GLP-1 Drugs and Autoimmune Disease: What They Are, How They Work, and Where the Research Stands
I can't remember the last week that's gone by without a patient asking me about Ozempic or Wegovy. Used to be strictly about weight. Now it's just as often about their joints. Could this help my RA too? My lupus? My psoriatic arthritis?
It's a fair question. Let's start from the beginning.
What GLP-1 drugs are and how they work
GLP-1 stands for glucagon-like peptide-1. It's a hormone your gut releases naturally after you eat. It tells your pancreas to release insulin and to ease off glucagon (a hormone that raises blood sugar), and it tells your brain you're full. It also slows down how fast food leaves your stomach, which is part of why people on these drugs feel full longer and sometimes get nauseated if they eat too much too fast.
The problem with natural GLP-1 is that your body breaks it down within minutes. GLP-1 drugs (semaglutide, liraglutide, tirzepatide, dulaglutide, and a few others) are engineered versions built to resist that breakdown, so they stick around for hours to a week depending on the drug. Same signal, held much longer.
What they're approved for right now
Not every drug in the class carries every approval, but as of today, GLP-1 drugs are FDA approved for:
Type 2 diabetes
Chronic weight management, for adults with obesity or overweight with a related condition
Reducing cardiovascular risk in adults with established heart disease and obesity or overweight (semaglutide)
Chronic kidney disease in type 2 diabetes (semaglutide)
MASH, metabolic dysfunction-associated steatohepatitis, with moderate to advanced liver fibrosis (semaglutide)
Moderate to severe obstructive sleep apnea in adults with obesity (tirzepatide)
That list keeps growing. But none of it is rheumatology yet. The immune system piece is why we're watching so closely.
The side effects
Common:
Nausea, vomiting, diarrhea, constipation
These hit hardest when starting the drug or moving up a dose, and they're the main reason people stop
Slower dose increases and smaller meals help
Rare but serious:
Pancreatitis
Gallbladder disease
Slowed stomach emptying that occasionally becomes severe
Not for people with a personal or family history of medullary thyroid cancer or a syndrome called MEN2, based on tumors seen in rodent studies
For my patients specifically:
Muscle loss. Rapid weight loss on these drugs takes muscle along with fat, and muscle is what stabilizes and protects the joints.
Bone loss. In one trial, a year of semaglutide lowered bone density at the hip and spine compared to placebo. That matters extra in rheumatology, where inflammation and steroid use already raise fracture risk. The reassuring part: fracture rates in the large diabetes studies haven't gone up, and in a separate trial, people who exercised during GLP-1 treatment lost no bone density at all.
So if you're on one of these drugs, resistance training and adequate protein aren't optional extras. They protect your muscle and your bones at the same time.
The proposed mechanism in autoimmune disease
Here's what got rheumatologists paying attention. A few immune cells carry GLP-1 receptors directly. Most don't. But these drugs still seem to calm inflammation body-wide, likely through indirect routes, including receptors in the nervous system that signal the immune system to settle down. In lab studies, the end result is lower levels of inflammatory signals, including TNF, the same signal blocked by RA drugs like adalimumab and etanercept.
Then there's the weight angle, which is separate from all of that. Obesity modestly raises your risk of developing RA in the first place, and it makes RA harder to control once you have it. Fat tissue itself produces inflammatory signaling. Patients on GLP-1 drugs can lose 15 to 20 percent of their body weight, so less fat tissue means one less source of that signaling.
Earlier this year, researchers in Denmark added another piece. In a small study (44 patients), they looked directly inside the joints of people with RA and spondyloarthritis and found natural GLP-1, along with the enzyme that breaks it down, in the synovial fluid itself. That doesn't prove anything about treatment on its own. It raises the possibility that these drugs, at high enough doses, could reach the joint and act there directly. For now that's a hypothesis. The prevailing view is still that any benefit runs through weight loss and those indirect immune effects.
So there are at least two, maybe three, separate biological reasons this might matter.
Where the science stands
The research falls into two separate questions.
Question one: if you already have an autoimmune disease, do these drugs help?
The early data here is encouraging, though it all comes from observational studies, not trials:
A 2025 chart review followed 173 RA patients who started semaglutide or tirzepatide, compared against 42 who were prescribed one but never took it. Over a year, the treated group had greater improvement in disease activity, pain, weight, cholesterol, and A1c.
In psoriatic arthritis, a large database study found lower mortality and fewer heart attacks among patients on GLP-1 drugs compared to matched patients who weren't.
In patients who have both lupus and type 2 diabetes, GLP-1 use was tied to lower rates of new lupus nephritis (kidney inflammation, one of lupus's most serious complications), fewer lupus flares, and lower mortality at one year.
Question two: do these drugs change your risk of developing an autoimmune disease in the first place?
Nobody knows yet. Several large database studies have looked at this and they disagree. Some found lower rates of new autoimmune diagnoses in people on GLP-1 drugs. Others found slightly higher rates, including for RA. The most rigorous study so far, published this year in Arthritis & Rheumatology, followed over 229,000 adults and found no meaningful difference in either direction.
The disagreement isn't shocking. These studies pull from different databases, follow patients for different lengths of time, and can't fully account for why one patient was prescribed one drug over another. This is what early research looks like before dedicated trials settle it.
My read of the evidence today: on question one, cautious optimism. On question two, no convincing signal either way.
What's in the pipeline
The field is moving from noticing this by accident in diabetic patients to testing it on purpose.
A proof-of-concept randomized trial has been registered to test semaglutide in non-diabetic patients with psoriatic arthritis, looking at vascular and inflammatory markers rather than just weight.
Tirzepatide is being tested in Crohn's disease, a different immune-mediated condition, in patients who also have obesity.
