By Grey Pepper Editorial
Knee osteoarthritis comes with a frustrating piece of standard advice: lose weight. The trouble is that painful knees make exercise difficult, and losing enough weight to change how a joint feels can be an enormous challenge.
On November 8, 2026, the New England Journal of Medicine published results from two Phase 3 trials testing whether an experimental obesity drug could help.
The retatrutide knee pain findings, from studies called TRIUMPH-4 and TRIUMPH-1-OA, showed that people receiving Eli Lilly’s investigational triple hormone receptor agonist experienced significantly greater reductions in knee pain than those receiving placebo.
In a pooled analysis described by Northwestern University, approximately two-thirds of retatrutide-treated participants experienced at least a 50% reduction in knee pain. That sounds like an extraordinary result.
But 38% of participants receiving placebo also achieved at least a 50% reduction.
Both figures matter. The treatment difference supports a meaningful benefit, while the substantial improvement in the placebo groups illustrates why an uncontrolled before-and-after comparison would tell only part of the story.
Across the two knee osteoarthritis study populations, 1,019 adults were randomized. The evidence comes from human, randomized, double-blind, placebo-controlled trials, not preliminary animal experiments.
The drug also produced major weight loss. But the results leave an important unanswered question: did retatrutide merely help painful knees carry less weight, or did it alter something about the osteoarthritis itself?
What the Retatrutide Knee Pain Trials Actually Tested
Retatrutide is a once-weekly injectable peptide designed to activate three hormone receptors: glucose-dependent insulinotropic polypeptide (GIP), glucagon-like peptide-1 (GLP-1) and glucagon.
Its three-receptor design distinguishes it from semaglutide, which targets GLP-1, and tirzepatide, which activates GLP-1 and GIP receptors. Grey Pepper has previously examined the scientific reasoning behind retatrutide’s triple-agonist design.
The two osteoarthritis analyses were conducted in adults without diabetes who had painful knee osteoarthritis and a body mass index of at least 27.
TRIUMPH-4 randomized 445 participants to once-weekly retatrutide at a maintenance dose of 9 mg or 12 mg, or placebo. Treatment continued for 68 weeks.
Its two primary outcomes were change in knee pain and percentage change in body weight.
TRIUMPH-1-OA evaluated 574 participants with knee osteoarthritis within the larger TRIUMPH-1 obesity program. Participants received retatrutide 4 mg, 9 mg or 12 mg, or placebo, over 80 weeks.
The primary osteoarthritis endpoint in this second study was change in knee pain at week 80. The parent TRIUMPH-1 trial enrolled 2,339 participants overall, but not all of them belonged to the knee osteoarthritis population.
Both studies used the Western Ontario and McMaster Universities Osteoarthritis Index, commonly called WOMAC, to measure knee pain.
The pain measure used a scale from 0 to 10, with higher numbers representing worse pain. Researchers also evaluated physical function, including the limitations people experience during ordinary activities.
The investigators used a prespecified hybrid-treatment estimand for pain outcomes, combining different statistical approaches to treatment interruptions and other events depending on their circumstances.
They also reported intention-to-treat analyses. Both approaches favored retatrutide over placebo, although the exact estimates differed.
This distinction matters because the larger percentage reductions from Lilly’s initial December 2025 announcement were based on a different efficacy analysis. They should not be presented interchangeably with the newer primary publication’s results.
What the Retatrutide Phase 3 Results Actually Showed
The newly published results provide a more complete picture than the initial company headline.
In TRIUMPH-4, average WOMAC knee pain scores fell by 3.96 points with retatrutide 9 mg and 3.74 points with retatrutide 12 mg, compared with 2.12 points on placebo at week 68.
The adjusted treatment differences versus placebo were reductions of 1.83 points and 1.61 points, respectively.
The 95% confidence interval for the 9 mg treatment difference ranged from 1.30 to 2.37 points of greater pain reduction. For the 12 mg dose, the corresponding interval ranged from 1.07 to 2.16 points.
Both comparisons were statistically significant, with reported p values below 0.001.
TRIUMPH-1-OA produced a similar pattern over a longer period.
At week 80, average pain scores fell by 3.22 points with 4 mg, 3.48 points with 9 mg and 3.64 points with 12 mg, compared with 1.87 points on placebo.
The adjusted differences versus placebo were 1.35, 1.61 and 1.77 points of additional pain reduction, respectively. All three active-dose comparisons significantly favored retatrutide.
The weight-loss results were substantial too.
In TRIUMPH-4, the primary publication reported average bodyweight reductions of 20.0% with retatrutide 9 mg and 23.7% with 12 mg, compared with 4.6% on placebo under its treatment-policy analysis.
These results establish that retatrutide reduced both weight and reported knee pain significantly more than placebo in the studied population.
They do not establish that every individual experienced a comparable benefit.
They also do not prove that the highest dose produces the greatest pain relief in every patient. In TRIUMPH-4, the 9 mg arm had numerically greater pain reduction than the 12 mg arm, while the pattern in TRIUMPH-1-OA favored progressively higher doses.
Those numerical differences should not be treated as evidence of superiority between active doses without appropriate statistical comparisons.
