Cessation or Reduction of Alcohol Consumption in Veterans: A Randomized Controlled Trial for Alcohol Use Disorder (CRAVE)

Background AUD is one of the leading causes of disability worldwide. The prevalence of AUD is high, affecting 10.9% of US adults and 5.1% of adults worldwide. Oral naltrexone, the most widely prescribed medication for AUD, has a number needed to treat (NNT) to prevent a return to heavy drinking of 12, and thus is only modestly effective. Indeed, less than 2% of adults with AUD receive medication in a given year. Though the Department of Veterans Affairs promotes pharmacotherapy as a best practice, there are over 400,000 Veterans within the Veterans Health Administration (VHA) who have a diagnosis of AUD, with only about 40,000 being actively treated with pharmacotherapy (source: VA Quality Dashboard accessed 1/31/2025). As there have been no new Food and Drug Administration (FDA) approved medications in nearly two decades, there is an urgent need for novel treatments for AUD with superior efficacy and higher patient appeal. Based upon very promising clinical experience, retrospective studies, preclinical data, and recent pilot clinical trial results, the proposed clinical trial is designed to provide definitive evidence regarding the efficacy of the GLP-1 RA, semaglutide, compared to placebo for the treatment of AUD. This research will offer urgently needed information on the efficacy of GLP-1 RAs in the treatment of AUD in a diverse sample and is directly in line with the strategic priorities (SP) for VA Research codified by the Office of Research and Development (ORD) and the National Institute on Alcohol Abuse and Alcoholism (NIAAA).

Study Design This is a randomized, double-blinded, intent-to-treat, two-arm, parallel, superiority multisite clinical trial in which 622 participants will be randomized into the semaglutide treatment or placebo arm in a 1:1 ratio. Randomization will be stratified according to BMI (<30 or ≥ 30) and participating site. Veterans with DSM-5-diagnosed moderate to severe AUD who are seeking treatment will be invited to participate in this trial. After eligibility screening, participants will be randomized to either the semaglutide group or the placebo group.

The study will be conducted in three phases. Phase 1 will be Recruitment, Consent and Screening, Phase 2 will be Randomization and Intervention (6 months of treatment), which includes dose escalation and the primary endpoint ascertainment period (the last 28 days of the intervention period), and Phase 3 is the Post Treatment Safety Assessment (4 weeks).

Study Objectives Primary Objective: To evaluate the efficacy of semaglutide 2.4 mg compared to placebo in achieving the World Health Organization's (WHO) two-level alcohol risk reduction (WHO-2LRR), from baseline risk level, in the WHO risk drinking level in the treatment of moderate to severe AUD. The primary outcome will be measured during the last 28 days of the intervention period.

Secondary Objectives:

• To compare the safety and tolerability of semaglutide 2.4 mg versus placebo.
• To evaluate the efficacy of semaglutide on reducing the proportion of subjects reporting no heavy drinking days (NHDD) during the last 28 days of the intervention period.
• To evaluate the efficacy of semaglutide on a WHO-2LRR during the last 56 days of the intervention period.
• To assess moderation effects of race, age, sex, BMI, impulsivity, psychiatric comorbidity, or presence of other psychiatric treatments (medication and/or psychotherapy) on semaglutide treatment effects compared to placebo.

Exploratory Objectives:

The following objectives aim to provide a comprehensive assessment of semaglutide's impact on various aspects of patients' well-being and recovery from AUD, compared to placebo:

• WHO-2LRR and NHDD: Evaluate the risk level change over the entire course of the intervention (monthly repeated measures). Similarly, assess the presence of no heavy drinking each month over the entire intervention period.
• Clinical Global Impression - Improvement (CGI-I): Assess overall improvement of AUD symptoms.
• Health-Related Quality of Life (HRQoL): Evaluate the effect on improvement in health-related quality of life as measured by the Veterans RAND 12-Item Health Survey (VR-12).
• Progress in Addiction Treatment: Measure progress in addiction treatment using the Brief Addiction Monitor-Revised (BAM-R).
• Alcohol Craving: Assess the improvement in alcohol craving as measured by the Penn Alcohol Craving Scale (PACS).
• Psychiatric Distress as measured by

• Depressive symptoms: Measure reductions in psychiatric distress related to depression using the Patient Health Questionnaire (PHQ-9).
• Anxiety symptoms: Measure reductions in psychiatric distress related to anxiety using the Generalized Anxiety Disorder-7 (GAD-7).
• Post-Traumatic Stress Disorder (PTSD): Measure reductions in psychiatric distress related to PTSD using the PTSD Checklist (PCL-5) among participants with PTSD.
• Alcohol-Related Problems and Consequences: Assess the improvement in alcohol-related problems and consequences as measured by the Short Inventory of Problems (SIP-2R).
• Health Care Utilization (HCU): Evaluate reductions in health care utilization.

Intervention and Masking Participants assigned to the semaglutide 2.4 mg group will undergo 24 weeks of treatment (Phase 2). Phase 2 will be used to initiate a dose of 0.25 mg with further dose escalation up to 2.4 mg weekly starting at week 5. Participants will be increased to their maximal tolerable dose. Increases will be considered only after the participant has been on the current dose for 4 weeks (dose escalation schedule is 0.25, 0.5, 1.0, 1.7, and 2.4 mg). Doses are not to exceed 2.4 mg weekly. Patients may have their dose reduced, maintain their current dose, or have a slower dose escalation to ensure tolerability and increase retention in the study. Doses are delivered via a pen, a cartridge-based device that calibrates medication delivery based on the desired dose. An extremely small needle (4-mm, 32-gauge needle - the size of 2 human hairs) is used to deposit the medication subcutaneously. Participants assigned to the placebo group will receive a placebo that mimics the same treatment procedure.

