GLP-1 Drugs and Rheumatoid Arthritis: One Study, and What It Cannot Tell You

In the only study to look, people with rheumatoid arthritis and a BMI of 27 or above who took a GLP‑1 had greater reductions in disease activity and in pain than people who were prescribed one and did not take it. Inflammatory markers (ESR and CRP) fell within the treated group too.^[1] It is a retrospective chart review of 215 people, it is the first of its kind, and nearly a third of the treated group stopped the drug before the year was out.

What was measured

Between 2018 and 2024, researchers identified 173 patients with rheumatoid arthritis and a BMI of at least 27 who were prescribed a GLP‑1 (semaglutide or tirzepatide) and took it, and 42 who were prescribed one and did not. Everyone was assessed at three-month intervals for up to a year. Changes were modeled with linear mixed effects and adjusted for baseline characteristics that differed between the groups.^[1]

The comparator is the unusual part, so understand it before reading anything into the result. The control group is not untreated patients; it is patients whose clinician made the same prescribing decision, who then did not take the drug. That removes one large source of bias, the decision to prescribe, and introduces another, because the reasons a person does not start a medication are rarely unrelated to how they are doing.

Why weight loss alone does not settle it

The obvious explanation is that losing weight reduces mechanical load on joints and lowers systemic inflammation, and that the GLP‑1 is incidental to both. That would still be a useful finding for someone with rheumatoid arthritis carrying excess weight. But it would mean the drug is doing nothing to the disease that any weight loss would not do, and this study cannot separate the two.

The falls in ESR and CRP are consistent with either reading. Inflammatory markers respond to weight loss. They also respond to disease-modifying therapy, which everyone in the study was presumably continuing. Nothing here supports substituting a GLP‑1 for a DMARD, and nobody has studied that.

And lupus, where the records are new

Lupus is routinely bracketed with rheumatoid arthritis in coverage of this question, but it has its own records, and they mostly ask something else. In a 2026 US study of 1,914 people with lupus and type 2 diabetes, those who started a GLP‑1 had fewer major heart events (hazard ratio 0.66) and less kidney decline (0.77) than those who started a DPP‑4 inhibitor.^[2] A second records study, of 9,386 people, found less lupus nephritis at one year (relative risk 0.59) and fewer recorded flares.^[3] Disease activity itself has barely been measured: a New York lupus cohort followed 18 patients and recorded one flare,^[4] and a 2026 scoping review identified no randomized trials.^[5]

If you have lupus, these are heart and kidney findings from people with diabetes, no trial has tested the drugs in your condition, and a general claim about “autoimmune disease” is not a substitute for a study in your condition.

What to take from it

• This is one retrospective chart review of 215 people. It is a reason for a proper trial, not a reason to expect a joint benefit.
• No GLP-1 is approved for rheumatoid arthritis, and none of this displaces disease-modifying therapy.
• The improvement may be weight loss rather than anything specific to the drug. The study cannot tell the difference.
• Gastrointestinal side effects drove a substantial share of people off the drug within a year, in a population already managing a chronic illness.
• Methotrexate, biologics and other DMARDs interact with a great deal. Adding any drug is a conversation with your rheumatologist.

For a different inflammatory condition where a randomized trial does exist, see a biologic plus tirzepatide for psoriasis. For what these drugs do to joints through weight alone, see GLP-1s and knee osteoarthritis.