Retatrutide and knee osteoarthritis: two phase 3 trials

Retatrutide, an experimental once-weekly injection from Eli Lilly, reduced knee pain in adults with osteoarthritis and obesity in two phase 3 trials. The full results were published on November 8, 2026 in the New England Journal of Medicine, in a paper titled “Retatrutide Once Weekly in Adults with Knee Osteoarthritis Pain and Obesity”. No regulator has approved the drug yet.

The data were presented the same day at the American College of Rheumatology meeting (ACR Convergence 2026) in Orlando, which runs from November 6 to 11. Philip Conaghan, professor of musculoskeletal medicine at the University of Leeds, presented the TRIUMPH-4 trial at 11:15 a.m. Eastern time, and the second trial, TRIUMPH-1-OA, was shown at a separate poster session the same day. The lead author of the paper is Thomas Schnitzer, a professor at Northwestern University’s Feinberg School of Medicine.

Until now the public had only the company’s brief report on TRIUMPH-4 from December 2025. This is the first time results by dose, safety data and the second trial, with 574 participants, are available.

How were the trials designed?

Retatrutide acts on three hormone receptors: GIP, GLP-1 and glucagon. Drugs in use today, such as semaglutide and tirzepatide, target one or two receptors and typically lead to a loss of 10 to 20 percent of body weight. An earlier trial, STEP 9, showed that semaglutide also eases knee pain in people with obesity.

TRIUMPH-4 enrolled 445 adults without diabetes, with a body mass index of at least 27 and painful knee osteoarthritis. They received 9 mg of the drug, 12 mg or placebo for 68 weeks. In TRIUMPH-1-OA, 574 participants received 4, 9 or 12 mg of the drug or placebo for 80 weeks. Together the trials included more than a thousand people, and about 80 percent of them had a body mass index above 35.

Pain was measured with the WOMAC scale, on which participants rate their pain from 0 to 10 at each visit. They also entered a daily score in an app.

How much did pain fall?

In TRIUMPH-4, pain dropped by 3.96 points at 68 weeks on the 9 mg dose and by 3.74 points on 12 mg. On placebo it dropped by 2.12 points. Body weight fell by 20.0 percent on the lower dose and 23.7 percent on the higher one, against 4.6 percent on placebo.

In TRIUMPH-1-OA, weight loss ranged from 17.5 percent (4 mg) to 24.8 percent (12 mg), against 4.3 percent on placebo. The difference in pain versus placebo after 80 weeks was 1.35 points for 4 mg, 1.61 for 9 mg and 1.77 for 12 mg. When the analysis is run on an intention-to-treat basis, the differences are smaller but remain statistically significant.

Pooled data from both trials show the following at week 68:

• pain fell by at least half in 69.6 percent of patients on 9 mg (189 of 248) and 63.2 percent on 12 mg (167 of 236), compared with 37.8 percent on placebo (86 of 185)
• knee function improved by at least half in 64.6 percent of patients on 9 mg, 63.5 percent on 12 mg and 33.1 percent on placebo

The researchers also report lower blood pressure, lower blood fats and less systemic inflammation in participants on the drug.

What does the lead author say?

Schnitzer explained in a Northwestern University statement how he interprets the link between weight loss and pain:

“What the findings really say is that when people who are overweight and have osteoarthritis lose a substantial amount of weight with this treatment, there is a good chance their knee pain will be significantly reduced.”

He added: “The main factor is that you are putting less load on the knees, so there is less pain and you can increase function.” He also described how participants responded: “Participants had remarkable responses. Many people in this study lost a substantial amount of weight, and some of them up to 100 pounds (45 kilograms) or more.”

The first step for obese patients with painful osteoarthritis, he said, remains diet, exercise and lifestyle change, which are cheaper and carry no known gastrointestinal side effects or loss of the muscle mass that stabilizes the knee. If approved by the US Food and Drug Administration, the drug would be considered for those who fail to lose weight and are still limited by pain.

What are the side effects?

The most common were nausea, diarrhea and constipation, mostly while the dose was being increased. In TRIUMPH-4, 12.8 percent of patients on 9 mg and 18.2 percent on 12 mg stopped treatment because of side effects, compared with 4.0 percent on placebo.

Serious side effects occurred in 9.5 percent of patients on 9 mg, 10.8 percent on 12 mg and 2.0 percent on placebo. Dysesthesia, an abnormal sensation on the skin, affected 6.3 to 20.9 percent of patients on the drug and 0.7 percent on placebo. One participant in each drug group of TRIUMPH-4 died, and according to the study those deaths were not related to the drug.

What do these trials not show?

Eli Lilly funded both trials and took part in their design, data analysis and the writing of the paper, and its scientists are among the co-authors. Patients who needed a knee replacement were excluded, and the trials were not designed to show whether the drug slows the progression of osteoarthritis or reduces the need for surgery. People who are not overweight were not enrolled either, so it is not known whether the drug eases knee pain in them.

Schnitzer sees this as the biggest unknown: “That would be the goal: to be able to say that you can reduce people’s pain and that you can stop the progression of their osteoarthritis.” On the mechanism, he said:

“These drugs are believed to have a significant anti-inflammatory effect, but the mechanism is not fully understood. Whether reducing inflammation matters for the joint, in addition to reducing the load, needs further research, so in my opinion this is only a first step. I have worked in the field of osteoarthritis for a long time, and I think this opens the way to many new studies that will help us better understand the condition.”

Because the pain relief came with large weight loss, it remains open how much of it comes from unloading the knee and how much from the drug’s own effect on inflammation in the joint.