Investigation of the combined effects of cannabidiol plus naltrexone on alcohol craving in alcohol dependence: study protocol of a phase II randomised, double-blind, placebo-controlled, parallel-group trial – ICONICplus Trial

Abstract

Introduction

Alcohol dependence (AD) is highly prevalent and has severe consequences on health and quality of life. However, the efficacy of approved pharmacotherapies such as naltrexone (NTX) remains limited, highlighting the need for novel pharmacotherapeutic approaches. Cannabidiol (CBD) is a promising candidate, which has shown potential to reduce craving and alcohol use by modulating brain circuits involved in craving and addiction. Preclinical studies suggest that CBD may enhance NTX’s therapeutic effects.

Methods and analysis

This is a three-armed, randomised, double-blind, placebo-controlled parallel group, multicentre phase II clinical trial. A total of 150 patients with AD will be randomised (1:1:1) to receive either 800 mg or 1200 mg CBD plus 50 mg oral NTX or placebo plus 50 mg oral NTX for 14 days. Alcohol craving will be assessed using the Obsessive Compulsive Drinking Scale (OCDS-G) where the primary endpoint is the change from baseline to the end of treatment. Secondary outcomes include craving during the entire study, quality of life, depressive symptoms, anxiety, patient-reported outcomes, neural brain activation, CBD plasma levels, time to relapse, alcohol use and treatment safety. For the comparison of each experimental group to the control group, a strata-adjusted (centre and baseline OCDS-G) van Elteren test is applied with adjustment for multiple testing by Bonferroni-Holm.

Ethics and dissemination

The trial has been approved by the Ethics Committee and the competent authority (ID: B_03510). All participants will provide written informed consent. An independent Data and Safety Monitoring Board will monitor safety. This trial complies with national and international regulations.

Trial registration number

NCT06845124; EU Trial Number: 2024-518164-12-00.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• First randomised controlled trial to examine the efficacy and safety of a novel combination of cannabidiol (CBD) and naltrexone in alcohol dependence.

• Inclusion of an active comparator treatment with naltrexone, which is approved for alcohol dependence and allows the assessment of the added value of CBD in treating alcohol craving and comparison of two CBD dosages (800 mg or 1200 mg CBD) plus oral naltrexone over a 14-day period, with treatment effects monitored over 196 days.

• Use of established experimental procedures validated for assessing pharmacological effects on alcohol craving.

• Eligibility criteria were defined to maximise the generalisability of the findings, reflecting the group of patients receiving pharmacological treatment.

• Limitations include restricted generalisability to other substance use disorders, exclusive focus on the CBD-naltrexone combination and potentially limited generalisability to other study sites due to the study being conducted only at two specific study sites.

