Impact of dextroamphetamine substitution on the use of illicit amphetamines in adults with amphetamine dependence: a study protocol for the multicentre double blind randomised controlled trial ATLAS4Dependence

Abstract

Introduction

There is limited evidence on how to effectively treat individuals from marginalised populations with dependence on amphetamine and/or methamphetamine (collectively referred to hereafter as amphetamine dependence). The disease burden is extremely high in this population, especially related to psychiatric comorbidities, cardiovascular complications, injection-related infections and poor social functioning. ATLAS4Dependence is a multi-centre randomised, placebo-controlled, double-blind trial that will investigate the effectiveness and safety of substitution treatment with dextroamphetamine compared with placebo in people with amphetamine dependence.

Methods and analysis

The trial will recruit 226 adult patients in several outpatient clinics in Norway. Inclusion criteria comprise individuals with amphetamine dependence, defined as use on three or more days per week during the past 28 days, who currently inject or have formerly injected drugs. This includes individuals both with and without comorbid opioid dependence, as well as those currently receiving or not receiving opioid agonist treatment. Participants will be randomly assigned 1:1 to receive either dextroamphetamine or placebo for 12 weeks. Flexible doses within the range of 30–120 mg daily will be provided based on individual assessments. The participants in both arms will be offered standard psychosocial and medical follow-up in accordance with current clinical practice. The endpoint assessments will be conducted at 12 weeks with weekly self-reports and safety assessments and a follow-up assessment at 52 weeks. The primary objective of the study is to assess the impact of 12 weeks daily prescribed oral dextroamphetamine versus placebo on the use of illicit amphetamines as well as on the total amount of amphetamines used (including both illicit and prescribed sources). Secondary outcomes are the differences between the groups at 12 weeks regarding psychological distress, symptoms of psychosis, quality of life, cardiovascular risk factors, injection-related infections, executive functioning, attention-deficit hyperactivity disorder-related symptoms, sleep, violence risk, fatigue, symptoms of craving and withdrawal, treatment retention, days of use of illicit amphetamines and use at 4 weeks and 8 weeks during the intervention period, use of other illicit substances and alcohol, as well as a cost-effectiveness analysis (using private economy, criminal activity and health service utilisation) and a qualitative approach to assess overall experiences with the study intervention. Analysis and reporting will follow the Consolidated Standards of Reporting Trials guidelines. All tests will be two-sided. Descriptive results and the estimated effectiveness will be presented with 95% CIs. The difference between the groups at the primary time point (at the end of the 12-week trial) will be assessed using χ² test (for use of illicit amphetamines measured by monthly urine tests) and Analysis of Covariance (ANCOVA) (for weekly self-reported total amount of amphetamines). Analyses for the primary endpoint will be undertaken on an intention-to-treat basis and reported on as such, but sensitivity analyses with per protocol analyses will also be presented.

Ethics and dissemination

The study is approved by European Medicines Agency, Clinical Trial Information System (CTIS). Written informed consent will be obtained from all patients. Study results will be published in international peer-reviewed medical journals.

Trial registration number

CTIS 2023-510404-44-00.

Introduction

Dependence on illicit amphetamine and/or methamphetamine (hereby referred to as amphetamine dependence) is one of the contributors to global burden of disease and health adversities such as infections, cardiovascular conditions and psychotic disorders in addition to increased mortality and impaired quality of life.^{1 2} An estimated 34 million people worldwide use amphetamines, and about 17 million are dependent on methamphetamine.³ Multiple and injecting substance use, homelessness and increased criminality and violent behaviour are some of the major characteristics of this patient group.^{1 3} The prevalence of amphetamine use among patients undergoing opioid agonist treatment (OAT) varies significantly between countries, but amphetamines remain a key concern, particularly in northern and eastern Europe and Australia. Around 8–24% of patients on OAT in the European Union report using amphetamines, with Sweden and Finland reporting a higher use (30–40%).⁴ National surveys show similar or even higher prevalence of about 30–45% among more than 8000 Norwegian patients undergoing OAT, with almost half of those having injecting drug use during the past year and about 15% using at least 3 days a week.^{5 6} Similarly, Australian research suggests about a quarter of patients entering OAT reported using amphetamines in the last 28 days.⁷ These patterns underscore the need for tailored approaches in managing amphetamine use among patients undergoing OAT and people who currently inject or have formerly injected drugs (PWID) regardless of OAT status. PWID is referred to as a vulnerable group with multiple drug dependencies who suffer from a high burden of disease with somatic and mental comorbidities. They are exposed to a high risk of harm and death. Most of the people in OAT belong to this population.^{3 6} In the present trial, the target group will be people with amphetamine dependence who currently inject or have formerly injected drugs including both those receiving or not receiving OAT.

