Extended-release naltrexone versus standard oral naltrexone versus placebo for opioid use disorder: the NEAT three-arm RCT

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 10/46/01. The contractual start date was in September 2014. The draft report began editorial review in October 2017 and was accepted for publication in May 2018. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

John Strang reports grants and others from Martindale Pharma, grants and others from Mundipharma, and grants and others from Braeburn Pharma, outside the submitted work. In addition, John Strang has a Euro-Celtique patent issued, and a King’s College London patent pending. Michael Kelleher reports grants from the National Institute for Health Research (NIHR) during the conduct of the study, others from South London and Maudsley NHS Foundation Trust, Public Health England, Braeburn Pharma and Reckitt Benckiser, personal fees from Mundipharma and non-financial support from Cephaid, outside the submitted work. Sarah Byford reports grants from NIHR Health Technology Assessment programme during the conduct of the study. John Marsden reports investigator-led, educational grant funding from Indivior (administered by Action-on-Addiction) for a study of personalised psychosocial intervention for non-response to opioid agonist treatment. He declares consultancy for the US National Institute on Drug Abuse Centre for Clinical Trials Network. In the past 3 years, he received honoraria from Merck Serono (2015; clinical oncology training), Martindale (2017; expert meeting on opioid use disorder), and Indivior (via PCM Scientific) as co-chairperson (2015, 2016) and chairperson (2017) for the conference on Improving Outcomes in Treatment of Opioid Dependence.

Copyright statement

© Queen’s Printer and Controller of HMSO 2019. This work was produced by Strang et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

2019 Queen’s Printer and Controller of HMSO

Trial summary

What is the clinical effectiveness and cost-effectiveness of enhanced naltrexone in the treatment of opioid use disorder (OUD)? The Naltrexone Enhanced Addiction Treatment for Opioid Use Disorder (NEAT) trial was the first Phase III UK study to coalesce antagonist medication and behavioural interventions for the treatment of patients with heroin addiction problems.

The study was implemented in two specialist NHS outpatient addiction clinics in London and Birmingham (recruitment centres), each with formal links for research trials with a local university.

A total of 300 recently detoxified, formerly dependent heroin users were to be randomised to one of three treatments, which were to be received on site under supervision:

1. thrice-weekly oral active naltrexone tablets plus placebo extended-release naltrexone at the start of treatment

2. oral placebo naltrexone plus active extended-release naltrexone

3. oral placebo naltrexone plus placebo extended-release naltrexone.

Each condition was delivered over 12 weeks. All participants received standard NHS psychological interventions (weekly individual counselling) and a behavioural protocol incentivising clinic attendance to receive trial medication and complete research assessments.

The primary outcome measure was the number of heroin-negative urine drug screens (UDSs) in treatment (taken thrice weekly during the 12-week treatment phase of the trial; 36 UDSs in total). In addition to societal-focused health-related cost-effectiveness, secondary objectives related to treatment retention/adherence, craving for heroin and cocaine, and monitoring of naltrexone and 6-β-naltrexol (the primary metabolite of naltrexone). Research worker-administered follow-up assessments were at 16, 24 and 36 weeks after the active 12-week treatment phase.

Background and rationale

The term ‘opioids’ refers to semisynthetic and synthetic analgesic compounds with similar properties to the group of psychoactive analgesic substances derived from the poppy plant, including opium, morphine and codeine. In England, ‘street’ heroin is the most harmful illegal opioid in the UK. 1 Illicit heroin makes the user feel intense euphoria. It has an aggressive dependence liability, the predominant symptom being compulsive drug use despite significant health and social harms. 2 Physiologically dependent users need to take heroin every day to avoid the onset of acutely unpleasant flu-like withdrawal symptoms. Users report experiencing intense feelings of wanting and needing to take heroin and find it very difficult to stop. Craving is considered a core symptom of addiction and a prime cause of relapse. 3 Untreated, opioid dependence is a persistent and debilitating condition, associated with high social costs arising from drug misuse. The lives of most heroin addicts are disadvantaged in many ways and there is a strong link between heroin use and acquisitive crime. There is also an associated major public health burden owing to the acquisition and transmission of blood-borne viral infections in this population. Consequently, tackling the problem of opioid dependence is a high priority for the government and the NHS.

Most individuals presenting to specialist NHS community treatment clinics have established harmful illicit OUDs, almost all of which are related to street heroin use. However, the addition of cocaine dependence adds considerable severity to individual cases, and this patient subgroup has a relatively poorer outcome than primary heroin users. 4 Prior to the study’s commission, there were 321,229 individuals in England with problems relating to heroin and/or crack cocaine use (corresponding to 9.4 per 1000 of the population aged 15–64 years in 2008/95). In 2008/9, the combined use of heroin and cocaine was reported in 29% of patients admitted for treatment. 6 Intravenous injection is the preferred route of administration for approximately one-third of heroin users, with associated risks of infection and overdose. There are also substantially elevated rates of mood disorders among heroin users compared with the general population,7 and after opioid detoxification, in addition to craving for heroin, patients often report a syndrome of anhedonia, including affective disorders (depression, dysphoria and anxiety). These symptoms may trigger recurrence in heroin use. 8

The chronic relapsing nature of drug dependence means that helping a patient to achieve stable abstinence is often difficult. In the NHS, the frontline clinical response to heroin dependence is the prescription of a substitute (full or partial) μ-opioid agonist (oral methadone, oral buprenorphine hydrochloride or oral buprenorphine–naloxone medication) taken once daily in the context of case management and general counselling support. If the patient receives an appropriate prescribed dose of opioid agonist maintenance therapy, physiological tolerance to opioids is medically managed and there are usually no breakthrough withdrawal symptoms before the next dosing. Opioid substitution therapy (OST) medication is administered directly under clinical supervision or by the patient at home.

Properly delivered OST creates a platform for patients to receive structured psychosocial interventions. All patients are supported by a keyworker and may also receive structured psychosocial treatment if needed. A specified keyworker (a physician, psychologist or, more commonly, a nurse, social worker or trained non-medical drugs worker) takes the lead role in co-ordinating a patient’s care. Through regular clinic appointments, the keyworker gives practical advice, applies psychological techniques to build motivation to reduce drug-related harms and organises access to community services as required. In around two-thirds of patients receiving substitution treatment, the prescribing physician maintains the patient on a stable daily dose for as long as is clinically indicated, and then supervises a gradual withdrawal to achieve opioid abstinence. In the remainder of cases, prescribing is of shorter duration, usually involving a gradual withdrawal of medication immediately following stabilisation.

However, not all patients derive a clinical benefit from OST. Some respond initially, then lapse to resumed heroin use during treatment; a minority deteriorate progressively during treatment. Some patients and clinicians prefer abstinence, rather than a maintenance approach from the outset, and some patients prefer to continue their personal recovery journey by withdrawing early from agonist therapy and receiving support for abstinence. Overall, reduced therapeutic engagement, ongoing or resumed street heroin and cocaine or amphetamine use, and variations in satisfaction with medication vary widely between programmes. Furthermore, some patients do not wish to receive OST. There are substantially elevated rates of mood disorders among heroin users compared with the general population,7 and, after opioid detoxification, in addition to craving for heroin, patients often report a syndrome of anhedonia, including affective disorders (depression, dysphoria and anxiety). These symptoms may trigger a recurrence in heroin use. 8 Unfortunately, psychological supports have been shown not to be particularly effective at helping patients to maintain abstinence, and the NHS currently has no significant alternative treatment options.

The NEAT trial addressed this need and evaluated an μ-opioid antagonist using naltrexone as part of a relapse prevention maintenance programme for formerly opioid-dependent individuals who were seeking abstinence treatment. Naltrexone blocks the effects of any subsequently ingested heroin and prevents physical dependence. Naltrexone is used as a treatment for alcohol dependence by reducing craving for alcohol and the subjective reinforcement effects of drinking. 9 For opioid dependence, naltrexone does not directly reduce the craving for heroin, but, in the absence of the physical effects of heroin, clinical studies of maintenance therapy indicate that the craving gradually attenuates. 10 This highlights the importance of combining naltrexone with behavioural therapies to maintain abstinence.

