Original Article
`
`Extended-Release Naltrexone to Prevent
`Opioid Relapse in Criminal Justice Offenders
`Joshua D. Lee, M.D., Peter D. Friedmann, M.D., M.P.H.,
`Timothy W. Kinlock, Ph.D., Edward V. Nunes, M.D.,
`Tamara Y. Boney, M.S., C.C.R.C., Randall A. Hoskinson, Jr., Donna Wilson, M.S.,
`Ryan McDonald, M.A., John Rotrosen, M.D., Marc N. Gourevitch, M.D., M.P.H.,
`Michael Gordon, D.P.A., Marc Fishman, M.D., Donna T. Chen, M.D., M.P.H.,
`Richard J. Bonnie, L.L.B., James W. Cornish, M.D., Sean M. Murphy, Ph.D.,
`and Charles P. O’Brien, M.D., Ph.D.
`
`ABS TR ACT
`
`BACKGROUND
`Extended-release naltrexone, a sustained-release monthly injectable formulation of the full
`mu-opioid receptor antagonist, is effective for the prevention of relapse to opioid depen-
`dence. Data supporting its effectiveness in U.S. criminal justice populations are limited.
`METHODS
`In this five-site, open-label, randomized trial, we compared a 24-week course of extended-
`release naltrexone (Vivitrol) with usual treatment, consisting of brief counseling and refer-
`rals for community treatment programs, for the prevention of opioid relapse among adult
`criminal justice offenders (i.e., persons involved in the U.S. criminal justice system) who
`had a history of opioid dependence and a preference for opioid-free rather than opioid
`maintenance treatments and who were abstinent from opioids at the time of randomiza-
`tion. The primary outcome was the time to an opioid-relapse event, which was defined as
`10 or more days of opioid use in a 28-day period as assessed by self-report or by testing
`of urine samples obtained every 2 weeks; a positive or missing sample was computed as
`5 days of opioid use. Post-treatment follow-up occurred at weeks 27, 52, and 78.
`RESULTS
`A total of 153 participants were assigned to extended-release naltrexone and 155 to
`usual treatment. During the 24-week treatment phase, participants assigned to extended-
`release naltrexone had a longer median time to relapse than did those assigned to
`usual treatment (10.5 vs. 5.0 weeks, P<0.001; hazard ratio, 0.49; 95% confidence interval
`[CI], 0.36 to 0.68), a lower rate of relapse (43% vs. 64% of participants, P<0.001; odds
`ratio, 0.43; 95% CI, 0.28 to 0.65), and a higher rate of opioid-negative urine samples (74%
`vs. 56%, P<0.001; odds ratio, 2.30; 95% CI, 1.48 to 3.54). At week 78 (approximately 1 year
`after the end of the treatment phase), rates of opioid-negative urine samples were equal
`(46% in each group, P = 0.91). The rates of other prespecified secondary outcome mea-
`sures — self-reported cocaine, alcohol, and intravenous drug use, unsafe sex, and rein-
`carceration — were not significantly lower with extended-release naltrexone than with
`usual treatment. Over the total 78 weeks observed, there were no overdose events in the
`extended-release naltrexone group and seven in the usual-treatment group (P = 0.02).
`CONCLUSIONS
`In this trial involving criminal justice offenders, extended-release naltrexone was asso-
`ciated with a rate of opioid relapse that was lower than that with usual treatment.
`Opioid-use prevention effects waned after treatment discontinuation. (Funded by the
`National Institute on Drug Abuse; ClinicalTrials.gov number, NCT00781898.)
`
`From the Departments of Population
`Health (J.D.L., R.M., M.N.G.), Medicine,
`Division of General Internal Medicine
`and Clinical Innovation (J.D.L.), and Psy-
`chiatry (J.R.), New York University, and
`the New York State Psychiatric Institute,
`Columbia University College of Physi-
`cians and Surgeons (E.V.N.) — both in
`New York; the Division of General Inter-
`nal Medicine, the Department of Medi-
`cine, Rhode Island Hospital and Alpert
`Medical School of Brown University, Prov-
`idence (P.D.F., R.A.H., D.W.); Friends Re-
`search Institute (T.W.K., M.G., M.F.), the
`University of Baltimore, School of Crimi-
`nal Justice (T.W.K.), and Maryland Treat-
`ment Centers (M.F.) — all in Baltimore;
`the University of Pennsylvania (T.Y.B.,
`J.W.C., C.P.O.) and the Philadelphia Vet-
`erans Affairs Medical Center (J.W.C.) —
`both in Philadelphia; the Center for Bio-
`medical Ethics and Humanities, School
`of Medicine (D.T.C.) and the School of
`Law (R.J.B.), University of Virginia, Char-
`lottesville; and Washington State Univer-
`sity, Spokane (S.M.M.). Address reprint
`requests to Dr. Lee at the Department of
`Population Health, New York University,
`227 E. 30th St., New York, NY 10016, or at
` joshua . lee@ nyumc . org.