Introduction

Despite the high prevalence of alcohol dependence (AD) and its substantial impact on health and quality of life,¹ only a limited number of medications are currently approved for its treatment. Randomised controlled trials (RCTs) have demonstrated modest overall efficacy and high relapse rates.² Although approved medications, such as naltrexone (NTX), can reduce craving and relapse risk to some extent,² novel pharmacotherapeutic approaches are needed to enhance existing pharmacotherapies. Cannabidiol (CBD) is a promising candidate for the treatment of AD, which has shown an exceptional pharmacological effect profile. CBD may be effective in targeting symptoms of AD and underlying disease processes, such as its anticraving, stress-reducing, anxiolytic, antidepressant and anticompulsive actions (for review see, Larsen and Shahinas³). Unlike existing treatments for AD, CBD targets multiple states associated with drug addiction and relapse risk, potentially expanding current treatment options. Preclinical studies have demonstrated that CBD (15–60 mg/kg) reduces the reinforcing properties of alcohol and decreases cue-induced and stress-induced alcohol self-administration,4,6 as well as impulsive choices.⁴ These effects have been linked to CBD’s ability to modulate the activation of brain circuits involved in drug craving and seeking,^{4 7} specifically through its effects on CB₁-receptors⁸ and mu-opioid receptors in the nucleus accumbens.⁵ This suggests that CBD could enhance the effects of NTX, as both substances act on mu-opioid receptors that regulate dopamine levels in the brain’s reward system, a neurobiological correlate of alcohol craving and consumption.⁶ Recent studies support the idea of synergistic effects of NTX and CBD in reducing relapse to alcohol consumption and anxiety in animal models of AD.⁶ The effect of CBD did not diminish with multiple administrations. Rather, both forms of reinstatement remained significantly reduced up to 138 days after CBD treatment. The potential of CBD to treating substance use disorders was further supported by two randomised placebo-controlled studies in patients with opioid use disorder (OUD) that showed significant craving-reducing effects of oral CBD in dosages between 400 and 800 mg daily, already a few hours after the first administration.^{7 9} In addition, a naturalistic observational study in 120 alcohol and cannabis using adults showed that ad libitum use of CBD-dominant cannabis (24% CBD, 1% THC) over 5 days (vs THC-dominant or THC/CBD-balanced cannabis) resulted in lower reported drinking days and drinks per drinking day.¹⁰ A randomised-controlled clinical trial over 28 days in individuals with cannabis use disorder also showed higher abstinence rates from cannabis in individuals receiving 400 mg and 800 mg oral CBD daily versus placebo, supporting CBD’s potential for treating substance use disorders.¹¹ Beyond this, studies (eg, in epilepsy) demonstrated an excellent safety profile of CBD.³ The profile of behavioural and pharmacological effects of CBD, as well as CBD’s effects within brain networks that mediate drug use, supports the potential of CBD to treating AD. As a first trial, this trial seeks to assess if CBD reduces alcohol craving—a hallmark symptom of AD that is insufficiently treated by existing medication—and enhances the effects of NTX.

This paper describes the study protocol and provides the rationale for a phase II randomised-controlled clinical trial designed to investigate the added benefit of CBD in combination with NTX in reducing alcohol craving. This clinical trial aims to strengthen the evidence base for pharmacological interventions targeting alcohol craving in patients with AD. This paper complies with the SPIRIT recommendations for protocol reporting^{12 13} and the Consolidated Standards of Reporting Trials guidelines.¹⁴

Methods

Clinical trial design

This RCT is designed as a three-armed, double-blind, placebo-controlled phase II trial to investigate the effects of either 800 mg or 1200 mg of CBD combined with oral NTX (50 mg) compared with a placebo plus NTX (50 mg) on alcohol craving in 150 patients with AD.

Potential eligible patients will be informed in detail about the study procedures and potential risks and provide informed consent before they will be carefully assessed for eligibility during the screening visit (day −7). After being considered eligible, baseline assessments will be performed during the baseline visit, that is, between day −2 and day 0. In addition, patients enrolled at the Central Institute of Mental Health (CIMH) will undergo a baseline functional MRI (fMRI) examination at visit 2 and blood will be drawn to determine CBD plasma levels at visit 3. At the end of the baseline visit, patients will be randomly assigned to one of the three treatment groups (800 mg CBD+NTX, 1200 mg CBD+NTX or placebo+NTX). On the morning of the third visit at the latest, which marks the beginning of the 14-day treatment phase (day 1 to day 14), the oral intake of 50 mg NTX will be started. The 14-day intake of either 800 mg or 1200 mg CBD or placebo, consisting of 6 CBD or placebo capsules per day in each group, also begins with the third visit. At visit 3 (day 1), visit 4 (day 7±2) and visit 5 (day 14), patient-reported outcomes (PROs) and different questionnaires will be assessed (see online supplemental table S1). At the fifth visit, patients enrolled at the CIMH will undergo a second fMRI examination and blood will be drawn to determine CBD plasma levels. After the last treatment visit (visit 5 on day 14), patients remain in in-patient treatment for at least 24 hours to observe and treat any potential acute side effects resulting from CBD intake. The aim of the 14-day treatment phase is to assess the impact of CBD compared with placebo in addition to NTX on alcohol craving and evaluate behavioural and neural markers associated with craving and relapse risk. After the treatment phase, the 182 day follow-up phase will begin, consisting of four follow-up visits. During these visits, data on alcohol craving, drinking behaviour, PROs and the patients’ quality of life will be assessed to evaluate the longer-term effects of the intervention. Throughout the study, all participants receive standard treatment with 50 mg NTX. The safety of the treatments, especially any potential side effects, is continuously monitored until the last visit (visit 9, day 196).

