This column is the second in a two-part series updating two earlier columns in which psychiatric drugs were classified according to their pharmacodynamic profiles.1,2 The purpose of this update is to help clinicians stay abreast of the pharmacology of new drugs with central nervous system effects and their potential to cause clinically relevant drug-drug interactions (DDIs).*
Of the 51 new drugs that entered the U.S. market between 2003 and 2005 (see Table 1 in Part I of this series4), only 4 drugs are formally classified as psychiatric medications: acamprosate (Campral) for alcohol dependence; duloxetine (Cymbalta) for major depression; eszopiclone (Lunesta) for insomnia; and ramelteon (Rozerem) for sleep onset insomnia. Two of the 51 drugs, memantine (Namenda) for the symptomatic treatment of moderate to severe Alzheimer's disease and apomorphine hydrochloride (Apokyn) for the symptomatic treatment of Parkinson's disease, while classified as neurological drugs, are likely to be prescribed by psychiatrists in substantial quantities or taken by patients being seen by psychiatrists. Due to the size of the therapeutic areas they target (clinical depression and Alzheimer's disease), duloxetine and memantine are arguably the most important of the 6 agents discussed in these columns in terms of the number of patients who will receive them and the diverse group of practitioners who are likely to prescribe them.
In Part I of this two-part series, I described the properties of acamprosate and memantine and discussed their potential to be either the perpetrator or victim of a clinically relevant DDI. In this column, I focus on apomorphine, duloxetine, eszopiclone, and ramelteon.
Apomorphine (Apokyn) is approved in a subcutaneous only formulation for the symptomatic treatment of recurring episodes of hypomotility (i.e., "end of dose wearing off" or "on/off" episodes) associated with advanced Parkinson's disease.5-9 It is a non-ergoline dopamine (D) agonist with its highest binding affinity for the D4 receptor (Ki = 4.4 nM), moderate affinity for D2, D3, and D5 receptors (Ki = 35, 26, and 15 nM, respectively) and adrenergic alpha 1D, 2B, and 2C receptors (Ki = 65, 66, and 36 nM, respectively), and low affinity for D1 (Ki = 370) and 5-hydroxytrptyramine (5-HT = serotonin) 1A, 2A, 2B, and 2C receptors (Ki = 120, 120, 130, and 100 nM, respectively), and essentially no affinity for adrenergic beta 1 or 2 or histamine-1 receptors (Ki > 10,000 nM).5,10,11 While the precise mechanism of action of apomorphine in Parkinson's disease is unknown, it is most likely due to stimulation of post-synaptic D2 receptors in the caudate-putamen.
Given its agonism at D2 receptors, apomorphine would have at least additive effects when combined with other D2 agonists, and it potentiates and extends the duration of activity of L-dopa. For the same reasons, apomorphine and D2 antagonists (e.g., virtually all antipsychotic medications) antagonize each other's motor and psychotropic effects.1,2
Dopamine agonists, including apomorphine, can cause orthostatic hypotension, especially during dose escalation.12-15 These effects may be increased by alcohol, anti-hypertensive medications of a wide variety of pharmacologic classes, and vasodilators, especially nitrates. For reasons that are not fully understood, the hypotensive effects of apomorphine are potentiated by ondansetron and probably other 5-HT3 antagonists (e.g., alosetron, dolasetron, granisetron, and palonosetron) and their concomitant administration is contraindicated.
Apomorphine causes dose-dependent increases in the QTc interval; however, these effects are minimal (i.e., mean of 1 msec) at the maximum recommended dose of 6 mg.5,16 The risk of torsade de pointes arrhythmias is clearest for drugs that produce mean increases of 20 msec or greater. Nevertheless, DDI studies examining the concomitant administration of apomorphine and other drugs capable of prolonging the QTc interval have not been done; hence, caution is advised when prescribing apomorphine to patients receiving such drugs.