`
`This article was updated on April 14, 2016,
`at NEJM.org.
`
`N Engl J Med 2016;374:1232-42.
`DOI: 10.1056/NEJMoa1505409
`Copyright © 2016 Massachusetts Medical Society.
`
`1232
`
`n engl j med 374;13 nejm.org March 31, 2016
`
`T h e ne w e ngl a nd jou r na l o f m e dicine
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 8, 2019. For personal use only. No other uses without permission.
`
` Copyright © 2016 Massachusetts Medical Society. All rights reserved.
`
`ALKERMES EXHIBIT 2043
`Amneal Pharmaceuticals LLC v. Alkermes Pharma Ireland Limited
`IPR2018-00943
`
`Page 1 of 11
`
`

`

`Extended-Release Naltrexone in offenders
`
`Opioid-use disorder is a chronic re-
`
`lapsing condition that has serious public
`health consequences. Opioid dependence
`disproportionately affects U.S. criminal justice
`system populations, and relapse and overdose
`deaths occur at high rates after release from
`incarceration.1 Evidence-based opioid-agonist
`maintenance therapies for opioid dependence
`(methadone and buprenorphine) are effective in
`prison, jail, and community reentry (i.e., parole)
`settings2-5 but have historically been unavailable
`or discouraged among criminal justice clients.6-8
`Extended-release naltrexone (Vivitrol, Alkermes),
`a sustained-release monthly injectable formula-
`tion of the full mu-opioid receptor antagonist,
`was approved by the Food and Drug Administra-
`tion in 2010 for the prevention of relapse to
`opioid dependence. Extended-release naltrexone
`may be particularly appealing and beneficial to
`patients and providers who are unlikely to access
`opioid-agonist maintenance treatment or who
`prefer a relapse-prevention medication. As a
`noncontrolled substance with no known abuse
`or diversion potential, extended-release naltrex-
`one has gained increasing acceptance in the
`criminal justice system despite limited data on
`effectiveness.
`Extended-release naltrexone gradually releases
`sufficient naltrexone to block the euphoric ef-
`fects of opioids for approximately 1 month after
`injection and is efficacious as compared with
`placebo.9-12 A pilot study that was performed at
`the same five sites that participated in this trial
`used a single-group observational cohort design
`and showed the feasibility of using Depotrex, an
`alternative formulation of extended-release nal-
`trexone, as a treatment option for outpatient
`parolees and probationers.13 We conducted a
`large, multisite, randomized trial to examine
`the effectiveness of extended-release naltrexone
`among community-dwelling criminal justice of-
`fenders who were at high risk for opioid relapse
`and related adverse outcomes.
`
`Methods
`
`Trial Design, Sites, and Oversight
`This open-label, randomized, controlled effec-
`tiveness trial compared six monthly injections of
`extended-release naltrexone (Vivitrol, Alkermes)
`with usual treatment (brief counseling and refer-
`rals for community treatment programs) for the
`prevention of opioid relapse among criminal
`
`justice offenders. We hypothesized that the like-
`lihood of an opioid-relapse event would be low-
`er, the time to relapse longer, and overall rates
`of opioid use lower with extended-release nal-
`trexone than with usual treatment.
`Five independently funded sites implemented
`a common collaborative protocol: University of
`Pennsylvania (Philadelphia), New York University
`School of Medicine and Bellevue Hospital Center
`(New York), Rhode Island Hospital and Brown
`University (Providence, Rhode Island), Columbia
`University Medical Center (New York), and Friends
`Research Institute (Baltimore). The University of
`Pennsylvania, which was the lead site, hosted
`the regulatory and data management cores and
`the data and safety monitoring board.
