This two-armed, parallel-group, double-blind, multi-center randomized control trial (RCT) aims to evaluate the efficacy of probiotic supplement compared with placebo in ameliorating alcohol craving and severity of alcohol use disorder (AUD) in patients diagnosed with AUD after 2 weeks of in-patient detoxification. In addition, this study also compares the efficacy of probiotic supplement and placebo to mitigate common comorbid of AUD (such as depression and anxiety symptoms); depreciate the serum level of pro-inflammatory cytokines, such as interleukin-1β (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α) indicating lowering of systemic inflammation.

In phase I of the study, 28 patients diagnosed with AUD (using Diagnostic and Statistical Manual for Mental Disorders 5th Edition or DSM-5) and 18 healthy controls will be recruited. The measured outcomes to be compared between patients with AUD and healthy non-AUD controls include ERP on EEG monitoring, serum levels of pro-inflammatory cytokines (IL-1β, IL-6, and TNF-α), and the fecal microbiota content. Then, in phase II of the study, 58 AUD patients will be randomized into two groups of intervention in a 1:1 ratio (Lactobacillus sp. probiotic and placebo group; n = 29 per group). The participants in probiotic and placebo groups will then consumed the Lactobacillus sp. Probiotic and placebo 1 sachet once a day of probiotic and placebo, respectively for 12 weeks. Outcome assessments will be performed across four time points, such as t0 = before intervention began, t1 = 8 weeks after intervention began, t2 = 12 weeks after intervention began, and t3 = 24 weeks after intervention began. The primary outcomes to be measured are the degree of alcohol craving, alcohol withdrawal, and severity of alcohol use disorder. While the secondary outcomes to be assessed are severity of comorbid depression and anxiety symptoms, serum levels of pro-inflammatory cytokines (IL-1β, IL-6, and TNF-α), and fecal microbiota content.

The study will be conducted in the Second Affiliated Hospital of Xinxiang Medical University (Henan Mental Hospital), Xinxiang, Henan, China and Psychiatric Outpatient Clinic in Advanced Medical and Dental Institute, Universiti Sains Malaysia.

The approval of the study was obtained from the Human Research Ethics Committee of Xinxiang Medical University (code: XYEFYLL-(research)-2022-28) and from the Human Research Ethics Committee of Universiti Sains Malaysia (code: USM/JEPeM/22080546). Prior to participation in the trial, participants will be informed of their right to withdraw from the trial at any time and the data collected will be discarded immediately and will not be used for the study. Participants will be reminded that all personal identifiable information will not be used and only group data will be analyzed and published. They will be assured of their anonymity for participating in the trial. The participants will be assured that the data collected will only be used for research purposes and the findings will not be recorded in their case files. All participants will be assigned a research number for identification purposes, for example "PROB 001". All the data collected from the participants will be stored in the research file which will be locked in a file cabinet, whereby the key will be kept by the principal investigator. Data collected may also be stored in thumb drive and lap top which is only accessible by the research team and the principal investigator.

In order to monitor adverse effect (AE) during the trial, all participants will be given a trial card which contains the name, contact number and email of the members of the research team. They are encouraged to contact the research team member in charge if there is any occurrence of AE. The occurrence of AE will then be reported to the "adverse effect" section of the trial's case report form. The AE which may lead to withdrawal of participants from the trial are: (1) adverse reaction which is not related to the study intervention but cause discomfort to the participants, (2) adverse reaction which is related to the study intervention (probiotic, ACT and/or placebo), and (3) any changes in behavior, temperament, personality, psychotic symptoms, and suicidal tendency that occurred after the study intervention began (probiotic, ACT and/or placebo). All AE should be reported to the Human Research Ethics Committee of Xinxiang Medical University and Universiti Sains Malaysia and investigation should be carried out.

The principal investigator will lead the trial center and will coordinate closely with the research and site coordinators in a day-to-day basis involving in the conduct of the clinical trial, subject recruitment, and data collection. A trial monitoring unit will also be set up which will be chaired by the principal investigator, whereby weekly meeting will be held to discuss on the conduct of the trial, auditing of the trial, and prepare report to be submitted to the Human Ethics Committee of Xinxiang Medical University and Universiti Sains Malaysia. The monitoring of trial data and auditing of the trial will be managed by the clinical trial coordination unit of the 2nd Affiliated Hospital of Xinxiang Medical University and Advanced Medical and Dental Institute, Universiti Sains Malaysia, which are independent of the funder. Interim analysis will be carried out if there is such necessity for premature termination of the trial based on assessment and auditing of the trial data.