Trials pairing GLP-1 drugs with existing biologics are underway in psoriatic disease, tracking what adding the GLP-1 does to outcomes.
A joint-targeted version already exists and is in human trials: a French company is testing liraglutide injected directly into the knee for osteoarthritis, now in a placebo-controlled Phase 2 trial with results expected in 2027. Osteoarthritis isn't an autoimmune disease, so this doesn't answer our question directly. But if injecting GLP-1 into a joint helps an inflamed OA knee, that finding won't stay confined to OA for long. For RA and its cousins, the joint-injection idea is still just that, an idea.
None of this is close to changing practice. But the question has moved from "we noticed something" to "let's go test it."
What I've seen in my own practice
I'll add my own experience here, with the usual caveat that stories aren't studies.
I've had numerous patients tell me their joints feel better after starting a GLP-1 drug. Some of that is surely the weight loss. Less load on the knees and feet is real relief on its own. But the reports keep coming, including in joints that don't bear weight, like the hands.
I should name the bias in this, though. Patients who feel better bring it up. Patients who feel no different don't walk in and announce that their GLP-1 isn't helping their joints. We call this reporting bias, and it means what I hear in clinic is skewed toward the wins. That's part of why anecdotes sit at the bottom of the evidence ladder, mine included.
One case stands out. A patient with seropositive rheumatoid arthritis (meaning the blood tests show the antibodies we associate with RA) is currently off immunosuppression entirely, with no evidence of active inflammation on exam. No swelling, no signs of active disease. They're on a low dose of a GLP-1 drug, and they've also made real lifestyle changes, so I can't hand the credit to any one thing. Seropositive RA staying quiet without immunosuppression is uncommon, and I don't know how much the GLP-1 is contributing. But it's the kind of case that keeps me paying attention.
What about microdosing?
Microdosing means taking a GLP-1 drug at a fraction of the standard dose, usually through a compounding pharmacy, with the pitch that you get the benefits without the nausea or the cost.
There are no published studies of microdosing for autoimmune disease, and no randomized trials of microdosing for anything. Every study in this post used standard doses. Some physicians, including rheumatologists, have described patients who improved on microdoses. Those are anecdotes, the same tier of evidence as my stories above.
What this means in my office
None of this makes GLP-1 drugs an RA, lupus, or PsA treatment. Not yet, and given how mixed the comparative data looks right now, maybe not for any of these diseases the way biologics or methotrexate are. Almost everything above comes from chart reviews and registry databases, not trials built to test whether these drugs treat autoimmune disease. Neither the ACR nor EULAR recommend GLP-1 drugs for RA, PsA, or lupus at this point, even though both groups have pushed weight loss as part of managing these diseases for years.
Here's how I'm using this with patients today. If you have RA, PsA, or lupus and you already qualify for a GLP-1 drug because of your weight, your diabetes, your heart disease risk, or now MASH or sleep apnea, this is one more piece of context for that conversation. If you don't have another reason to be on one, insurance won't cover a GLP-1 drug for an autoimmune condition, because the current evidence doesn't support that use.
The good news is that this question is being actively studied. Over the next few years, the trials in the pipeline should tell us whether these drugs have a place in treating autoimmune disease on its own, not just alongside diabetes or weight loss. When those results come in, I'll write about them here.
Sources
Chart review of GLP-1 drug use in RA patients with overweight or obesity, from UCLA. Kellner DA, et al. ACR Open Rheumatology, 2025.
Large database comparison of new RA and gout diagnoses in diabetic patients on GLP-1 drugs versus an older diabetes medication. Presented by Betul Ibis, MD, at ACR Convergence 2025 (abstract 2657).
Study on GLP-1 drug use, lupus nephritis, and flares in patients with lupus and type 2 diabetes. Hanif M, et al. The American Journal of Medicine, 2026.
Database study on mortality and cardiovascular events in psoriatic arthritis patients on GLP-1 drugs. Presented by Nanuka Tsibadze, MD, at ACR Convergence 2025 (abstract 0849).
Population-based case-control study finding a short-term rise in new RA diagnoses after starting GLP-1 drugs. Israel A, Hassan F, Merzon E, et al. Therapeutic Advances in Musculoskeletal Disease, 2026.
Registry study finding higher rates of several autoimmune conditions, including RA, with GLP-1 drugs compared to an older diabetes medication. Reported in European Medical Journal, 2025.
Study comparing autoimmune disease risk between GLP-1 drugs and two other diabetes drug classes. Mahajan A, et al. ACR Open Rheumatology, 2026.
Comparative study on autoimmune rheumatic disease risk with GLP-1 drugs versus other diabetes drug classes, 229,300 adults. Karacabeyli D, et al. Arthritis & Rheumatology, 2026.
Detection of natural GLP-1 and its breakdown enzyme in joint fluid of RA and spondyloarthritis patients. Broksø AD, et al. The Lancet Rheumatology, 2026. DOI: 10.1016/S2665-9913(26)00074-3.
Trial showing reduced hip and spine bone density after one year of semaglutide versus placebo. Hansen MS, et al. eClinicalMedicine, 2024.
Trial showing exercise preserved bone density during GLP-1 treatment for weight loss. Jensen SBK, et al. JAMA Network Open, 2024.
Registered proof-of-concept trial of semaglutide in non-diabetic psoriatic arthritis patients. ClinicalTrials.gov, NCT07251556.
Phase 2a trial of intra-articular liraglutide (4P004) in knee osteoarthritis with synovitis. ClinicalTrials.gov, NCT07225829.
Review of GLP-1 drug studies in psoriasis and psoriatic arthritis, including the small dose-level trials mentioned in the microdosing section. Frontiers in Immunology, 2026.