Naltrexone is rapidly absorbed, metabolised by the liver and excreted in the urine with an elimination half-life of 4 hours (13 hours for the principal metabolite 6-β-naltrexol). Behaviourally, naltrexone blocks the euphoric effects of opioids. It has no psychoactive effect of its own and tolerance and dependence do not develop. 11 Clinical studies indicate that 50 mg of oral tablet naltrexone hydrochloride will block the pharmacological effects of 25 mg of intravenously administered heroin for a period of at least 24 hours. Doubling this dose provides a blockade for around 48 hours, and tripling the dose provides a pharmacological opioid blockade for approximately 72 hours. Depending on whether 1, 2 or 3 days elapse before a patient’s next clinic visit to receive medication, a dose of 50 mg, 100 mg or 150 mg of oral naltrexone is prescribed. An open-ended and flexible approach to the dosing regimen and the duration of treatment is usually used in routine NHS practice with this medication. Patients may receive 50 mg of oral naltrexone each weekday, with a 100 mg dose on Saturday, or patients may receive 100 mg every other day, or 150 mg every third day. Several clinical trials have used the following dosing regimen: 100 mg on Monday, 100 mg on Wednesday and 150 mg on Friday. This schedule is acceptable to patients, balances the level of attendance at the clinic required to collect research assessments, and was, therefore, used in this study.

Oral naltrexone has an excellent pharmacological profile as an opioid blocker. However, as a relapse prevention pharmacotherapy it has produced disappointing results. The main reason for this is that patients who succumb to cravings (or are otherwise motivated to use heroin) can relatively easily discontinue their medication and then return to heroin use. Consequently, retention has been shown to be poor in all but the most motivated or socially supported patients. There have been several meta-analyses. In 2006, Berglund et al. 12 reported on 10 studies of oral naltrexone versus control (seven placebo) and six studies of psychosocial/psychopharmacological interventions involving 1071 patients randomised to oral naltrexone maintenance therapy for OUD or a control. This review pointed to retention as the key variable in explaining the effectiveness of naltrexone. The studies with the highest retention in the experimental group had better results than the control group for differences in retention, opioid-positive UDSs, psychiatric symptoms and craving for heroin during the experimental period. Among these were those studies that incentivised clinic attendance for each dose by offering vouchers that could be exchanged for recovery-appropriate goods or services. In these trials, there was increased retention and a greater reduction in the number of opioid-positive UDSs. 13

In 2007, Adi et al. 14 reported on 26 studies with 940 participants. They concluded that the methodological quality of the reviewed trials was poor to moderate. The results suggested that oral naltrexone may be better than placebo in terms of retention in treatment, but overall this was not statistically significant. Among the trials including a contingency management element, the mean length of time patients stayed on naltrexone was 7.4 weeks, compared with 2.3–5.6 weeks on naltrexone treatment alone. Nevertheless, based on evidence and clinical experience, according to the National Institute for Health and Care Excellence (NICE) ‘naltrexone is recommended as a treatment option in detoxified formerly opioid-dependent people who are highly motivated to remain in an abstinence programme’. 15

Against the background of clinical evidence for oral naltrexone, there is a clear logic for a sustained-release formulation of naltrexone: it removes the need for the patient to remember to take the medication (usually either daily or thrice weekly). Medical products variously described by prefix as controlled, modified, slow, extended, sustained or prolonged-release (extended-release is the term used herein) are designed to reduce the frequency of dosing by modifying the rate of release and absorption of an active substance. Such products have been available for some time and have been used effectively to treat a wide range of clinical indications. First-generation products achieved modified release through intramuscular or subcutaneous injections of suspensions of insoluble drug complexes.

Trial design

The NEAT trial was initiated as a 3-year, definitive, two-centre, three-arm, parallel-group, placebo-controlled, double-blind, double-dummy, Phase III randomised controlled trial. Its objective was to evaluate and compare the clinical effectiveness of oral naltrexone with implanted extended-release naltrexone as relapse prevention therapy for participants with OUD. After a literature review and discussion with experts, 12 weeks was selected as the optimum duration over which to deliver medication, the psychological intervention and the incentivised clinical attendance protocol. Primary and secondary outcomes were assessed after 12 weeks, with follow-up interviews after 16, 24 and 36 weeks.

The trial was designed as double blind. Active and placebo oral medication were produced and encapsulated identically. Active and placebo implant devices were produced and packaged identically. Clinicians and research workers completing baseline assessments, clinic attendance assessments and all follow-ups were blind to group allocation, as were patients and pharmacists. This design ensured that the study had a high level of both treatment integrity (delivery of the treatment as intended) and treatment differentiation (treatment conditions differed from one another in the intended manner). The trial had three groups (Figure 1):

1. group A – active extended-release naltrexone and placebo oral naltrexone

2. group B – placebo extended-release naltrexone and active oral extended-release naltrexone

3. group C – placebo extended-release naltrexone and placebo oral naltrexone.

FIGURE 1.

Participants: patients. ADAPT, Addiction Dimensions for Assessment and Personalised Treatment; ADSUS, Alcohol & Drig adapted Adult Service Use Schedule; BIS, Barratt Impulsiveness Scale; BPAQ-SF, Buss–Perry Aggression Questionnaire-Short Form; DSM-V, Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; ECG, electrocardiography; EQ-5D, EuroQol-5 Dimensions; EQ-5D-5L, EuroQol-5 Dimensions, five-level version; GAD-7, Generalised Anxiety Disorder-7; HRBS, HIV Risk-taking Behaviour Scale; MCCS, Minnesota Cocaine Craving Scale; MoCA, Montreal Cognitive Assessment; O-NTX, oral naltrexone; PHQ-9, Patient Health Questionnaire-9 items; PIL, patient information leaflet; SCID-V, Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; TOP, treatment outcomes profile; WSAS, Work and Social Adjustment Scale; XR-NTX, extended-release naltrexone.

Research governance

The trial was co-sponsored by King’s College London (KCL) and South London and Maudsley (SLaM) NHS Foundation Trust, with regulatory compliance oversight by the King’s Health Partners Clinical Trial Office (KHP-CTO).

An ethics application was submitted to the Dulwich National Research Ethics Service (NRES) Committee London on 21 August 2014 and the NEAT trial received a favourable opinion on 22 September 2014 (reference number: 14/LO/1615). The protocol is available at www.journalslibrary.nihr.ac.uk/programmes/hta/104601/#/.

A submission was filed by the KHP-CTO with the Medicines & Healthcare products Regulatory Agency (MHRA) on 8 August 2014. The NEAT trial gained MHRA approval on 2 October 2014 and was given the reference number 28482/0014/001–0001.

To ensure transparency, the trial was registered with the International Standard Randomised Controlled Trial Number (ISRCTN) Registry on 12 January 2015 (ISRCTN95809946).

The National Institute for Health Research (NIHR) Clinical Research Network (CRN) Portfolio details high-quality clinical research studies that are eligible for support from the NIHR CRN in England. The trial was adopted on to the NIHR CRN Portfolio on 21 September 2015 and was issued the NIHR CRN Portfolio number 17950.

Global NHS permissions were obtained through the Integrated Research Application System on 10 November 2014 and local NHS permissions were obtained from each participating NHS trust. A clinical trial site agreement, based on the model agreement for non-commercial research in the health service, was signed by each participating NHS trust and the sponsor (KCL and SLaM).

Following guidelines from the NIHR, a Trial Steering Committee (TSC), with a majority of independent members, was convened to oversee the trial on behalf of the funder (NIHR) and the sponsor (KCL and SLaM).