Lilly’s earlier efficacy analysis of TRIUMPH-4 reported WOMAC pain reductions of 75.8% with 9 mg, 74.3% with 12 mg and 40.3% with placebo. Those percentages were derived in a post-hoc calculation and belong to that specific analysis.
The newer peer-reviewed findings provide the stronger basis for describing the primary treatment effect.
The treatment effect is real in these randomized trials. The magnitude of the placebo-group improvement is equally important for understanding what the results mean.
Why 38% of Placebo Patients Also Halved Their Knee Pain
The most revealing number in Northwestern University’s summary may be the one that does not involve the active drug.
Approximately 38% of placebo recipients experienced at least a 50% reduction in knee pain.
That does not mean the treatment was ineffective. Nor does it mean every improvement in the placebo group was caused by a psychological placebo effect.
Osteoarthritis symptoms fluctuate. People may enter clinical trials when pain has become unusually troublesome, then experience some improvement as symptoms return toward their typical levels.
Trial participation can also change behavior. Participants receive regular assessments and may pay closer attention to activity, diet and other aspects of daily care.
These factors can contribute to improvements in both treatment groups, alongside expectations and the inherent variation in patient-reported pain.
Pain is a legitimate clinical outcome, but it is not a direct anatomical measurement. A person can experience substantial pain relief without necessarily having repaired the underlying cartilage.
The pooled responder figures suggest an advantage of roughly 28 percentage points for retatrutide if approximately two-thirds is interpreted as 66%, compared with 38% on placebo.
That is a meaningful difference. However, the published summary uses an approximate figure, so it should not be treated as an exact pooled effect estimate with a known confidence interval.
There is also a distinction between a percentage reduction in the average WOMAC pain score and the percentage of individual participants whose pain fell by at least half.
Those are different statistics. The first describes the average change in a continuous pain score. The second describes the proportion of people who crossed a predefined improvement threshold.
Both are useful, but they cannot be substituted for each other.
About two-thirds halved their pain on retatrutide. More than one-third did so on placebo. The difference between those groups is the evidence of additional benefit.
Did Retatrutide Improve the Knee or Just Reduce Its Load?
The trials also raise a biological question that is more difficult than simply measuring weight loss.
Excess body weight increases mechanical loading on weight-bearing joints such as the knee. Losing substantial weight can make walking, climbing stairs and standing from a chair less painful.
That explanation is consistent with the large weight reductions seen in the retatrutide studies.
But it may not be the entire explanation.
Weight loss can affect inflammatory signaling, and GLP-1-based treatments have been investigated for biological effects beyond appetite suppression.
The study also reported improvements in cardiometabolic measures, including markers related to blood pressure, lipids and inflammation.
Whether changes in inflammatory pathways independently contributed to knee pain relief remains uncertain.
To establish such a mechanism, researchers would need analyses capable of separating the effects of reduced body weight from other pharmacological effects.
The greater unanswered question concerns the joint’s structure.
These trials primarily measured symptoms and physical function. They did not establish that retatrutide regenerates cartilage, reverses osteoarthritis or prevents structural deterioration.
Northwestern researcher Thomas Schnitzer identified reduced progression of osteoarthritis and fewer knee replacements as important questions for future research.
Participants who already needed knee replacement were excluded, so the results cannot be generalized directly to people with severe disease requiring surgery.
A treatment that reduces pain could still be clinically valuable without reversing cartilage damage. But those are different kinds of benefit, and they require different kinds of evidence.
For now, the most defensible conclusion is that retatrutide reduced reported pain and improved functional outcomes in the population studied.
The question of whether it changes the underlying course of knee osteoarthritis remains unanswered.
Retatrutide Side Effects Were Not a Small Detail
The efficacy results need to be considered alongside tolerability.
The November 8 NEJM publication reported that serious adverse events were more common with retatrutide than with placebo in both trials. Gastrointestinal events were the most frequently reported adverse effects.
Lilly’s detailed TRIUMPH-4 topline disclosure from December 2025 helps illustrate what that meant for participants.
Nausea was reported by 38.1% of participants receiving retatrutide 9 mg and 43.2% receiving 12 mg, compared with 10.7% on placebo.
Vomiting occurred in 20.4% of the 9 mg group and 20.9% of the 12 mg group, compared with no reported vomiting in the placebo group.
Diarrhea occurred in 34.7% and 33.1% of the two active-dose groups, compared with 13.4% receiving placebo.
Constipation was also common, affecting 21.8% of participants receiving 9 mg and 25.0% receiving 12 mg, compared with 8.7% on placebo.
One less familiar adverse event deserves particular attention: dysesthesia.
Dysesthesia describes abnormal or unpleasant sensations, which may include tingling, burning or altered skin sensitivity.
In TRIUMPH-4, it was reported by 8.8% of participants receiving 9 mg and 20.9% receiving 12 mg, compared with 0.7% on placebo.
Lilly characterized these events as generally mild and rarely responsible for stopping treatment. However, the higher frequency at 12 mg makes the finding relevant to the overall safety assessment.
Discontinuation because of adverse events occurred in 12.2% of the 9 mg group and 18.2% of the 12 mg group, compared with 4.0% on placebo.