`The rationale, protocol development, design,
`and methods of this trial are described in full
`elsewhere.14 All sites obtained approval from the
`local institutional review board and the U.S. Office
`for Human Research Protections for the com-
`mon trial protocol. The authors alone designed
`and implemented the trial; collected, accessed,
`and analyzed the data; and vouch for the com-
`pleteness and accuracy of the data and for the
`fidelity of the trial to the protocol, which is
`available with the full text of this article at
`NEJM.org. The first author wrote the initial draft
`of the manuscript, and all the authors partici-
`pated in revisions and approved the final draft.
`The sponsor (National Institute on Drug Abuse)
`and the manufacturer of extended-release nal-
`trexone (Alkermes) did not have editorial control
`or access to trial data. The manufacturer con-
`tributed Vivitrol in kind through an investigator-
`initiated trial contract, which allowed for review
`of and comment on the manuscript before sub-
`mission for publication.
`
`Participants
`We recruited community-dwelling adult volunteers
`who were criminal justice offenders and who had
`a history of opioid dependence. Eligibility crite-
`ria were current (within the previous 12 months)
`or lifetime (any previous) opioid dependence (as
`defined by the Diagnostic and Statistical Manual of
`Mental Disorders, 4th edition [DSM-IV])15; a stated
`goal of opiate-free treatment rather than opioid-
`agonist or partial-agonist maintenance therapy;
`an opioid-free status as confirmed by negative
`urine toxicologic screening for all opioids before
`randomization; residence in the community and
`receipt of an adjudicated sentence that included
`
`n engl j med 374;13 nejm.org March 31, 2016
`
`1233
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 8, 2019. For personal use only. No other uses without permission.
`
` Copyright © 2016 Massachusetts Medical Society. All rights reserved.
`
`Page 2 of 11
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dicine
`
`supervision (e.g., parole, probation, outpatient
`drug-court programs, or other court-mandated
`treatment) or, in the previous 12 months, release
`from jail or prison, a plea-bargain arrangement,
`or any community supervision as above; general
`good health as determined by history and physi-
`cal examination; an age of 18 to 60 years; and
`the ability to provide written informed consent.
`Exclusion criteria were other drug or alcohol
`dependence requiring a level of care that would
`interfere with trial participation; pregnancy or a
`plan to conceive during the 24-week treatment
`phase, lactation, or an inability to use adequate
`contraceptive methods; an untreated psychiatric
`disorder or medical condition that might make
`participation hazardous, including liver-enzyme
`levels more than three times the upper limit of
`the normal range and a body-mass index (BMI, the
`weight in kilograms divided by the square of the
`height in meters) of more than 40; allergy to nal-
`trexone, polylactide-co-glycolide, carboxymethyl-
`cellulose, or other components of the diluent; a
`current diagnosis of chronic pain for which opioids
`were prescribed; or a drug overdose in the previ-
`ous 3 years requiring inpatient hospitalization.
`We recruited participants by standard out-
`reach to community-dwelling at-risk populations
`through print, radio, and online publicity and
`provider detailing (e.g., letters to clinic direc-
`tors); in an effort to minimize potential coer-
`cion, we did not recruit through direct referrals
`from criminal justice authorities, including de-
`partments of corrections, probation, or parole
`and drug courts or other diversion programs.
`Prescreening questionnaires were used to briefly
`evaluate potential participants and to schedule
`an in-person screening visit at which written
`informed consent was obtained; participants
`were required to establish their comprehension
`of consent information by passing an informed-
`consent quiz.
`
`Trial physicians or nurses administered
` extended-release naltrexone by injection and
`provided medication-management counseling.
`Extended-release naltrexone, at a dose of 380 mg,
`was administered by intramuscular injection once
`every 4 weeks during medical management visits,
`the first of which occurred at the time of ran-
`domization. A standard naloxone challenge (i.e.,
`administration of >0.8 mg of naloxone intrave-
`nously, intramuscularly, or subcutaneously and
`assessment of opioid-withdrawal symptoms) had
`to be negative before the initial injection. At one
`site (Friends Research Institute), 12.5 mg of oral
`naltrexone was also administered as a low-dose
`challenge, followed by a 2-hour observation peri-
`od, before injection of extended-release nal-
`trexone; this additional challenge reflected the
`preferences of the local site and institutional
`review board. Medical management counseling
`focused on medication side effects, support for
`recovery and treatment participation, and coun-
`seling to reduce the risk of relapse and over-
`dose.18 Participants in the usual-treatment group
`received similar counseling that was focused on
`adverse events, the prevention of relapse and
`overdose, and support for community treatment
`involvement from the same trial personnel.