The TSC met at least annually during the trial and comprised an independent chairperson, an independent lay member (representing patient perspectives), independent clinicians (specialising in addictions), the chief investigator (JS) and the lead investigator (JM) representing the Trial Management Group (TMG).

Management of the trial

The trial manager was responsible for day-to-day management of the trial with support from the data manager and trial statistician. The TMG was responsible for overseeing day-to-day management of the trial and comprised the chief investigator (JS), the lead investigator (JM) and statisticians (JH and ER). The TMG met regularly throughout the trial to ensure adherence to the trial protocol and monitor the conduct and progress of the trial.

Design and development of the protocol

Clinicians (including doctors, nurses and key workers from the addictions field across the UK), as well as people who previously used drugs, were invited to discuss the trial protocol. This feedback was utilised by the applicants when designing and developing the protocol.

Amendments to the trial protocol

Following receipt of a favourable opinion of the trial protocol from the NRES on 22 September 2014, four substantial amendments were submitted and received favourable opinion.

In summary, these were as follows.

1. Amendment 1 (23 July 2015):

   • When reviewing the patient visit timeline the TMG realised that certain measures were not needed as often or at all. Electrocardiograms (ECGs) were deemed necessary only at baseline and the treatment outcomes profile and both the outcome rating scale and the session rating scale were not needed.

   • The TMG decided that comparing the patterns of heroin relapse between treatment arms among the extended-release naltrexone and the extended-release naltrexone and placebo conditions was needed, and this was added as a secondary end point.

   • It was noted in the investigator brochure that severe renal impairment was included in the list of contraindications but not in the protocol, so this was added as an exclusion criterion.

   • A new formula of placebo from the manufacturer was released so changes from the new investigational medicinal product dossier were added to the protocol. They included an amended version of the undesirable effects, which was added to the protocol and Patient Information Leaflet. The new description of the drug was added to the protocol too.

   • The telephone number of the emergency unblinding service was added to the protocol.

   • It was noted that the dosing level for the ‘take away’ period was incorrect and was deleted from the protocol.

   • It was no longer the case that, if a patient missed six consecutive appointments, the 14-day rule would apply, and this rule was deleted from the protocol.

2. Amendment 2 (13 August 2015):

   • The pharmacy indicated to us that a dosing leaflet was needed to instruct patients on the correct dosing instructions for the oral tablets taken in the trial.

3. Amendment 3 (8 December 2015):

   • Additional text was added to the ‘Economic’ section of the ‘Primary Endpoints’ section to reflect the clinical end points.

   • The ‘Investigational Medicinal Product (IMP), Dosage and Route Of Administration’ section had the words ‘immediately following randomisation’ deleted as there was no clinical reason for the implant to be given immediately after randomisation.

   • ‘Plasma naltrexone’ was added to the secondary end points as it was mentioned in the text of the ‘plasma monitoring’ section but not in the secondary objectives.

   • Reference to ‘antipsychotics, anticonvulsants, antidepressants and anxiolytics’ was removed in the ‘Concomitant Medication’ section as there was no evidence to support those types of drugs being prohibited medication.

   • The 48-hour minimum time period was removed from the consenting period to ensure that patients – particularly those at risk of drug overdose – were able to access the study more quickly.

   • The text in inclusion criterion 5 – ‘a Morphone 2000 [opioid]’ (Alere, Loughton, UK) was removed, as that type of UDS cup was no longer being used in the study.

   • The note about failing the naloxone challenge in criterion 6 of the inclusion criteria was changed from having a 1-month minimum time period during which a patient can return to a patient being able to return when it was clinically indicated. This was because there was no evidence that a 1-month minimum time period was required.

   • Exclusion criterion 2 was amended to make it clearer that patients were not to be excluded from the study if they fail the breathalyser test.

   • The text in exclusion criterion 4 – ‘a Morphone 2000 [opioid]’ was removed, as that type of UDS cup was no longer being used in the study.

   • In exclusion criterion 5, the word ‘opiate’ was replaced with the word ‘opioid’ as it was the more correct word to use in this instance.

   • Exclusion criterion 18 was amended as a result of concern about potentially vulnerable patients not having naltrexone to carry them through the screening period. Some potential patients needed naltrexone during the screening period as they would have recently been released from prison and the normal clinical practice was to give these patients naltrexone to prevent overdose. If they were not given naltrexone, the risk of overdose increases greatly.

   • Exclusion criterion 20 was amended to make the wording was more in line with clinical practice.

   • There was the addition of new exclusion criterion, number 22. This exclusion criterion was to stop patients entering this study if they were on other studies.

   • A ‘Patient Identification Centres’ section was added as to better reflect recruitment practices.

   • In the ‘Primary Effectiveness Parameters’ the analysis model was changed from an analysis of covariance model to a regression model.

4. Amendment 4 (11 March 2016):

   • The primary care sites [general practitioner (GP) practices] in the West Midlands trust were added as non-recruiting sites as they covered the GP practices that were administering the insertion of the naltrexone implant.

   • There was a change in the language of the exclusion criteria to allow for a holding dose of naltrexone if there was a length of time between randomisation and the start of treatment.

   • The recruitment flyer was added to help increase recruitment to the study.

Support costs

Trials with participants with addictions are challenging. Experienced staff were needed to be able to deal with the complicated issues of the population group. To this end, resource equivalent to 1.0 whole-time equivalent (WTE) band 7 research nurse, a 1.0 WTE band 8 research nurse and a 0.05 WTE clinician were budgeted into the cost of the trial.

Patient and public involvement

The original study proposal was reviewed and endorsed by a member of the public through INVOLVE-related activities. The former service user was a member of the Mental Health Research Network and became an independent member of the TSC. He provided input into the conduct of the trial, for example by reviewing literature to be given to patients and their families (e.g. patient information sheets and patient newsletters).

Site management

Inclusion criteria

Inclusion criteria for the study were intended to be as close to clinical practice as possible. Each participant in the trial had to meet all the following criteria:

• be aged ≥ 18 years

• be able to demonstrate a verbal understanding of the study patient information material, able to provide written consent and able to understand and confirm willingness to comply with the protocol

• have a diagnosis of OUD based on the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) in the past 12 months, as assessed at baseline

• be completing or have recently completed an inpatient or outpatient treatment for opioid detoxification or has been completely and continuously abstinent from all opioids for at least 7 days

• have no tolerance to opioids, as verified by a negative urine toxicology screening test prior to randomisation (using an instant result immunoassay device)

• pass a naloxone challenge test (to confirm zero opioid tolerance by demonstrating no clinical sign or subjective report of opioid withdrawal before randomisation and prior to implant procedure) (individuals failing screening were allowed to enter screening as clinically indicated)

• be voluntarily seeking opioid antagonist treatment for OUD

• live in stable/secure accommodation in the community

• have a personal (mobile/cellular) phone and are able to nominate at least one locator individual (e.g. a family member, friend or recovery mentor) with a verifiable address and a telephone number to assist with the arrangement of follow-up appointments as required

• if female, not be pregnant or breastfeeding and agree to use a birth control method [oral hormonal contraceptives, barrier (condom or diaphragm) or Nexplanon implant (Merck, Kenilworth, NJ, USA)] for the duration of the study.

