Commentary on Heinzerling et al. (2014): A growing methamphetaminedependence therapeutics graveyard

Methamphetamine remains one of the most widely usedillicit substances in the world. This highly addictivepsychostimulant has multiple sites of action in the brain,acting as a potent indirect agonist of dopamine,noradrenaline, and serotonin [1]. In addition to itsintense euphoric effects, the acute hyperdopaminergicaction of methamphetamine often results in heightenedrisk-taking behavior, severe aggression, and psychoticsymptoms. Chronic methamphetamine abuse results inneurotoxic effects leading to psychological problems,including depression, anxiety, social isolation, psychosis,and neurocognitive deficits [2]. Despite an urgent need,there are currently no widely accepted psychological orpharmacological treatments for methamphetaminedependence, and certainly no medications approved bymajor regulatory bodies [3].

In recent years, several indirect dopamine agonistsand partial agonists (e.g. aripiprazole, d-amphetamine,methylphenidate, and modafinil), opioid antagonists (e.g.naltrexone), and serotonin/norepinephrine re-uptakeinhibitors (e.g. ondansetron and mirtazapine) haveadvanced to early-stage clinical trials [4]. Despite positivepre-clinical findings, most of these clinical studies havefailed to provide any convincing results. At best, theyshow effectiveness in treating only specific subgroups ofpatients and highlight the necessity to collect thoroughphenotypic information during screening in order todetermine the patient groups most likely to respond topharmacotherapeutic intervention. Furthermore, of thesmall number of drugs that have shown promise in treat-ing methamphetamine dependence, many maintain sig-nificant abuse potential. Perhaps unsurprisingly, thesedrugs (e.g. d-amphetamine and methylphenidate) aremembers of the amphetamine-type stimulant family andact on similar, if not identical, neural pathways to meth-amphetamine itself [3]. Thus, the identification of effec-tive therapeutics with low abuse potential is of criticalimportance.

In this issue, Heinzerling et al. [5] outline results froma double-blind, randomized, placebo-controlled 12-weektreatment trial of bupropion for methamphetamine-dependent patients with less than daily use. This studyaimed to replicate findings from two previously publishedclinical trials that showed efficacy of bupropion in treat-ing a subgroup of patients with lower frequency of meth-amphetamine use at baseline [6,7]. Bupropion is apopular antidepressant agent with additional approvalfor aiding smoking cessation; it inhibits the re-uptake ofdopamine and noradrenaline, increases dopamine in thesynaptic cleft by blocking presynaptic dopamine trans-

porter activity, and also targets noradrenaline transport-ers and nicotinic receptors. Importantly, althoughbupropion shares some of the pharmacodynamic proper-ties of methamphetamine, it is far less potent andmaintains substantially lower abuse potential thanamphetamine-type stimulants [8]. It has been proposedthat bupropion may treat withdrawal symptoms and cog-nitive deficits associated with methamphetamine depend-ence by increasing depleted cytoplasmic monoamine (e.g.dopamine and noradrenaline) concentrations [9].

Unfortunately, Heinzerling et al.’s study provided noevidence for an effect of bupropion in treating metham-phetamine dependence as measured by end-of-treatmentabstinence. It is disappointing to consider that these find-ings may result in bupropion being added to the growingmethamphetamine-dependence therapeutics graveyardalongside once-promising drugs such as aripiprazole,topiramate, and baclofen. However, Heinzerling et al.’sstudy provides opportunities for understanding some ofthe obstacles faced by scientists when designing clinicaltreatment trials for this patient population, and raisesquestions about how best to proceed with drug develop-ment in this field. The recruitment of methamphetamineusers into tightly controlled clinical treatment trials pre-sents many difficulties, including high rates of infectiousdisease, significant cognitive impairment, co-occurringpsychiatric disorders, and polysubstance abuse amongpatients [10]. These challenges are evident in the sizeableexclusion rate in Heinzerling et al.’s study; that is, of the294 participants screened only 84 were randomized to atreatment group. An additional challenge in the assess-ment of novel therapeutics is the selection of effectiveoutcome measures. Heinzerling et al. used end-of-treatment abstinence—defined as having no drugmetabolites in urine at weeks 11 and 12, as well as havingmissed no more than one urine screen per week through-out the study—as their primary outcome measure.Unsurprisingly, a significant amount of data was dis-counted due to a large number of participants (56%) notshowing up consistently for urine sampling and the major-ity of patients being classified as medication non-adherent(68%). It is clear from this research that any possiblebenefits from bupropion would be impossible to determinewith confidence due to the extreme difficulties in main-taining patient compliance to what is essentially a verystraightforward clinical trial protocol.

