The Protease Inhibitor Lopinavir-Ritonavir May Produce Opiate Withdrawal in Methadone- Maintained Patients

Transcription

1 MAJOR ARTICLE The Protease Inhibitor Lopinavir-Ritonavir May Produce Opiate Withdrawal in Methadone- Maintained Patients Elinore F. McCance-Katz, 1 Petrie M. Rainey, 2 Gerald Friedland, 3 and Peter Jatlow 3 1 Medical College of Virginia, Virginia Commonwealth University, Richmond; 2 University of Washington, Seattle; and 3 Yale University School of Medicine, New Haven, Connecticut This study examines the pharmacokinetic/pharmacodynamic interactions between (1) lopinavir-ritonavir (L/ R), a fixed combination of protease inhibitors used for the treatment of HIV disease, and (2) ritonavir alone at the same dosage as that in the L/R formulation, with methadone, an opiate frequently used in substance abuse pharmacotherapy for opioid (heroin) dependent injection drug users, many of whom are infected with HIV. L/R was associated with significant reductions in the methadone area under the concentration-time curve ( P!.001), maximum concentration ( P!.001), and minimum concentration ( P!.001), as well as increased methadone oral clearance ( P!.001) and increased opiate withdrawal symptoms ( P p.013), whereas ritonavir use alone modestly and nonsignificantly increased methadone concentrations. Lopinavir is a potent inducer of methadone metabolism, and treatment with L/R requires clinical monitoring and increased methadone doses in some patients, whereas ritonavir has no significant effect on methadone metabolism. An important consideration in the treatment of opiate (principally heroin) addicted individuals with HIV disease is the potential for harmful drug interactions between medications used for the treatment of opiate addiction and antiretrovirals. Several antiretroviral agents have been shown to have interactions of potential clinical significance with methadone [1 5], the opiate therapy most frequently used for the treatment of heroin addiction and currently the substance abuse Received 17 December 2002; accepted 17 March 2003; electronically published 1 August Financial support: National Institute on Drug Abuse (grants KO2-DA and RO1-DA13004) and National Institutes of Health (General Clinical Research Center grant MO1-RR12248) awarded to the Albert Einstein College of Medicine of Yeshiva University. Reprints or correspondence: Dr. Elinore F. McCance-Katz, Virginia Commonwealth University, Medical College of Virginia, P.O. Box , Richmond, VA (emccancekatz@vcu.edu). Clinical Infectious Diseases 2003; 37: by the Infectious Diseases Society of America. All rights reserved /2003/ $15.00 pharmacotherapy of choice for opioid-addicted patients with HIV infection. Lopinavir-ritonavir (L/R) is the newest of the licensed protease inhibitors and the first HIV medication to combine 2 protease inhibitors in 1 gelatin capsule [6]. The protease inhibitors are a potent class of drugs active against HIV and a recommended component of HAART. L/R was formulated as a combination drug because ritonavir has been shown to inhibit CYP450 3A [7], thereby slowing lopinavir s pharmacokinetic disposition and producing lopinavir levels that exceed inhibitory concentrations for HIV. CYP450 3A has been reported to be the principal enzyme responsible for the metabolism of methadone, albeit not the only one [8, 9]. Given the reported effect of ritonavir on lopinavir pharmacokinetics, it might be supposed that coadministration of L/R with methadone would increase methadone concentrations. However, the converse has been demonstrated in 2 studies in which methadone concentrations were significantly decreased when L/R was administered [10, 11]. One of these studies reported that, despite substantial decreases 476 CID 2003:37 (15 August) McCance-Katz et al.