Exclusion criteria

Otherwise eligible individuals who met any of the following criteria were excluded from the study:

• have a clinically significant medical condition or observed abnormalities on physical examination or laboratory investigation, including but not limited to:

  • uncontrolled hypertension

  • significant heart disease (including angina and myocardial infarction in past 12 months)

  • any ECG/cardiovascular abnormality that, in the investigator’s judgement, is clinically significant

• suffer from severe alcohol dependence and/or alcohol withdrawal (by clinical assessment)

• currently undergoing opioid withdrawal syndrome

• test positive for presence of opioids in urine (i.e. indicating current opioid use) prior to randomisation (using an instant result immunoassay device)

• have a clinical diagnosis of opioid dependence syndrome (F11.2 on the DSM-5) with current physical dependence such that an antagonist medication (e.g. naloxone, naltrexone) could precipitate a withdrawal syndrome

• receive a positive naloxone challenge test at randomisation (confirming opioid use) or having an absence of a recorded result from a naloxone provocation test

• have acute hepatitis taken as clinical jaundice on examination and/or blood bilirubin level above the normal range for local reference criteria or aspartate aminotransferase or alanine aminotransferase levels (more than three times the upper limit of the normal range)

• have hepatic insufficiency (taken as more than three times the upper limit of the normal range of aspartate aminotransferase or alanine aminotransferase levels)

• have severe renal impairment evaluated by clinical decision

• have known Icenko-Cushing syndrome or require investigation if suspected Cushingoid features/symptoms

• have systemic mycoses

• have a clinical history of glaucoma

• have a clinical history of osteoporosis

• be pregnant, have a positive or unclear pregnancy test result, or intend to try to become pregnant during the study period or be sexually active without using a birth control method [oral hormonal contraceptives, barrier method (condom or diaphragm) or Nexplanon implant] for the duration of the trial

• be currently breastfeeding

• have a history of hypersensitivity to opioid receptor blockers (naloxone and naltrexone formulations) and other components of the formulation

• have a history of hypersensitivity to triamcinolone or related compounds

• be currently taking oral or depot naltrexone therapy or have been enrolled in any form of naltrexone therapy within 90 days prior to study screening, apart from treatment given by trial team between screening and the start of treatment

• be undergoing current criminal justice involvement with legal proceedings (not including current probation supervision) and, in the opinion of the clinical worker, an expectation that the participant would fail to complete the study protocol owing to reincarceration or relocation from the centre’s catchment area

• be currently (past 30 day) suicidal planning or have recently (past 6 months) attempted suicide

• have an active, uncontrolled severe mental illness (e.g. psychosis, bipolar I disorder, schizoaffective disorder) and/or a history or evidence of organic brain disease or dementia that would compromise the participant’s ability to comply with the study protocol

• be currently participating in any interventional trial or completed participation in any interventional trial (which in the view of the chief investigator might interfere with the NEAT trial) within the past 3 months.

Screening and recruitment

Following attendance at a site initiation meeting, screening and recruitment were commenced at participating units once the clinical trial site agreement had been signed and all necessary approvals were in place.

Potentially eligible patients were identified and approached by authorised members of staff about taking part in the trial. Information about the trial was provided to the patient, which included the purpose of the trial, the consequences of taking part or not, data security and funding of the trial. This information was also provided in a patient information sheet, along with the name and contact details of the local PI, which was given to the patient to read before making their decision to take part, or not, in the trial.

Informed consent

Potential participants were approached by a member of the clinical team. Each screening procedure was overseen by the centre PI or a medical officer reporting to the PI. Individuals failing screening were allowed to enter screening again as clinically indicated.

A study doctor or trial nurse implemented the enrolment procedure and obtained informed consent. If the taking of consent had been appropriately delegated to a non-physician, patients were offered the opportunity to speak with the study doctor, who documented that they had confirmed the patient’s eligibility in the medical notes before the patient was randomised. The study information sheet was read to the potential participant and discussed to ensure that he/she fully understood the purpose and key conditions of the trial, what is required and the risks and benefits arising from taking part. Each interested participant received an informed consent document with participant information and was asked to read the information and ask questions.

If the patient wanted to participate, he or she was required to sign the informed consent document prior to the conduct of any study-specific clinical procedures. This document was witnessed and signed by the clinical worker.

In addition (and not a requirement of participation in the trial), participants were asked if they were also willing to participate in the collection of venous blood samples at intervals over the course of their treatment to enable the study of blood levels of naltrexone and its metabolites. If participants agreed to participate, a written record of their consent was additionally collected (as earlier). This was considered separately.

Randomisation and allocation procedure

Randomisation was requested by study sites online using a bespoke web-based randomisation system hosted at the King’s Clinical Trials Unit (KCTU).

Once baseline assessments were complete, the individuals were randomised to one of the three treatment arms. Randomisation was at the patient level and was performed using an online randomisation system set up by the KCTU at the Institute of Psychiatry, Psychology & Neuroscience at KCL in London. Recruiting centre research staff randomised participants to one of the three arms of the study (ratio 1 : 1 : 1), stratifying by clinical centre, prison or community referral and recent cocaine use (yes or no), using randomly varying block sizes via the online randomisation system based at the KCTU.

Only study site staff authorised by the trial manager were given login details to the randomisation system. Authorised staff were allocated a username and password for the randomisation system. Once a patient was consented, all baseline data were collected and eligibility was confirmed, the staff member would log in to the randomisation system and enter the patient’s details. Once randomised, the system automatically generated confirmation e-mails to key staff, with or without treatment allocation information depending on their role in the study.

Screening log

To enable full and transparent reporting for the trial, brief details of all patients who met the eligibility criteria or who met all of the inclusion criteria plus one or more of the exclusion criteria were recorded in the screening log. The reasons for eligible patients not being recruited were recorded, which included the patient declining the invitation to take part, the patient being excluded by the treating clinician and logistical reasons. No patient identifiers were recorded in the screening log.

Treatment groups

Three hundred recently detoxified, formerly dependent heroin users were to be randomised to one of three conditions to receive (on site and supervised) one of the following:

1. thrice-weekly oral active naltrexone tablets plus placebo extended-release naltrexone at the start of treatment; or

2. oral placebo plus active extended-release naltrexone; or

3. oral placebo naltrexone plus placebo extended-release naltrexone.

Each condition was delivered over 12 weeks. All participants received standard NHS psychological interventions (weekly individual counselling) and a behavioural protocol incentivising clinic attendance to receive trial medication and complete research assessments.

Dosing regimen

Oral medication was administered under direct supervision in the outpatient clinics on Mondays (100 mg), Wednesdays (100 mg) and Fridays (150 mg, a higher dose to last until Monday) for the first 4 weeks. Oral medication during weeks 1–4 was directly observed. Small doses were given as takeaway medication if clinic attendance was impossible (e.g. owing to court appearances or urgent hospital appointments). Contingent on good adherence during the first month, patients were able to self-administer oral medication in weeks 5–12, dispensed on a week-by-week basis and contingent on attendance at the clinic three times a week to complete research measures and return packaging and report dosing. If there were any adherence problems, the patient was supervised for 2 weeks and would return to self-administration if adherence improved.

The single iGen/Atral-Cipan device (iGen/Atral-Cipan, Castanheira do Ribatejo, Portugal) was administered on a day-patient basis by a centre doctor (a local GP or hospital physician) appropriately experienced in general practice minor surgical procedures. A clinical consultant with extensive experience in these procedures was secured to guide the training programme. The implant procedure took approximately 30 minutes in an appropriate clinical facility attached to each centre with one of the two trial nurses assisting. A single-use minor surgical pack was used for each procedure.

Each participant was scheduled to receive the following study interventions:

• 1 × iGen/Atral-Cipan (extended-release naltrexone) implant (765 mg) or matching placebo at day 0 of study week 1

• 3 × oral naltrexone tablets (2 × 50 mg, Monday and Wednesday; 3 × 50 mg, Friday) or matching placebo at day 0 of study week 1 (for 12 weeks), directly observed for the first 4 weeks and then patient reporting self-consumption for next 8 weeks when attending clinic to complete research measures (the higher dose given on Fridays was to cover the weekend period).

The oral placebo tablet had the same excipients as the active medication. The tablet core contains lactose anhydrous, lactose monohydrate, microcrystalline cellulose and magnesium stearate. Each tablet was film-coated with Opadry^® II Yellow (Colorcon Ltd, Dartford, UK) and purified water phEUR.