Few randomised clinical trials have assessed amphetamine substitution treatment for patients with amphetamine dependence.8,10 Some systematic reviews suggest that withdrawal symptoms and use of illicit amphetamines might be reduced by such treatment.11,14 However, the results are inconsistent, and the researchers conclude the need for more evidence in this field using high-quality designs and higher dosages, larger sample sizes and longer intervention and follow-up periods. Importantly, studies in populations receiving OAT are lacking despite the high prevalence of amphetamine dependence and a high burden of disease in this group. Trials on amphetamine substitution treatment should also carefully assess safety due to the potential risks of overdose, psychosis, cardiovascular events or diversion of the prescribed medication. Based on existing research, the safety of such interventions seems acceptable8,14 although the effectiveness and safety within key groups such as people receiving OAT or PWID is uncertain.¹⁵

Clinical experiences indicate that lowering the thresholds for treatment access is crucial to reach the patients with severe amphetamine dependence. Requiring inpatient or long-term residential treatment is a barrier for many, compared with outpatient settings, with the latter providing an effective and long-term platform for treatment delivery, in line with the emphasis on recovery from substance use disorders. Thus, the design of the current trial combines low threshold for treatment access and comprehensive assessment of patient safety by integrating the added amphetamine substitution treatment within the existing outpatient OAT or other substance use care clinics and corresponding follow-up. This naturalistic approach and effectiveness design further increases the generalisability of the findings to everyday clinical practice.

Our trial proposes amphetamine substitution treatment given to patients with amphetamine dependence undergoing OAT or PWID not receiving OAT, using individually assessed higher daily doses (up to 120 mg) of dextroamphetamine (DXA) than what is recommended for the treatment of attention-deficit hyperactivity disorder (ADHD) in adults.¹⁶ Clinical and peer experience suggest that short-acting DXA is a more desired medication compared with long-acting preparations among patients with ADHD and comorbid amphetamine dependence. This can be explained by the rapid onset and shorter duration of its pharmacological effect, more like what is experienced from the use of street amphetamines. In contrast, the use of slow onset and longer acting amphetamine preparations (eg, lisdexamfetamine, depot formulations) may have lower risk of inappropriate use and be more compatible with the typical goals set for the treatment of ADHD without dependence on street amphetamines.¹⁰ Thus, the hypothesis of this study is that substitution treatment with DXA in people with amphetamine dependence may reduce the use of illicit amphetamines and the related harms.

Objectives

The overall aim of this study is to investigate the effectiveness and safety of substitution treatment with prescribed DXA added to the standard interventions ie, psychosocial approaches in patients with dependence to amphetamines including those undergoing OAT for comorbid opioid dependence or other PWID not undergoing OAT.

The primary objective of the study is to assess the impact of 12 weeks’ daily prescribed oral DXA versus placebo (PBO) on the use of illicit amphetamines and the total amount used.

The secondary objectives are to assess the impact of DXA at 12 weeks on mental and physical health, including psychological distress, risk of psychosis and violence, suicidality, ADHD symptoms, sleep, executive functioning, fatigue, health-related quality of life, cardiovascular risk and injection-related infections, treatment retention, use of illicit and total amphetamines at 4 weeks and 8 weeks, use of other illicit substances or alcohol, symptoms of craving and withdrawal and cost-effectiveness of the treatment (regarding criminal activity, private economy and health service utilisation). Additionally, we will use a qualitative approach to assess participants’ overall experiences with the study intervention.

Methods and analysis

Trial design

This project is an investigator-led multicentre randomised, PBO-controlled, double blind, parallel group trial without pharmaceutical industry funding (figure 1).

Figure 1. Study design. DXA: Dextroamphetamine; PBO: Placebo.

Settings and participants

This study will be conducted in outpatient OAT clinics in four Norwegian cities; Bergen (Vestland County), Tønsberg (Vestfold County), Stavanger (Rogaland County) and Alta (Finnmark County). Potential participants will include the patients with amphetamine dependence, including those who are undergoing OAT or other PWID not receiving OAT. If potentially eligible, they will be invited to attend a formal screening by a research nurse and the final eligibility assessment by a physician in line with good clinical practice and the trial protocol (online supplemental file 2).

The Department of Addiction Medicine, Haukeland University Hospital (Vestland county, city of Bergen, Norway), as the sponsor and coordinating site, is responsible for the treatment and follow-up of more than 1000 patients with opioid dependence receiving OAT, of which almost 60% receive buprenorphine preparations, while the remaining receive methadone or, more seldom, long-lasting morphine formulations. All medical interventions are integrated with psychosocial care provided in multidisciplinary outpatient clinics as part of specialist healthcare system. The outpatient OAT clinics are staffed by physicians specialised in addiction medicine in addition to nurses, social workers and psychologists. Dependent on the overall functioning level and decisional capacity, the follow-up of the patients ranges from directly observed treatment and consultations to weekly take-home doses. All the clinical measurements and laboratory data will be stored in the hospital medical record system. A mainly similar OAT model but with some variations in organisation is also applied in the three other participating sites as part of specialist healthcare system in Norway. The county of Finnmark represents a geographically challenging region for the participants to reach the involved clinic due to longer distances. However, this may be considered beneficial for the feasibility of the intervention regarding the use of different delivery platforms. Totally, these additional sites include more than one thousand patients with similar clinical and sociodemographic characteristics as the main site.

PWID with amphetamine dependence not receiving OAT are offered multidisciplinary approaches in specialised outpatients and inpatients clinic at the Department of Addiction Medicine in the city of Bergen, as well as the other participating sites. Many of these patients are also followed by the municipality care centres or other low-threshold facilities in close collaboration with peer groups and voluntary organisations.

