sparteine/ debrisoquine oxidation polymorphism. Eur. J. Clin. Pharmacol. 36, 537 – 547. Caplehorn, J.R.M., Bell, J., 1991. Methadone dosage and retention of patients in maintenance treatment. Med. J. Aust. 154, 195–199. D’Aunno, T., Vaughn, T.E., 1992. Variations in methadone treatment practices — Results from a national study. J. Am. Med. Assoc. 267, 253 – 258. Dole, V.P., 1988. Implications of methadone maintenance for theories of narcotic addiction. J. Am. Med. Assoc. 260, 3025 – 3029. Dyer, K.R., Foster, D.J.R., White, J.M., Somogyi, A.A., Menelaou, A., Bochner, F., 1999. Steady-state pharmacokinetics and pharmacodynamics in methadone maintenance patients: comparison of those who do and do not experience withdrawal and concentration – effect relationships. Clin. Pharmacol. Ther. 65, 685–694. Eap, C.B., Finkbeiner, T., Gastpar, M., Scherbaum, N., Powell, K., Baumann, P., 1996. Replacement of (R)-methadone by a double dose of (R,S)-methadone in addicts: interindividual variability of the (R)/(S) ratios and evidence of adaptive changes in methadone pharmacokinetics. Eur. J. Clin. Pharmacol. 50, 385 – 389. Eap, C.B., Bertschy, G., Powell, K., Baumann, P., 1997. Fluvoxamine and fluoxetine do not interact in the same way with the metabolism of the enantiomers of methadone. J. Clin. Psychopharmacol. 17, 113 – 117. Eap, C.B., Bertschy, G., Finkbeiner, T., Gastpar, M., Scherbaum, N., Baumann, P., 1998. High interindividual variability of methadone enantiomer blood levels to dose ratios. Arch. Gen. Psychiatry 55, 89 – 90. Foster, D.J.R., Somogyi, A.A., Bochner, F., 1999. Methadone Ndemethylation in human liver microsomes: lack of stereoselectivity and involvement of CYP3A4. Br. J. Clin. Pharmacol. 47, 403– 412. Holmstrand, J., A8 nggard, E., Gunne, L.-M., 1978. Methadone maintenance: plasma levels and therapeutic outcome. Clin. Pharmacol. Ther. 23, 175 – 180. Horns, W.H., Rado, M., Goldstein, A., 1975. Plasma levels and symptom complaints in patients maintained on daily dosage of methadone hydrochloride. Clin. Pharmacol. Ther. 17, 636–649. Iribarne, C., Berthou, F., Baird, S., Dre´ano, Y., Picart, D., Bail, J.P., Beaune, P., Me´nez, J.F., 1996. Involvement of cytochrome P450 3A4 enzyme in the N-demethylation of methadone in human liver microsomes. Chem. Res. Toxicol. 9, 365 – 373. Kell, M.J., 1995. Utilization of plasma and urine methadone concentration measurements to limit narcotics use in methadone maintenance patients: II. Generation of plasma concentration response curves. J. Addict. Dis. 14, 85 – 108. Ketter, T.S., Flockhart, D.A., Post, R.M., Denicoff, K., Pazzaglia, P.J., Marangell, L.B., George, M.S., Callahan, A.M., 1995. The emerging role of cytochrome P450 3A in psychopharmacology. J. Clin. Psychopharmacol. 15, 387 – 398. Kristensen, K., Christensen, C.B., Christrup, L.L., 1995. The mu1, mu2, delta, kappa opioid receptor binding profiles of methadone stereoisomers and morphine. Life Sci. 56, 45 – 50. Kronig, M.H., Munne, R.A., Szymanski, S., Safferman, A.Z., Pollack, S., Cooper, T., Kane, J.M., Lieberman, J.A., 1995. Plasma clozapine levels and clinical response for treatment-refractory schizophrenic patients. Am. J. Psychiatry 152, 179 – 182. Liu, S.-J., Wang, R.I.H., 1984. Case report of barbiturate-induced enhancement of methadone metabolism and withdrawal syndrome. Am. J. Psychiatry 141, 1287 – 1288. Loimer, N., Schmid, R., 1992. The use of plasma levels to optimize methadone maintenance treatment. Drug Alcohol Depend. 30, 241 – 246. Loimer, N., Schmid, R., Gru¨nberger, J., Jagsch, R., Linzmayer, L., Presslich, O., 1991. Psychophysiological reactions in methadone maintenance patients do not correlate with methadone plasma levels. Psychopharmacology 103, 538 – 540.