Those figures are from TRIUMPH-4 specifically and should not be presented as pooled safety rates across both knee trials.
They are also not proof that every discontinuation involved a medically severe reaction. Lilly reported that the category included some participants who perceived their weight loss as excessive.
Even so, treatment persistence matters for a chronic condition. A drug may deliver substantial pain relief while still proving difficult for some patients to continue.
For people with knee osteoarthritis, maintaining muscle mass, strength and physical function is also important. Large reductions in body weight should not automatically be treated as equivalent to improvements in all aspects of musculoskeletal health.
Nearly one in five participants receiving the 12 mg dose in TRIUMPH-4 discontinued because of adverse events. That belongs beside the pain-relief headline.
What the Trials Cannot Tell Us and What Comes Next
The evidence is stronger now that the two knee osteoarthritis analyses have appeared in a peer-reviewed publication. But several limitations remain.
First, the participants had painful knee osteoarthritis alongside overweight or obesity, and they did not have diabetes.
These results cannot establish whether retatrutide provides similar knee pain relief in people with normal body weight, advanced joint damage or different metabolic conditions.
Second, the trials compared retatrutide with placebo rather than with another approved obesity medicine.
They therefore cannot establish whether retatrutide provides better osteoarthritis outcomes than semaglutide, tirzepatide or another medication when tested in comparable patients under identical conditions.
Third, the WOMAC outcomes describe pain and function rather than joint structure. Evidence that a treatment reduces symptoms is not the same as evidence that it slows cartilage loss or prevents surgery.
Fourth, the published follow-up periods were 68 and 80 weeks. That provides important medium-term safety and efficacy data, but it cannot answer what happens after several years of treatment or after the medication is discontinued.
Whether patients regain weight and whether knee pain returns following treatment withdrawal are particularly important questions for a therapy that may need to be continued long term.
Finally, both trials were funded by Eli Lilly, the developer of retatrutide. Company scientists participated in the research and were listed among the authors.
Industry sponsorship does not invalidate randomized trial results. It does make transparent reporting, independent scrutiny and replication important.
Retatrutide also remains investigational. As of November 10, 2026, it has not received FDA approval for obesity, knee osteoarthritis or another indication.
Lilly announced in July that it planned to submit a US regulatory application for retatrutide in the first quarter of 2027, supported by its broader Phase 3 program.
That anticipated filing is not an approval, and it does not establish which indications regulators will eventually authorize.
For people with painful knee osteoarthritis, existing management options still matter, including appropriate physical activity, strengthening exercises, weight management when indicated, physical therapy and established pain treatments.
Retatrutide should not be treated as an available clinical option or sourced from unapproved suppliers on the basis of these trial findings.
The broader significance is that obesity treatment and osteoarthritis treatment may be more closely connected than traditional clinical pathways have allowed.
If a treatment produces enough weight loss to meaningfully reduce knee pain and improve mobility, it may create opportunities for more integrated management of these conditions.
But treatment cost, tolerability, muscle health, long-term safety and the question of structural joint progression all remain relevant.
Grey Pepper’s earlier coverage of retatrutide’s TRIUMPH-1 Phase 3 obesity program examines the broader weight-loss evidence.
For now, the conclusion is more precise than the original headline alone might suggest.
Retatrutide produced substantial pain relief in two randomized Phase 3 study populations. Many participants receiving placebo also improved. The drug’s additional benefit was statistically significant, but adverse events were common, and the trial results cannot establish whether the underlying knee disease was changed.
The drug has shown that a painful knee may feel considerably better after major weight loss. It has not yet shown that the damaged joint itself has been repaired.
Sources and Further Reading
- Schnitzer TJ et al. Retatrutide Once Weekly in Adults with Knee Osteoarthritis Pain and Obesity. New England Journal of Medicine, November 8, 2026. DOI: 10.1056/NEJMoa2610676.
- Northwestern University. Experimental Weight-Loss Drug Eases Painful Knee Osteoarthritis in Two Clinical Trials. November 8, 2026.
- American College of Rheumatology. TRIUMPH-4: Retatrutide Once Weekly for the Treatment of Knee Osteoarthritis Pain and Obesity. ACR Convergence 2026, Abstract 0871.
- American College of Rheumatology. TRIUMPH-1-OA: Three Doses of Once-Weekly Retatrutide in Adults with Knee Osteoarthritis and Obesity. ACR Convergence 2026, Abstract 2573.
- Eli Lilly and Company. Retatrutide TRIUMPH-4 Phase 3 Topline Results, Including Safety and Tolerability. December 11, 2025.
- Jastreboff AM et al. Retatrutide, a Triple Hormone Receptor Agonist, for Treatment of Obesity. New England Journal of Medicine, September 29, 2026.
- Eli Lilly and Company. Additional Retatrutide Phase 3 Results and Planned US Regulatory Submission. July 23, 2026.
- ClinicalTrials.gov. TRIUMPH-4 Phase 3 Study Registration, NCT05931367.
- ClinicalTrials.gov. TRIUMPH-1 Phase 3 Study Registration, NCT05929066.