2 in methadone concentrations after L/R administration, no evidence of opiate withdrawal was observed [11]. Ritonavir alone has also been reported to significantly decrease methadone concentrations [12], and a case report describes opiate withdrawal in a methadone-maintained individual with HIV disease treated with ritonavir [13]. The seemingly contradictory reports to date led us to design the current study, in which we examined the pharmacodynamics and pharmacokinetics of methadone before and after treatment either with L/R or with ritonavir alone at the same dosage received in the L/R combination product. METHODS Table 1. Characteristics of patients enrolled in a study of pharmacologic interactions between methadone and either lopinavirritonavir or ritonavir alone. Characteristic Lopinavirritonavir group Ritonavir alone group No. of patients Age, mean years SE Weight, mean kg SE Female sex 5 8 Race African American 5 7 White 3 3 Hispanic 7 5 Substance use Opioid dependence Cocaine 6 7 Alcohol abuse 0 1 Hepatitis C seropositivity 4 3 AST level, mean U/L Before therapy After therapy ALT level, mean U/L Before therapy After therapy NOTE. Data are no. of patients, unless otherwise indicated. ALT, alanine aminotransferase; AST, aspartate aminotransferase. Study population. Fifteen healthy, methadone-maintained individuals participated in each of 2 within-subject pharmacokinetics studies to determine the effect on methadone disposition of treatment with (1) L/R (400 mg/100 mg twice per day) and (2) ritonavir alone (100 mg twice per day). The study had the approval of the Institutional Review Board of the Albert Einstein College of Medicine, Montefiore Medical Center (Bronx, NY), and voluntary, written, informed consent was obtained before any study procedures were undertaken. The US Department of Health and Human Services guidelines for human experimentation were followed in the conduct of this clinical research. Men and women were enrolled in the study if they were found to be HIV seronegative by ELISA (current standards preclude treatment of HIV-positive individuals with a single antiretroviral agent), were 18 years of age, were not being treated with medications that might alter hepatic cytochrome P450 function, and were without clinically significant medical conditions, as determined by medical history and physical examination, complete blood cell count, liver function tests, glucose level, urea nitrogen level, creatinine level, pregnancy tests (for women), and urinalysis. Urine toxicology screens were completed for all participants. Subject characteristics are summarized in table 1. Participants met the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) [14], for opioid dependence, were enrolled in a certified opiate treatment program, and had been maintained with a methadone dose that had been unchanged for 2 weeks. Other substance-use disorders were identified by clinical assessment and administration of the Mini International Neuropsychiatric Interview [15]. No participants met diagnostic criteria for physiological alcohol dependence. No participants met DSM-IV [14] criteria for current mental disorders. Upon determination of study eligibility, participants were admitted to an inpatient facility to participate in a bloodsampling study over 24 h to determine methadone pharmacokinetics. On the morning of the study, a urine toxicology screen was completed to determine recent use of opiates (samples containing opioids other than methadone would result in termination of the study session) or other drugs, and a Breathalyzer test for detection of recent alcohol use was also performed. Opiate withdrawal symptoms were assessed by experienced clinicians using the Objective Opiate Withdrawal Scale (OOWS) [16] before starting each pharmacokinetic study. The usual daily oral dose of methadone was given within 1 h of the participant s usual dosing time. Five-milliliter blood samples were obtained immediately before methadone dosing and at the predetermined time points of 15 min, 30 min, 45 min, 60 min, 90 min, 2 h, 4 h, 6 h, 8 h, 10 h, 12 h, and 24 h after methadone dosing. Participants received a breakfast that included a choice of foods with moderate to high fat content at the time of study drug administration. Cigarette smoking was permitted 6 h after the patient initiated the pharmacokinetic study. After completion of the initial pharmacokinetic study, participants completed an inventory of medicationrelated adverse events and were administered the first dose of L/R (400 mg/100 mg) or ritonavir (100 mg) at the research clinic (female subjects underwent another urine pregnancy test Lopinavir-Ritonavir and Opiate Withdrawal CID 2003:37 (15 August) 477