Study objectives and primary estimand

The main purpose of the clinical trial is to assess the effect of CBD plus NTX on alcohol craving in patients with AD. Specifically, the study aims to assess the impact of the additional administration of CBD (at two doses, 800 mg and 1200 mg) on reducing alcohol craving compared with placebo plus NTX. The attributes of the primary estimand are depicted in box 1.

Box 1. Primary estimand.

The attributes of the primary estimand are defined as follows:

• Treatment: (1) 800 mg cannabidiol (CBD)+50 mg naltrexone (NTX), (2) 1200 mg CBD+50 mg NTX and (3) placebo (PLC)+50 mg NTX.

• Population: The population is defined by the inclusion and exclusion criteria.

• Variable: Difference in Obsessive Compulsive Drinking Scale (OCDS-G) scores (alcohol craving) at day 14 and day 0.

• Intercurrent events: (1) Omission of medication dosages, (2) relapse to alcohol with and without following interventions to treat withdrawal symptoms and (3) discharge from inpatient treatment for any reason will be handled using the treatment-policy strategy.

• Summary measure: The summary measure is the common language effect size which is the probability of a random OCDS-G score difference value in the experimental group being greater than a random value in the control group.

Secondary objectives are depicted in box 2.

Box 2. Secondary objectives.

Comparison of the effects of either 800 mg or 1200 mg cannabidiol (CBD) plus 50 mg oral naltrexone (NTX) versus placebo plus 50 mg NTX on:

• Changes in alcohol craving:

  From baseline (visit 2) to visit 4 during the early intervention phase, to detect early effects of the active intervention on craving, and during the follow-up phase after completion of the intervention, to assess whether the effects on craving are sustainable.

• Changes in quality of life:

  From baseline (visit 2) to visit 5 at the end of the intervention phase, to detect the effects of the active intervention on quality of life, and during the follow-up phase, to assess whether these effects are sustainable.

• Changes in depressive symptoms:

  From baseline (ie, visit 2) to visit 5 at the end of the intervention phase, to detect effects of active intervention on depressive symptoms.

• Changes in state and trait anxiety:

  From baseline (ie, visit 2) to visit 5 at the end of the intervention phase, to detect effects of active intervention on state and trait anxiety.

• Patient-reported outcomes (PROs):

  To assess the effects of the active intervention on PROs (subjective burden and benefit of treatment, alcohol craving, perceived stress, mood, anxiety, well-being, daily functioning and confidence in abstinence) at visits 3, 4 and 5, and during the follow-up phase, to assess whether these effects are sustainable.

• CBD plasma levels:

  At visit 3 and visit 5, to detect effects of active intervention on blood plasma levels (only for patients enrolled at Central Institute of Mental Health (CIMH) study site).

• Time to relapse, cumulative alcohol use and percent heavy drinking days:

  To assess the effects of the active intervention on relapse-related outcomes (time to relapse, cumulative alcohol use and per cent heavy drinking days) at visit 5, and during the follow-up (visits 6–9), to determine whether these effects are sustainable after the intervention phase.

• Average weekly alcohol consumption and maximum weekly craving:

  Alcohol consumption and craving will be assessed every seventh day via a smartphone-based e-diary during the follow-up phase (from visit 5 to visit 9) to evaluate the sustainability of the active intervention’s effects (optional, only for patients that agree to use the study-specific app).