`Incentives, follow-up visit schedules and pro-
`cedures, and community treatment referrals were
`the same in the two groups; all participants were
`encouraged by the same trial staff to access
`appropriate community treatment and relapse-
`prevention resources, including buprenorphine
`or methadone treatment if preferred or indicated
`during the trial and after the treatment phase
`(no extended-release naltrexone was provided
`after the end of the 24-week treatment phase).
`All participants were compensated for atten-
`dance at individual visits; total cash or voucher
`compensation across 17 visits varied according
`to site ($385 to $820).
`
`Randomization and Trial Treatments
`Participants were randomly assigned, in a 1:1 ratio,
`to extended-release naltrexone or usual treatment
`for opioid-relapse prevention. An urn random-
`ization procedure ensured balance with respect
`to trial site, sex, and status regarding the need
`for opioid detoxification.16 An independent, cen-
`tralized, automated telephone system made the
`treatment assignments after eligibility of the
`participants was confirmed.17
`
`Clinical Assessments
`Follow-up and assessment procedures were the
`same in the two groups. Visits occurred at
`screening, randomization, and then every 2 weeks
`for 24 weeks during the treatment phase. Post-
`treatment follow-up assessments occurred at
`weeks 27, 52, and 78 (three visits only). The visits
`occurring every 2 weeks and at weeks 27, 52, and
`78 included urine toxicologic screening and self-
`report of opioid, cocaine, alcohol, and intrave-
`
`1234
`
`n engl j med 374;13 nejm.org March 31, 2016
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 8, 2019. For personal use only. No other uses without permission.
`
` Copyright © 2016 Massachusetts Medical Society. All rights reserved.
`
`Page 3 of 11
`
`

`

`Extended-Release Naltrexone in offenders
`
`nous drug use with the use of the Timeline
`Followback calendar method for count data.19
`Urine samples were tested for opiates (a level
`>300 ng per milliliter was considered to indicate
`a positive test), oxycodone, methadone, buprenor-
`phine, and cocaine metabolites. Data on unsafe
`sex were captured every 6 months with the use
`of the Sex Risk subscale of the Risk Assessment
`Battery (on which scores range from 0 to 18,
`with higher scores indicating a greater risk of
`contracting and spreading human immunodefi-
`ciency virus [HIV] infection through sexual be-
`haviors).20 Self-reported information about crim-
`inal activity, rearrests, and days of reincarceration
`was collected every 2 weeks with the use of the
`Timeline Followback method for days in con-
`trolled environments, monthly with the use of
`the Crime and Legal Activities Report,21 and
`every 6 months with the use of the legal-status
`items in the Addiction Severity Index Lite.22
`
`Outcomes
`The primary outcome was the time (in weeks) to
`an opioid-relapse event during the 24-week treat-
`ment phase. A relapse event was defined as 10
`or more days of opioid use in a 28-day (4-week)
`period as assessed by self-report or by testing of
`urine samples obtained every 2 weeks; a positive
`or missing sample was computed as 5 days of
`opioid use. Relapse was considered to be a one-
`time event and corresponded to the loss of per-
`sistent opioid abstinence after randomization,
`when all participants had been opioid-free and
`had endorsed a goal of opioid abstinence. Re-
`lated opioid-use outcomes were rates of opioid-
`negative (vs. opioid-positive or missing) urine
`samples, the percentage of 2-week intervals with
`no opioid use as assessed by self-report or by
`testing of urine samples (confirmed abstinence),
`the percentage of days with self-reported opioid
`use, and post-treatment rates of opioid use as
`assessed by self-report (percentage of 2-week
`intervals with any opioid use vs. no opioid use)
`and by testing of single urine samples at weeks
`52 and 78. The primary relapse outcome used
`during the treatment phase, defined by assess-
`ments performed every 2 weeks, self-report, and
`testing of urine samples, was not available dur-
`ing long-term follow-up because visits were
`scheduled only at weeks 52 and 78. Secondary
`outcomes of interest were rates of alcohol and
`nonopioid drug use, HIV risk behaviors, rearrests
`
`and reincarcerations, and adverse events includ-
`ing opioid overdose.