Patients with a problematic and long-lasting amphetamine use receiving OAT or PWID not receiving OAT from the included outpatient clinics or municipality care centres will be considered the reference target population and informed about the study by the caregiving staff. Participants will be recruited regardless of motivation to change drug use habits. Those potentially eligible will be registered in a prescreening list by the staff in collaboration with a research nurse. The consent process will involve summarising benefits and risks of study participation (online supplemental file 4). Those giving written informed consent will be further screened with extended clinical and laboratory assessments as specified in the schedule of activities (table 1), conducted by a research nurse and a physician at the related OAT clinic or other participating care centres. The screening step will involve assessment of the inclusion and exclusion criteria for study participation. Those who meet the inclusion criteria and do not have any of the exclusion criteria will go through baseline assessments and then be randomised to one of the two study arms (figure 2).

Table 1. Study plan and procedures during the study period.

Procedure stages (weeks)	Screening- (1–3)	Baseline (4–5)	Dose titration (6)	12-week trial (7–18)	Endpoint (19)	Dose tapering (20–21)	Safety follow-up (22–25)	52 weeks assessment
Informed consent (RN/D)	X
Assess potential eligibility (RN/D)	X
Inclusion/exclusion criteria (D)	X
Current medical conditions (D)	X
Physical examination (D)	X							(X)
Vital signs: BP, P, T, BMI (RN)	X			X X	X		(X)	X
12-lead ECG (RN)	X			X X	X		(X)	X
Laboratory blood tests* (RN)	X			X	X		(X)	X
Substance use history (RN/D)	X
Demography (RN/D)	X
Pregnancy urinary test (RN)		X		X X	X		(X)	(X)
Urine samples (RN)		X		X X X	X
Saliva samples (RN)		X		X X	X
Hair samples (RN)		X			X
Randomisation (TP)		X
Study interventions (RN/CS)			X	X X X X X X X X X X X X	X	X
Mental health symptoms (RN)		X			X			X
Health related quality of life (RN)		X			X			X
Treatment satisfaction (RN)		X			X			X
Treatment retention (RN)		X			X			X
Injection-related infection (RN)		X			X			X
Fatigue (RN)		X			X			X
Cognitive function (RN)		X			X			X
Cognitive function (RN) if indicated	(X)
ADHD screening (RN)		X			X			X
Cost-effectiveness† (RN)		X			X			X
Sleep (RN)		X			X			X
Agitation and violence risk‡ (RN)	X	(X)		X X	X		(X)	X
Psychosis risk§ (RN)	X	(X)		X X	X		(X)	X
Suicidality¶ (RN)	X	(X)			X		(X)	X
AE/SAE review (RN/CS)		X	X	X X X X X X X X X X X X	X	X	X
Co-medication review (RN/CS)		X	X	X X X X X X X X X X X X	X	X	X
Self-reported drug use (RN/CS)		X	X	X X X X X X X X X X X X	X	X	X	X
Craving and withdrawals (RN/CS)		X	X	X X X X	X	X		X

X: an 'X' indicates that the procedure will be performed, and the number of Xs denotes how many times the procedure will be conducted during the specified period.

()The use of parentheses signifies that the procedure is conditional on clinical indication.

^*Blood test for the assessment of inclusion/exclusion criteria at screening, safety and pharmacological analyses.

^†Cost-effectiveness assessments using private economy, criminal activities and health service utilisation.

^‡Agitation risk will be assessed for inclusion/exclusion criteria at screening, as a secondary outcome at baseline and at the endpoint, and as a safety measure every 4–5 weeks.

^§Psychosis risk will be assessed for inclusion/exclusion criteria at screening, as a secondary outcome at baseline and at the endpoint, and as a safety measure every 4–5 weeks.

^¶Suicidality risk for inclusion/exclusion criteria at screening, as a secondary outcome at baseline and at the endpoint.

ADHD, attention deficit hyperactivity disorder; AEs/SAEs, adverse events/serious adverse events; BMI, body mass index; BP, blood pressure; CS, clinical staff; D, doctor; P, pulse; RCT, randomised controlled trial; RN, research nurse/other trained clinical staff; T, temperature; TP, third part.

Figure 2. The flow chart of the study procedures. DXA: Dextroamphetamine; PBO: Placebo.

Inclusion criteria

The eligibility will be determined by clinical assessment with a medical doctor.

Participants are eligible to be included in the study only if all the following criteria apply:

• Being aged ≥18 years and ≤65 years at the time of signing the informed consent.

• Being overtly healthy as determined by medical evaluation (see exclusion criteria below).

• Being dependent on amphetamines according to International Classification of Diseases, 10th revision (ICD-10) with an average frequency of use of ≥3 days a week during the last 28 days.

• Being concurrently dependent on opioids according to ICD-10, receiving a stable dose of OAT medication the last 28 days (no dose change) or being among PWID not receiving OAT (formerly or currently injecting drugs regardless of frequency).

• If being female in fertile age, using contraceptive consistent with local guidelines regarding the methods of contraception for those participating in clinical studies (non-pregnancy will be confirmed by human choriogonadotropin test if indicated).

• Willing to give signed informed consent.

Exclusion criteria

• Any contraindications for the use of DXA¹⁶ including:

  • Concomitant use (or less than 14 days since the last use) of monoamine oxidase inhibitors.

  • Symptomatic cardiovascular diseases or related risk factors (Corrected QT Interval (QTc)≥500 ms, persistent moderate to severe hypertension, ie, uncontrolled blood pressure≥140/90 mm Hg not responded to a single antihypertensive agent, prior myocardial infarction or stroke, advanced atherosclerosis).

  • Glaucoma.

  • Diagnosed hyperthyroidism or biochemical profiles indicating it.

