Preliminary Population PK of Methadone in a High-Dose Patient Population

Background

Methadone is a racemic mixture of the two enantiomers, d- methadone and l-methadone (d,l-6-dimethylamino-4,4-diphenyl-3- heptanone). The l-enantiomer is largely responsible for the pharmacodynamic effects.(1,2) The drug has found widespread use in the treatment of narcotic addiction, a chronic, relapsing disorder.(3) Nevertheless, the optimal amount of orally administered methadone associated with positive outcome parameters such as decreased illicit drug use, increased retention time in the program, and increased social stability continues to be the subject of controversy.(4 ,5)

In many studies, daily methadone dose is inversely correlated with illicit drug use and positively correlated with retention time in treatment. (6,7,8,9,10) The risk of HIV transmission is also lower with higher methadone doses, probably as a result of decreased illicit drug use and needle-sharing.(11,12)

Factors which influence the pharmacokinetics of methadone have not been well defined, due to the small number of patients studied. Three studies of methadone kinetics abstracted from the literature involved a total of 40 individuals, all but 8 sampled from a methadone maintained population. Methadone kinetic parameters were as follows: Mean Clearance (Cl) was 7.53 L/hr, range of 7.16 L/hr to 8.53 L/hr; Mean Volume of Distribution (Vdb) was 275.3 L, range of 258.7 L to 319.2 L; Mean terminal half-life (T1/2) was 28.5 hr, range of 25.8 hr to 30.2 hr.(13,14,15)

Approximately 25% of patients enrolled in a methadone maintenance treatment program (MMTP) were found to require greater than 100 mg methadone daily for stability. Stability was evidenced by daily attendance at the clinic, attendance at weekly individual counseling sessions and abstinence from opiate use as evidenced by random urine toxicology testing. We therefore investigated the kinetics of methadone in this group of patients hypothesizing that such patients might have a decreased T1/2 for the drug.

Subjects

The setting for the study was an urban MMTP treating 625 patients. Blood was sampled for methadone concentrations from 169 patients between August 1, 1992 and June 1, 1994. By policy, the program has no ceiling dose for methadone. To remain eligible for the outpatient MMP, patients were expected to abstain from use of illicitly obtained opiates, attend biweekly group therapy and attend once weekly individual counseling. Urine toxicology screens were randomly performed so that each patient had approximately 52 screens per year. Patients enrolled in the MMTP could request to see clinical staff for dose evaluations at any time. At the time of evaluation, the patient's subjective report as well as objective evidence of opiate withdrawal or opiate effects were evaluated, and the dose of methadone was readjusted, if necessary.

Figure 1 below shows the frequency distribution of doses for n=625 patients enrolled in the program. The distribution is log-normal, as evidenced by the bottom graph.

Figure 2 below shows the frequency distribution of methadone doses for n=169 high-dose patients who were sampled for methadone levels.

Procedures

Blood samples for methadone concentration were requested from patients receiving any dose of methadone who had no evidence of illicit opiate use within two weeks of the request yet continued to voice subjective complaints of opiate withdrawal without objective symptoms noted on physical examination performed prior to the day's dose (n=21). Blood samples were also requested from patients who displayed objective evidence of opiate withdrawal prior to the day's dose, and who at the time of evaluation were receiving > 90 mg methadone (n=148). Blood was sampled in both groups prior to the morning dose. The time of the dose prior to the blood level as well as the time of the blood drawing were recorded. Serum concentrations of methadone were determined by an HPLC method using photodiode array detection.(16)

Demographic characteristics are shown in Table 1. In addition to HIV status, patients also were asked for a list of currently prescribed medications, since it is well known that anticonvulsant treatment with diphenylhydantoin, phenobarbital, carbamazepine or primidone may accelerate methadone clearance by induction of P450.