3 before the dose of antiretroviral medication was administered) and were monitored for 2 h for the occurrence of immediate adverse events or side effects. L/R was administered at standard clinical dosages (400 mg/100 mg twice per day) and ritonavir was administered at dosages equivalent to those received in L/ R therapy (i.e., 100 mg twice per day) over a 7-day period, the amount of time calculated to produce 75% of maximum hepatic microsomal induction. Each morning dose of L/R or ritonavir (with the exception of the Sunday dose, because the clinic was closed) was administered in the drug abuse treatment program by nursing staff, who also administered the usual methadone dose and conducted a brief evaluation to determine whether participants were experiencing adverse events. Evening and the 2 Sunday doses were provided in unit (single, individual) doses. Participants were instructed to take the medication within the hour before (for doses received at the clinic) or at the time of the morning or evening meal. Subjects returned to the research clinic on day 3 or 4 of the 7-day treatment period for a clinical assessment, recording of any adverse events, urine toxicology screening, and adherence query. After completion of 7 days of administration of methadone and either L/R or ritonavir, participants were readmitted to the hospital. The second pharmacokinetic study was completed as for study day 1. Upon completion of the second pharmacokinetic study, participants underwent a physical examination and a brief clinical assessment, were queried regarding adverse events, and had a blood sample obtained for determination of blood urea nitrogen and creatinine levels and for performance of liver function tests. Participants were then discharged to continue their usual activities and substanceabuse treatment programs. Drug sampling. On pharmacokinetic study days, blood samples were obtained by venipuncture and drawn into Vacutainer tubes (Becton Dickinson) immediately before dosing and at the predetermined time points listed above. Blood samples were allowed to clot before centrifugation. All blood samples were centrifuged at 3000 g for 10 min. Serum was then transferred into prelabeled tubes and stored at 70 C until the time of analysis. Methadone assay. Serum concentrations of methadone were determined by high-performance liquid chromatography (HPLC). After addition of 200 ng of an internal standard (propoxyphene), 1 ml of serum was alkalinized and extracted with hexane containing 2% isobutanol. The solvent layer was backextracted with dilute phosphoric acid and submitted to reversedphase HPLC on a C-18 column (5-mm Spherisorb ODS-2 column [Alltech]) with UV detection at 210 nm. Between-day coefficients of variation were 7.5% at 40 mg/l and 5.1% at 400 mg/l. Pharmacokinetic analysis. The pharmacokinetic parameters for methadone were evaluated for each subject after oral ingestion of methadone only and again after ingestion of the same dose of methadone after 7 days of treatment with either L/R or ritonavir (100 mg twice daily). Standard noncompartmental methods were used to estimate methadone pharmacokinetic parameters [17]. These parameters included the area under the interdose serum concentration-time curve (AUC 0 24h ), oral clearance, and observed minimum serum concentration (C min ), maximum serum concentration (C max ), and time of C max (T max ). The interdose sampling period, relative to methadone s reported half-life, was too short to allow accurate estimation of elimination half-life and volume of distribution. Statistical analysis. Student s paired t test was used to test the significance of the within-subject differences in pharmacokinetic parameters for methadone versus methadone plus L/ R or methadone plus ritonavir and for ratings of severity of opiate withdrawal. The Wilcoxon signed rank test was used for the within-subject comparison of the values of T max. Comparisons of subject characteristics were made by single-factor analysis of variance. RESULTS Study participants. All participants completed the study. Table 1 summarizes the characteristics of study participants. Participants in each group did not differ by age or weight. Female subjects constituted 33% of the L/R group and 53% of the ritonavir group. In the L/R group, 40% of the participants met diagnostic criteria for a cocaine use disorder, and 47% of the participants in the ritonavir sample met criteria for a cocaine use disorder. Urine toxicology screening showed that 6 (40%) of 15 patients in the L/R group and 8 (53%) of 15 in the ritonavir group to be positive for cocaine metabolite; 1 (7%) of 15 patients in each group tested positive for cannabis, and 1 L/R recipient (7%) tested positive for benzodiazepines. No participants met diagnostic criteria for physiological dependence on alcohol or had an acute mental disorder diagnosed. Evidence of hepatitis C virus infection was present in 26% of the L/R recipients and in 20% of the ritonavir recipients, but indices of hepatic function (alanine and aspartate aminotransferase levels) remained within normal range throughout the study period. L/R (400 mg/100 mg twice per day) substantially altered the pharmacokinetics of methadone, with a 26% decrease in AUC 0 24 (AUC 0 24 for methadone alone vs. methadone plus L/ R, 9199 vs mg h/l; P!.001). The oral clearance of methadone proportionately increased in conjunction with L/R treatment (before vs. after L/R therapy, 9.97 vs L/h; P!.001; figure 1; table 2). C max and C min each decreased by 28% ( P!.001 and P p.002, respectively) after L/R treatment, but no effect was observed on T max. The effect of L/R on meth- 478 CID 2003:37 (15 August) McCance-Katz et al.

