Best Peptides for Joint Pain: What Human Evidence Exists

The short answer

For knee osteoarthritis pain, the strongest human evidence among the peptides we reviewed comes from randomized trials of weight-loss drugs in people with obesity: STEP 9 for semaglutide and a knee subgroup of TRIUMPH-1 for retatrutide, which is investigational. Neither can separate the pain benefit from the weight loss. Controlled knee trials of liraglutide and oral salmon calcitonin did not show a clear benefit. Collagen peptides have a small meta-analysis that found pain relief, but all four trials in it were rated at high risk of bias. BPC-157 has one small retrospective report in knee injections and no controlled study of joint pain.

We reviewed peptides only. We did not review exercise, physical therapy, weight loss without a drug, injections or other approved osteoarthritis treatments, which are usually the first steps. Ask a clinician.

On this page

Joint pain is where “peptides” covers a prescription drug, a food supplement and an unapproved research compound, with very different evidence behind each. The evidence is mostly about one condition, knee osteoarthritis.

The ranking, by strength of evidence

Ordered by the strength of the evidence that it helps with this goal: the type of study first, then its quality and its result. A rank isn’t a recommendation.
Peptide	U.S. status	Strongest evidence
1. Semaglutide	FDA-approvedFor other usesCompounded versions: not FDA-approved	Human trials
2. Retatrutide	Investigational	Human trials
3. Collagen peptides	Not FDA-approved	Limited human data
4. BPC-157	Compounded versions: not FDA-approved	Limited human data

How do we decide what is “best”?

“Best” means best evidence, not best marketing. We rank by the type of evidence first, then its quality and its result. A rank isn’t a recommendation. See how we rate peptides. For joint pain we look for a controlled trial in people with the condition that measured pain, not only a lab marker, and we note who the people were.

	Semaglutide	Retatrutide	Collagen peptides	BPC-157
What it is	FDA-approved drug (Wegovy, Ozempic, Rybelsus), not approved for joint pain	Investigational; not FDA-approved for any use	Dietary supplement; FDA does not approve supplements	Not FDA-approved; named in an interim HHS enforcement-discretion policy, with conditions
Best human evidence for knee pain	One randomized trial (STEP 9), 407 adults with obesity and knee osteoarthritis	A subgroup of 574 people in one randomized trial (TRIUMPH-1)	Meta-analysis of four randomized trials, 507 people	One retrospective series of knee injections, with no control group
Main limit	Only in people with obesity; knee osteoarthritis is not among the uses we found on the label	Only in people with obesity; a subgroup, with substantial weight loss; company-funded	All four trials rated at high risk of bias; results varied between trials	No controlled comparison; part of the group also received another peptide

Does semaglutide help knee osteoarthritis pain?

In STEP 9, 407 adults with obesity and knee osteoarthritis were followed for 68 weeks. Weight fell 13.7% on semaglutide vs 3.2% on placebo, and the WOMAC pain score improved by 41.7 points vs 27.5 points. The trial was funded by Novo Nordisk, which makes semaglutide. We read the abstract.

What to take from that:

It is a randomized, placebo-controlled trial that measured pain, which is the strongest design here, along with TRIUMPH-1.
The people had obesity. Their weight fell substantially, so the result does not show a pain benefit separate from weight loss, and it does not tell us about people without obesity.
It is not confirmed by a trial of the same drug that we found. Other drugs have knee trials, below, with different results.
Osteoarthritis is not a labeled use. Knee osteoarthritis pain is not among the uses we found on the Wegovy label, which covers weight, cardiovascular risk and fatty liver disease, so this would be an off-label use. FDA approval for one use does not extend to another.

Semaglutide carries a boxed warning for thyroid C-cell tumors and other warnings. See our semaglutide guide for the risks and for how compounded versions differ from the approved drug.

What about other peptides with knee trials?

We read these through their abstracts.