`
`Statistical Analysis
`We calculated that a sample size of 164 partici-
`pants per group, with an assumed loss to attri-
`tion of approximately 5% per month, would
`provide the trial with 80% power to detect a
`hazard ratio for relapse with usual treatment of
`1.53 or higher, equivalent to an estimated differ-
`ence in relapse rates of 15 percentage points
`(45% vs. 30%).9,13 The primary outcome analysis
`tested whether extended-release naltrexone re-
`sulted in a longer time to relapse than that with
`usual treatment, with the use of Cox proportional-
`hazards regression models. We compared over-
`all relapse rates using intention-to-treat mixed-
`effects logistic-regression models and compared
`rates of positive urine tests and days with self-
`reported opioid use using a linear mixed-effects
`model for count data. Missed visits and missing
`data on urine samples were counted as positive
`for opioid use; thus, dropouts contributed to a
`relapse event. Missing data for secondary out-
`comes (cocaine, alcohol, and intravenous drug
`use; score on the Sex Risk subscale of the Risk
`Assessment Battery; and reincarceration) were
`estimated from available data only.
`At the week 52 and week 78 visits, participants
`provided 6 months of self-reports on opioid use
`and a single urine sample. We analyzed the per-
`centage of opioid-negative (vs. opioid-positive or
`missing) urine samples at both visits using
`mixed-effects logistic models. We used a logistic
`mixed-effects model for repeated measures to
`analyze self-reported opioid use from week 1 to
`week 78 and to test for differences in the treat-
`ment groups over time with the use of an inter-
`action term between group and time.
`
`R esults
`
`Screening and Randomization
`Recruitment began in February 2009 and contin-
`ued through November 2013. The five sites ob-
`tained consent from and screened 437 persons,
`of whom 308 underwent randomization; 153
`were assigned to extended-release naltrexone
`and 155 to usual treatment (Fig. 1). Common
`reasons for exclusion were an incomplete screen-
`ing visit, incomplete detoxification or a lack of
`opioid abstinence before randomization, and
`
`n engl j med 374;13 nejm.org March 31, 2016
`
`1235
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 8, 2019. For personal use only. No other uses without permission.
`
` Copyright © 2016 Massachusetts Medical Society. All rights reserved.
`
`Page 4 of 11
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dicine
`
`437 Potential participants were
`assessed for eligibility
`
`129 Were excluded
`57 Had incomplete screening
`25 Were not abstinent from
`opioids
`19 Had medical or psychiatric
`reasons
`3 Had recent drug overdose
`2 Had a body-mass index >40
`23 Had other reasons
`
`308 Underwent randomization
`
`153 Were assigned to receive
`extended-release naltrexone
`146 Received intervention
`7 Declined to receive intervention
`
`155 Were assigned to receive
`usual treatment
`155 Received intervention
`
`119 Completed 24-wk treatment phase
`follow-up
`7 Were lost to follow-up
`8 Withdrew consent
`1 Had an adverse event
`1 Changed residence
`13 Were incarcerated
`4 Had other reasons
`
`126 Completed 24-wk treatment phase
`follow-up
`4 Were lost to follow-up
`1 Withdrew consent
`2 Died
`5 Changed residence
`15 Were incarcerated
`2 Had other reasons
`
`153 Were included in primary analysis
`
`155 Were included in primary analysis
`
`Figure 1. Screening, Randomization, and Follow-up.
`
`serious medical or psychiatric coexisting condi-
`tions. Two potential participants were excluded
`because of a BMI of more than 40 and therefore
`a potentially elevated risk of a severe injection-
`site reaction.
`
`Participants
`The characteristics of the trial groups were
`similar at baseline. The mean age was 44 years;
`85% of the participants were male, 77% were
`black or Hispanic, 74% were on parole or proba-
`tion, and 65% had not used heroin or other
`opioids in the previous 30 days (Table 1). All
`reported a history of DSM-IV opioid dependence.
`A total of 88% of the participants reported
`heroin use and 41% reported injection-drug use
`
`during their lifetimes; 34% reported any opioid
`(heroin or other) use in the previous 30 days. A
`total of 9% of the participants required opioid
`detoxification to enter the trial.