  • Moderate to severe renal failure (ie, glomerular filtration rate <60 mL/min/1.73 m²).

• Hypersensitivity to any ingredients in the interventional medical product or the PBO.

• Already receiving prescribed centrally acting stimulants (eg, for ADHD)

• Concurrent participation in other clinical studies when undergoing medication interventions.

• Having at least three substance-related overdose episodes during the past 3 months (F10.0–19.0) based on medical records.

• Having started assessment or treatment during the last 6 months for a psychosis spectrum disorder (ICD-10 diagnoses F20–29), substance induced psychosis (F10.5–F19.5) or affective psychosis (F30, F31, F32.3) based on medical records or having a score of ≥4 on the following psychosis symptoms items using the Positive and Negative Syndrome Scale (PANSS) instrument¹⁷: P1 (delusions), P3 (hallucinatory behaviour), A9 (unusual thought content) based on the interview conducted at the time of screening.

• Having/suspecting seizures during the last 6 months.

• Having or having had a high risk of agitation and violent behaviour during the past 6 months based on medical records or having a score of ≥4 on the following psychosis symptoms items using the PANSS instrument¹⁷: P4 (agitation) and P7 (hostility).

• Having an intellectual disability based on clinical assessments or using the Hayes Ability Screening Index instrument if indicated.¹⁸

• Having or having had suicidal plans or attempts during the past 3 months based on self-reports and medical records.

• Challenges related to the ability to understand, consent or willingness to collaborate in follow-up of the study and its protocol.

• Inability to complete study procedures (eg, travel plans or planned incarceration)

• Current pregnancy or breastfeeding.

Randomisation and blinding

All the participants (n=226) will be randomised 1:1 (blocked, site-stratified randomisation) at baseline for the 12-week trial period. Each participant will be assigned to receive either DXA or PBO in this period (figure 1). On commencing the study, participants will be assigned a unique number (randomisation number) in an ascending numerical order at each study site. The randomisation number encodes the participant’s assignment to one of the two study arms.

All the participants, care providers, investigators and analysts are blinded to study intervention. To maintain this blind, an otherwise uninvolved third party will be responsible for supply of study medication to the participating clinics. However, the medication will be further dispensed to the participants by blinded research nurses or other clinical staff members. PBO will be identically labelled as the active medication. In the event of a quality assurance audit, the auditors will be allowed access to unblinded study intervention records at the sites to verify that randomisation and dispensing have been conducted accurately.

In case of an emergency or a severe adverse event (SAE), the investigator can determine if unblinding of a participant’s intervention assignment is warranted. Participant safety will always be the first consideration in making such a decision. The date and reason for the unblinding will be recorded.

Study medications

The study-coated versions of the medications are not commercially available and will be produced as a special order from the sponsor (Department of Addiction Medicine, Haukeland University Hospital, Bergen, Norway). Each active tablet will contain DXA 30 mg, which will be ingested orally to provide the total prescribed study dosage. Trial medications will be obtained, prepacked in blister packs, stored and transported to participating sites by a contracted good manufacturing practice registered facility (Kragerø Tablettproduksjon AS) in accordance with the current regulations and requirements. The active drug and the PBO will be provided in identical tablets to ensure the study blindness. The medication packages will be delivered with a unique number encoded at the producing manufacture containing predefined dosages/tablet counts as for the two trial groups. Blinding and the linking code between each participant and the package code will be performed by an independent party and held at a secure database.

Interventions

Enrolment was started in the autumn of 2024 and will be completed when all the required and eligible participants are recruited to the trial (planned within 2027). Eligible participants will be randomised to one of the two trial’s arms. The intervention is substitution treatment with an individually assessed flexible dose of 30–120 mg/d DXA provided for 12 weeks. Participants in the control arm will receive 1–4 tablets with identical appearance containing PBO for the same duration of 12 weeks, with similar assessments to determine tablet counts. The initial dosage decision will be based on the degree of dependence, the dose of illicit amphetamines used prior to study entrance and the individual’s clinical condition. Given that street amphetamines can usually contain about 40% pure amphetamine (with a range of 5–67%),⁴ the participants reporting to use at least 500 mg street amphetamines as a typical daily dose during the past 28 days can be considered to receive the maximum daily dosage of 120 mg DXA after a 1-week titration period based on the patient’s clinical condition, for example, the presence of signs or symptoms such as withdrawals, craving, intoxication or adverse effects. Similarly, the decision-making regarding the lower daily dosages of 90 mg, 60 mg or 30 mg will be based on the dose of street amphetamines currently used and clinical symptoms. Accordingly, equal numbers of participants will receive DXA and PBO oral tablets during the 12-week intervention period (figure 1). Dose titration (30 mg at each step up to desired dose) will be conducted on the first trial week after completing the baseline assessments and measurements according to table 1 for all the participants (online supplemental table s1). Similarly, a dose tapering will be conducted at the end of the 12-week trial for all the participants (online supplemental table s2).

For both arms, treatment initiation and follow-up will be conducted at the outpatient clinics where the participants undergoing OAT already receive standard treatment and relevant care including voluntary psychosocial interventions. The participants not receiving OAT may be referred to a participating OAT clinic or other involved care units to start the intervention and receive further follow-up. Two-thirds of the daily dose will usually be taken in the early morning and one third about 6–8 hours after the first dose. In the case of using 30 mg/day or only one tablet/day, the whole dose will be taken in the morning.