Table 1 Demographics and Subject Characteristics

Data Analysis

Population pharmacokinetic parameters were determined using the software NONMEM 77 Version IV Level 2.0 developed and programmed by Stuart Beal and Lewis Sheiner. (17,18) The program was compiled on an IBM 9121-440 mainframe in double-precision format with the IBM VS Fortran VS 2 compiler with optimization. NONMEM, "Nonlinear Mixed Effects Model" can be used to analyze population data with regression type statistical models containing both fixed (e.g., drug concentration data) and random effects (unexplainable inter- and intra-subject variation). Given the sparse data available on a relatively large number of individuals in this study, this approach allowed us to estimate the oral clearance parameter for methadone.

Since subject plasma methadone concentration was measured during the dispositional phase of the dose-time concentration curve, at 23.2 ± 2.0 hours, a simple one-compartment system was modeled at steady-state, C(t) = (D/Vd)e^(-kt) where k=Cl/Vd. C(t) is methadone level at time t in a particular subject, D (mg) is the dose of methadone, Vd (L) is the volume of distribution of methadone for the patient, t (hours) is the interval between the last methadone dose and the sample time, and Cl is the oral clearance of methadone in L/hr.

Only one blood sample was obtained for each subject, Vd could not be estimated from the data. However, based on a consensus of estimates for Vd presented in the literature, a value of 3.68 L/kg was utilized for this population parameter.

Gender, HIV status and anticonvulsant use were entered into the model. The decision to retain these factors was based on the Aikaike Information Criterion (AIC) where AIC = la - lb + 2(pa - pb), where la and lb represent the minimum values of the objective function for Models A and B respectively, and pa and pb are the number of parameters in the models. Model A is chosen for AIC < 0 and Model B is chosen for AIC > 0.

Results

The final model incorporated HIV status and anticonvulsant use as explanatory variables. The resulting population values for the clearances are found in Table 2.

A value of 11.2 L/kg ± 0.5 (99% CI 10.0-12.4) was obtained for the population oral clearance of methadone in those patients belonging to the first group not receiving anticonvulsants and who were HIV negative. This contrasts with the value of 7.4 ± 1.8 reported from the literature.

Patients who were receiving anticonvulsants had a clearance value of 23.9 ± 3.1 (99% CI 16.0 - 31.8), higher than the first group at p < 0.01, and higher than those patients who were HIV positive, whose clearance value was 8.9 ± 1.7 (99% CI 4.59 - 13.31) at p < 0.01. There was no difference between HIV positive patients and patients in the first group.

We have calculated the terminal half-life of methadone in these three groups of patents using the following relationship: t1/2 = 0.693*Vd/Cl The results are also presented in Table 2 below:

Discussion

The hypothesis that patients enrolled in a MMP receiving higher doses of methadone have a decreased T1/2 for the drug is supported by the results of the present study. The decreased T1/2 is likely due to an increased oral clearance for the drug, though changes in volume of distribution may also have played a role. The design of the study, whereby a single sample of blood for methadone concentration at steady state was obtained in a large number of patients, did not allow for estimation of volume of distribution. In spite of this limitation, the results suggest that for some patients, higher methadone doses may be required to achieve opiate abstinence due to faster metabolic disposition of the drug.

The high value for clearance we have obtained for patients who are neither HIV positive nor are using anticonvulsants suggests that there may be a subset of patients receiving methadone, as previously suggested (19), who may have different dispositional kinetics, resulting in a lower value for half-life. This particular study, which measured volume of distribution, found that the difference in half-life between therapeutic failures, (half-life 24.5 ± 2.6 hours, Vd = 3.09 ± 0.96 L/kg) vs the comparison group (half-life 34.0 ± 7 hours, Vd = 4.56 ± 1.00 L/kg) could be accounted for by a smaller volume of distribution in the therapeutic failures. In the present study, we have not excluded this possibility, as we could not directly estimate Vd. We used the value 3.68 L/kg for Vd, which lies between these two values, whereas our estimate for half-life is 16.46 hours, considerably smaller than 24.5 hours. It is unlikely that such a large difference in half-life could be accounted for by Vd, suggesting that clearance of the drug may also be faster in this population of patients requiring higher doses of methadone for stability.