In terms of pharmacokinetic DDIs, the metabolism of apomorphine is not well characterized, making definitive statements difficult.5,17-20 Nevertheless, in vitro studies indicate that CYP enzymes play a minor role in its metabolism and hence apomorphine is unlikely to be a victim of CYP-enzyme-mediated DDIs. The fact that it is only available in a subcutaneous dosing formulation also reduces the likelihood of apomorphine being the victim of such DDIs. Apomorphine is also not metabolized by cathechol-O-methyltransferase (COMT) and hence is unlikely to be affected by the COMT inhibitors that are also used to treat Parkinson's disease.21-23 Based on in vitro studies, apomorphine is also unlikely to be a perpetrator of CYP-enzyme-mediated DDIs in terms of either induction or inhibition.
Duloxetine (Cymbalta) is approved for the treatment of clinical depression.24 Like venlafaxine, duloxetine belongs to the class of serotonin-norepinephrine reuptake inhibitor (SNRI) antidepressants. Readers are referred to an earlier column for an in-depth discussion of the preclinical and clinical pharmacology of duloxetine.25 In this column, I focus only on the pharmacodynamic and pharmacokinetic profile of this relatively new antidepressant.
Pharmacodynamically, duloxetine, being an SNRI, can interact with other drugs via its ability to potentiate the effects of both norepinephrine and serotonin by inhibiting their neuronal uptake pump or transporter protein.25
As discussed in several previous columns, an important-although unintended-feature that distinguishes antidepressants from one another is their pharmacokinetic profile, specifically, whether they inhibit certain cytochrome P450 (CYP) enzymes and can therefore have clinically meaningful DDIs with other drugs whose clearance is dependent on biotransformation by one or more of these human drug metabolizing enzymes.26,27Table 1 summarizes the differential effects of currently available antidepressants on cytochrome P450 (CYP) enzymes.
In terms of pharmacokinetic DDIs, duloxetine is a substrate of CYP 2D6 and CYP 1A2. Paroxetine, a known CYP 2D6 inhibitor, increases the concentration of duloxetine by about 60%.28 Similar effects would be expected to occur with other potent CYP 2D6 inhibitors (e.g., fluoxetine, quinidine). Coadministration of duloxetine with fluvoxamine, a strong CYP 1A2 inhibitor, increases the peak plasma concentration (Cmax) of duloxetine 2.5-fold and the area under the plasma concentration versus time curve (AUC) 5-fold.24 Combinations of duloxetine and other CYP 1A2 inhibitors (e.g., fluoroquinolones) should therefore be avoided.24 At a dose of 120 mg/day, duloxetine is a moderately potent inhibitor of CYP 2D6 and a mildly potent inhibitor at a dose of 60 mg/day, using the definition outlined in Table 1.28,29 Duloxetine at 120 mg/day increases the Cmax and AUC of desipramine, a drug almost exclusively metabolized by CYP 2D6, by 1.7- and 2.9-fold, respectively.28 Duloxetine at 60 mg/day increases the Cmax and AUC of metropolol, another drug almost exclusively metabolized by CYP 2D6, by 1.0- and 1.8-fold, respectively.29 These results demonstrate the usual dose-dependent nature of the inhibition of CYP enzyme function. Given these results, drugs that are principally metabolized by CYP 2D6 and have potentially clinically significant dose-dependent adverse effects should be co-administered cautiously with duloxetine. Such drugs include tricyclic antidepressants, phenothiazines, and type 1C antiarrhythmics such as propafenone and flecainide.24
Parenthetically, as can readily be seen in Table 1, the antidepressants with the greatest effects on CYP enzymes are fluoxetine and fluvoxamine. As discussed in earlier columns, these two agents would likely not be approved today for this reason and should be used cautiously, if at all, in patients who are taking more than one medication or who may be prescribed more than one medication.30 In other words, use of fluoxetine and fluvoxamine should be severely curtailed because they pose a significant risk of being a perpetrator of a clinically significant DDI. Nefazodone and paroxetine do not affect as many CYP enzymes as fluoxetine and fluvoxamine but nevertheless, at usual clinical doses, they produce substantial inhibition of one CYP enzyme (CYP 2D6 in the case of paroxetine and CYP 3A3/4 in the case of nefazodone). For this reason, restricted use of these two antidepressants is also prudent.