The morning dose will usually be taken under the observation of the research nurse, OAT clinic staff or other health workers especially during the dose titration period. The frequency of observed dose intake will be assessed individually, but at least once a week. The duration of observation is almost 1 hour in this period until the stabilisation is achieved (approximately 1 week from the start of medication). The afternoon dose can be self-administered at home.

The participants will receive most of the doses observed by the staff at outpatient clinics or other care centres. Take-home doses will be delivered in zip-locked bags and will be administered orally by the participants. Individually assessed, the frequency of observed and take-home doses will follow the same agreement that applies to OAT medications (for those undergoing OAT) and any changes will be assessed by the prescribing physician in agreement with the trial investigator. In any case, all participants should attend the outpatient clinic at least weekly for clinical observations. Every participant should, regardless of the delivery agreement, have daily or frequently observed intakes for the first 1–2 weeks after starting the treatment (in both study arms) to ensure treatment safety. After this period, they follow the delivery agreement for OAT as before or based on individual assessments. Some patients will receive OAT through pharmacies and will also need to attend the OAT outpatient clinic or the pharmacy at least once a week to be given DXA/PBO during the project period. The take-home dosages will typically be stored by the patient for 1–2 days, exceptionally longer but not for longer than 7 days (the maximum take-home tablet count may be four tablets per day for 7 days (ie, 28 tablets). A participant who is receiving three tablets a day and usually meeting OAT clinic 3 days a week (typically Monday, Wednesday and Friday), will, in addition to six tablets taken during observation, be dispensed a total of 15 tablets as the take-home dosage (to be used on Tuesday, Thursday, Saturday and Sunday and as afternoon doses on Monday, Wednesday and Friday).

Clinical and biological follow-up of participants and systematic report of potential adverse effects will be organised according to international good clinical practice (GCP) guidelines. Participants will be followed up by staff in outpatient clinics during regular visits and, if needed, by home-based visits. The participants will be assigned to the scheduled assessments by the research team in line with the study protocol (online supplemental file 2). Assessment will be performed at baseline and at 12 weeks in addition to a follow-up assessment 52 weeks post randomisation. The clinical staff and/or research nurses will follow-up the participants at least once a week by consultations including self-reports on adherence to the study medication, the use of illicit amphetamines and other substances, adverse events and other safety measurements, as well as monthly urinary tests during the trial period. Following the tapering period and discontinuation of the study medication, all participants will be followed up for an additional period of 4 weeks, with safety assessments conducted if medically indicated (table 1; figures1 2).

Sampling of biological material

Urinary testing for the analysis of illicit amphetamines and other substances or alcohol will take place at the OAT clinics or other planned units under the observation of a staff member. The samples will be collected randomly in one of the 4 weeks each month, and one extra sample will be obtained at the end of the 12-week trial (4 samples in total for the primary outcome measure). Randomisation will be programmed by an independent statistician and is aimed to reduce the effect of any variation in the intake of illicit amphetamines between the intervention and the standard arm in relation to the time of sampling. The patients will be informed on the given Monday of the week that they need to provide an observed urinary test before Friday of the given week. In the case of a no-show for urinary sampling, the patient will receive a warning to deliver a sample the subsequent week (up to once). In the event of a missing urine sample or two delays, study medication will be stopped.

The analytical method is based on an enantiomeric high-performance liquid chromatography-tandem mass spectrometry technology, enabling us to distinguish between the various amphetamine types as well as the prescribed DXA from racemic amphetamine acquired illicitly.¹⁹

It is also aimed to investigate the use of various biological matrices to monitor both DXA concentrations (blood samples for pharmacological purposes) and the intake of other substances, including racemic (illicit) amphetamine, alcohol, opioids, benzodiazepines, cannabis and other stimulants. Analysing these compounds alongside DXA may help identify differences in substance use patterns and provide valuable insight into DXA’s pharmacokinetics and concentrationeffect relationships in the study population. Furthermore, employing alternative matrices, that is, saliva and hair in addition to urine and blood, offers a less invasive and more tamper-resistant approach to sample collection and may enable a longer window of detection. If urine collection is not viable, saliva can be used as an alternative specimen (table 1).

The blood samples obtained for safety purposes will be used for the measurement of haemoglobin, Mean Corpuscular Volume (MCV), erythrocytes, leucocytes with diff, thrombocytes, Thyroid-Stimulating Hormone (TSH), Free Thyroxine (FT4), folate, creatinine, Glomerular Filtration Rate (GFR), glycated hemoglobin (HBA1c), Alanine Aminotransferase (ALAT), Aspartate Aminotransferase (ASAT), cholesterol, total Low-Density Lipoprotein (LDL) and High-Density Lipoprotein (HDL). Also, Hepatitis C Virus (HCV) RNA, Hepatitis B Virus (HBV) and HIV will be investigated in the blood samples for the injected-related infections as one of the secondary outcomes. The frequency of the samplings during the trial period is presented in table 1.

