The global burden of disease and injury attributable to alcohol is considerable: worldwide, 5.3% of deaths are directly related to alcohol. In France, 41 000 deaths were attributable to alcohol in 2015. Alcohol can cause more than 200 diseases. There is an important social cost attached to alcohol, which is estimated at 120 billion euros in France in 2010.

There is a scientific and clinical rationale for promoting alcohol reduction in alcohol use disorder (AUD). Alcohol harm reduction strategies have consistently shown promising results. Reducing alcohol intake is an established treatment goal with significant impact on morbimortality and associated socio-economic burden. This is possible because it reduces direct alcohol damage, and because it represents an attractive goal for patients who are not interested in abstinence.

For alcohol consumption reduction, two pharmacological agents are currently available on the market: nalmefene and baclofen. Nalmefene's approved indication is the "reduction of alcohol consumption in adult patients with alcohol dependence who have a high drinking risk level, without physical withdrawal symptoms and who do not require immediate detoxification." The therapeutic effects of nalmefene on alcohol consumption reduction remain modest: a recent metaanalysis of pharmacological treatments for alcohol use disorder found no effect on the number of heavy drinking days (RR 0.74; 95% CI [0.44-1.15]) (Bahji 2022). In addition, adverse events such as nausea, vomiting, dizziness, sleep disturbances, and decreased appetite are commonly reported (van den Brink 2014). A recent Cochrane review reported a very modest effect of baclofen on the number of abstinent days (mean difference 9.07%; 95% CI [3.30 to 14.85]) and no significant effect on the number of heavy drinking days (SMD -0.18; 95% CI [-0.48 to 0.11]) or on the number of drinks per day (mean difference -0.45; 95% CI [-1.20 to 0.30]) (Agabio 2023). Moreover, the use of baclofen requires cautious and gradual dose titration due to frequent adverse effects, particularly sedation, dizziness, and vigilance disturbances, which may limit its tolerability and adherence in clinical practice.

Given that currently available treatment options for alcohol consumption reduction display suboptimal efficacy and tolerability profiles, other approaches, such as neuromodulation interventions for alcohol use disorder, are currently under investigation.

Transcranial direct current stimulation (tDCS) is a non-invasive and safe cortical excitability modulation method. It involves passing a weak direct current through the brain between two electrodes, anode and cathode.

Anodal tDCS has been shown to improve numerous cognitive processes, and it has been hypothesized that anodal tDCS enhances executive function and improves cognitive control in AUD, reducing the probability of relapse. Current evidence allows for the formulation of a Level B recommendation (probable efficacy) of repeated tDCS in fibromyalgia, major depressive episodes and addiction. This recommendation was formulated for a specific electrode montage with anode over the right dorsolateral prefrontal cortex (DLPFC) and cathode over the left DLPFC.

Overall, the results of the REDSTIM trial are encouraging, with statistically significant effects in favor of active stimulation compared with sham. However, the magnitude of the clinical effects remains limited, as the reduction in alcohol consumption appears modest.

Since the initiation of the REDSTIM protocol in 2014, intensified tDCS protocols have been investigated in psychiatry and neurology, including so-called "clustered" designs. The principle of clustered tDCS relies on the repeated, sequential administration of acute phases of tDCS treatment.

In line with this approach, REDSTIM 2 will reproduce the same technical characteristics as REDSTIM, with the exception of the tDCS administration schedule. In the present protocol, three series (clusters) of 10 sessions will be delivered over a 12-week period, with each treatment cluster separated by a three-week interval. We hypothesize that this clustered design will increase the clinical potential of the original tDCS protocol.

The decision to intensify the REDSTIM protocol through repeated tDCS clusters is based on the following considerations:

• From a neuroscientific perspective
• From a clinical perspective
• From the patient perspective
• From the research team perspective In REDSTIM 2, tDCS treatment clusters will be repeated across three phases over a 12-week period, representing a significant intensification of the REDSTIM protocol while remaining compatible with participant constraints and safety, and while potentially improving adherence to the protocol.

Inclusion Criteria:

• Age ≥ 18 years old
• Meeting at least two criteria for Alcohol Use Disorder (DSM-5) as determined by the structured clinical interview for DSM
• Having an average alcohol consumption at least at medium risk level (WHO criteria)
• Willing to reduce their alcohol consumption
• Not protected by law
• Covered by national health insurance
• Able to complete the TLFB
• Able to give a written informed consent
• Accepting to sign the "safety" agreement

Exclusion Criteria:

• Meeting DMS-5 criteria for current substance use disorder with substance other than nicotine and caffeine
• Breath-alcohol test > 0 mg/L at Visit 1 (informed consent signature). In this case the inclusion day will be scheduled later during the following weeks)
• Withdrawal Assessment (CIWA) score ≥ 10 at randomization, history of Delirium tremens or preDelirium tremens
• Concomitant treatment with disulfiram, acamprosate, topiramate, baclofene, naltrexone, and nalmefene
• Acute psychiatric/somatic disorders and severe chronic psychiatric/somatic disorders
• Recent change in psychotropic medication (< 1 month)
• Patients having already experienced tDCS treatment
• Patients with a history of seizures;
• Patients with unexplained episodes of loss of consciousness, since such condition could be related with brain alterations or epilepsy;
• Patients with unstable or non-controlled neuropsychiatric illness;
• Patients having implanted brain medical devices;
• Patients with implanted pacemakers;
• Patients having any electrically, magnetically or mechanically activated implant;
• Patients having cardiac, neural or medication implants;
• Patients having vascular clips or any other electrically sensitive support system in the brain;
• Patients with serious brain injury;
• Patients showing damage of skin at sites of stimulation (the device can only be used in healthy skin without wounds, otherwise the resistance to current can be altered);
• Patients suffering from skin problems, such as dermatitis, psoriasis or eczema;
• Patients suffering from severe or frequent headaches;
• Patients with any serious life-threatening disease
• Pregnancy/lactation, positive pregnancy test at randomization
• Concurrent participation in other trial