Outcome measures

Safety monitoring

In each clinical recruitment site, patients were asked to give blood samples for liver function (hepatotoxicity) testing during screening and monthly over the course of active treatment (weeks 4, 8 and 12). Samples were analysed at the local hospital pathology service, with the following clinical biochemistry assays conducted: albumin, bilirubin, liver transaminases (aspartate aminotransferase/alanine aminotransferase/serum glutamic oxaloacetic transaminase/serum glutamic pyruvic transaminase) and transaminases.

Safety reporting followed the requirements described in The Medicines for Human Use (Clinical Trials) Regulation 2004: SI 2004/103117 and the EU Directive 2001/20/EC. 16 Each participant was given a study identification (ID) card, which described the trial and provided the following information: some cough and cold medicines containing opiates may not work as well as they should and alternatives are recommended, emergency pain relief following an accident may not be achieved using opiates, taking an extremely large dose of heroin to overcome naltrexone blockade could result in serious overdose, and there may be sensitivity to small doses of opiates after discontinuing naltrexone. The patient ID card listed telephone contact information to enable emergency unblinding. A 24-hour emergency code break was available through Emergency Scientific and Medical Services, London, UK.

It is possible that the naltrexone implant procedure led to local site infection or other complications. Prophylactic antibiotic medication was able to be used post surgery and participants were monitored and checked by the trial nurse on each clinic visit and by each centre PI (physician) each month. Site inflammation was man-managed on a case-by-case basis and potentially involved steroidal anti-inflammatory treatment. All SAEs and non-SAEs were identified.

All adverse events (AEs) were recorded in the eCRF and reported, as part of routine reporting throughout the trial, to the Data Monitoring and Ethics Committee and Research Ethics Committee. AEs that were assessed to be serious (i.e. prolonging hospitalisation or resulting in persistent or significant disability/incapacity), life-threatening or fatal – collectively termed SAEs – were reported to the KHP-CTO and reviewed by the PI. SAEs that were unspecified and considered to be possibly, probably or definitely related to the trial treatment were reported to the Research Ethics Committee within 15 calendar days of the event being reported.

Data management

Data management was an ongoing process. Data entered by sites on to the eCRF were monitored and checked throughout the recruitment period to ensure that data were as complete and accurate as possible.

Two levels of data validation were incorporated into the eCRF. The first was to prevent obviously erroneous data from being entered, for example entering a date of birth that occurred after the date of randomisation. The second level involved checks for data completeness and any unusual data entered, for example a physiological variable, such as blood pressure, which was outside the predefined range. The KCTU could generate data validation reports, listing all outstanding data queries, at any time via the eCRF. The site PI was responsible for ensuring that all the data queries were resolved. Ongoing data entry and validation at sites was closely monitored by the trial manager and any concerns were raised with the site PI.

Sample size

Data analysis plan

Data description

For progress reports to the Data Monitoring and Ethics Committee and the TSC, the Consolidating Standards of Reporting Trials (CONSORT) flow diagram was constructed (Figure 2 and see Figure 27). 2 This included the number of eligible patients, the number of patients agreeing to enter the trial, the number of patients refusing, and, by treatment arm, the number of patients not/inadequately/adequately treated or compliant/non-compliant, the number of patients continuing through the trial, the number of patients withdrawing, the number of patients lost to follow-up and the numbers of patients excluded/analysed.

FIGURE 2.

Template CONSORT flow diagram for NEAT trial. O-NTX, oral naltrexone; XR-NTX, extended-release naltrexone.

Data analysis performed

Owing to the small number of patients recruited into the NEAT trial (six patients), the statistical analyses performed were exploratory/descriptive rather than inferential, and no statistical tests were performed. A CONSORT flow diagram was constructed (see Figures 2 and 27). This included the number of:

• patients screened

• eligible patients

• patients agreeing to enter the trial

• patients refusing to give samples by treatment arm

• patients not/inadequately/adequately treated or compliant/non-compliant

• patients continuing through the trial

• patients withdrawing

• patients lost to follow-up

• patients excluded/analysed.

Owing to the small numbers, baseline variables were described overall, rather than by treatment arm. The following baseline variables were described: age, sex, ethnicity, duration of addiction history, previous therapies and duration of treatment. The stratification variables, clinical centre and whether or not patients had used cocaine in the past 30 days were described. The following clinical measures were also described: DSM-IV criteria for major depressive disorder, post-traumatic stress disorder (PTSD), panic disorder or generalised anxiety disorder (GAD), medical history and concomitant medications.

The number of negative urine samples collected in the 12-week treatment period was used to calculate the proportion of negative urine samples for each participant, using a denominator of 36. For patients who DNA or refused clinic visits or to give urine samples, it was assumed that these samples were positive.

The lapse and relapse patterns of each patient (based on urine samples and self-report) were summarised numerically and presented graphically over the 36 urine samples taken during the study period. Owing to the small number of patients, Kaplan–Meier curves could not be used here. Self-reported heroin use was also examined over the 12-week treatment period. Adherence and lapse–relapse were summarised by trial arm.

The first heroin lapse was summarised for each patient who had a heroin lapse–relapse, in terms of the main reason that they used heroin and whether or not they used other opioids, cocaine, amphetamine-type stimulants, alcohol or benzodiazepines on that day. Heroin use post treatment was also summarised when available.

Adherence to the oral study medication was ascertained via the dosing schedule and clinic visits for research measures. A patient is classified as being ‘compliant’ if they take at least 50% of their oral study medication. Adherence to treatment was described by the median and range of the number of clinic visits attended. The number of therapy sessions attended by each patient was summarised.

Adverse events were summarised by intervention arm and time point, in particular implant site infections.

Patient treatment guess and the clinician’s guess of the treatment for each patient (for both the implant and the tablets) were also summarised.

Economic evaluation

Given that implanted extended-release naltrexone is currently an unlicensed medication in Europe, and is more expensive than oral naltrexone, the relative cost-effectiveness of extended-release naltrexone and oral naltrexone treatments was to be assessed by including opportunity costs for stakeholders and comparing ratios of incremental opportunity costs (for all stakeholders) and incremental outcome (health-related quality of life). Costs of the study interventions and external health services and expenditure by the social and criminal justice sectors were to be combined with the primary clinical outcome measure and quality-adjusted life-years to produce incremental ratios that will determine relative cost and cost-effectiveness.

Economic outcome assessments were to be carried out after the 12-week treatment period and at the 36-week follow-up. The a priori primary economic outcome measure was to be quality-adjusted life-years using the EuroQol-5 Dimensions. The economic evaluation was to take a broad policy perspective, including costs borne by hospital and community health and social services and the criminal justice sector, plus the costs of criminal activity. Detailed information on the resources associated with the treatments, including study medications, equipment, dispensing services, urine tests, nurse time, and contact with key workers, medical, nurse and psychology staff, was to be collected from clinical records. Resources external to the clinics, including staffed/supported accommodation, hospital contacts, community health and social services, criminal justice sector resources and crimes committed, were to be collected in interviews with study participants at baseline, after the 12-week treatment period and at the 36-week follow-up.

Acquisition of the ultra-long-acting naltrexone implant/depot injection and naltrexone oral tablet

Change in environment owing to retendering of trusts to third-sector providers

General practitioner implant insertion training, contract negotiation and approval

Lack of referrals led to decreased recruitment

Patients

Problems in the prison setting

Explanatory statement

Following the Health Technology Assessment meeting held on 2 December 2016, in which the difficulties of recruiting participants to the trial were discussed, a decision was made that the NEAT trial should be closed with immediate effect, apart from the continuation of provision of trial treatment and data collection for study participants already recruited into the trial plus analysis and write-up. A later decision was made to conduct a qualitative investigation of the perspectives of potential and recruited study participants (patients) and other key stakeholders, including clinical staff, at the study sites, either as a separate report or as a publication of these findings, with conclusions being made available later.

A brief final report is shown below. No statistical tests were performed owing to the small number of patients recruited into the trial. A brief description of the participants and their outcomes is provided below.

