Authors' reply: "Risk reduction with buprenorphine-naloxone and methadone: patient's choice".

L ETTERS Risk Reduction With Buprenorphine– Naloxone and Methadone: Patient’s Choice To the Editors: A recent study by Woody et al 1 demonstrated “. marked and approxi- mately equal reductions in injection related risk” among subjects who applied for methadone maintenance but agreed to be randomized to receive either buprenor- phine or methadone, and who remained in their assigned treatment condition for 24 weeks. It is respectfully suggested that this trial raises serious ethical issues while contributing little if anything to guide either clinicians or prospective patients. Although the outcome of the 2 groups with regard to virtually all of the parameters measured was remarkably sim- ilar, a major distinction was observed with regard to retention: at the end of 24 weeks, 74% of methadone recipients were still enrolled compared with only 46% of those assigned to the buprenorphine arm. This finding is not unexpected when research protocol rather than applicant preference determines the treatment regimen to be provided. It is of particular concern given the clear risk of both morbidity and mortality associated with opioid depen- dence, and the very strong and consistent evidence of the favorable outcomes asso- ciated with methadone, which was the treatment of choice of all these subjects. Furthermore, in this trial, buprenor- phine recipients were subject to the same inflexible demands for observed “dosing” as are required by federal regulation of all those who receive methadone. Thus, for roughly half the 24 weeks of this trial, they were obliged to make visits to the OTP (not to a primary care provider) at least 6 days per week, and for the balance of the investigation could receive no more than 2 “take-home” doses weekly. 2 The author has no funding or conflicts of interest to disclose. e142 | www.jaids.com TO THE E DITOR The protocol was thus the antithesis of what prevails in the “real world,” where patients can receive prescriptions for as much as a month’s supply of buprenor- phine from the outset, to be dispensed by a community pharmacy. The bottom line would seem to be that this study underscores the rationale for allowing those who need and seek care for opioid dependence to “vote with their feet,” and for researchers and clini- cians to respect that vote. Robert G. Newman, MD, MPH Department of Preventive Medicine, Icahn School of Medicine at Mount Sinai, New York, NY REFERENCES 1. Woody G, Bruce D, Korthuis P, et al. HIV risk reduction with buprenorphine-naloxone or meth- adone: findings from a randomized trial. J Acquir Immune Defic Syndr. 2014;66:288–293. 2. Federal opioid treatment standards. Fed Regist. 2002;1(42):65–68. To be codified at 42 CFR 2.1 Authors’ Reply: “Risk Reduction With Buprenorphine– Naloxone and Methadone: Patient’s Choice“ To the Editors: We agree with the observation 1 that this trial did not represent the way patients are treated with buprenorphine–naloxone in clinical practice; however, it is impor- tant to point out that the main purpose of Supported by NIDA grants: U10-DA013714 (D. Donovan); U10 DA-13043, KO5 DA-17009 (G.E.W.). The authors have no conflicts of interest to disclose. These HIV data have not been presented; however, they will be presented at AMERSA, November 8, 2014, San Francisco, CA, and at the ACNP, December 2014. this study was to evaluate if buprenor- phine–naloxone causes hepatotoxicity. Several previous case reports suggested that it does, and the FDA asked NIDA to conduct a trial to explore this question. The request was that the trial includes at least 300 patients who were treated with each medication for 6 or more months and had 4 or more postbaseline blood draws to test for liver enzyme changes. A random- ized trial, as the gold standard for compar- ing interventions, was judged to be the best way to explore this question. It was necessary to do the study in methadone programs since both medications would available and the daily dosing minimized the risk of medication nonadherence due to diversion. All potential participants received thorough informed consent including the fact that they could drop out of the study at any time, agreed to be randomized before entering the study, and were not asked for their preferred treatment. The protocol was written at a time that patient requests for buprenor- phine–naloxone treatment in primary care were expanding and before publication of the Cochran review 2 showing better reten- tion with methadone than buprenorphine. Thus, the magnitude of the differential dropout that occurred was unexpected. Study findings showed no evidence of bu- prenorphine–naloxone induced transami- nase elevations 3 thus addressing concerns that led the FDA to request the study, and methadone maintenance and other local treatment options were available to pa- tients who dropped out of their assigned medication condition. HIV risk was one of several measures included for secondary analyses. We hope that these clarifications are responsive to the comments raised by this letter. George Woody, MD*† Douglas Bruce, MD‡ P. Todd Korthuis, MD§ Sumedha Chhatre, PhD*† Maureen Hillhouse, PhDk Petra Jacobs, MD† James Sorensen, MD¶ Andrew J. Saxon, MD# Sabrina Poole, PsyD* David Metzger, PhD*† Walter Ling, MDk J Acquir Immune Defic Syndr Volume 67, Number 5, December 15, 2014