Adverse events

Potential adverse events will be managed according to standard treatment guidelines.²⁰ Some adverse effects and minor toxicity are expected with the applied intervention medication. Complete lists of the reported adverse effects described in the summaries of product characteristics for the DXA medicinal products available in Norway will be used as the reference safety information.¹⁶ All SAEs reported to the sponsor will be assessed against the reference safety information to consider whether the event is unexpected (suspected unexpected serious adverse reaction, SUSAR) or not. Regular clinical observations by the clinical staff and physicians will secure prompt identification of potential adverse events. Any observed side effects or suspicious clinical conditions such as symptoms of agitation, psychosis, chest pain or intoxication will promptly be reported to the responsible clinic physicians and the local and national trial investigators. Emergency intoxication care will be available at the involved clinics, and further transport to an emergency unit will be secured when close and continuous clinical monitoring is needed. Clinical and biological safety will be assessed according to the standard adults scaling instrument to grade the severity of adverse events.²¹ All unexplained grade III or IV events will lead to temporary interruption of the study medication before a new assessment is made by clinic physician and study investigators. A rapid report system for the management of SAEs and SUSARs will be available.

Stop criteria

The stop criteria for the individual participant are based on non-compliance to the study protocol especially non-attendance for dosing, providing urinary tests and other critical study appointments such as endpoint assessments and unexpected adverse events. Medication can be withheld in the event of missed urinary tests as described above. Participants who have discontinued protocol treatment will be motivated to continue to participate in all remaining research interviews and assessments. For those participants who revoke their consent for the entire study, no further data will be collected. At the end of the trial period, each participant will be individually assessed to receive further clinical follow-up as needed.

Outcomes

Primary outcome measures at the end of the 12-week intervention will include:

The differences between the intervention (DXA) and control (PBO) arm during the 12 weeks (cumulative data) regarding:

1. The proportion of urinary tests positive on illicit amphetamines.

2. The mean total self-reported quantity (mg/day) of amphetamines (dosage of illicit amphetamines will be converted to prescribed dexamphetamine dosage and added to the prescribed dosage) used.

Secondary outcome measures include the differences between the study arms assessed at the end of the trial period regarding:

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The double-blinding of the trial and use of placebo will increase the reliability of the results.

• The use of a biological primary outcome will ensure an objective measure of difference.

• The proposed sample size is powered to detect a medium effect size in the difference between the groups.

• A close follow-up of participants will reduce the dropout risk and increase the safety.

• Not fully observing the dose intake and infrequent urine sampling may be some of the limitations of this study.

• Psychological distress (measured using the Hopkins Symptom Check list with 10 items).²²

• Psychosis symptoms (measured with items P1, P3 and A9 on PANSS¹⁷/Launay-Slade Hallucination Scale-Extended).²³

• Suicidality (measured using semistructured questionnaire based on self-reports and medical records).

• ADHD-related symptoms (measured using Adult ADHD Self-Report Scale V.1.1).²⁴

• Health-related quality of life (measured using the European Five Dimension and Five Level tool from Euro QoL).²⁵

• Executive performance (measured using Trail Making Test B).²⁶

• Fatigue (measured using Fatigue Symptom Scale with three items).²⁷

• Treatment satisfaction (measured with a Visual Analogue Scale).

• Treatment retention (measured with days of drop-out relative to days in treatment).

• The use of illicit amphetamines (measured based on proportion of positive urinary tests) and the total quantity of amphetamines used (measured based on self-reports and DXA dosages used) at 4 weeks and 8 weeks during the study interventions.

• Days of use of illicit amphetamines (measured based on self-reports).

• Craving and withdrawals to amphetamines (measured based on self-reports and Amphetamine Withdrawal Questionnaire).²⁸

• Days of use and proportion of other illicit substances and alcohol use (measured based on self-reports and urinary tests, respectively).

• Injection-related infections (measured based on self-reports, lab tests and medical records).

• Violent behaviour (measured using Brøseth Violent Checklist²⁹ and PANSS items P4 and P7¹⁷).

• Quality of sleep (measured using Insomnia Severity Index).³⁰

• Cost-effectiveness assessments (using private economy,³¹ criminal activity and health service utilisation)

In addition, we will use a qualitative approach to explore underlying reasons for study dropout, identify facilitators and barriers to participation and assess overall experiences with the study intervention.

We will conduct individual interviews with approximately 30 participants, including those who completed the intervention, those who discontinued and those who declined participation. These interviews are intended to provide indepth, rich and nuanced insights into individual experiences and rationales. A sample of approximately 10 participants per group is considered sufficient to ensure analytic depth. An interview guide will be developed by the responsible investigator in collaboration with the peer group involved in the project. After obtaining informed consent, the patients will be recruited and interviewed at the participating clinics where the study is being conducted.

Data collection and management

Data will be collected according to GCP. All participating clinics will use electronic case report forms (CRFs) that will be entered online in a study database through the data collection software (Viedoc). For exceptions such as home visits or internet connection abruption, temporary paper-based CRFs could be used and later entered into the electronic CRF. This software is accredited by Haukeland University Hospital in Bergen for health research and clinical integration in Norway. A central data manager at this centre will assist in designing the CRF, train data collectors in use of the CRF and aid data export. Only authorised study personnel, including the principal investigators, research nurses and coordinating research staff, will have access to CRFs and supporting documents. Data capture and storage will be undertaken using computer systems compliant to GCP. To ensure correct operation according to standard operating procedures, all system users will be trained and evaluated on a regular basis. The research data will be stored in an encrypted data server for research with access limited to the principal investigators and coordinating research nurse.