Description of the participants in the Naltrexone Enhanced Addiction Treatment trial

Six patients were recruited into the NEAT trial. All patients were recruited from the London site by community referral. The first patient was recruited on 2 February 2016 and the last patient was recruited on 26 October 2016. The patients were recruited in February 2016, April 2016, August 2016 and October 2016 (three patients were recruited in October 2016). Patients received their implants and first dose of oral medication 1–5 days after they were randomised. The last date of patient follow-up was 27 February 2017 (a 16-week visit).

Four males and two females were enrolled in the study. The patient ages were 28, 30, 32, 38, 44 and 55 years. One patient was white and five patients were mixed race.

The duration of addiction history ranged from 2 to 18 years. Two patients had previously had methadone maintenance therapy, five patients had previously had buprenorphine maintenance therapy and one patient had previously used Suboxone (Reckitt Benckiser, Slough, UK; sublingual buprenorphine–naloxone combination tablet) (Table 1). Two patients had used cocaine in the past 30 days at the time of randomisation but were not taking cocaine in a significantly problematic or dependent manner such as to constitute criteria for exclusion. No patients had previously had naltrexone relapse prevention medication.

None of the patients met the Stuctured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders-IV criteria for major depressive episode, PTSD, panic disorder or GAD. Medical history and concomitant medications of the patients are summarised in Appendix 1.

Treatment allocation

Three patients (patients 1, 3 and 6) were allocated to extended-release naltrexone placebo and oral naltrexone placebo (double placebo). Two patients (patients 2 and 5) were allocated to active extended-release naltrexone and oral naltrexone placebo (oral placebo). One patient (patient 4) was allocated to extended-release naltrexone placebo (implant placebo) and active oral naltrexone.

Summary of patient behaviour during trial

Patient 1

Patient 1 was on the double placebo and had 28 out of 36 negative UDSs (for heroin), no positive UDSs (for heroin) and eight DNAs (Figures 3 and 4; see also Figures 15 and 16). This gives a proportion of 0.78. This patient did not disclose using heroin in the 12-week treatment period (or any other substance). This patient did not have any positive urine samples for any other substances during the 12-week treatment period. However, this patient disclosed using heroin for 6 days in weeks 12–16, 28 days in weeks 17–20, 23 days in weeks 20–24 and 28 days in weeks 25–32 (as well as other substances in the follow-up period). They did not disclose using heroin in weeks 32–36. Patient 1 reported having their first lapse during week 16 and used heroin ‘to feel less sad or bored’. They also took cocaine at this time (after heroin).

FIGURE 3.

Patient 1’s urine screen results for heroin, where 0 = negative for heroin, 1 = positive urine sample for heroin. DNAs are shown as blanks.

FIGURE 4.

Patient 1’s urine screen results for heroin, where 0 = negative for heroin. DNAs are shown as positive urine samples.

Patient 1 took all their study medication up to the end of week 8, missed two doses in week 9 and did not take their tablets in weeks 10–12.

At week 16, patient 1 had a positive urine sample for benzodiazepine. At week 24 patient 1 had a positive urine sample for heroin, cocaine and methadone. At week 36 patient 1 had a positive urine sample for methadone.

Patient 1 overdosed on heroin 17 weeks after randomisation. This patient also experienced redness, swelling and pain at the implant site, reported 1 week after the date of implant. Patient 1 also had pus reported at the implant site 2 weeks after implantation.

Neither the patient nor the clinician correctly guessed the study participant’s treatment allocation.

Patient 2

Patient 2 was on the active implant and oral placebo and had 34 out of 36 negative urine (for heroin), no positive urine (for heroin) and two DNAs (Figures 5 and 6; see also Figures 17 and 18). This gives a proportion of 0.94 for the negative urine. This patient did not disclose using heroin in the 12-week period (or any other substance) and did not fill out a ‘first heroin lapse’ form. This patient did not have any positive urine samples for any other substances. This patient did not disclose using heroin in the follow-up period and did not have any positive urine samples in the follow-up period. Patient 2 took of all their study medication for all 12 weeks.

FIGURE 5.

Patient 2’s urine screen results for heroin, where 0 = negative for heroin, 1 = positive urine sample for heroin. DNAs are shown as blanks.

FIGURE 6.

Patient 2’s urine screen results for heroin, where 0 = negative for heroin. DNAs are shown as positive urine samples.

Patient 2 experienced intermittent burning and pain at the site of the implant, reported at 2 weeks, as well as at 26 days after the date of implant. The implant was reported to be unchanged at 5 weeks after implant date.

Neither the patient nor the clinician correctly guessed the patient’s treatment allocation.

Patient 3

Patient 3 was on the double placebo and had 14 out of 36 negative urine samples (for heroin), 8 out of 36 positive urine samples for heroin and 14 DNAs (Figures 7 and 8; see also Figures 19 and 20). This gives a proportion of 0.39 negative urine samples. Patient 3 reported their first heroin lapse to have occurred during week 4 (30 days after randomisation) but did not have any positive urine samples or self-report of substance use. They stated that they used heroin at this time ‘because they had already used something else’ and reported consuming alcohol that day.

FIGURE 7.

Patient 3’s urine screen results for heroin, where 0 = negative for heroin, 1 = positive urine sample for heroin. DNAs are shown as blanks.

FIGURE 8.

Patient 3’s urine screen results for heroin, where 0 = negative for heroin. DNAs are shown as positive urine samples.

In week 5, patient 3 tested positive for heroin in the first two visits but did not self-report heroin use, and had a positive cocaine sample (they did not disclose cocaine use). They did disclose that they had used amphetamines later in the week (after the positive samples had occurred). In week 8 the patient did not attend the first two visits and had a positive urine sample for heroin (and cocaine) at the third visit, and disclosed using heroin, other opioids and cocaine on days 4–7 of that week. In week 9 the patient did not attend twice for the visits, had one positive urine sample for heroin and cocaine, and self-reported using heroin every day that week and using cocaine on day 7. In week 12 all three visits had positive urine samples for heroin and the patient self-reported using heroin every day. The second and third visits had positive methadone urine samples and the patient self-reported using other opioids on days 5 and 6. Patient 3 used heroin, cocaine, crack cocaine, cannabis, benzodiazepine, methadone and alcohol in the follow-up period (only data up to week 24 available). Patient 3 took all their medication in weeks 1–4, missed two doses in week 5, missed four doses in week 6 and did not take any study medication for the remainder of the trial.

Both the patient and the clinician correctly guessed the patient’s treatment allocation.

Patient 4

Patient 4 was on placebo implant and active oral medication. They had 31 out of 36 negative urine samples (for heroin), one positive urine sample (for heroin, in week 7) and four DNAs (Figures 9 and 10; see also Figures 21 and 22). This gave a proportion of 0.86 negative urine samples. This patient did not disclose using heroin in the 12-week period (or any other substance) and did not fill out a ‘first heroin lapse’ form. At screening, patient 4 had a positive urine screen for buprenorphine (sample 1). The remainder of their urine samples at screening were negative.

FIGURE 9.

Patient 4’s urine screen results for heroin, where 0 = negative for heroin, 1 = positive urine sample for heroin. DNAs are shown as blanks.

FIGURE 10.

Patient 4’s urine screen results for heroin, where 0 = negative for heroin. DNAs are shown as positive urine samples.

This patient did not have any other positive urine samples for any other substances. This patient did not disclose using heroin in the follow-up period and did not have any positive urine samples in the follow-up period (data only up to week 16 were available). Patient 4 took all of their study medication for all 12 weeks.

Neither the patient nor the clinician correctly guessed the patient’s treatment allocation.

Patient 5

Patient 5 was on active implant and placebo oral medication. This patient had 20 out of 36 negative urine samples for heroin, two positive urine samples for heroin and 14 DNAs (Figures 11 and 12; see also Figures 23 and 24). This gave a proportion of 0.56 negative urine samples. At screening, patient 5 had a positive urine screen for heroin and cocaine.