Retatrutide (investigational, not FDA-approved): in a subgroup of 574 people with knee osteoarthritis in TRIUMPH-1, pain scores improved more than with placebo (differences of −1.4 and −1.6 on a 1-to-10 scale, in the intention-to-treat analysis for two dose arms). The trial was funded by Eli Lilly. See our retatrutide guide.
Liraglutide: in a randomized trial of 156 people who had first lost weight on a diet, weight fell more than with placebo over 52 weeks, but knee pain did not differ.
Oral salmon calcitonin: the authors of two randomized, placebo-controlled trials (1,176 and 1,030 people) concluded it did not provide reproducible clinical benefits.

Do collagen peptides help joint pain?

Collagen peptides are hydrolyzed collagen taken by mouth. A 2023 meta-analysis of four randomized trials (507 people with knee osteoarthritis) found pain relief compared with placebo, with a standardized mean difference of −0.58 (95% CI −0.98 to −0.18). Two cautions from the analysis itself: the studies varied a lot from one to the next, and all four were rated at high risk of bias. The authors ask for better trials. This summary relies on the abstract.

Adverse events: in that analysis, adverse events did not differ significantly from placebo (odds ratio 1.66, 95% CI 0.99–2.78), and the quality of that evidence was very low.
Not the same as UC-II. Undenatured type II collagen (UC-II) is a different product that has been studied separately.
Regulation: FDA does not have authority to approve dietary supplements before they are sold. The manufacturer is responsible for safety and labeling. Our collagen peptides guide has more on the evidence, including for skin.

Does BPC-157 help joint pain?

We found no controlled study of BPC-157 for joint pain. The human evidence is one retrospective series of knee injections with no control group: of 17 patients, 16 were reached by phone and 14 reported relief, and 4 of the 16 also received thymosin beta-4. No specific tools measured function, and the authors worked from a one-year chart review. FDA found no human studies of BPC-157 by oral, subcutaneous, nasal or transdermal routes. A scoping review of six peptides, including BPC-157, concluded that their claimed benefits remain unsubstantiated by current human trials. It says human data showed “modest improvements at best” for degenerative knee pain and metabolic bone health, across those peptides. Animal studies of tendon, muscle, bone and ligament described effects as “promising but variable,” which does not show a benefit in people. Read our BPC-157 guide for the full detail.

What did HHS change?

On Oct. 30, 2026, HHS published an interim policy saying it does not intend to take enforcement action against certain compounding of BPC-157 acetate under conditions. The policy lists subcutaneous, intramuscular, oral and topical forms; injection into a joint is not on that list. It is not FDA approval, and it does not mean BPC-157 works for joint pain. FDA does not review compounded drugs for safety, effectiveness or quality before they are marketed.

Before the policy was issued, boards in Alabama and Mississippi told licensees they may not compound, administer or dispense non-FDA-approved or research-grade peptides. We found no update from them as of Oct. 31, and it isn’t clear how those notices apply to compounding under the HHS policy.

BPC-157 is prohibited in sport at all times (WADA 2026 list).

What are the risks?

Semaglutide: boxed warning for thyroid C-cell tumors, plus warnings for pancreatitis, gallbladder disease, kidney injury from dehydration and others (see the semaglutide guide).
Retatrutide: investigational, with no FDA label; long-term safety is unknown (see the retatrutide guide).
Collagen peptides: a 2026 review of oral and topical peptides reports they were well tolerated. We did not systematically assess contaminants or allergies, and FDA does not approve supplements before sale.
BPC-157: FDA cites “insufficient clinical safety information”, animal signals potentially related to clotting and the liver, and a potential for immune reactions. Long-term safety in humans is unknown.

Which claims are not supported?

“BPC-157 repairs cartilage and joints.” No controlled human study shows that.
“Collagen rebuilds your joints.” The meta-analysis found pain relief in four trials rated at high risk of bias; its abstract reports pain and adverse events, not rebuilt cartilage.
“Semaglutide treats arthritis.” In STEP 9, people with obesity and knee osteoarthritis had less pain, alongside substantial weight loss. It is not a labeled use. The same applies to retatrutide, which is not approved for any use.
“BPC-157 is FDA-approved” or “legal now.” Neither. It is under interim enforcement discretion, with conditions.