54

C.B. Eap et al. / Drug and Alcohol Dependence 61 (2000) 47–54

Maxwell, S., Shinderman, M., 1999. Optimizing response to methadone maintenance treatment: Use of higher dose methadone. J. Psychoact. Drugs 31, 95–102. Meresaar, U., Nilsson, M.-I., Holmstrand, J., A8 nggard, E., 1981. Single dose pharmacokinetics and bioavailability of methadone in man studied with a stable isotope method. Eur. J. Clin. Pharmacol. 20, 473 – 478. Moody, D.E., Alburges, M.E., Parker, R.J., Collins, J.M., Strong, J.M., 1997. The involvement of cytochrome P450 3A4 in the N-demethylation of L-a-acetylmethadol (LAAM), norlaam, and methadone. Drug Metab. Dispos. 25, 1347–1353. Nilsson, M.-I., Meresaar, U., A8 nggard, E., 1982. Clinical pharmacokinetics of methadone. Acta Anaesthesiol. Scand. 74 (Suppl.), 66– 69. Payte, J.T., Khuri, E.T., 1993. Principles of methadone dose determinations. In: U.S. Department of Health and Human Services (Eds.), State Methadone Treatment Guidelines. Rockwall, Rockville, MD, pp. 101–124. Schottenfeld, R.S., Pakes, J.R., Oliveto, A., Ziedonis, D., Kosten, T.R., 1997. Buprenorphine vs methadone maintenance treatment for concurrent opioid dependence and cocaine abuse. Arch. Gen. Psychiatry 54, 713 – 720. Schweikl, H., Taylor, J.A., Kitareewan, S., Linko, P., Nagorney, D., Goldstein, J.A., 1993. Expression of CYP1A1 and CYP1A2 genes in human liver. Pharmacogenetics 3, 239–249.

.

Sullivan, H.R., Due, S.L., 1973. Urinary metabolites of DL-methadone in maintenance subjects. J. Med. Chem. 16, 909 – 913. Tennant, F.S., Jr, 1987. Inadequate plasma concentrations in some high-dose methadone maintenance patients. Am. J. Psychiatry 144, 1349 – 1350. Torrens, M., Castillo, C., San, L., del Moral, E., Gonzalez, M.L., de la Torre, R., 1998. Plasma methadone concentrations as an indicator of opioid withdrawal symptoms and heroin use in a methadone maintenance program. Drug Alcohol Depend. 52 (7), 193 – 200. Wilkinson, G.R., Guengerich, F.P., Branch, R.A., 1989. Genetic polymorphism of S-mephenytoin hydroxylation. Pharmacol. Ther. 43, 53 – 76. Wolff, K., Hay, A.W.M., Raistrick, D., 1992. Plasma methadone measurements and their role in methadone detoxification programs. Clin. Chem. 38, 420 – 425. Yasar, U8 ., Eliasson, E., Dahl, M.-L., Johansson, I., 1999. Validation of methods for CYP2C9 genotyping: frequencies of mutant alleles in a Swedish population. Biochem. Biophys. Res. Commun. 254, 628 – 631. Yue, Q.Y., Svensson, J.O., Bertilsson, L., et al., 1995. Racemic methadone kinetics in relation to the debrisoquine hydroxylation polymorphism in man. Abstracts of the 1st Congress of the European Association for Clinical Pharmacology and Therapeutics 27 – 30 September 1995 [Abstract]. Therapie (Suppl.), 88.