FIGURE 11.

Patient 5’s urine screen results for heroin, where 0 = negative for heroin, 1 = positive urine sample for heroin. DNAs are shown as blanks.

FIGURE 12.

Patient 5’s urine screen results for heroin, where 0 = negative for heroin. DNAs are shown as positive urine samples.

Patient 5 did not self-report using heroin during the 12-week treatment period (according to the self-report form for substance use in past 7 days), but did fill out a first heroin lapse form, which indicated a first lapse at the start of week 5. The reason the patient gave for this first lapse was ‘to manage the effects of other drugs’, and they used cocaine and alcohol before taking heroin on that occasion.

This patient did disclose using cocaine, amphetamines, other opioids and benzodiazepine during the 12-week treatment period. Patient 5 had 15 positive urine samples for cocaine (not including DNAs) and one positive sample for amphetamines. Patient 5 had a positive urine sample for cocaine at the 16-week follow-up visit, and disclosed using alcohol, crack, cocaine, amphetamines, other opioids and diazepam at the 16-week follow-up visit (no additional follow-up data were available). Patient 5 took all the required doses of study medication in the first 2 weeks, missed three doses in week 3, missed two doses in week 4, took all doses in week 5, missed two doses in week 6, took all doses in weeks 7–9, did not take any doses in week 10 and took complete medication in weeks 11 and 12.

Patient 5 reported swelling at the implant site 3 weeks after the date of implant.

Neither the patient nor the clinician correctly guessed the patient’s treatment allocation.

Patient 6

Patient 6 was on double placebo, had 29 out of 36 negative urine samples for heroin, 6 out of 36 positive urine samples and one DNA (Figures 13 and 14; see also Figure 24). This gave a proportion of 0.81 negative urine samples. Patient 6 had a positive urine for heroin and cocaine at the first visit in week 1 but did not self-report substance use at this time. Their first heroin lapse form indicated that they first lapsed at the beginning of week 5. They did not have any positive urine samples in week 5, but self-reported using heroin on day 2 of this week. The reason they provided for this first lapse was to determine their treatment allocation. They did not take any other substances on that occasion.

FIGURE 13.

Patient 6’s urine screen results for heroin, where 0 = negative for heroin, 1 = positive urine sample for heroin. DNAs are shown as blanks.

FIGURE 14.

Patient 6’s urine screen results for heroin, where 0 = negative for heroin. DNAs are shown as positive urine samples.

In week 9 patient 6 had two positive urine samples for heroin – they did not report using heroin at the first visit but reported using heroin on day 6, which coincided with the second positive sample of that week. In week 10 the patient self-reported using heroin on days 3 and 7, but had a positive urine sample only at the third visit. Patient 6 continued using heroin until the end of the treatment period.

Patient 6 had 19 positive urine samples for cocaine during the 12-week treatment period. Patient 6 reported using crack, cannabis and alcohol in the follow-up period (week 16 and 24 data; week 36 data were not available) and had a positive urine sample for cocaine at week 16 (no data were available for weeks 24 and 36). Patient 6 took all their study medication for all 12 weeks.

Neither the patient nor the clinician correctly guessed the patient’s treatment allocation.

Tables displaying the urine screen results for heroin at each visit are given in Appendix 2. Urine screen results for substances other than heroin are given in Appendix 3.

Self-reported heroin use for weeks 1–12

Self-report of other substance use is displayed in Appendix 3.

Combining urine screen log results with self-report for heroin and other substance use

FIGURE 15.

Patient 1’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits (provided patient did not already have evidence of using heroin) are shown as blanks.

FIGURE 16.

Patient 1’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits are shown as positive for using heroin.

FIGURE 17.

Patient 2’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits (provided patient did not already have evidence of using heroin) are shown as blanks.

FIGURE 18.

Patient 2’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits are shown as positive for using heroin.

FIGURE 19.

Patient 3’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits (provided patient did not already have evidence of using heroin) are shown as blanks.

FIGURE 20.

Patient 3’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits are shown as positive for using heroin.

FIGURE 21.

Patient 4’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits (provided patient did not already have evidence of using heroin) are shown as blanks.

FIGURE 22.

Patient 4’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits are shown as positive for using heroin.

FIGURE 23.

Patient 5’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits (provided patient did not already have evidence of using heroin) are shown as blanks.

FIGURE 24.

Patient 5’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of each of the week’s three visits are shown as positive for using heroin.

FIGURE 25.

Patient 6’s heroin use results over the 12-week treatment period (combining urine screen and self-report), where 0 = did not use heroin, 1 = used heroin. DNAs in any of the week’s three visits are shown as positive for using heroin.

Substance use in follow-up

At each follow-up point, the questionnaire asked about substance use in the past 4 weeks.

First heroin lapse

Adherence to oral tablet

Full details of the oral dosing record are provided in Appendix 1.

Three of the participants took all of their oral study medication over the 12 weeks. All participants took all of their study medication for the first 2 weeks. One participant dropped to 250 mg at week 9 (takeaway dose) and did not take any of their study medication for weeks 10–12 (takeaway week 10 and DNA in weeks 11 and 12). One participant dropped to 250 mg at week 5, then 150 mg at week 6 and then did not take any study medication from week 7 onwards (takeaway week 7 and DNA in remaining weeks). One participant dropped to 200 mg at week 3 and 250 mg week 4, took all doses in week 5, dropped to 250 mg week 6, took all doses in weeks 7–9, missed week 10 and took all doses in weeks 11 and 12 (Figure 26).

FIGURE 26.

Total weekly dose received by each patient (mg).

Withdrawal

No patients withdrew from the trial.

Adverse events

All AEs are reported in Appendix 1. Here, only implant-related AEs are reported.

Patient treatment guess

Only one patient correctly guessed their treatment.

Clinician treatment guess

The clinician(s) correctly guessed only one patient’s allocation.

The CONSORT flow diagram

FIGURE 27.

The CONSORT flow diagram for the NEAT study. a, Breakdown of exclusions (prior to consent): not abstinent from all opioids for 7 days/not completing treatment for opioid detoxification (n = 1081); not seeking opioid antagonist treatment for opioid use disorder (n = 10); opiate dependence syndrome (n = 33); currently taking naltrexone therapy (n = 3); severe alcohol dependence/withdrawal (n = 4); current/recent suicidal ideation/plan/attempt (n = 2); does not live in stable/secure accommodation in the community (n = 1); current criminal justice involvement (n = 1); declined participation (n = 13); unable to consent (n = 1); other (n = 19). O-NTX, oral naltrexone; XR-TX, extended-release naltrexone.

Limitations

Firm conclusions cannot be reached from the observations on the small number of study participants who entered the study, but it is still possible to reach conclusions on some elements of the work planned and undertaken. Thus, although this does need to be acknowledged as a major limitation within this report, we are nevertheless able to reach tentative conclusions about some aspects and gain an understanding of the obstacles encountered. We are also currently conducting qualitative interviews with study participants recruited, patients who considered participation, clinical staff involved in the trial and key informants in the overall trial conduct, and so it will also be possible to report on the lived experience of participation in the conduct of the trial from the perspective of potential patients approached and interested in participation. This qualitative study comprises an investigation of the knowledge, attitudes and behaviour of the six study participants, or potential study participants, all involved clinical staff and other key individuals. These interviews will explore participants’ perspectives of the trial design and trial validity.

The double-blind, double-dummy trial design was recognised from the outset as new to the particular clinical field (the addictions treatment field) because it involved introducing this trial design to a treatment field that had not previously encountered it. It was also challenging in terms of the necessary pre-trial arrangements to produce the supply of active and placebo treatments. However, once established, it was broadly understood and acceptable to many potential study participants as well as to staff.