Eszopiclone (Lunesta) is approved as a nonbenzodiazepine hypnotic.31 As the name implies, eszopiclone is the S-entantiomer of the racemic drug, zopiclone. As with virtually all drugs that affect the central nervous system (CNS), the mechanism of action that accounts for the efficacy of eszopiclone is not known but is believed to be its interaction with GABA receptor complexes at binding domains close to or allosterically coupled to the benzodiazepine binding sites. While chemically unrelated to benzodiazepines, barbiturates, or other drug with known hypnotic properties, co-administration of these drugs can result in additive or enhanced sedative effects. When combining eszopiclone with other sedating agents, clinicians should keep this caution in mind.
The metabolism of eszopiclone is mediated by CYP 3A4 and CYP 2E1. Co-administration of eszopiclone and ketoconazole, a substantial CYP 3A4 inhibitor, results in a 2.2-fold increase in exposure to eszopiclone.31 The dose of eszopiclone should therefore be reduced when it is administered with ketoconazole or other substantial CYP 3A4 inhibitors, such as itraconazole, clarithromycin, and ritonavir.32 Racemic zopiclone exposure is decreased by 80% with concomitant use of rifampin, a substantial CYP 3A4 inducer. A similar effect would be expected with eszopiclone, potentially reducing its efficacy.31,32
Ramelteon (Rozerem) is approved for the treatment of sleep-onset insomnia.33 (For a review of clinical trial data on ramelteon, readers are referred to an article in the July 2006 issue of this journal.34)
Ramelteon has a chiral center but the marketed product is the S-enantiomer. The maintenance of the normal circadian sleep-wake cycle is believed to be due to the action of melatonin (MT) on MT1 and MT2 receptors. Ramelteon acts like melatonin at these same two receptors and that is believed to underlie its efficacy in sleep-onset insomnia. In contrast to its high affinity for MT1 and MT2 receptors, ramelteon has low affinity for MT3 receptors and is thus considered a selective MT1 and MT2 agonist. Ramelteon also has no meaningful binding affinity for the GABA receptor complex or any other known neuropeptide, cytokine, opiate, biogenic amine, or acetylcholine receptors.33
Ramelteon is principally a substrate of CYP 1A2. The CYP 2C and CYP 3A4 enzymes are also involved in its metabolism but to a substantially smaller degree. Fluvoxamine, a strong CYP 1A2 inhibitor, increases the AUC and Cmax of ramelteon approximately 190-fold and 70-fold, respectively. Other CYP 1A2 inhibitors (e.g., fluoroquinolones) would be expected to have similar effects.35 For this reason, combined use of ramelton with substantial CYP 1A2 inhibitors should be avoided. Ketoconazole, a strong CYP 3A4 inhibitor, increases the plasma concentrations (i.e., AUC0-inf and Cmax) of ramelteon by approximately 84% and 36%, respectively. Ramelteon should be administered with caution in subjects taking other CYP 3A4 inhibitors (e.g., itraconazole, erythromycin, clarithromycin, ritonavir).33 Total and peak systemic exposure (AUC0-inf and Cmax) of ramelteon increases by approximately 150% when it is administered with fluconazole, a CYP 2C9 inhibitor. Ramelteon should be administered with caution in subjects taking other CYP 2C9 inhibitors, such as fluconazole and fluvoxamine.33 Administration of rifampin, a CYP 3A4 and CYP 2C9 inducer, results in a mean decrease of approximately 80% (40%-90%) in total exposure to ramelteon (both AUC and Cmax).35 For this reason, the efficacy of ramelteon may be reduced when it is used in combination with CYP 1A2, CYP 3A4, or CYP 2C9 inducers (e.g., phenobarbital, carbamazepine, phenytoin).
Other Agents of Interest
While atazanavir (Reyataz), and fosamprenavir (Lexiva) are drugs used in combination with other antiretrovirals to treat human immunodeficiency virus (HIV) and are not psychiatric medications, they deserve mention here because they are substantial CYP 3A inhibitors and, thus, can be the perpetrators of DDIs with a wide variety of psychiatric medications that are principally dependent on this CYP enzyme for their clearance.36,37 For example, the concomitant use of atazanavir or fos-amprenavir with drugs dependent on CYP 3A for their clearance (e.g., midazolam, triazolam, pimozide) is contraindicated.