Throughout the study, the investigators will maintain double-blind conditions regarding the medication condition. To enhance the ability to maintain blinded assessment of the primary outcome, the assessment will be collected by a Central Assessment Center.

Sample Size and Study Duration This study plans to randomize 622 Veterans, with 311 participants assigned to each group. The study duration will be minimally 47 months for study start-up, approximately 32 months for recruitment, dose escalation, endpoint assessment and follow-up for safety, three months for data cleaning, and nine months for data analysis and reporting.

Inclusion criteria

• Veteran
• WHO risk drinking level of Very High or High in the 28 days prior to screening (based on screening TLFB)
• Current diagnosis of moderate or severe AUD (i.e., meeting at least 4 of 11 DSM-5 AUD criteria) based on semi-structured diagnostic exam (MINI)
• Able and willing to provide informed consent
• Has a desire to reduce their alcohol consumption

Exclusion criteria

• Medical History (medical history form)

• Type 1 diabetes
• History of acute or chronic pancreatitis
• History of diabetic ketoacidosis
• History of proliferative diabetic retinopathy
• History of ascites, advanced liver fibrosis, compensated cirrhosis with portal hypertension, decompensated cirrhosis, variceal bleeding, hepatic encephalopathy, spontaneous bacterial peritonitis, or hepatocellular carcinoma (HCC)
• History of advanced fibrosis or cirrhosis, including (but not limited to) transient elastography (liver stiffness) of >12 kPa, FIB-4 ≥2.67, ELF ≥9.8, MRE ≥3.63 kPa
• History of stage 3 fibrosis or stage 4 cirrhosis from a liver biopsy
• History of esophageal varices on endoscopy or imaging
• History of nodular liver, cirrhosis, splenomegaly, varices or splenic venous shunting or collaterals on prior imaging
• History or acute alcohol hepatitis (by liver biopsy or elevated bilirubin > 1.5 times the upper limit of normal)
• History of primary biliary cholangitis
• History of primary sclerosing cholangitis
• Current drug-induced liver disease
• History of alpha1 antitrypsin deficiency related liver disease
• History of autoimmune liver disease
• History of hemochromatosis
• History of Wilson's disease
• Presence of gastroparesis
• History of acute gallbladder disease in the prior 6 months
• Personal or family history of medullary thyroid carcinoma (MTC) or multiple endocrine neoplasia syndrome type 2 (MEN 2)
• Unstable body weight defined as >5% change in body weight (documented or self-report; intentional or not) in the 90 days prior to randomization
• Recent major cardiovascular event in the 90 days prior to randomization (myocardial infarction, stroke, New York Heart Association class IV heart failure, transient ischemic attack (TIA), or unstable angina
• Known history of prior hypersensitivity reaction to semaglutide, any of the product components, or any other GLP-1 analogue
• Concurrent Treatments (medical history form):

• Current (within the past 30 days) use of pharmacotherapy for AUD (including oral or intramuscular naltrexone, acamprosate, disulfiram, topiramate)
• Current (within the past 30 days) use of the following medications with glucose-lowering properties: GLP-1 analogues; sulfonylurea; insulin and insulin products; dipeptidyl peptidase-4 (DPP-4) inhibitors; sodium-glucose cotransporter-2 (SGLT-2) inhibitors, meglitinides, thiazolidinediones or other medications that may interact with semaglutide
• Recent changes in dose (within 2 months of randomization) of psychiatric medications (i.e., antidepressants, antianxiety, mood stabilizing)
• Psychiatric diagnosis (MINI)

• Current serious psychiatric illness (any psychotic disorder, bipolar 1 disorder, psychotic major depression, antisocial personality disorder, bulimia, or anorexia)
• Current DSM-5 diagnosis of a SUD (other than moderate-to-severe alcohol, any nicotine, or mild cannabis use disorders)
• Other assessments (local site)

• At the time of randomization, moderate-to-severe alcohol withdrawal (Clinical Institute Withdrawal Assessment for Alcohol (CIWA-AR) >8)
• BMI <21 kg/m2
• Acute high risk of suicide requiring hospitalization at the time of screening or randomization
• Medical, psychiatric, behavioral, or logistical conditions which, in the judgment of the Local Site Investigator (LSI) or Co-Investigator (Co-I), make it unlikely the participant can participate in or complete the 24-week active phase of the study
• Pregnant, actively breastfeeding, or female of childbearing potential who is unwilling to use a highly effective method of contraception as defined by the NIH
• Currently enrolled in another therapeutic or investigational clinical trial without a preexisting dual enrollment agreement
• Participant is incarcerated
• Laboratory

• Hemoglobin A1c (HbA1c)>10
• Estimated glomerular filtration rate (eGFR) <30 mL/min
• Albumin < 3.5 g/dl
• Aspartate aminotransferase (AST) >3 the Upper Limit of Normal (ULN)
• Alanine aminotransferase (ALT) >3 the ULN
• Lipase > 2 times the upper limit of normal
• Alkaline phosphatase > 1.5 times the ULN
• Total bilirubin > 1.5 times the ULN except with documented Gilbert's syndrome
• International Normalized Ratio (INR) > 1.3 unless due to anticoagulation therapy
• Platelet count <150,000/ L unless consistent with baseline and reflects the participant's habitual thrombocyte level, and there was no presence of portal hypertension
• Hepatitis B surface antigen positive
• Hepatitis C virus RNA positive - participants treated and cured of hepatitis C must have at least 2 years of negative testing
• Anti-HIV antibody positive test with uncontrolled or unstable treatment
• Positive urine drug screen for substances other than cannabis and prescribed medications
• Positive urine pregnancy test at screening in those considered of childbearing potential