• Safety:

  Determined by assessing adverse events (AEs) and serious AEs at visits 3, 4 and 5 to detect the effects of the active intervention on safety, and during the follow-up phase to determine whether the treatment’s safety is sustained.

• Changes in neural brain activation and differences in alcohol craving, response times, rates of errors, correct responses and omission rates during functional MRI paradigms:

  From baseline (ie, visit 2) to visit 5 at the end of the intervention phase, the effects of the active intervention on the blood oxygenated level dependent response and behavioural outcomes (craving, response times, rates of errors, correct responses and omission rates) will be assessed during alcohol cue presentation, natural reward cue presentation, response inhibition, presentation of emotional faces and neutral shapes, and functional connectivity during resting state (only for patients enrolled at CIMH study site).

Study population

Patients of both sexes diagnosed with AD and currently receiving standard inpatient treatment at one of the study sites (the CIMH in Mannheim, Germany and at the Psychiatric Center North Baden (PZN) in Wiesloch, Germany) will be enrolled. This population represents the target group for pharmacologically supported relapse prevention and craving reduction according to current national treatment guidelines. The inclusion and exclusion criteria (see table 1) were selected in accordance with previous trials¹⁵ and define a group of AD patients that is representative of those receiving pharmacological treatment, thereby allowing for generalisation to this group.

Table 1. Inclusion and exclusion criteria.

Inclusion criteria	Exclusion criteria
Age between 18 and 70 years	Current psychotic or bipolar disorder or current severe depressive episode with suicidal ideations
Diagnosis of an AD (ICD-10)	Positive drug screening (amphetamines/ecstasy, opiates, cocaine, barbiturates)
Reporting alcohol craving as symptom of AD according to the ICD-10 symptom definition	Pregnancy, lactation or breastfeeding
Ability to understand character and individual consequences of the clinical trial	Current severe somatic comorbidities: severe liver cirrhosis (stage: CHILD B or C) or epilepsy determined by medical history or prolonged or shortened QT intervals (determined by ECG)
Written informed consent	Patients with elevated transaminase levels (GOT or GPT) above three times the upper limit normal (ULN) value with elevated bilirubin levels above twice the ULN value
Women with childbearing potential and males with partners with childbearing potential, use of a highly effective birth control method until 1 month after last IMP administration and negative pregnancy test at the time of enrolment	History of hypersensitivity to the investigational medicinal product CBD and/or naltrexone (trade names: Adepend, Naltrexon-Hcl neuraxpharm, Naltrexonhydrochlorid Accord) or to any drug with similar chemical structure or to any excipient present in the pharmaceutical form of the investigational medicinal product CBD and/or naltrexone
Consent to random assignment	Participation in other clinical trials or observation period of competing clinical trials, respectively.
	Acute suicidal tendency or acute endangerment of self and others
	Current treatment with any of the following substances: Any investigational medicinal product, opioid-containing analgesics, anti-obesity drugs, anticonvulsants, opioid-containing antidiarrhoeal agents, antineoplastics, antipsychotics (exception: episodic use of melperone, prothipendyl, pipamperone, promethazine and quetiapine are allowed), antidepressants (exception: allowed, when being taken in stable dose for a minimum of 14 days prior to enrolment and/or doxepine in low doses (max. 75 mg daily)), opioid-containing cough/cold agents, systemical steroids, other anti-craving (eg, acamprosate) or aversive medication (eg, disulfiram), THC-containing or CBD-containing medication, antiretroviral medication (eg, efavirenz), xanthines (eg, theophylline), general anaesthetics (eg, propofol), hypericum perforatum, antibiotics (eg, rifampin, clarithromycin, erythromycin).*

^*See table 2 for an overview of non-permitted concomitant medication.