The antibiotic, telithromycin (Ketek), is another substantial CYP 3A inhibitor38 and should therefore be used cautiously with other CYP 3A substrates (e.g., midazolam, triazolam). Telithromycin may increase pimozide plasma levels by inhibition of CYP 3A4 pathways. In addition, telithromycin, like pimozide, can pro-long the QTc interval. Hence, there is the potential for both a pharmacodynamic and pharmacokinetic DDI that could result in a fatal arrhythmia.
The goal of this two-part series of columns was to update an earlier series on the classification of neuropsychiatric medications as a means of helping clinicians avoid unintended and deleterious DDIs.1,2 This column presents information on neuropsychiatric drugs approved between 2003 and 2005.
1. Preskorn SH. Classification of neuropsychiatric medications by principal mechanism of action: A meaningful way to anticipate pharmacodynamically mediated drug interactions (Part I). J Psychiatr Pract 2003;9:376-84.
2. Preskorn SH. Classification of neuropsychiatric medications by principal mechanism of action: A meaningful way to anticipate pharmacodynamically mediated drug interactions (Part II). J Psychiatr Pract 2004;10:177-81.
3. Preskorn SH, Flockhart D. 2006 guide to psychiatric drug interactions. Primary Psychiatry 2006;13:35-64.
4. Preskorn SH, Borges S, Flockhart D. Clinically relevant pharmacology of neuropsychiatric drugs approved over the last three years: Part I. J Psychiatr Pract 2006;12:244-9.
5. Apokyn package insert. In: Physicians' desk reference. Montvale, NJ: Thomson PDR; 2006:2132-6 (www.apokyn.com/pdf
, accessed July 1, 2006).
6. Nyholm D. Pharmacokinetic optimisation in the treatment of Parkinson's disease: An update. Clin Pharmacokinet 2006;45:109-36.
7. Stacy M. Apomorphine: North American clinical experience. Neurology 2004;62(Suppl 4):S18-21.
8. Factor SA. Literature review: Intermittent subcutaneous apomorphine therapy in Parkinson's disease. Neurology 2004; 62(Suppl 4):S12-7.
9. van Laar T. Levodopa-induced response fluctuations in patients with Parkinson's disease: Strategies for management. CNS Drugs 2003;17:475-89.
10. Morishita H, Shibata K, Sakata N, et al. A new approach to finding specific dopamine D4
receptor agonists. Eur J Pharmacol 2005;516:145-50.
11. Hirota S, Kawashima N, Chaki S, et al. Neuropharmacological profile of an atypical antipsychotic, NRA0562. CNS Drug Rev 2003;9:375-88.
12. Obering CD, Chen JJ, Swope DM. Update on apomorphine for the rapid treatment of hypomobility ("off") episodes in Parkinson's disease. Pharmacotherapy 2006;26:840-52.
13. Bonuccelli U, Pavese N. Dopamine agonists in the treatment of Parkinson's disease. Expert Rev Neurother 2006;6:81-9.
14. Tyne HL, Parsons J, Sinnott A, et al. A 10 year retrospective audit of long-term apomorphine use in Parkinson's disease. J Neurol 2004;251:1370-4.
15. Montorsi F. Tolerability and safety of apomorphine SL (Ixense (TM) Int J Impot Res 2003;15(Suppl 2):S7-9.
16. Hurst RS, Higdon NR, Lawson JA, et al. Dopamine receptor agonists differ in their actions on cardiac ion channels. Eur J Pharmacol 2003;482:31-7.
17. LeWitt PA. Subcutaneously administered apomorphine: Pharmacokinetics and metabolism. Neurology 2004;62(Suppl 4):S8-11.
18. Argiolas A, Hedlund H The pharmacology and clinical pharmacokinetics of apomorphine SL. BJU Int 2001;88(Suppl 3): 18-21.
19. Vietri M, Vaglini F, Cantini R, et al. Quercetin inhibits the sulfation of r(−)-apomorphine in human brain. Int J Clin Pharmacol Ther 2003;41:30-5.