There is no denying that methamphetamine usersrepresent one of the most difficult-to-treat patient groupsin all fields of medicine. We propose that the majority ofdrugs tested in early clinical studies, including

COMMENTARY

bs_bs_banner

© 2014 Society for the Study of Addiction Addiction, 109, 1887–1888

bupropion, have failed due to fundamental flaws in thetranslational approach taken for drug development inthis field. Specifically, progress in the identification ofnovel therapeutics is hindered by overly strict primaryoutcome measures, lack of effective biomarkers (e.g.genetic and protein-based) for predicting treatmentresponse, and overly selective drug targets for a disorderwith a somewhat ‘dirty’ pharmacotherapeutic targetprofile (see [11] for an elegant review from Shoptaw’slaboratory). Future studies would benefit from selectingless strict primary outcome measures aimed at determin-ing the effects of drug therapies to improve associatedpsychological problems, reduce the number of heavyusing days (as opposed to complete abstinence), andimprove surrogate measures, including biomarkers andphysical health outcomes. That said, we anxiously wait,with bated breath, as emerging therapeutics with noveltargets such as methamphetamine vaccines, oxytocinagonists, and cannabinoid antagonists make their waytowards clinical trials.

Declaration of interests

None.

Keywords Biomarkers, bupropion, clinical trials,methamphetamine, outcome measures, translationalresearch.

DEAN S. CARSON1 & EMILY R. TAYLOR2

1Department of Psychiatry and Behavioral Sciences,

Stanford University School of Medicine, Stanford, CA, USA2Department of Neurobiology, Stanford University School of

Medicine, Stanford, CA, USA.

E-mail: dcarson@stanford.edu

References

1. Carson D. S., Guastella A. J., Taylor E. R., McGregor I. S. Abrief history of oxytocin and its role in modulatingpsychostimulant effects. J Psychopharmacol 2013; 27: 231–47.

2. Meredith C. W., Jaffe C., Ang-Lee K., Saxon A. J. Implicationsof chronic methamphetamine use: a literature review. HarvRev Psychiatry 2005; 13: 141–54.

3. Carson D. S., Hermens D. F., Guastella A. J., Bosanquet D. P.,McGregor I. S. Neurocognitive Effects of Amphetamine-typeStimulants and Implications for Intervention. Perspectives onAmphetamine-type Stimulants. Melbourne, Australia: IPCommunications; 2012.

4. Karila L., Weinstein A., Aubin H.-J., Benyamina A.,Reynaud M., Batki S. L. Pharmacological approaches tomethamphetamine dependence: a focused review. Br J ClinPharmacol 2010; 69: 578–92.

5. Heinzerling K. G., Swanson A. N., Hall T. M., Ba Y. Y., Wu Y.,Shoptaw S. J. Randomized, placebo-controlled trial ofbupropion in methamphetamine-dependent participantswith less than daily methamphetamine use. Addiction 2014;doi: 10.1111/add.12636.

6. Shoptaw S., Heinzerling K. G., Rotheram-Fuller E., StewardT., Wang J., Swanson A.-N. et al. Randomized, placebo-controlled trial of bupropion for the treatment of metham-phetamine dependence. Drug Alcohol Depend 2008; 96:222–32.

7. Elkashef A. M., Rawson R. A., Anderson A. L., Li S.-H.,Holmes T., Smith E. V. et al. Bupropion for the treatment ofmethamphetamine dependence. Neuropsychopharmacology2008; 33: 1162–70.

8. Mori T., Shibasaki M., Ogawa Y., Hokazono M., Wang T.-C.,Rahmadi M. et al. Comparison of the behavioral effects ofbupropion and psychostimulants. Eur J Pharmacol 2013;718: 370–5.

9. Rau K. S., Birdsall E., Hanson J. E., Johnson-Davis K. L.,Carroll F. I., Wilkins D. G. et al. Bupropion increases striatalvesicular monoamine transport. Neuropharmacology 2005;49: 820–30.

10. Blackstone K., Iudicello J. E., Morgan E. E., Weber E., MooreD. J., Franklin D. R. et al. Human immunodeficiency virusinfection heightens concurrent risk of functional depend-ence in persons with long-term methamphetamine use. JAddict Med 2013; 7: 255–63.

11. Brensilver M., Heinzerling K. G., Shoptaw S. Pharmaco-therapy of amphetamine-type stimulant dependence: anupdate. Drug Alcohol Rev 2013; 32: 449–60.

1888 Commentary

© 2014 Society for the Study of Addiction Addiction, 109, 1887–1888