A patient at an addiction treatment center in Pennsylvania said it first. After starting Ozempic for weight loss, she noticed something unexpected. The mental noise had gone quiet. The constant thoughts about her next drink, her next pill, her next cigarette had faded. She wasn’t trying to quit anything. She just wanted to lose weight. But something else had happened in her brain. That observation, repeated by patient after patient, is now reshaping addiction medicine.
Three things to know before you read further
-
GLP-1 drugs (Ozempic, Wegovy, Mounjaro, Zepbound) appear to reduce cravings across nearly every addictive substance. Recent studies show signal in alcohol, opioid, nicotine, cocaine, and cannabis use disorders.
-
They are not yet FDA-approved for addiction. Use today is off-label, often prescribed at addiction treatment centers as part of a broader recovery plan.
-
They are not a cure. The most experienced clinicians in this space describe GLP-1 medications as buying patients time, lowering the volume on cravings long enough for therapy and recovery to do the deeper work.
A Brief Word From the Gila Monster That Started All This
The original GLP-1 molecule was discovered in the saliva of the Gila monster, a desert lizard. With a few amino acids added to extend its half-life from minutes to roughly 24 hours, it became biologically usable as a medication. The first wave of these drugs, exenatide and liraglutide, was approved in the early 2000s for type 2 diabetes.
Dr. Steven D. Klein, MD/PhD, FAAP, an addiction medicine physician at Caron Treatment Centers in Wernersville, Pennsylvania, walked NatCon attendees through this history because of one important fact patients often miss: these drugs are not new.
“These medications have been game changers in type two diabetes, but did not really get the attention until their effects on obesity. So the effects on weight really changed everything.” — Dr. Steven D. Klein, MD/PhD, Addiction Medicine, Caron Treatment Centers
The second wave, semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro, Zepbound), reached the public eye through landmark weight loss data published in the New England Journal of Medicine. The same molecules act on three different organ systems. In the pancreas, they raise insulin and lower glucagon, which helps with diabetes. In the stomach, they slow gastric emptying, which produces fullness. In the brain, they appear to dampen appetite, increase satiety, and quite possibly do something else: they appear to dampen cravings.
Why a Diabetes Drug Might Treat Cravings
To understand how, patients need a quick tour of the brain’s reward circuit. Klein walked the audience through it in plain language for a non-specialist crowd.
The brain’s reward system is called the mesolimbic system. Several regions matter:
-
Ventral tegmental area (VTA): the engine room. The starting point of dopamine release. Klein’s analogy: the boiler room of the Titanic, where coal is shoveled into the furnaces.
-
Nucleus accumbens: where reward and reinforcement are felt. Goal-directed behavior and compulsions live here.
-
Prefrontal cortex: the brake pads. Executive function. Notably, this region is not fully developed until roughly age 25.
-
Hippocampus: memory and contextual cues, the “people, places, and things” that trigger cravings.
-
Lateral dorsal tegmental nucleus: what Klein calls the captain’s quarters, setting the brain’s overall “hedonistic tone.”
“Reward is not a glitch in the system. It’s the entire system. It’s why we do everything. It’s why we procreate, seek shelter, seek warmth, seek food, and get up in the morning.” — Dr. Steven D. Klein
In a non-addicted brain, a stimulus produces a dopamine response, then returns to baseline. In an addicted brain, the response drops below baseline. That undershoot is craving. The person now seeks the substance not for pleasure but to feel normal. With each use, the brain rewires itself a little more, and the floor drops a little lower. Klein calls this region of the curve “misery.”
The key fact for patients: every single one of those brain regions has a GLP-1 receptor on it. When semaglutide or tirzepatide binds to those receptors, multiple things happen. In the VTA, dopamine release is dampened. In the nucleus accumbens, reward gets uncoupled from the stimulus. In the prefrontal cortex, top-down control is restored. In the hippocampus, there is even early evidence of increased neuronal plasticity, which may help patients form new associations during recovery.