AD, alcohol dependence; CBD, cannabidiol; GOT, aspartate aminotransferase; GPT, alanine aminotransferase; ICD-10, International Classification of Diseases; IMP, investigational medicinal product; THC, Tetrahydrocannabinol.

Table 2. List of non-permitted concomitant medication.

Drug class	Disallowed prior to the screening visit	Disallowed during the trial for
Chronic use	Episodic use	Comments or exceptions
Any investigational medicinal product	Within 30 days or 5 half-lives (whichever is longer)	X	X
Opioid-containing analgesics	X	X	Non-opioid analgesics are allowed
Anti-obesity drugs	X	X
Anticonvulsants	X	X
Opioid-containing antidiarrhoeal agents	X	X
Antineoplastics	Within 30 days or 5 half-lives (whichever is longer)	X	X
Antipsychotics	Within 14 days	X	X	Use of melperone, prothipendyl, pipamperone, promethazine and quetiapine is allowed
Antidepressants	X	X	Antidepressants are allowed, when being taken in stable dose for a minimum of 14 days prior to enrolment and episodic use of doxepine in low doses (max. 75 mg daily) is permitted
Opioid-containing cough/cold agents	X	X
Steroids Systemic Topical Inhalant	X	X	Systemic corticosteroids are not allowed. Topical and inhalant use is allowed.
Other anti-craving (eg, acamprosate) or aversive medication (eg, disulfiram)	Within 7 days	X	X
THC-containing or CBD-containing medication	Within 7 days	X	X
Antiretroviral medication (eg, efavirenz)	X	X
Xanthines (eg, theophylline)	X	X
General anaesthetics (eg, propofol)	X	X
Hypericum perforatum	X	X
Antibiotics (eg, rifampin, clarithromycin, erythromycin)	X	X

CBD, cannabidiol; THC, Tetrahydrocannabinol.

Subjects meeting all of the inclusion criteria and none of the exclusion criteria will be considered eligible for enrolment in the clinical trial.

Recruitment and consent

Before being enrolled, patients must provide written informed consent (a model consent form is provided in the online supplemental file 1) to participate after being fully informed by the investigator or a designated member of the investigating team about the nature, importance, risks and individual consequences of the clinical trial and their right to terminate the participation at any time. The informed consent to participate in the clinical trial may be withdrawn by the subject verbally in the presence of, or in written form directed to, the investigator or a member of the investigating team at any time during the clinical trial without any disadvantage for the participant.

Randomisation and blinding

Patients will be sequentially allocated a unique identifying number, which will be used for all subsequent study documentation, ensuring pseudonymisation. Eligible patients will be randomised in a concealed manner to one of the three treatment arms in a 1:1:1 ratio at the end of visit 2 (baseline). Block randomisation will be applied stratified by centre and three groups of baseline Obsessive Compulsive Drinking Scale (OCDS-G) scores (0–9/10–24/25–40). Randomisation and emergency unblinding will be carried out using a GCP-compliant, centralised web-based tool (www.randomizer.at). Study-medication blisters (CBD/Placebo) are prepared by the pharmacy and blinded to patients and investigators. For all trial personnel, including the biometricians, patient treatment remains blinded from the time of randomisation until final database lock.

Intervention

The 14-day treatment phase of this RCT involves the administration of either CBD or placebo plus NTX for the treatment of AD and consists of three treatment arms:

• Arm 1: 800 mg CBD and 50 mg NTX daily.

• Arm 2: 1200 mg CBD and 50 mg NTX daily.

• Arm 3: (Control group): Placebo and 50 mg NTX daily.

After the 14-day intervention phase starting with visit 3 (day 1) and ending with visit 5 (day 14), there is a follow-up period of 182 days with follow-up visits on days 28, 42, 105 and 196.