`
`Attendance at Scheduled Visits
`and Adherence to Medication
`Most of the visits that were scheduled every
`2 weeks (3096 of 4004, 77%) were attended;
`participants assigned to extended-release nal-
`trexone attended 79% of the scheduled visits,
`and those assigned to usual treatment attended
`75%. A total of 75% of the participants com-
`pleted an end-of-treatment-phase visit at week 27.
`Overall, participants assigned to extended-release
`naltrexone completed 711 of the 918 planned
`monthly injections (77%). Seven participants
`(5%) declined any injections after randomiza-
`tion; 146 (95%) completed the first injection,
`132 (86%) the second injection, 119 (78%) the
`third injection, 111 (73%) the fourth injection,
`100 (65%) the fifth injection, and 93 (61%) the
`sixth injection.
`
`Primary Outcome and Related Opioid-Use
`Outcomes
`During the 24-week treatment phase, the time to
`relapse was significantly longer in the extended-
`release naltrexone group than in the usual-
`treatment group: 10.5 weeks versus 5.0 weeks
`(P<0.001; hazard ratio for relapse, 0.49; 95%
`confidence interval [CI], 0.36 to 0.68) (Fig. 2).
`A relapse event was detected in 66 participants
`assigned to extended-release naltrexone (43%) as
`compared with 99 assigned to usual treatment
`(64%) (P<0.001; odds ratio, 0.43; 95% CI, 0.28 to
`0.65); this finding is consistent with the higher
`rate of opioid-negative urine samples, the lower
`percentage of days with self-reported opioid use,
`and the higher percentage of 2-week intervals
`with confirmed abstinence that we observed
`with extended-release naltrexone than with usual
`treatment (Table 2). An alternative analysis of
`missing urine data, in which only two consecutive
`confirmed positive urine screening results or self-
`reports of opioid use contributed to a “con-
`firmed relapse outcome,” also favored extended-
`release naltrexone (rate of relapse, 15% vs. 37%;
`P<0.001; hazard ratio, 0.33; 95% CI, 0.21 to 0.54).
`The treatment effect did not differ significantly
`according to site.
`
`1236
`
`n engl j med 374;13 nejm.org March 31, 2016
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 8, 2019. For personal use only. No other uses without permission.
`
` Copyright © 2016 Massachusetts Medical Society. All rights reserved.
`
`Page 5 of 11
`
`

`

`Extended-Release Naltrexone in offenders
`
`Table 1. Baseline Characteristics of the Participants.*
`
`Characteristic
`
`Age — yr
`
`Male sex — no. (%)
`
`Race or ethnic group — no./total no. (%)†
`
`White
`
`Black
`
`Hispanic
`
`Years of education
`
`Current employment — no. (%)
`
`Status with respect to supervision by criminal justice system — no. (%)
`
`Current supervision‡
`
`Probation
`
`Parole
`
`Other
`
`No supervision§
`
`Health insurance
`
`Any
`
`Medicaid
`
`Opioid use during lifetime — no./total no. (%)
`
`Opioid dependence¶
`
`Heroin use
`
`Other, non-heroin, opioid use
`
`Injection-drug use
`
`Opioid use in past 30 days — no./total no. (%)
`
`Heroin use
`
`Other, non-heroin, opioid use
`
`Any opioid use
`
`Needed opioid detoxification to enter trial — no. (%)
`
`Cocaine use in past 30 days — no./total no. (%)
`
`Heavy alcohol use in past 30 days — no. (%)
`
`Extended-Release
`Naltrexone
`(N = 153)
`
`44.4±9.2
`
`129 (84.3)
`
`31/152 (20.4)
`
`81/152 (53.3)
`
`37/152 (24.3)
`
`11.5±2.2
`
` 26 (17.0)
`
`121 (79.1)
`
` 55 (35.9)
`
` 57 (37.3)
`
` 9 (5.9)
`
` 32 (20.9)
`
`109 (71.2)
`
` 70 (45.8)
`
`153/153 (100)
`
`135/152 (88.8)
`
` 77/152 (50.7)
`
` 64/152 (42.1)
`
` 32/152 (21.1)
`
` 31/152 (20.4)
`
` 47/152 (30.9)
`
`13 (8.5)
`
`Usual
`Treatment
`(N = 155)
`
`43.2±9.4
`
`132 (85.2)
`
`30/155 (19.4)
`
`74/155 (47.7)
`
`45/155 (29.0)
`
`11.5±1.8
`
` 29 (18.7)
`
`124 (80.0)
`
` 62 (40.0)
`
` 54 (34.8)
`
` 8 (5.2)
`
` 31 (20.0)
`
`111 (71.6)
`
` 65 (41.9)
`
`155/155 (100)
`
`137/155 (88.4)
`
` 74/155 (47.7)
`
` 62/155 (40.0)
`
` 43/155 (27.7)
`
` 26/155 (16.8)
`
` 59/155 (38.1)
`
`14 (9.0)
`
` 30/152 (19.7)
`
` 29/155 (18.7)
`
` 18 (11.8)
`
` 19 (12.3)
`
`* Plus–minus values are means ±SD. There were no significant differences between the two groups at baseline.
