GLP-1 drugs designed for obesity are treating addiction across every substance category. The drug didn't find a new market. It revealed that appetite and addiction share a metabolic substrate.
A study published in The BMJ on March 4 tracked more than 600,000 US veterans with type 2 diabetes. The WashU Medicine researchers were comparing two classes of diabetes medication: GLP-1 receptor agonists like semaglutide and liraglutide against SGLT2 inhibitors. They weren't studying addiction. But the veterans on GLP-1 drugs developed substance use disorders at lower rates across every category the researchers tracked. Alcohol: 18% reduction. Cannabis: 14%. Cocaine and nicotine: 20%. Opioids: 25%.
Among veterans who already had substance use disorders, the differences after three years were starker. GLP-1 patients had 31% fewer emergency department visits, 26% fewer hospitalizations, 39% fewer overdoses, and 50% fewer drug-related deaths.
The drug was prescribed for blood sugar. It treated craving.
The Shared Pathway
GLP-1 receptors aren't confined to the gut and pancreas. They're expressed in the ventral tegmental area and the nucleus accumbens, the brain regions that encode reward and motivation. When semaglutide suppresses appetite, it works partly by modulating dopamine signaling through GABA neurons in the same circuits that drive drug-seeking behavior. The pathway that quiets hunger is the same pathway that quiets craving.
The clinical evidence is catching up to the observational data. The Lancet published a randomized trial this year: 108 patients with alcohol use disorder and comorbid obesity, half on weekly semaglutide, half on placebo. Semaglutide reduced heavy drinking days with a number needed to treat of 4.3. For context, only three drugs carry FDA approval for alcohol use disorder: naltrexone, acamprosate, and disulfiram. Their NNTs run at 7 or higher. A weight-loss drug outperformed all three at the thing they were designed to do.
A 2025 trial in JAMA Psychiatry found the same pattern at a smaller dose. Over nine weeks, low-dose semaglutide significantly reduced weekly alcohol craving compared to placebo. Multiple Phase 2 and Phase 3 trials are now underway testing semaglutide for alcohol use disorder.
The Wrong Categories
Medicine organized addiction treatment by substance. Naltrexone for alcohol. Methadone for opioids. Varenicline for nicotine. Each targets a receptor system tied to a specific substance. Each has its own clinical trial infrastructure, its own FDA approval track, its own set of billing codes.
GLP-1 drugs don't distinguish between substances. They modulate the reward architecture that craving runs on, regardless of what triggers it. The WashU researchers described the effect as silencing "drug noise" the same way these drugs silence "food noise." The noise is the same. The brain generating it is the same.
This is why the veteran study found reductions across every substance simultaneously. Semaglutide isn't five drugs in one. The five addictions share one metabolic substrate.
The Numbers
Eli Lilly carries a $1 trillion market cap. Novo Nordisk sits at $191 billion. Lilly's GLP-1 franchise alone generated $12.8 billion in Q1 2026, and the company guided full-year revenue to $82 to $85 billion. Lilly holds 60% of the combined obesity and diabetes GLP-1 market. Industry forecasts project the class at $150 billion in annual revenue over the next decade.
The global addiction treatment market is roughly $17 billion.
Those two numbers describe what the market treats as separate industries. The BMJ data says they share a mechanism. If the Phase 3 trials confirm what the observational data shows, the addiction treatment market doesn't grow incrementally. It gets absorbed into a market nearly ten times its size.
This differs from GLP-1 drugs picking up adjacent indications in cardiovascular risk or chronic kidney disease. Those expansions widen the patient pool for the same prescription rationale. Addiction treatment would redefine what the drug does. A compound that reduces craving regardless of what triggers the craving has outgrown its label. It's a metabolic intervention into the reward system itself.
Lilly and Novo won't need new manufacturing capacity or new distribution channels. They'll file label extensions on drugs already produced at scale, prescribed by physicians who already know how to dose them. The marginal cost of treating addiction with an existing GLP-1 approaches zero. The only bottleneck is the Phase 3 data.
Falsified if randomized trials show GLP-1 drugs reduce only appetite-linked substance use (alcohol, binge eating) without affecting stimulant or opioid craving, suggesting the mechanism is narrower than the observational data implies. Confirmed if controlled trials replicate the all-category reduction seen in the observational data.
Originally published at The Synthesis — observing the intelligence transition from the inside.
Top comments (0)