CBD doses were chosen based on studies showing significant effectiveness at 800 mg CBD, with 1200 mg as an additional dose for potentially enhanced effects as dose-dependent effects have been observed.¹⁶

The dosage range and administration of CBD (ie, 800 mg/day and 1200 mg/day) was chosen in accordance with previous trials, which indicated superiority of a dosage of 800 mg CBD versus lower dosages (400 mg, 600 mg) on drug craving^{7 17 18} and considering further studies, which indicated that a dosage of 1200 mg CBD might yield additional effects.³ Our own study indicated dose-dependent effects.¹⁶ Due to the fact that the optimal dosage in AD has not been established yet, we will compare two dosages that reflect the effective and safe dosage range observed in clinical trials in other substance use disorders and preclinical studies in AD,¹⁹ omitting low dosages (<800 mg), due to data indicating limited effects of such dosages. The treatment duration (14 days) was chosen in accordance with previous studies on CBD¹⁹ and follows the design of previous clinical trials in AD.²⁰

All patients will receive 50 mg NTX daily, irrespective of participation in the trial. Treatment with NTX will be initiated at any time before, but not later than the day of visit 3 (day 1) and will be administered before IMP/placebo administration, regardless of randomisation group, to treat the indication which is the object of the study, that is, AD. NTX is considered standard care for treating AD according to the German S3 guideline.²¹ The treatment with NTX should run at least from visit 3 (or earlier) to visit 5 (or longer), covering a period of 14 days in parallel to the treatment with CBD.

Data collection and outcome measures

The OCDS-G^{22 23} will be used to assess the difference in alcohol craving between baseline (visit 2) and visit 5 after the 14-day treatment period, and serve as primary outcome. The OCDS-G is a suitable tool for measuring craving. It shows high internal consistency and test–retest reliability and was validated in multiple pharmacological trials in AD patients and shows close associations with patient outcome.^{22 23}

Secondary outcomes will be assessed by different validated measures at various time points throughout the entire study. Box 3 gives an overview of the questionnaires and scales that will be assessed during this trial and online supplemental table S1 provides an overview of the schedule and the outcome measures recorded at each visit.

Box 3. Questionnaires and psychometric scales that are included in the study.

• Screening for psychiatric disorders via Structured Clinical Interview.^{28 29}

• Obsessive Compulsive Drinking Scale.^{22 23}

• Timeline Follow-Back interview of alcohol consumption in past weeks (Form 90).³⁰

• Alcohol Dependence Scale.³¹

• Fagerström Test for Nicotine Dependence.³²

• Edinburgh Inventory of Handedness.³³

• Beck Depression Inventory-II.³⁴

• State Trait Anxiety Inventory.³⁵

• WHO Quality of Life questionnaire.³⁶

• Questionnaire on the use of aftercare.

• Patient-reported outcomes.

• e-Diary based assessment of alcohol use and craving (optional).

Functional and structural MRI data will be collected using a 3 Tesla SIEMENS Magnetom PRISMA MRI scanner (Siemens Healthcare, Erlangen, Germany), which are located at the CIMH Mannheim. Only patients enrolled at the CIMH will receive functional and structural MRI measurements at baseline (visit 2) and at the end of the treatment phase (visit 5). fMRI data will be analysed using a general linear model (GLM) within the (SnPM13) toolbox for Matlab, with non-parametric statistical testing. fMRI measurements will assess neural reactivity to alcohol cues, emotional processing, reward processing, inhibitory control and functional/structural connectivity. In addition, only in patients enrolled at the CIMH, blood plasma is taken at visit 3 and visit 5 to determine CBD levels.

Adverse events (AEs) will be assessed from visit 3 to visit 9. New pathological and clinically relevant findings, or worsening of pre-existing symptoms, will be documented as AEs or serious AEs (SAEs).