`† Race and ethnic group were self-reported.
`‡ Current supervision by the criminal justice system was defined as parole, probation, or other involvement (drug court,
`diversion, or alternative sentencing program) at baseline.
`§ No supervision by the criminal justice system was defined as recent criminal justice involvement (i.e., arrest, incarcera-
`tion, conviction, or plea bargain) 0 to 12 months before baseline, with no current supervision as above.
`¶ Opioid dependence was defined according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition.
`
`Secondary Outcomes
`Several important secondary outcomes did not
`differ significantly between the groups: rates of
`cocaine, alcohol, and intravenous drug use; the
`score on the Sex Risk subscale of the Risk As-
`
`sessment Battery; and self-reported reincarcera-
`tion (percentage of participants with any reincar-
`ceration and total days of incarceration) (Table 3).
`Participation in other, nontrial treatment in the
`community for opioid-use disorders did not dif-
`
`n engl j med 374;13 nejm.org March 31, 2016
`
`1237
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 8, 2019. For personal use only. No other uses without permission.
`
` Copyright © 2016 Massachusetts Medical Society. All rights reserved.
`
`Page 6 of 11
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dicine
`
`Table 2. Opioid Relapse and Related Outcomes during the 24-Week Treatment Phase.*
`
`Outcome
`Primary outcome: median time to relapse — wk†
`
`Opioid-relapse event — no. (%)
`
`Extended-Release
`Naltrexone
`(N = 153)
`10.5
`
`66 (43.1)
`
`Percentage of 2-wk intervals with confirmed abstinence
`
`Percentage of opioid-negative urine samples
`
`Percentage of days with self-reported opioid use
`
`71.1
`
`74.1
`
`4.6
`
`Usual
`Treatment
`(N = 155)
`5.0
`
`99 (63.9)
`
`49.5
`
`55.7
`
`12.7
`
`Hazard Ratio, Odds Ratio, or
`Incidence-Density Ratio
`(95% CI)
`0.49 (0.36–0.68)‡
`
`0.43 (0.28–0.65)§
`
`2.50 (1.66–3.76)¶
`
`2.30 (1.48–3.54)¶
`
`0.35 (0.21–0.59)‖
`
`P Value
`<0.001
`
`<0.001
`
`<0.001
`
`<0.001
`
`0.02
`
`* CI denotes confidence interval.
`† Shown is the median time to relapse if a relapse event occurred.
`‡ Shown is the hazard ratio for relapse, calculated with a Cox proportional-hazards model.
`§ Shown is the odds ratio calculated with a generalized estimating equation (GEE) mixed-effects logistic-regression model, with trial site as
`the repeated measure.
`¶ Shown is the odds ratio calculated with a GEE mixed-effects logistic-regression model for repeated measures, with participant as the repeat-
`ed measure and with adjustment for trial site and week.
`‖ Shown is the incidence-density ratio calculated with a GEE mixed-effects Poisson regression model, with trial site as the repeated measure
`and with adjustment for log days at risk.
`
`not widely available in public-sector facilities
`during the treatment phase.