Related authorised clinical study-monitoring organs will visit the study sites on a regular basis to ensure proper handling of informed consent procedures, reporting of relevant safety data including adverse events, adherence to protocol, maintenance of required regulatory documents, facilities and data completion on the CRFs including source data verification. Additionally, a data monitoring committee (DMC) comprising two independent clinicians and researchers, and a statistician will ensure the safety and well-being of the trial participants and will assist and advise the coordinating and principal investigators to protect the validity and credibility of the trial.

Statistical methods

Sample size is estimated based on clinical experiences and published randomised controlled trials.8,14 Previous analysis of the Bergen OAT cohort (n=1060) shows that around 40% used amphetamines regularly, with around 15% using ≥3 days a week.^{5 6} We have calculated the range of sample size estimates (power 90%, two-sided alpha error of 2.5%) by using low, medium and high levels of variations in effects (SD: 0.1, 0.2 and 0.3, respectively).

Due to the nature of the disease and clinical experience, we expect the group to be heterogeneous and that there may be variations in effects. If the expected effects are >0.1, that is, at least 10% lower use of illicit amphetamines in the intervention arm compared with the control arm, all three scenarios will have sufficient power with n=222. However, the sample size of 222 must be adjusted to 226 to include the number of patients required for interim analysis.

The total self-reported quantity of amphetamines used will be the dosage of illicit amphetamines converted to prescribed DXA (by a conversion ratio of 0.3) and added to the dosage of the prescribed DXA. Street amphetamines contain on average 40% pure racemic amphetamine being equal to about 25–30% DXA.⁴

We aim to have less than 10% loss-to-follow-up to reduce the risk of selection bias. This seems realistic based on experience from similar trials and considering the strong motivation patients have for seeking care and showing up regularly for existing treatment in the participating clinics. In addition, we expect the effectiveness to be higher than 0.1 and therefore a sample size of 226 will be sufficient, even with extreme variations in the effects³² (online supplemental file 1, figures s1a and b).

Analysis and reporting will follow the Consolidated Standards of Reporting Trials guidelines.³³ All tests will be two-sided. Descriptive results and the estimated effectiveness will be presented with 95% CIs. Categorical variables will be summarised as percentages and continuous variables as medians with IQRs or means with SD for variables with a Gaussian distribution. Analyses for the primary endpoint will be undertaken on an intention-to-treat basis and reported on as such, but sensitivity analyses with per protocol analyses will also be presented. All the randomised and eligible participants will be analysed based on initial group allocation. Participants who withdraw from the study treatment will not be censored, as treatment discontinuation is likely to be related to allocation. Deaths will be censored at the last outcome measurement.

The primary endpoint is the difference between the DXA (n=113) and PBO (n=113) groups at the end of the 12-week trial regarding:

1. The intake of illicit amphetamines which measures the use of illicit amphetamines (as assessed by monthly supervised urinary tests³⁴) during the 12-week trial. The primary endpoint is therefore defined as the proportion of urine samples positive on illicit amphetamines at the end of the 12-week trial showing either ≤50% use of illicit amphetamines or not (dichotomous outcome measure).

2. The index for total weekly self-reported quantity (mg/day) used, which is defined as the total dosage of prescribed and illicit amphetamines (converted to prescribed DXA) during the 12-week trial and is considered continuous.

The difference between the groups at the primary time point (at the end of the 12-week trial) will be assessed using χ² test (for use of illicit amphetamines measured by monthly urine tests) and Analysis of Covariance (ANCOVA) (for weekly self-reported total amount of amphetamines). Risk difference with 95% CIs will be reported.

Analyses for the secondary outcomes will also be conducted with intention-to-treat analysis, as well as per protocol analyses for sensitivity purposes. We will compare the two study groups at the end of the 12-week trial. Depending on the type of outcomes, appropriate analysis methods will be used including: ANCOVA, t-test, cross table, risk difference and OR with CIs as well as the χ² test, asymptotic or exact depending on the marginal distributions. Additionally, we will estimate a linear mixed effects model for the continuous variables for the outcome depending on time, randomisation group and the interaction between time and randomisation group, adjusted by using an individual random intercept. For all regression models, we will present the regression table; for the linear mixed effects model, we will provide the graphics with mean and 95% CI for the follow-up time point. For the dichotomous variables, we will provide a bar plot of the proportions for each available time point. We will also have a sensitivity analysis using the method last observation carried forward, that is, if a urinary test is missing by random, the last recorded value is imputed at future time points where it is missing. The results will be interpreted considering the linear mixed effects to reduce the risk of bias. Robustness of the primary outcome will be checked with sensitivity analyses considering censoring and adjusting for potential baseline imbalances.  

Subgroup analyses for the primary endpoint will be made to assess the consistency of the investigational intervention effect across the subgroups such as:

• Age groups.

• Females versus males.

• Having vs not having comorbid opioid dependence.

• Receiving vs not receiving OAT.

• The type of OAT medication: methadone, buprenorphine and other prescribed or non-prescribed strong opioid agonists.

• Having versus not having a recorded ADHD diagnosis.

• Receiving versus not receiving prescription benzodiazepines.

If the number of participants in a subgroup is too small (less than 10% of the total study population), the subgroup categories may be redefined prior to unblinding the study. The analyses will be conducted using a test for heterogeneity and results will be presented in forest plots displaying the estimated study arm difference and 95% CIs.

The cost-effectiveness analysis in this study will adopt a prospective economic evaluation approach. To measure effectiveness, we will use quality-adjusted life years, drawing on primary and secondary results in conjunction with a prospective Markov model. The collection of cost data will be conducted alongside the clinical trial, encompassing all pertinent costs linked to the intervention and its outcomes. This includes both direct healthcare costs and indirect costs borne by patients and society.