A January 2026 systematic review in *Frontiers in Pharmacology* from the Mayo Clinic confirmed the breadth of this signal across alcohol, nicotine, cocaine, and opioid models, both preclinical and clinical.
The Salience Shift, or Why Cravings Stop Feeling So Loud
Salience is how the brain assigns importance to stimuli. Right now, you are not thinking about the fan in the room until someone mentions it. Then suddenly, you can hear it.
In addiction, drug cues become the highest-priority signal in the system. Higher than food. Higher than relationships. Higher even than self-preservation.
“Salience and addiction is very important to the addicted person. Their drug cravings are the most important thing in their hierarchy, much more important than self-preservation, than relationships, than anything else in that equation.” — Dr. Steven D. Klein
Patients on GLP-1s describe a quieting of cravings. fMRI studies show reduced prefrontal activation in response to food cues, suggesting the brain is no longer assigning those cues such high importance. This may explain why GLP-1s are showing effect across so many different substances. Unlike naltrexone (which blocks opioid receptors) or acamprosate (which acts on glutamate), GLP-1 medications appear to recalibrate the brain’s reward weighting itself.
Klein offered patients a simple linguistic frame. Replace the word appetite with words like craving, urge, longing, yearning, hunger, obsession, compulsion, or drive. Replace the word satiety with calmness, contentment, stillness, peacefulness. The same biology that helps a person stop reaching for a second helping of pasta may also be what helps them stop reaching for a second drink.
What the Evidence Actually Shows
The most important study to date came out just weeks before the NatCon panel.
In March 2026, researchers at Washington University School of Medicine in St. Louis published a landmark cohort study in *The BMJ* analyzing the records of more than 600,000 U.S. veterans with type 2 diabetes. Some had been started on a GLP-1 medication. Others had been started on a different diabetes drug class, SGLT-2 inhibitors. The patients were followed for up to three years.
Among veterans who did not have a substance use disorder at baseline, GLP-1 use was tied to lower rates of new diagnoses of:
-
Alcohol use disorder
-
Cannabis use disorder
-
Cocaine use disorder
-
Nicotine use disorder
-
Opioid use disorder
Among veterans who already had a substance use disorder, the numbers were even more striking:
-
25% fewer drug-related hospitalizations
-
40% fewer overdoses
-
50% fewer substance-related deaths
-
25% fewer suicide attempts and suicidal ideation events
Earlier in 2025, the Hendershot et al. randomized controlled trial in JAMA Psychiatry established that low-dose semaglutide reduced alcohol cravings, drinking quantity, and the frequency of heavy drinking days in adults with alcohol use disorder. Nearly 40% of the semaglutide group had no heavy drinking days by week 9, compared to roughly 20% on placebo. This was the first gold-standard trial in the field.
A January 2026 systematic review in *Addictive Behaviors Reports* identified 33 active GLP-1 trials registered for substance use disorders, including ongoing Phase 2 work for opioid use disorder at Penn State Hershey Medical Center. The pipeline is real, and it is full.
Why GLP-1s Are Not a Magic Bullet
Klein, who has spoken openly about his own long-term recovery and his personal experience taking a GLP-1, was direct on the panel about this point: there is no silver bullet for addiction.
“I don’t think there’s a silver bullet that treats addiction. There is a treatment to addiction. That treatment is recovery. And I think these medications facilitate recovery capital.” — Dr. Steven D. Klein
The closest analogy comes from weight loss research. Patients who pair GLP-1s with healthy lifestyle change regain weight more slowly than those who do not. The same logic likely applies to substance use. The medication holds the door open. The therapy, the support groups, the sponsor relationships, the rebuilding of family and community: those are what walk through it.
For some patients, that broader infrastructure is not yet accessible. Klein described the population he serves through the Open Doors program in Rhode Island, a transitional housing program for women exiting the justice system. Many of those women cannot afford the luxury of full recovery work yet. Their priorities are housing, family reunification, and basic stability. For them, GLP-1s “hold the bookmark,” keeping cravings manageable until the rest of the support system can catch up.