One area of particular challenge warrants more detailed description. Considerable problems were encountered with the stipulated requirement of a validated ‘detoxified’ status prior to the initiation of the study naltrexone, and also with the requirement for a ‘cooling-off’ period for the consent that had been provided by the participant for participation in the trial. A further area of challenge concerned the additional delay awaiting these surgical implant procedure. Ways around these obstacles were developed during the preparation and early stages of conduct of the trial, at least to some extent, which were able to mitigate the problem.

The possibility of direct recruitment of study participants on release from prison (because they represent a population with a particular likelihood of having achieved a state of general abstinence which, in view of their imminent release, might at least be considered fragile) was also explored. However, the operational aspects of securing organisational commitment to participation, alongside the lack of previous significant experience in trial participation, led to problems that could not be resolved in the time available to add within-prison participation to the NEAT trial processes. For future work, we recommend consideration of this potentially important patient population.

Consideration of the findings

From the findings described above, we select the following as warranting particular consideration.

We note the success of the double-blind procedure, with the finding of a remarkable failure on the part of participating patients, and also on the part of clinical staff, to identify correctly the random allocation. It was particularly noteworthy that this observation was applicable not only for the active implant and active oral, but also for the two study participants in the double-placebo condition.

It is noted that at least one patient for whom trial participation, and the provision of an antagonist blocking medication with unknown status as either active or placebo, appears to have triggered the first instance of heroin use in order to test the blockade and thereby learn whether they were in the placebo or active condition. This unintended consequence needs to be borne in mind when considering trial designs and their potential advantages and disadvantages.

We also note that, even though an opiate antagonist medication was being taken, there was self-report and urine analysis evidence of a considerable extent of other drug use (e.g. amphetamine, alcohol). Furthermore, there seemed to be evidence from self-report that these instances of use of other drugs often directly preceded instances of use of heroin/opiates.

It is also interesting to look at the clinical course of the six study participants on the basis of whether they received the active implant, active oral or placebo. There is evidence of observed benefit and good continuing recovery with patient 2 (active implant), whose the good progress continued into the follow-up period, and also of the continuing recovery with patient 4 (active oral), whose the good progress similarly continued into the follow-up period. In contrast, for two of the three patients who received double-placebo there was evidence of considerable drug use, including use of heroin.

Implications for future research

A potentially important observation is the difference between lapse and relapse – we suggest that this needs to be explored more carefully. For several study participants, instances of ‘lapse’ (i.e. instances of heroin use) occurred but were not necessarily followed by immediate clinical ‘relapse’. This is potentially a particularly interesting area when we consider the likely beneficial effect of naltrexone. Let us give this some consideration. Instances of ‘lapse’ might occur as harbingers of major relapse, or they may alternatively be isolated instances that have little bearing on longer-term outcomes – the difference is clearly important. Indeed, at least one of the study participants who participated in the NEAT trial specifically attributed their first episode of use to a wish to test their double-blind assignment and their presumed opiate blockade.

Following on from this consideration, it also needs to be borne in mind that it could reasonably be presumed that it is precisely this lapse-to-relapse likelihood that is the area in which naltrexone might be expected to deliver benefit (i.e. that naltrexone might reduce the likelihood that an instance of lapse might occur). If this is correct, then it has profound implications for the selection of study design and choice of outcome measures that should be deployed in future trials.

Overall conclusion

Lessons need to be learnt from the difficulties encountered by the NEAT trial team so as to ensure that these do not occur again (whether for the NEAT team members or for another group of investigators). Some of the problems encountered relate to the choice of trial elements and their conduct, whereas others relate to the severely disrupted and disturbed nature of the addictions treatment field at this time, particularly in the light of the introduction of aggressive competitive tendering processes, with their focus not only on value for money but also on baseline cost specifically, and also the move of commissioning out of health care into local authorities. We consider that, given wider variations in practice and operational uncertainties in the addictions treatment field, it is particularly important that attention is paid to reduce the potential for practitioner and observer bias. We also conclude that, on the basis of our modest experiences in the NEAT trial, a mixture of self-report, urine analysis and subsequent independent description of lapse episodes has blended value in the understanding of extent of resumption of heroin/opioid and other drug use. It is noted that it has been possible to produce placebo medications and to use these in a trial setting with good resulting blinding, and that specially produced medications can effectively be used in such a trial. There remains a need to assess the value of the new ultra-long-acting formulation of naltrexone (depot naltrexone or naltrexone implant) in a properly constructed randomised trial informed by a external pilot.

Contributions of authors

Professor Sir John Strang (Professor of Addiction Psychiatry) conceived and designed the trial, contributed to the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr Michael Kelleher (Consultant Psychiatrist) contributed to the design of the trial, the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr Soraya Mayet (Consultant Psychiatrist) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr Ed Day (Senior Clinical Lecturer in Addiction Psychiatry) contributed to the design of the trial, the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Ms Jennifer Hellier (Biostatistician) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Professor Sarah Byford (Professor of Health Economics) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Ms Caroline Murphy (Director, KCTU) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Mr Blair McLennan (Trial Manager) managed the trial, contributed to the acquisition, analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr James Shearer (Lecturer in Health Economics) contributed to the design of the trial, the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Dr Elizabeth Ryan (Biostatistician) contributed to the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Professor John Marsden (Professor of Addiction Psychology) conceived and designed the trial, contributed to the analysis and interpretation of the data, and drafted and critically reviewed the manuscript.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to available anonymised data may be granted following review.

Patient data

This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data is vital to improve health and care for everyone. There is huge potential to make better use of information from people’s patient records, to understand more about disease, develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone’s privacy, and it’s important that there are safeguards to make sure that it is stored and used responsibly. Everyone should be able to find out about how patient data are used. #datasaveslives You can find out more about the background to this citation here: https://understandingpatientdata.org.uk/data-citation.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the PHR programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the PHR programme or the Department of Health and Social Care.

Patient 1

Patient 1 had positive urine samples for other substances only in follow-up (after treatment period):

• At week 16, patient 1 was positive for benzodiazepine.

• At week 24, patient 1 was positive for heroin, cocaine and methadone.

• At week 36, patient 1 was positive for methadone.

Patient 2

No positive urine samples for other substances.

Patient 3

• Patient 3 had a positive urine sample for cocaine at the week-5 (sample 1) visit, week-8 (sample 3) visit and week-9 (sample 2) visit.

• They had a positive methadone urine sample at week 12, samples 2 and 3.

• At week 16, patient 3 was positive for heroin, cocaine, benzodiazepine and methadone.

• At week 24, patient 3 was positive for heroin, cocaine and benzodiazepine.

Patient 4

No positive urine samples for other substances.

Patient 5

Patient 5 had positive cocaine samples at the following visits:

• week 1, sample 3

• week 2, sample 1

• week 2, sample 2

• week 2, sample 3

• week 3, sample 2

• week 3, sample 3

• week 4, sample 1

• week 4, sample 2

• week 5, sample 1

• week 5, sample 3

• week 6, sample 2

• week 7, sample 1

• week 8, sample 1

• week 9, sample 2

• week 11, sample 1.

They were also positive for amphetamines at the week 5 (sample 3) visit.

Patient 5 was positive for cocaine at the week 16 visit; there were no data available for the week 24 and week 36 visits.

Patient 6

Patient 6 had positive samples for cocaine at the following visits:

• week 1, sample 1

• week 1, sample 2

• week 1, sample 3

• week 2, sample 2

• week 3, sample 1

• week 4, sample 1

• week 4, sample 2

• week 6, sample 1

• week 6, sample 2

• week 8, sample 2

• week 9, sample 1

• week 9, sample 3

• week 10, sample 1

• week 10, sample 2

• week 10, sample 3

• week 11, sample 3

• week 12, sample 1

• week 12, sample 2

• week 12, sample 3.

Patient 6 also had a positive sample for cocaine at week 16; there were no data available for weeks 24 and 36.