All patients enrolled in the study will be evaluated with respect to safety-related outcomes including adverse events and SAEs. Safety data will be analysed according to the treatment that the patients receive. Safety analyses will include summaries of the incidence of all adverse events and SAEs that are possibly or probably related to the study intervention and occur during the study treatment period or within 30 days of the last dose of study treatment. Safety analysis will be specified by DMC. 

An interim analysis for effectiveness of the primary endpoint will be done when 50% of planned sample size is assessed at the end of 12-week trial. We will use a group sequential design without futility and the O’Brien-Fleming alpha spending approach. The interim analysis will be performed by the DMC. The DMC will give a recommendation on whether the study should be continued or stopped. Even if not in the model, the DMC can recommend stopping the study for futility.³⁵ 

Ethics and dissemination

The study is approved by European Medicines Agency, Clinical Trial Information System (CTIS), Regional Ethical Committees (REK KULMU, reference number CTIS 2023-508291-11-00). This randomised controlled trial will be conducted in strict accordance with the protocol, the Declaration of Helsinki and other international conventions (such as Convention of the Council of Europe on Human Rights and Biomedicine, UN Convention on the Rights of the Child, Universal Declaration on the human genome and human rights), with GCP and good laboratory practice standards^{20 36} and with current national legislations and regulations. Regarding informed consent and assent, the necessary information on the study will be provided to the participants at the recruiting visits. Data collection, clinical interviews and biological samplings will be conducted by the trained research and clinical staff and medical professionals to minimise the risk of any complications. All the investigators and involved staff have been GCP trained in addition to further training and follow-up of the data collectors to ensure that those concerned are competent to execute the trial protocol (online supplemental file 2). Should serious concerns arise, the trial steering committee will discuss termination of the trial with the sponsor. The participants who complete the trial assessments will receive a voucher with a value equal to €85 at the end of the 12-week trial.

The results from the study will be published in articles through peer-reviewed scientific journals in line with the recommendations by the International Committee of Medical Journal Editors guidelines.³⁷ Open access journals indexed in PubMed/Medline will be opted for (or will be paid for if this is not default in the journal).

Patient involvement

The project has a strong user involvement and close collaboration with the peer organisation proLAR Nett (http://prolar.no) and The Norwegian Association for Humane Drug Policies, both of which represent a significant number of patients on OAT and PWID, including those living with amphetamine dependence. Each site will have the possibility to involve local user representatives. This collaboration will contribute to a broader understanding of the patients’ multimorbid conditions and treatment needs and will enable us in more targeted planning of future research and establishment of more holistic treatment approaches for this population. Representatives from the involved user organisations will also play an active role in the development of an interview guide for the qualitative component of the study.

Discussion

To the best of our knowledge, this is the first randomised controlled trial of DXA substitution treatment for amphetamine dependence within the context of OAT and PWID combined. We will investigate the effectiveness of substitution treatment with individually assessed flexible doses of DXA compared with PBO and accordingly the well-being of patients with amphetamine dependence including those undergoing OAT and PWID not receiving OAT. The research project will provide knowledge on the impact of such intervention on patient outcomes. We will also conduct a cost-effectiveness analysis of the trial using the collected data on private economy, criminal activity and health service utilisation, as well as a qualitative substudy to investigate the overall experiences with the study intervention.

Our trial involves some limitations and several strengths. Some of the strengths include that the trial is PBO-controlled and blinded for participants, investigators, clinicians and analysts; it is funded by public sources to ensure independence, and it will use a biological primary outcome to ensure objective measure of the differences. Thus, substantial information biases are considered unlikely. Additionally, the study is individually randomised, which minimises potential confounding. Another strength of this study is its sample size, powered to detect a medium effect size in the difference between groups for the primary outcome using a flexible dose of the substitution medication. Participants will be closely monitored with weekly outpatient visits where they already receive usual treatment and follow-up. This will contribute to increased retention and safety of the study. The study protocol (online supplemental file 2) proposes to deliver the medication in an outpatient setting, allowing participants’ take-home dosing to mimic the service delivery situations in which the medication may be used, ensuring a relatively naturalistic design and increasing generalisability. Not fully observed daily dosing to confirm the doses taken, possible risk of drop-out, using several questionnaires and measurements and relatively infrequent urinary sampling may be some of the trial limitations. However, as this study also includes a qualitative component, this may complement the quantitative data, thereby contributing to a more comprehensive understanding of facilitators and barriers to amphetamine substitution and participants’ experiences of this intervention.

If the intervention is found to be safe and efficacious, it may potentially influence the standard of clinical practice in the treatment of dependence to amphetamines in marginalised populations.

Trial status

Recruiting started in September 2024. The current protocol is version 3.0 of 14-8-2025 (Supplementary file 2). Currently (14th of October 2025), we have included 25 patients. Patient recruitment is estimated to be completed around December 2027.

Reporting checklist for protocol of the clinical trial

We have used the SPIRIT reporting guidelines (Chan A-W, Tetzlaff JM, Gøtzsche PC, et al. SPIRIT 2013. Explanation and Elaboration: Guidance for protocols of clinical trials. BMJ 2013;346:e7586). The checklist is attached this protocol.

Data availability statement

Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.