What should you ask your doctor?

What is causing my joint pain, and what are the standard treatments for it?
Does my weight play a role, and is an FDA-approved weight-management drug appropriate for me? Which uses are on its label?
If I want to try collagen, what would tell us it is helping, and does it interact with anything I take?
If a peptide such as BPC-157 is suggested, what human evidence supports it for my joint, who prepares it, and do I compete in a tested sport?

We do not give doses, cycles or preparation instructions, and we do not recommend products or sellers.

Frequently asked questions

What is the best peptide for joint pain?

None has strong evidence for joint pain on its own. The best-controlled trials are for semaglutide and retatrutide in people with obesity and knee osteoarthritis, and they cannot separate the pain benefit from the weight loss. Retatrutide is not FDA-approved.

Does collagen work for osteoarthritis?

A 2023 meta-analysis of four knee osteoarthritis trials found pain relief, but the studies varied and all four were rated at high risk of bias.

Does BPC-157 heal joints?

No controlled human study shows that it does.

Is it safe to combine them?

We did not review combinations. Ask your prescriber about your own situation.

How did we rank them?

We rank by the type of evidence first, then its quality and its result. If no peptide has good evidence for a goal, we say so.

Human trialsControlled trials in people, published in a journal, that tested a clinical outcome and showed a benefit. The strongest kind of evidence we report.
Limited human dataSmall, short or inconclusive human studies, including observational studies, uncontrolled trials and controlled trials that did not show a clinical benefit.
Animal studiesResults in animals often don’t carry over to people.
Lab studiesCells or tissue in a dish. An early signal, not an answer.

Sources we checked

STEP 9 trial: semaglutide in obesity and knee osteoarthritis (PubMed)PubMed. Accessed October 30, 2026.
Wegovy prescribing information (DailyMed, rev. 06/2026)FDA. Accessed October 30, 2026.
Collagen peptides in knee osteoarthritis, meta-analysis, 2023 (PubMed)PubMed. Accessed October 30, 2026.
Undenatured type II collagen (UC-II) in knee osteoarthritis (PubMed)PubMed. Accessed October 30, 2026.
Oral and topical peptides for skin aging, systematic review of 19 RCTs, Front Med 2026 (PubMed)PubMed. Accessed October 30, 2026.
FDA: questions and answers on dietary supplementsFDA. Accessed October 30, 2026.
FDA presentations to the Pharmacy Compounding Advisory Committee (July 23, 2026)FDA. Accessed October 30, 2026.
FDA briefing document on BPC-157FDA. Accessed October 30, 2026.
Lee & Padgett 2021, BPC-157 knee injections (PubMed)PubMed. Accessed October 30, 2026.
Peptide Supplements and Their Therapeutic Applications in Sports Medicine, Am J Sports Med 2026 (PubMed)PubMed. Accessed October 30, 2026.
HHS: Enforcement Discretion Policy, with Appendices A and BOther. Accessed October 31, 2026.
Federal Register: HHS enforcement discretion policy on 503A peptide compounding (Oct. 30, 2026)Other. Accessed October 31, 2026.
Alabama Board of Medical Examiners: notice on prescribing non-FDA-approved research-grade peptidesOther. Accessed October 31, 2026.
Mississippi nursing, pharmacy and medical licensure boards: joint statement on peptides (Aug. 19, 2026)Other. Accessed October 31, 2026.
WADA 2026 Prohibited ListOther. Accessed October 30, 2026.
TRIUMPH-1 trial, retatrutide in obesity, NEJM (PubMed)PubMed. Accessed October 31, 2026.
Liraglutide in knee osteoarthritis after diet weight loss (PubMed)PubMed. Accessed October 31, 2026.
Oral salmon calcitonin in knee osteoarthritis, two trials (PubMed)PubMed. Accessed October 31, 2026.

Written and maintained by the Best Peptides For Editorial Team. How we rate evidence.