The unique dose-concentration relationship in the group of HIV - ACV - patients presented in Table 3 is evident when these dose-concentration values are compared to other studies. These patients appear to require approximately twice the average daily methadone dose reported in previous studies to achieve comparable blood concentrations. One study of 21 methadone maintained patients receiving 60 mg methadone showed mean trough concentrations of 217 ng/ml. (20) Another study of 9 stable methadone maintained patients receiving a mean daily dose of 74.4 mg methadone revealed trough concentrations of 235.5 ng/ml. (21)

Using our value for the clearance of this high-dose population, 11.2 L/hr, we can calculate the dose needed to maintain a 70 kg male with a Vd of 266 liters above a trough level of 300 ng/ml using the following relationship (22): DM = {Vd*Css,min*(1-e-kt)}/e-kt k = Cl/Vd where DM is the maintenance dose, Css,min is the minimum concentration at steady state, k is the elimination rate constant, and t is the dosing interval, which is 24 hours. This is computed to be 137 mg. This amount of methadone is higher than generally used in MMTP.

If these results are substantiated by larger studies, where Vd for this type of population can be estimated, such as contained in the present proposal, then a case can be made for twice daily dosing of some patients in order to maintain methadone levels within a defined therapeutic range. Alternately, such patients may benefit from L-acetyl-methadone, which having a longer half-life, minimizes between dose fluctuation of blood levels.

The population clearance for patients on anticonvulsant medications resulting in a half-life of 7.9 hours raises the issue of whether such patients can be adequately dosed once daily, without running into the risks of post-dose sedation due to high initial levels, followed by withdrawal symptoms as levels rapidly drop over time before the next dose.

HIV positive patients in this high-dose group tended to have a clearance which approached that reported in the literature. The reasons for this are not obvious, but may relate to possible inhibitory actions of prescribed drugs on methadone disposition. Many of these patients were using multiple medications such as zidovudine, ketoconazole, fluconazole, and tricyclic antidepressant drugs. From other preliminary data, we suspect that some of these, especially ketoconazole, may affect methadone disposition. In addition, such drugs may affect changes in serum protein and tissue binding of methadone, which may have influenced our estimate of Vd. In summary, we have shown preliminary evidence that there may be a population of patients with faster metabolic clearance for methadone, who may require higher than usual doses of methadone for stabilization. The reasons for this altered disposition will need to be explored with a larger and more comprehensive study. The identification of a subset of patients with faster methadone metabolism than previously appreciated would result in a reassessment of the adequacy of methadone dosing in outpatient methadone maintenance programs.