Primary objective:

To determine the differences in the severity of alcohol dependence, alcohol withdrawal and craving among patients on Lactobacillus sp. and placebo at 4 timelines (baseline, 8 weeks after starting intervention, 12 weeks after starting intervention, and 24 weeks after starting intervention which is 12 weeks post termination of intervention).

Secondary objectives:

1. To determine the differences in severity of depression and anxiety symptoms, and serum levels of pro-inflammatory cytokines among patients on Lactobacillus sp. and placebo, at 4 timelines (baseline, 8 weeks after starting intervention, 12 weeks after starting intervention, and 24 weeks after starting intervention which is 12 weeks post termination of intervention).
2. To assess the differences in gut microbiota profiles of patients on Lactobacillus sp. and placebo via the use of fecal samples, at 3 timelines (baseline, 2 weeks after starting intervention, and 8 weeks after starting intervention).

In this study, the participants will be recruited from the source population. The source population is all the patients who registered under the 2nd Affiliated Hospital of Xinxiang Medical University (XXMU), China and Outpatient Psychiatry Clinic of Advanced Medical and Dental Institute, Universiti Sains Malaysia with alcohol-related mental illness. Advertisement with posters will be placed on display in the 2nd Affiliated Hospital of XXMU, the Department of Psychiatry, XXMU, and Advanced Medical and Dental Institute, Universiti Sains Malaysia to help in the recruitment of subjects. In this study, patients with alcohol dependence who met the diagnostic criteria of "alcohol use disorder" in DSM-5 were selected as the experimental group. Healthy people who met WHO healthy drinking standards or did not drink were used as the healthy control group. Then, baseline assessment (t0) will commence followed by admission of the experimental group subjects for initial 2 weeks for detoxification (routine treatment). The assessment for experimental group subjects include Clinical Institute Withdrawal Assessment for Alcohol-Revised (CIWA-Ar), Penn Alcohol Craving Scale (PACS), Alcohol Use Disorder Identification Test (AUDIT), Hamilton Depression Rating Scale (HAMD), and Hamilton Anxiety Rating Scale (HAMA) scores serum pro-inflammatory cytokine levels, and fecal microbiota analysis. While the assessment for healthy controls include serum pro-inflammatory cytokine levels, and fecal microbiota analysis. During detoxification, treatment of withdrawal symptoms mainly with benzodiazepine replacement therapy, while adequate supplementation of B vitamins, strengthening symptomatic supportive treatment, and corresponding antipsychotic drugs, antidepressants or emotional stabilizers were given according to the condition. Benzodiazepines are gradually discontinued after withdrawal symptoms resolved. Then, subjects in the experimental group will be randomized into two groups, such as probiotics + routine treatment group, and placebo + routine treatment group. Specific intervention will be administered in the respective groups.

The estimated sample size needed for comparing the fecal microbiota, the serum cytokines, and the EEG and physiological indicators between subjects with alcohol use disorder and healthy controls was computed using the G*Power calculator 3.1.9.7 for calculating mean differences between two independent groups, whereby type I error = 0.05, power = 0.8, allocation ratio = 1:1, and effect size = 0.4 [De Giani et al. (2020)]. The estimated sample size needed was 120 subjects per group (inclusive of 20% of drop out). The estimated sample size needed for the experimental subject was calculated using the G*Power calculator 3.1.9.7 for calculating sample size for repeated measure ANOVA, within-between interaction, wherein type I error = 0.05, power = 0.8, number of groups = 3, number of measurements = 3, effect size = 0.15 [Kazemi et al. (2019)]. The estimated sample size needed was 120 subjects, 40 subjects per group (inclusive of 30% drop out).

A double-blind randomized controlled trial design has been chosen. The eligible AD patients will be randomized 1:1 ratio to the two arms of the study according to a computer generated, blocked randomization list. The eligible patients will be assigned to the probiotic group and placebo group with a treatment code concealed in a closed envelope. The randomization will be performed by the study statistician, who had no contact with the patients and not involve in the research project. The statistician is trained to randomize subjects who first enrol in the study into the two groups using computer system for randomization purpose. The allocation sequence will not be available to any member of the research team until databases had been completed and locked. Data collection will be performed by research assistants who were not involved in the study and do not know the objectives of the study. While data analysis will be carried out by the statisticians not involved in the study. Hence, this study is blinded to the researchers but not blinded for the subjects as the ACT group has no parallel control group.