Sample size calculation

To date, no study investigated the combined effects of CBD plus NTX on craving in AD. Hence, no firm prediction of the expected effect size can be made. However, based on previous clinical trials in AD, we assume that only moderate to large effects on craving reflect a clinically meaningful reduction of alcohol craving^{22 23} that warrants further confirmatory trials. The current study is designed to detect such effects, with a focus on the OCDS-G score difference. Based on data from a previous study investigating CBD mono-treatment (800 mg and 400 mg) in OUD,⁷ we assume a further craving reduction of 3 points on the OCDS-G in each of the experimental groups compared with the control group, resulting in a uniform distribution on (−14; 0) with a median OCDS-G score difference of −7. This corresponds to a common language effect size²⁴ of 0.68 and a Cohen’s d of 0.69. Under these assumptions and balanced randomisation, 34 patients per group are needed to obtain a power of 80% to reject at least one of the two null hypotheses at the local two-sided level of 0.025 (Bonferroni corrected). Considering a 32% dropout rate, we plan to enrol 50 patients per group. A gain in power can be expected by using the non-parametric strata-adjusted van Elteren test and multiple imputation of missing values.

Statistical analysis

The analysis of the primary and secondary efficacy endpoints will be conducted based on the full analysis set (FAS) with patients analysed in the group as randomised according to the intention-to-treat principle. All statistical analyses will be described in detail in the statistical analysis plan, which will be finalised before data base closure.

Primary analysis

The primary analysis describes the analysis of the primary estimand and will be conducted based on the FAS. All intercurrent events (ICEs) are handled by a treatment policy strategy, that is, all values are used as observed. The primary outcome will be compared between each dosage group and the placebo group. As two dosages of CBD are investigated, two null hypotheses are tested:

H₀,_800mg: F_800mg(z) = F_placebo(z) and H₀,_1200mg: F_1200mg(z) = F_placebo(z).

Each hypothesis will be tested using the van Elteren test, stratified by centre and baseline OCDS-G score (0–9/10–24/25–40). The Bonferroni-Holm method will be used to control the error rate at a two-sided significance level of 0.05. The summary measure within the strata is the relative treatment effect according to Brunner and Munzel²⁵ and the global summary measure is the weighted mean of the stratum-specific relative treatment effect sizes. A 95% CI will be given for the treatment effect using a t-distribution approximation.²⁵ Missing OCDS-G scores at visit 5 after an ICE will be imputed aligned with the treatment policy strategy. Otherwise, missing values for those who did not experience an ICE are imputed under missing at random. Values are imputed using multiple imputation using the fully conditional specification method²⁶ with predictive mean matching with the predictors OCDS-G score at baseline (visit 2) and at visit 4, treatment group, sex, centre and AD severity at visit 2. A sensitivity analysis will be done to verify the robustness of the primary estimate using a pattern mixture model applied under the assumption of a ‘missing not at random’ mechanism. For the complementary estimand, the ICE ‘omission of medication dosages’ will be handled by the hypothetical strategy so that data after the ICE are not used in estimating the treatment effect and missing data will be imputed aligned with the respective strategy.

Analysis of the secondary endpoints

Secondary outcomes will be evaluated descriptively. Continuous variables will be described using number of missing values, non-missing values, mean, SD, median, Q1, Q3, minimum and maximum. For binary or categorical variables, absolute and relative frequencies will be provided. Effect estimates and 95% CIs will be reported. For all outcomes, all comparisons will be done once for 800 mg vs placebo and once for 1200 mg versus placebo.

Change from baseline of the OCSD-G craving score to visits 4, 6, 7, 8 and 9 will be compared between groups using the van Elteren test stratified by centre and baseline OCDS-G categories. Quality of life (WHO-QOL-BREF scores) change from baseline to visits 5, 6, 7, 8 and 9 will be compared by fitting a linear mixed model. State Trait Anxiety Inventory and Beck Depression Inventory-II changes from baseline to visit 5 will be compared using analysis of covariance. Cumulative alcohol use, per cent heavy drinking days, as well as average weekly alcohol use and maximum weekly craving during the follow-up period, are compared between groups using the van Elteren test stratified by centre. Time to relapse from visit 2 will be analysed descriptively by providing Kaplan-Meier estimates together with 95% log-log-type CIs and by conducting a (descriptive) log-rank test.