References

  1. Isbell H, Eisenman AJ. The addiction liability of some drugs of the methadone series. Journal of Pharmacology and Experimental Therapeutics 1948; 93:305-313.
  2. Scott CC, Robbins EB, Chen KK. Pharmacologic comparison of the optical isomers of methadone. Journal of Pharmacology and Experimental Therapeutics 1948; 93:282-286.
  3. Dole VP. Implications of methadone maintenance for theories of narcotic addiction. Journal of the American Medical Association 1988; 260:3025-3029 .
  4. Horns WH, Rado M, Goldstein A. Plasma levels and symptom complaints in patients maintained on daily dosage of methadone hydrochloride. Clinical Pharmacology and Therapeutics 1975; 17:636-649.
  5. Hargreaves WA. Methadone dose and duration for maintenance treatment. In: Cooper Jr, Altman F, Brown BS, Czechowicz D, eds. Research on the treatment of narcotic addiction: state of the art. Rockville, Maryland: National Institute on Drug Abuse, 1983; US Dept of Health and Human Services publication ADM 83-1281.
  6. Ball JC, Lange WR, Myers CP, Friedman SR. Reducing the risk of AIDS through methadone maintenance treatment. Journal of Health and Social Behavior 1988; 29:214-226.
  7. Loimer N, Schmid R. The use of plasma levels to optimize methadone maintenance treatment. Drug and Alcohol Dependence 1992; 30:241-246. D'Aunno T, Vaughn TE. Variations in methadone treatment practices. Journal of the American Medical Association 1992; 267:253-258.
  8. D'Aunno T, Vaughn TE. Variations in methadone treatment practices. Journal of the American Medical Association 1992; 267:253-258.
  9. Dole VP. Methadone treatment and the acquired immunodeficiency syndrome. Journal of the American Medical Association 1989; 262:1681-1682.
  10. Caplehorn JRM, Bell J. Methadone dosage and retention of patients in maintenance treatment. The Medical Journal of Australia 1991; 154:195-199.
  11. Cooper JR. Methadone treatment and acquired immunodeficiency syndrome. Journal of the American Medical Association 1989; 262:1664-1681.
  12. Cooper JR. Ineffective Use of psychoactive drugs. Journal of the American Medical Association 1992; 267(2):281-282.
  13. Nilsson MI, Anggard E, Holmstrand J, Gunne LM. Pharmacokinetics of methadone during maintenance treatment: adaptive changes during the induction phase. European Journal of Clinical Pharmacology 1982; 22(4):343-349.
  14. Nilsson MI, Gronbladh L, Widerlov E, Anggard E, Pharmacokinetics of methadone in methadone maintenance treatment: characterization of therapeutic failures. European Journal of Clinical Pharmacology 1983; 25(4):497-501.
  15. Inturrisi CE, Colburn WA, Kaiko RF, Houde RW, Foley KM. Pharmacokinetics and pharmacodynamics of methadone in patients with chronic pain. Clinical Pharmacology and Therapeutics 1987; 41(4):392-401.
  16. Greizerstein HB, McLaughlin IG. Sensitive method for the determination of methadone in small blood samples. Journal of Chromatography. 264(2)312-315. 1983.
  17. Beal S, Sheiner L. The NONMEM system. Am Statistician. 34:118. 1980.
  18. Beal SL, Sheiner LB. Estimating Population Kinetics. CRC Critical Reviews in Biomedical Engineering. 8:195-222. 1982.
  19. Nilsson MI, Gronbladh L, Widerlov E, Anggard E, Pharmacokinetics of methadone in methadone maintenance treatment: characterization of therapeutic failures. European Journal of Clinical Pharmacology 25(4):497-501. 1983.
  20. Holmstrand J, Anggard E, Gunne L-M. Methadone maintenance: plasma levels and therapeutic outcome. Clin Pharmacol Ther. 1978; 23(2):175-180.
  21. Loimer N, Schmid R, Grunberger J, Jasch R, Linzmeyer L, Presslich O. Psychophysiological reactions in methadone maintenance patients do not correlate with methadone plasma levels. Psychopharmacology 1991.103:538-540.
  22. Rowland M, Tozer TN. Clinical Pharmacokinetics. Lea & Febiger, London. 1989.

Author's Note

The data and results presented were included in a grant submitted to NIDA on February 1, 1995 entitled "Population Kinetics of Methadone: Understanding the Variance in Response" [1 RO1 DA10097-01] PI: Paolo B. DePetrillo, MD, Assistant Professor, Department of Medicine, Brown University School of Medicine, Providence, RI 02912, USA. Current Contact Page

Consultant: David A. Flockhart, MD, PhD; Assistant Professor, Department of Medicine and Pharmacology, Georgetown University, Division of Clinical Pharmacology, 3900 Reservoir Road NW, Washington DC, 20007, USA.

Web-published on 12/4/95

Acknowledgments

The author wishes to thank Dr. Darrell R. Abernethy, who suggested using NONMEM to analyze population kinetics in methadone-maintained patients while providing his much appreciated critiques of the work.