4 Figure 1. Effect of lopinavir-ritonavir therapy on methadone concentration in a cohort of methadone-maintained subjects. adone exposure was associated with a significant increase in the number of opiate withdrawal symptoms in methadonemaintained participants ( P p.013 ; figure 2). OOWS scores before L/R administration were 0 2; after completion of 7 days of L/R, OOWS scores were 0 5. Four (27%) of 15 participants had opiate-withdrawal severity increases of 2 on the OOWS, and all of these individuals showed subtherapeutic trough methadone levels (!200 mg/l) after administration of L/R. Furthermore, these individuals showed the greatest decrement in methadone trough level after L/R administration (42% decrease in trough concentration of methadone), compared with those who had modest changes in opiate-withdrawal symptoms (i.e., an increase in the OOWS of 1; 32% decrease in trough concentration of methadone) and those who had no increase in opiate-withdrawal symptoms (12% decrease in methadone trough concentration). Ritonavir administration had no significant effect on methadone pharmacokinetics (table 2; figure 3). Methadone concentrations after 7 days of ritonavir administration were modestly but nonsignificantly increased (figure 3), and ritonavir did not produce increases in opiate-withdrawal symptoms (figure 2). DISCUSSION Administration of L/R in standard clinical dosages used in the treatment of HIV disease was associated with reduced methadone concentrations that were presumably a consequence of induction of methadone metabolism and, in some methadonemaintained participants, production of opiate-withdrawal symptoms after 7 days of treatment. Because L/R is a combination protease inhibitor, a second set of methadone pharmacokinetic data was obtained to determine whether ritonavir or lopinavir or a combination of their actions explained the outcome. Ritonavir was found to modestly and nonsignificantly increase methadone concentrations, indicating that lopinavir is most likely responsible for our observations. The pharmacokinetic findings in this study are consistent with those of Bertz et al. [10] and Clark et al. [11], who reported that L/R decreased methadone exposure. However, the pharmacodynamic findings in the present study are in contrast to those of Clark et al. [11], who reported that opiate-withdrawal symptoms were not observed in their sample. In the current study, some participants (27%) developed opiate-withdrawal symptoms expressed as scores on the OOWS as high as 5. It should be noted that this scale was developed as a clinical tool to measure the severity of opiate abstinence related symptoms during the withdrawal process and to standardize the administration of medications to relieve withdrawal symptoms. A score of 3 on this scale indicates the need to administer medication to alleviate withdrawal symptoms. Thus, the findings in this study have direct clinical relevance. One explanation for the difference in findings for this study and that by Clark et al. [11] is that the mean trough methadone level ( SD) after L/R administration for the participants in the current study who developed opiate-withdrawal symptoms was mg/l, well less than the minimum therapeutic trough methadone concentration of 200 mg/l. It has been suggested that the clinical consequences of the decrease in methadone levels seen in conjunction with protease inhibitor use may be mitigated by changes in protein binding and disparate effects on the 2 enantiomers of methadone [18]. However, in this instance, the significant increase in withdrawal symptoms was consistent with the observed reductions in the AUC and trough concentrations of methadone after L/R therapy. The findings of this study are also in contrast to those of Hsu et al. [12], who reported that ritonavir treatment (500 mg q12h) significantly decreased the methadone AUC. The latter study involved healthy volunteers who did not receive methadone as part of their usual clinical care. As such, much lower doses of methadone (20 mg of methadone in the control study and only 5 mg of methadone on day 14 of ritonavir treatment) were administered than those generally used in methadone maintenance treatment. The findings of the present study underscore the need to conduct drug interaction studies that involve doses likely to be used in the population of interest. In vitro experiments have indicated that methadone is principally metabolized via N-demethylation by CYP450 3A enzymes, but several lines of evidence suggest that CYP450 2D6 may also play a significant role in methadone clearance in vivo, possibly involving a metabolic pathway other than N-demethylation [19]. In vitro, where only in- Lopinavir-Ritonavir and Opiate Withdrawal CID 2003:37 (15 August) 479