`
`Safety and Long-Term Opioid-Use Outcomes
`Adverse events, including medication-related ad-
`verse events, were more common among par-
`ticipants assigned to extended-release naltrexone
`than among those assigned to usual treatment
`(Table 4); however, significantly more serious
`adverse events occurred in the usual-treatment
`group than in the extended-release naltrexone
`group. All recorded overdose events, fatal or non-
`fatal, occurred among participants assigned to
`usual treatment (0 events in the extended-release
`naltrexone group vs. 5 in the usual-treatment
`group from week 0 to 25, P = 0.10; 0 vs. 7 events
`from week 0 to 78, P = 0.02); no overdoses oc-
`curred in the extended-release naltrexone group
`after discontinuation of the agent. The percent-
`age of participants with opioid-negative urine
`samples was similar in the two groups after
`completion of the treatment phase: 49% in the
`extended-release naltrexone group and 46% in the
`usual-treatment group at week 52 (P = 0.61), and
`46% in both groups at week 78 (P = 0.91). Self-
`reported opioid use showed an interaction be-
`tween treatment and time: participants assigned
`to extended-release naltrexone reported less use
`than those assigned to usual treatment during
`the active treatment phase, but the increase after
`the active treatment phase was greater in the
`extended-release naltrexone group, with the re-
`sult that opioid use was similar in the two groups
`
`fer significantly between the groups (62% of the
`participants in the extended-release naltrexone
`group and 65% of the participants in the usual-
`treatment group, P = 0.63). More participants in
`the usual-treatment group than in the extended-
`release naltrexone group pursued opioid-agonist
`treatments during the trial (37% vs. 11%,
`P<0.001), primarily after resumed illicit opioid
`use and relapse. No participants reported con-
`tinuing extended-release naltrexone after the
`treatment phase; the drug was approved for the
`prevention of opioid relapse mid-trial and was
`
`Extended-release naltrexone
`
`Usual treatment
`
`1
`
`3
`
`5
`
`7
`
`9
`
`11
`
`13
`Week
`
`15
`
`17
`
`19
`
`21
`
`23
`
`25
`
`1.0
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`0.0
`
`ProbabilityofRelapse-freeSurvival
`
`No.atRisk
`Extended-release
`naltrexone
`Usual treatment
`
`153
`
`144
`
`139
`
`129
`
`121
`
`117
`
`112
`
`110
`
`104
`
`100
`
`92
`
`155
`
`116
`
`104
`
`96
`
`84
`
`76
`
`72
`
`67
`
`65
`
`61
`
`59
`
`87
`
`56
`
`87
`
`56
`
`Figure 2. Kaplan–Meier Curves for Relapse-free Survival.
`
`1238
`
`n engl j med 374;13 nejm.org March 31, 2016
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 8, 2019. For personal use only. No other uses without permission.
`
` Copyright © 2016 Massachusetts Medical Society. All rights reserved.
`
`Page 7 of 11
`
`

`

`Extended-Release Naltrexone in offenders
`
`Table 3. Secondary Outcomes.*
`
`Outcome
`Percentage of days with cocaine use†
`
`Heavy drinking in past 30 days at
`wk 27 — no. (%)§
`
`Any intravenous drug use — %‖
`
`Mean score on Sexual Risk subscale
`of RAB at wk 27**
`
`Any reincarceration — no. (%)
`
`Total days of reincarceration
`
`Days incarcerated
`
`Among participants with any
`reincarceration
`
`Extended-Release
`Naltrexone
`(N = 153)
`3.9
`
`16 (13.9)
`
`Usual
`Treatment
`(N = 155)
`4.3
`
`17 (14.9)
`
`7 (5.9)
`
`2.75
`
`35 (22.9)
`
`1651
`
`10 (8.6)
`
`2.86
`
`45 (29.0)
`
`2628
`
`Odds Ratio or
`Incidence-Density Ratio
`(95% CI)
`0.91 (0.56–1.48)‡
`
`0.89 (0.43–1.87)¶
`
`0.67 (0.25–1.82)¶
`
`0.71 (0.33–1.52)‡‡
`
`0.63 (0.32–1.23)‡
`
`P Value
`0.71
`
`0.77
`
`0.43
`
`0.68††
`
`0.38
`
`0.22
`
`47.2±41.8
`
`58.4±51.4
`
`0.30††
`
`Among all participants
`
`11.1±28.4
`
`17.6±38.9
`
`0.10††
`
`*
`†
`
` Plus–minus values are means ±SD.
` Data are based on self-report for weeks 1 to 25 with the use of the Timeline Followback calendar method. No
`Timeline Followback data were available for 4 participants in the extended-release naltrexone group an