What Patients and Families Should Know Before Asking
A short, practical checklist if you are considering whether to bring this up with a doctor:
-
GLP-1 medications for substance use disorder are off-label today. Insurance will not typically cover them for that indication unless the patient also has obesity or type 2 diabetes.
-
The standards of care for substance use disorder remain in place: naltrexone or acamprosate for alcohol use disorder; buprenorphine, methadone, or naltrexone for opioid use disorder. Most addiction physicians will consider these first.
-
Side effects matter. Nausea and vomiting are common, usually self-limiting, and tend to improve with time. Patients should be screened for restrictive eating disorders before starting, since the appetite suppression can worsen those patterns.
-
Dose matters too. The doses used for addiction tend to be lower than those used for weight loss, often around 0.5 mg of semaglutide.
-
Cost is a real barrier. Some clinics offer compounded versions or direct-pay programs at lower prices. Caron’s program runs roughly $250 per month delivered.
Patients with a personal or family history of medullary thyroid cancer, multiple endocrine neoplasia type 2, or active pancreatitis should not take GLP-1 medications. Pregnant patients should not start these medications. Always consult a physician before starting or stopping any prescription.
Frequently Asked Questions From the NatCon Audience
Q: Will patients have to take these medications for life? Klein’s answer: probably yes, for many patients. “Chronic brain diseases require chronic brain treatments.” But he emphasized that what really matters is whether the patient is using the time on the medication to do the deeper recovery work.
Q: How does the molecule know to turn off the addiction without blunting normal pleasure, like sex or other rewards? This was the question of the day, asked from the audience by Dr. Bruce Bassi, an addiction psychiatrist and the founder of TelepsychHealth. Klein’s answer: “The fire that’s closest to the house is the one that gets the water.” In his clinical experience, the compulsive behaviors are what get affected. Patients on GLP-1s are not reporting the kind of broad anhedonia (loss of pleasure) seen with some other medications. Klein cited his own experience: he used to stop at a casino on his drive to the Jersey Shore. After starting a GLP-1, he simply did not feel pulled to anymore.
Q: What are the side effects to look out for? Most common: nausea and vomiting, usually self-limiting. Serious but rare: pancreatitis (the fear of which has not been borne out in large studies). Watch carefully for restrictive eating disorder triggering. Avoid in pregnancy.
Q: Does the patient need to be overweight to benefit for addiction? No. Klein routinely uses GLP-1s in normal-weight patients, often at lower doses. He coordinates with primary care providers to monitor weight, especially in patients with a BMI under 27.
Q: How much do these medications cost out of pocket? Cost is one of the biggest barriers patients face. Brand-name semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro, Zepbound) typically run between $900 and $1,300 per month at the pharmacy counter without insurance. When prescribed for substance use disorder specifically, insurance coverage is rare. Klein noted that even getting approval for the obesity indication has become much harder. Approval rates at his clinic have dropped from roughly 50% a few years ago to between 5% and 10% today, in part because most payers now require documentation of six months of failed weight loss attempts before approving coverage. To work around the cost barrier, many addiction clinics now offer compounded versions of these medications at significantly lower prices. Caron’s Bridge Initiative program, for example, offers a flat $250 per month direct-to-patient model that includes the medication and home delivery. For patients without insurance coverage and without access to a clinic-based program, that price gap (roughly $250 to over $1,000 per month) is often the difference between starting treatment and not.
Q: Why isn’t this FDA-approved for addiction yet? “Directly, the holdup is randomized controlled trials.” Industry funding has flowed primarily into obesity and diabetes. The first FDA approval for an addiction indication, Klein predicts, will likely be for alcohol use disorder.
For decades, addiction medicine has had a small toolbox. Naltrexone, buprenorphine, methadone, acamprosate, disulfiram. All useful, all limited. GLP-1 receptor agonists are the first new class of medications in years that may treat addiction at the level of the brain’s underlying reward system itself, not just one drug at a time. The data are early. The trials are running. But for patients and families who have run out of options, this is real reason for hope.