All data analysis will be performed using SPSS software version 26.0. Descriptive statistics for socio-demographic and clinical characteristics, CIWA-Ar), PACS, AUDIT, HAMD, and HAMA scores, and serum pro-inflammatory cytokine levels will be computed. All categorical variables will be presented in frequency and percentage, while all continuous variables will be reported in mean and standard deviation or median and interquartile range, depends on normality. Initially, the fecal microbiota content, and serum levels of pro-inflammatory cytokines (IL-1β, IL-6, and TNF-α) will be compared between the experimental subjects and the healthy controls. The serum levels of pro-inflammatory cytokines (IL-1β, IL-6, and TNF-α) will be compared using independent t-test (if data normally distributed) or Mann-Whitney U test (if data non-normally distributed). The mean difference in the primary outcome (CIWA-Ar, PACS, and AUDIT scores) for the two randomized groups (probiotic, and placebo groups) at each specific time point (pre-intervention, 2 weeks at completion of conventional treatment [t0], post-treatment at 8 weeks [t1], post-treatment at 12 weeks [t2], and post-treatment at 24 weeks [t3]) will be assessed using one way analysis of variance (ANOVA) followed by false discovery rate adjustment. For the main pool analysis, mixed ANOVA is used to determine the interaction between the groups (probiotic, and placebo groups) and time points on the primary outcome; interaction = intervention × time; where time is a within-subject variable and intervention effect is a between-subject variable). The main effects of intervention in the groups and time points will be presented as estimated marginal mean and standard error of mean. The study's primary analysis will follow the intention-to-treat (ITT) principle. The data analysis for the study's secondary outcomes (severity of anxiety and depressive symptoms, pro-inflammatory cytokines, and fecal microbiota) will be conducted similarly to the primary outcomes' calculation. Statistical significance will be two-tailed and set to p < 0.05.

To handle any missing data, if the missing data represent less than 5% of the study's total collected data, they will be ignored. If the missing data represent more than 5% but less than 40% of the total collected and are assumed to be randomly missing, then multiple imputation (restricted maximum likelihood estimation) will be performed using Stata 15. However, if the missing data represent more than 40% of the total collected data or are assumed to be missing either not randomly or completely randomly, then only the collected data will be used for the study's analysis, and the missing data will be explained as a research limitation in any publications of the study's findings.

Inclusion Criteria (AUD patients):

• Hospitalized patients diagnosed with untreated alcohol use disorder (confirmed by the diagnostic criteria of DSM-5).
• Male, age 18 to 55 years old, Han nationality, junior high school education or above, right-handed.
• Those with normal eyesight, including corrected vision(screened clinically).

Exclusion Criteria (AUD patients):

• The patient suffer from other mental illnesses (screen by DSM-V criteria).
• The patient has allergy reaction to medications use for conventional treatment of alcohol use disorder.
• The patient had a history of organic brain disease, a pacemaker, gastrointestinal surgery, or serious health problems.
• The patient had a history of seizures.
• The patient is complicated with severe physical disease.
• Patient has other drug dependence (except for nicotine dependence).
• The patient took drugs affecting intestinal flora 30 days before and during admission.
• The patient has participated in any other alcohol-related studies or trials within the past 30 days.
• Patients had use any prescription or over-the-counter drugs in the past 30 days that may affect mood or alcohol cravings.

Inclusion Criteria (controls):

• Male, age 18 to 55 years old, Han nationality, junior high school education or above, right-handed.
• No history of psychoactive substance abuse, except tobacco (screen with urine dipstick test).
• Those with normal eyesight, including corrected vision (screen clinically).
• No alcoholic beverages in the last 2 weeks.
• According to the WHO healthy alcohol consumption standard, the average intake of pure alcohol per week is less than 210 grams, or do not drink at all (screen clinically).

Exclusion Criteria (controls):

• Those who suffer from other mental illnesses (screen by DSM-V criteria).
• Those who has history of organic brain disease, a pacemaker, gastrointestinal surgery, or serious health problems.
• Those who has history of seizures.
• Those complicated with severe physical disease.
• Those with drug dependence, including alcohol dependence (exception for nicotine dependence).
• Those who took drugs affecting intestinal flora 30 days before and during admission.
• Those who has participated in any other alcohol-related studies or trials within the past 30 days.
• Those who use any prescription or over-the-counter drugs in the past 30 days that may affect mood or alcohol cravings.

Male gender

The randomization will be performed by the study statistician, who had no contact with the patients and not involve in the research project. The statistician is trained to randomize subjects who first enrol in the study into the three groups using computer system for randomization purpose. The allocation sequence will not be available to any member of the research team until databases had been completed and locked. Data collection will be performed by research assistants who were not involved in the study and do not know the objectives of the study. While data analysis will be carried out by the statisticians not involved in the study.