5 Table 2. Effects of use of lopinavir-ritonavir or ritonavir alone on methadone disposition. PI administered, methadone disposition parameter Before therapy After therapy P Lopinavir-ritonavir AUC 0 24, mg-h/l 9199 ( ) 6819 ( )!.001 Cl/F, L/h 9.97 ( ) ( )!.001 C max, mg/l 567 ( ) 410 ( )!.001 T max, h 1.8 ( ) 3.3 ( ) NS C min, mg/l 291 ( ) 210 ( ).002 Ritonavir alone AUC 0 24, mg-h/l 8485 ( ) 8808 ( ) NS Cl/F, L/h ( ) ( ) NS C max, mg/l 507 ( ) 525 ( ) NS T max, h 2.2 ( ) 2.2 ( ) NS C min, mg/l 269 ( ) 278 ( ) NS NOTE. Data are mean (95% CI). AUC 0 24h, area under the interdose serum concentrationtime curve; Cl/F, oral clearance; C max, maximum serum concentration; C min, observed minimum serum concentration; NS, not significant; PI, protease inhibitor; T max, time of C max. hibitory effects are readily demonstrated, ritonavir is a powerful inhibitor of CYP 3A and also inhibits CYP 2D6. Its lack of a net effect when administered alone in vivo suggests offsetting inhibition and induction of methadone metabolism. Lopinavir also inhibits both CYP 3A and CYP 2D6 in vitro, but less powerfully than ritonavir [20]. Thus, the overall effect of L/R to decrease methadone exposure could be attributable to net induction of metabolic clearance, involving either or both CP450 3A and CYP450 2D6. Induction of other enzymes, such as intestinal glycoprotein P-450, could also contribute. The consequences of unsuspected and clinically significant drug interactions between methadone and antiretroviral agents are of concern. The production of opiate-withdrawal syndromes or other drug toxicities can lead to nonadherence to antiretroviral regimens, with the risks of HIV treatment failure and selection of medication-resistant HIV. Increased abuse of other illicit substances or alcohol in an attempt to alleviate the discomfort resulting from withdrawal symptoms may also occur and represent additional risks to patients. These risks include the development of additional substance use related disorders and toxicities related to substance abuse, as well as the risk that methadone maintenance may be terminated because of noncompliance with substance abuse treatment program policies. Figure 2. Effect of therapy with lopinavir-ritonavir or ritonavir on ratings of opiate-withdrawal severity in a cohort of methadone-maintained subjects 480 CID 2003:37 (15 August) McCance-Katz et al.

6 Figure 3. Effect of ritonavir on methadone concentration in a cohort of methadone-maintained subjects Effective methadone maintenance treatment in persons with HIV disease requires that health care providers be aware of important drug interactions between antiretroviral medications and opioids. However, reports of reduced methadone exposure do not necessarily correlate with the need for increased doses of methadone [2, 18, 21]. The administration of antiretrovirals shown to induce methadone metabolism will require careful clinical assessment. In an analysis to determine whether there were any demographic variables that predicted subsequent opiate withdrawal upon L/R treatment, several findings may be of potential clinical concern. The study participants who developed opiate-withdrawal syndromes were receiving lower methadone doses (mean dose, 70 vs. 91 mg per day; P p.089 [1- tailed]), had higher body weight (mean weight, vs kg; P p.064 [1-tailed]), and were older (mean age, 51.6 years vs years; P p.077 [2-tailed]). In addition, it is possible that moderate alcohol use (mean number of drinks per week, 4.8 for those who developed opiate withdrawal vs. 1.2 for those who did not; P p.089 [1-tailed]) may contribute to the likelihood of experiencing an opiate-withdrawal syndrome during L/R treatment. Although it is reasonable to assume that weight and methadone dose might be predictive of the possibility of opiate withdrawal in those receiving treatment with a drug that induces methadone metabolism, the findings related to age and alcohol use should be examined in studies with larger sample sizes to replicate this finding. On the basis of the results of this study, a course of action that would assist in the treatment of patients needing both methadone maintenance and HIV care would be to determine both a trough methadone concentration and opiate-withdrawal symptom severity score (such as the clinician-administered OOWS) before the initiation of antiretroviral medications that may induce methadone metabolism and then to observe these patients closely for evidence of development of opiate-withdrawal symptoms in the initial weeks after addition of the antiretroviral agent to the treatment regimen. Upon observation of opiate-withdrawal symptoms, a trough methadone level should be determined. For those showing evidence of a significant decrease in trough methadone concentration (either a level of!200 mg/l or a decrease of 40% in the trough value), rapid increases in the methadone dose should be immediately instituted to restabilize the individual and to prevent the development of a full opiate-withdrawal syndrome. The findings of this study speak to the importance of drug interactions in the clinical care of HIV disease and substance disorders. Our research group is pursuing the elucidation of important drug interactions between frequently prescribed antiretroviral agents and other substance abuse pharmacotherapies with the goal of enhancing the clinical care and outcomes of drug abusers with HIV disease. Acknowledgments We thank Corinne Rogers, Julie Sarlo, Peter Cottone, Yolanda Smith, and Lauren Hellew, for expert technical assistance; and the participants in this study, for their willingness to assist with improving the clinical care of those with HIV disease. References 1. McCance-Katz EF, Jatlow P, Rainey P, Friedland G. Methadone effects on zidovudine (AZT) disposition. J Acqir Immune Defic Syndr Hum Retrovirol 1998; 18: Hsyu PH, Lillibridge JH, Maroldo L, et al. Pharmacokinetic (PK) and pharmacodynamic (PD) interactions between nelfinavir and methadone [abstract 87]. In: Program and abstracts of the 7th Conference on Retroviruses and Opportunistic Infections (San Francisco). Alex- Lopinavir-Ritonavir and Opiate Withdrawal CID 2003:37 (15 August) 481

7 andria, VA: Foundation for Retrovirology and Human Health, 2000: Altice FL, Friedland GH, Cooney EL. Nevirapine-induced opiate withdrawal among injection drug users with HIV infection receiving methadone. AIDS 1999; 13: Clarke SM, Mulcahy FM, Tija J, et al. The pharmacokinetics of methadone in HIV-positive patients receiving the non-nucleoside reverse transcriptase inhibitor efavirenz. Br J Clin Pharmacol 2001; 51: Rainey PM, Friedland G, McCance-Katz EF, et al. Interaction of methadone with didanosine (ddi) and stavudine (d4t). J Acquir Immune Defic Syndr 2000; 24: Lopinavir/ritonavir: a protease-inhibitor combination. Med Lett Drugs Ther 2001:43: Lea AP, Faulds D. Ritonavir. Drugs 1996; 52: Iribarne C, Berthou F, Baird S, et al. Involvement of cytochrome P450 3A4 enzyme in the N-demethylation of methadone in human liver microsomes. Chem Res Toxicol 1996; 9: Foster DJ, Somogyi AA, Bochner F. Methadone N-demethylation in human liver microsomes: lack of stereoselectivity and involvement of CYP3A4. Br J Clin Pharmacol 1999; 47: Bertz R, Hsu A, Lam W, et al. Pharmacokinetic interactions between Kaletra and other non-hiv drugs [abstract 438]. In: Program and abstracts of the 5th International Congress on Drug Therapy in HIV Infection (Glasgow, Scotland). Macclesfield, UK: Gardiner-Caldwell Group, Clarke S, Mulcahy F, Bergin C, et al. Absence of opioid withdrawal symptoms in patients receiving methadone and the protease inhibitor lopinavir/ritonavir. Clin Infect Dis 2002; 34: Hsu A, Granneman GR, Carothers L, Dennis S, Chiu Y-L, Valdes J, Sun E. Ritonavir does not increase methadone exposure in healthy volunteers [abstract 342]. In: Program and abstracts of the 5th Conference on Retroviruses and Opportunistic Infections (Chicago). Alexandria, VA: Foundation for Retrovirology and Human Health, Geletko SM, Erickson AD. Decreased methadone effect after ritonavir initiation. Pharmacotherapy 2000; 20: American Psychiatric Association. Diagnostic and statistical manual of mental disorders, fourth edition, text revision. Washington, DC: American Psychiatric Association, Sheehan DV, Lecrubier Y, Sheehan HK, et al. The mini international neuropsychiatric interview (M.I.N.I): the development and validation of a structured diagnostic interview for DSM-IV and ICD-10. J Clin Psychiatry 1998; 59: Handlesman L, Cochrane KJ, Aronson MJ, Ness R, Rubinstein KJ, Kanof PD. Two new rating scales for opiate withdrawal. Am J Drug Alcohol Abuse 1987; 13: WinNonlin 1.0 TM. Apex, NC: Scientific Consulting, Gerber JG, Rosenkranz S, Segal Y, et al. Effect of ritonavir/saquinavir on stereoselective pharmacokinetics of methadone: results of AIDS Clinical Trials Group (ACTG) 401. J Acquir Immune Defic Syndr 2001; 27: Eap CB, Buclin T, Baumann P. Interindividual variability of the clinical pharmacokinetics of methadone: implications for the treatment of opioid dependence. Clin Pharmacokinet 2002; 41: Kumar GN, Dykstra J, Roberts EM, et al. Potent inhibition of the cytochrome P-450 3A mediated human liver microsomal metabolism of a novel HIV protease inhibitor by ritonavir: a positive drug-drug interaction. Drug Metab Dispos 1999; 27: McCance-Katz EF, Rainey P, Smith P, et al. Drug interactions between opioids and antiretroviral medications: interaction between methadone, LAAM, and nelfinavir. Am J Addictions (in press). 482 CID 2003:37 (15 August) McCance-Katz et al.