Safety will be assessed based on the number and percentage of patients having any AE. If the number of SAEs is sufficiently large, the same descriptive analysis is done for SAEs. Otherwise, they will only be listed.

fMRI data (CIMH site only)

fMRI data will be analysed by conducting a GLM to compare BOLD responses between error rates, correct response rates and omission rates will be analysed using descriptive χ² tests. Alcohol craving (visual analogue scale) and response times will be analysed using descriptive Mann-Whitney U test.

Project timeline

Participant recruitment has not yet started. The clinical phase of the study is planned to last 34 months, with the first subject expected in Q3 2025 and the last subject’s last visit in Q2 2028. The database lock is scheduled for Q3 2028, with statistical analyses to be completed by Q4 2028.

Patient and public involvement

A trial advisory board (TAB) has been established composed of two representatives of the CIMH advisory board, representing the perspective of affected individuals and relatives, one representative of self-help and one representative of an addiction counselling centre to balance and diversify perspectives. The tasks of the TAB are to consult with the clinical trial team and solicit input into continuous improvement opportunities with areas of focus, such as design of patient-facing materials, communication (eg, potential alternative treatment, risks, results) with patients and the scientific field, clinical trials procedures, patient recruitment procedure, trial execution simulations, recommendations on reducing strain on participants.

Ethics and dissemination

Ethical considerations

This clinical trial will be conducted according to national and international rules and regulations (Good Clinical Practice, Declaration of Helsinki and Data Protection Laws, Clinical Trials Regulation). All study documents have been submitted to the Ethics Committee of Saxony-Anhalt in Dessau and the competent authority, which approved the conduct of the trial (ID: B_03510). Before trial admission, written informed consent will be obtained from each patient, after the risks of treatment and alternative therapies have been explained in detail. Patients will be insured for trial-specific risks, and an independent Data and Safety Monitoring Board will monitor safety approximately every 6 months throughout the trial.

Previous trials indicated minimal side effects of CBD in dosages up to 6.000 mg³ and safety of NTX, which is already approved for treating AD.²⁷ All trial procedures abide by Good Clinical Practice (GCP) and Clinical Trial Regulation (CTR) standards and are validated clinical methods that can be regarded as low-risk and intensive safety monitoring is established, so that we expect minimal risks for the enrolled patients. The intensity and invasiveness of the trial-specific procedures were reduced to a minimum and the trial design was developed together with patients and their representatives to minimise patient burden. All participants will undergo physical examination by trained physicians and blood sampling, as well as careful consideration of contraindications to further minimise risks. All patients receive state-of-the-art care according to current treatment guidelines and receive NTX, such that no effective treatment is withheld. Individual patients could benefit from craving reduction and the expected gain in knowledge is substantial. Thus, the potential patient benefit significantly outweighs the risks.

Confidentiality

Data obtained during the trial will be handled pursuant to the EU General Data Protection Regulation and national regulatory requirements. To ensure confidentiality of records and personal data, only pseudonymised data will be transferred to the sponsor by using a unique subject identification number. The investigators will maintain a subject identification list at each site. Clinical trial data stored on a computer will be stored in accordance with local data protection law and will be handled in strictest confidence. Data will be analysed by subject identification number only. Records will be retained for 25 years after study completion.

Dissemination

Individual deidentified data, that underlie results concerning primary or secondary endpoints reported in a published scientific article, will be made available with a respective data dictionary as far as legally possible. fMRI data or other secondary outcome measures will be made available on an aggregated group level (eg, for the purpose of meta-analyses). Data will be available upon publication of the results until 3 years after that. Individual data will be shared with researchers who provide a methodologically sound proposal (sent to the Principal Investigator of the Trial).