20. El-Bacha RS, Leclerc S, Netter P, et al. Glucuronidation of apomorphine. Life Sci 2000;67:1735-45.
21. Widnell KL, Comella C. Role of COMT inhibitors and dopamine agonists in the treatment of motor fluctuations. Mov Disord 2005;20(Suppl 11):S30-7.
22. Zijlmans JC, Debilly B, Rascol O, et al. Safety of entacapone and apomorphine coadministration in levodopa-treated Parkinson's disease patients: Pharmacokinetic and pharmacodynamic results of a multicenter, double-blind, placebo-controlled, cross-over study. Mov Disord 2004;19:1006-11.
23. Ondo WG, Hunter C, Vuong KD, et al. The pharmacokinetic and clinical effects of tolcapone on a single dose of sublingual apomorphine in Parkinson's disease. Parkinsonism Relat Disord 2000;6(4):237-40.
24. Cymbalta package insert. In: Physicians' desk reference. Montvale, NJ: Thomson PDR; 2006:1729-35. (www.cymbalta.com
, accessed July 1, 2006).
25. Preskorn SH. Duloxetine. J Psychiatr Pract 2004;10:375-85.
26. Preskorn SH. Reproducibility of the in vivo effect of the selective serotonin reuptake inhibitors on the in vivo function of cytochrome P450 2D6: An update (Part I). J Psychiatr Pract 2003;9:150-8.
27. Preskorn SH. Reproducibility of the in vivo effect of the selective serotonin reuptake inhibitors on the in vivo function of cytochrome P450 2D6: An update (Part II). J Psychiatr Pract 2003;9:228-36.
28. Skinner MH, Kuan HY, Pan A, et al. Duloxetine is both an inhibitor and a substrate of cytochrome P4502D6 in healthy volunteers. Clin Pharmacol Ther 2003;73:170-7.
29. Preskorn SH, Baker B, Klick-Davis A, et al. The effect of duloxetine, escitalopram, and sertraline on CYP 2D6 function. Clin Pharmacol Ther 2006;79:52 (abstract).
30. Preskorn SH. Modern drug development and the human genome project. Series of columns published in the Journal of Psychiatric Practice (can be accessed at www.preskorn.com/column4.html
31. Lunesta package insert. In: Physicians' desk reference. Montvale, NJ: Thomson PDR; 2006: 3139-43. (www.lunesta.com/PostedApprovedLabelingText.pdf
, accessed July 1, 2006).
32. Eszopiclone (Lunesta), a new hypnotic. Med Lett Drugs Ther 2005;47:17-9
33. Rozerem package insert. In: Physicians' desk reference. Montvale, NJ: Thomson PDR; 2006: 3228-31(www.rozerem.com/images/PI.pdf
, accessed July 1, 2006).
34. Bellon A. Searching for new options for treating insomnia: Are melatonin and ramelteon beneficial? J Psychiatr Pract 2006;12:229-43.
35. Ramelteon (Rozerem) for insomnia. Med Lett Drugs Ther 2005;47:89-91.
36. Reyataz package insert. In: Physicians' desk reference. Montvale, NJ: Thomson PDR; 2006: 948-57 (www.reyataz.com/managehiv/reyataz/dtc/index.jsp?BV_UseBVCookie=Yes
, accessed July 1, 2006.
37. Lexiva package insert. In: Physicians' desk reference. Montvale, NJ: Thomson PDR; 2006: 1473-9 (us.gsk.com/products/assets/us_lexiva.pdf
, accessed July 1, 2006.
38. Ketek package insert. In: Physicians' desk reference. Montvale, NJ: Thomson PDR; 2920-5 (http://products.sanofiaventis.us/ketek/ketek.pdf
, accessed July 1, 2006).
*Readers are referred to the first of those earlier columns for a table of psychiatric drugs organized according to mechanism of action(s)1 and to the second column for a table summarizing potential pharmacodynamic DDIs based on those mechanisms of action,2 as well as to Table 19 in Preskorn and Flockhart,3 for a summary of information on different drugs and cytochrome P450 (CYP) enzymes.