STEP 9 shows semaglutide reduces knee osteoarthritis pain

The New England Journal of Medicine published results on Oct. 30, 2024 from STEP 9, a 68-week phase 3 trial of once-weekly semaglutide 2.4 mg in 407 adults who had obesity and moderate knee osteoarthritis with at least moderately severe pain ^[2]. People on semaglutide lost more weight and reported larger reductions in knee pain than those on placebo, and both groups received counseling on a reduced-calorie diet and physical activity ^[1].

What the trial measured

Participants were randomly assigned 2:1 to semaglutide 2.4 mg or placebo at 61 sites in 11 countries, with enrollment between October 2021 and March 2022 ^[1][2]. That worked out to 271 people on semaglutide and 136 on placebo ^[3]. Mean age was 56, 81.6% were women, mean BMI was 40.3, and mean starting body weight was 239.5 pounds (108.6 kg) ^[1][2][3]. Everyone had a clinical diagnosis of knee osteoarthritis by American College of Rheumatology criteria plus moderate X-ray changes (Kellgren–Lawrence grade 2 or 3) and a WOMAC pain score of at least 40 out of 100 ^[2].

There were two primary endpoints: percentage change in body weight and change in the WOMAC pain score, a 0-to-100 scale where higher numbers mean worse pain ^[1]. At week 68, mean weight change was –13.7% with semaglutide versus –3.2% with placebo (95% CI –12.3 to –8.6; P<0.001) — a gap of about 10.5 percentage points ^[2][3]. Mean WOMAC pain score fell from a baseline of 70.9 by 41.7 points with semaglutide versus 27.5 points with placebo (95% CI –20.0 to –8.3; P<0.001) ^[2]. Physical function on the SF-36 improved by 12.0 points versus 6.5 points (95% CI 3.1 to 8.0; P<0.001) ^[1][2]. One independent summary put the number needed to treat for a clinically meaningful improvement in WOMAC function at roughly 5 over 68 weeks ^[3].

Safety and study limits

STEP 9 used a narrow safety data collection plan, recording only serious adverse events, events leading to discontinuation, events requiring invasive knee procedures, medication errors, acute pancreatitis, COVID-19 and pregnancy-related events ^[2]. Serious adverse events were similar between groups (10.0% semaglutide, 8.1% placebo) ^[2]. Adverse events leading to permanent discontinuation occurred in 6.7% on semaglutide and 3.0% on placebo, most often gastrointestinal problems (2.2% versus 0%) ^[1][2]. Neoplasms — benign, malignant and unspecified — led to discontinuation in 1.9% versus 1.5% ^[2].

An independent review by RxFiles noted several caveats: the trial was not designed for a full safety analysis; weight loss and side effects made it likely that participants and investigators could guess who was on the drug, which can inflate subjective pain reporting; and the entry criteria were strict, limiting how broadly the results apply ^[3]. It also flagged that the study was not built to show whether semaglutide relieves pain through any mechanism beyond weight loss ^[3]. Earlier work suggests each 1% drop in body weight corresponds to about a 2% drop in WOMAC scores, and that a 25% weight reduction would be needed for a 50% score improvement ^[3]. A separate 52-week trial of liraglutide 3 mg produced little weight loss versus placebo and no statistically significant pain improvement ^[3].

Why it matters for patients

Knee osteoarthritis is common in people with obesity, and the risk of developing it is more than four times higher, according to lead author Henning Bliddal of the Parker Institute at Copenhagen University Hospital ^[2]. Weight loss and exercise are already first-line recommendations for knee osteoarthritis, but sticking with them is hard for many people ^[2][3]. STEP 9 adds randomized evidence that a GLP-1 drug used alongside lifestyle changes can produce both weight loss and pain relief in this group.

Importantly, semaglutide is not approved in the U.S. to treat knee osteoarthritis, and Novo Nordisk states plainly that safety and efficacy for that use are not established ^[2]. Using it for arthritis would be off-label ^[3]. The trial also did not follow people long-term after stopping, so whether pain and function benefits last — or fade with weight regain — is not yet known ^[3].

What happens next

Novo Nordisk said it looks "forward to working with regulatory authorities on next steps based on these findings," but the company did not announce a filing date or a specific regulatory pathway ^[2]. RxFiles concluded the results support further investigation in larger, longer trials ^[3]. No such trials were announced in these sources.