Peptides for Injury Healing: Rotator Cuff Tears, Muscle Tears and Joint Pain: What do we know?
Evidence reviewed on 5 October 2026.
If you have been searching for peptides to heal rotator cuff injuries, peptides to heal muscular tears or peptides for joint pain, you have probably encountered impressive claims in articles, podcasts and social media. Faster recovery, stronger tendons and less pain are understandably appealing when an injury is keeping you away from training, work or everyday life.
This blog offers an honest overview of what peptides are, what we know, and where they may or may not be useful for injuries such as rotator cuff injuries and muscle tears. Our aim is to support informed conversations and decisions, separating promising ideas from benefits and risks that have actually been established in people.
As clinicians working with injury and pain, we would welcome treatments that reliably make recovery faster and easier. If recovery peptides demonstrated substantial benefits and reassuring long-term safety, we would be enthusiastic about incorporating them into appropriate medical and rehabilitation care.
Why interest in recovery peptides has increased
We understand why interest in recovery peptides has grown. Rehabilitation can feel slow, particularly when you are following advice and still struggling, so stories about faster recovery are understandably appealing and memorable. Articles, podcasts and social media make those stories highly visible. Clinicians who are enthusiastic about a new treatment may sincerely want to help, and a persuasive biological explanation can make a treatment feel established even when the patient research is still at an early stage. Professional caution is about how confidently we can interpret that evidence, rather than questioning those intentions.
One factor that may be driving interest in “recovery peptides” for injuries is the recent success of GLP-1 medicines. These medicines, including semaglutide, mimic a naturally occurring hormone involved in appetite and blood-sugar regulation. They can help people feel fuller and eat less, and clinical trials have shown substantial weight loss in the populations studied. For some people, the benefits have been life-changing. Semaglutide is known as Ozempic for type 2 diabetes and Wegovy for weight management in eligible patients. These medicines also have recognised risks, including gastrointestinal side effects that cause some people to stop treatment. [1,14,40]
That success may create the impression that peptides as a whole are effective treatments. Psychology research suggests that favourable impressions can influence how we judge benefits and risks, although this transfer of confidence has not been demonstrated specifically for GLP-1 medicines and recovery peptides. [32] Products such as BPC-157 and TB-500 are different compounds, proposed for different purposes and measured against different outcomes. The success of GLP-1 medicines should therefore not determine our willingness to adopt recovery peptides: that decision needs to rest on evidence of their own effectiveness and safety.
What peptides are and why the name can be misleading
Peptides are chains of amino acids, the building blocks from which proteins are made. Their effects depend on their sequence, structure, dose, delivery and biological targets. Describing something as a peptide tells us about its chemistry; it does not tell us whether it will treat an injury.
The term encompasses very different products:
GroupExamplesWhat the distinction meansEstablished peptide medicinesSemaglutide and other medicines developed for specific indicationsBenefits and risks must be judged for the particular medicine and patient group studiedExperimental products marketed for recoveryBPC-157, TB-500, injectable GHK-CuA proposed repair mechanism does not establish clinical effectivenessCompounds intended to stimulate hormone releaseCJC-1295, ipamorelinChanging a hormone concentration is different from demonstrating better injury recoveryNutritional peptidesHydrolysed collagen supplementsThese have a separate evidence base and should not be confused with experimental injections
Our bodies naturally make peptides. These endogenous peptides can act as chemical messengers: naturally occurring GLP-1, for example, helps regulate the insulin response to food. Synthetic peptides are manufactured rather than produced within the body; some reproduce a natural sequence, while others are modified or experimental compounds. A naturally occurring signalling molecule and a manufactured product sold for injection are not interchangeable simply because both are called peptides. [33]
This is why “natural” is not a safety guarantee. Dose, route, purity and duration of exposure matter. A substance resembling something involved in normal physiology can behave differently when concentrated and administered as a treatment. Equally, being synthetic does not itself make a medicine unsafe.
Throughout this article, “recovery peptides” refers principally to experimental products marketed for musculoskeletal repair, rather than all medicines or supplements containing peptides.
How is BPC-157 proposed to work?
BPC-157 is an experimental synthetic, or lab-made, peptide containing 15 amino acids. Its reputation for injury recovery comes mainly from cell and animal research. The proposed mechanisms are scientifically interesting, but they are frequently presented with more certainty than the experiments justify. [2–6,10]
Angiogenesis and blood-vessel signalling
Angiogenesis means the formation of new blood vessels. Laboratory and animal experiments have linked BPC-157 with VEGFR2 and the downstream Akt–eNOS signalling pathway. In plain language, these are parts of the machinery involved in blood-vessel responses and nitric oxide production. Improved blood supply is a proposed way it might support injured tissue. The studies do not establish faster human tendon recovery. [2]
Fibroblast movement and FAK–paxillin signalling
Fibroblasts help produce and organise the supporting material around cells. Experiments using rat tendon tissue and cells found increased cell movement and outgrowth with BPC-157. FAK and paxillin are proteins involved in how cells attach and move, and their signalling appeared to be involved. This could be relevant to repair, but it does not demonstrate that a repaired human tendon will tolerate sport better. The same study did not show a direct increase in cultured tendon-cell proliferation. [3]
Growth-hormone-receptor findings
A separate study using rat tendon fibroblasts found increased growth-hormone-receptor expression and a greater proliferative response when growth hormone was added. Receptors help cells respond to signals; more receptors may change that response. This was a laboratory observation, not evidence that BPC-157 increases useful tendon strength or speeds return to sport in people. [4]
Inflammation and tissue-repair responses
Rat studies involving muscle transection and crush injury reported improvements in measures of repair and function, with less swelling and haematoma in the crush-injury model. These findings support further investigation of tissue-repair responses; they are not proof of an effective treatment for an ordinary sporting strain. Inflammation is part of repair, so changing it is only useful if the overall result is better recovery without unacceptable harm. [5,6]
How we evaluate a treatment: meaningful benefit and safety
The questions are straightforward: does it make a worthwhile difference, and are its risks acceptable for the person considering it?
To establish benefit, we usually need sufficiently large, well-designed controlled trials. Random allocation helps create comparable groups. Blinding patients and assessors, where feasible, reduces the influence of expectations and measurement bias. A placebo or sham comparison can help isolate the treatment’s specific contribution when it is suitable and ethical; comparison with good usual care also matters.
This is especially important in injuries because people may improve through natural recovery, rehabilitation, changed training, better sleep or other treatment. Symptoms also fluctuate. An individual can genuinely feel much better after an injection without that story establishing what caused the improvement. The controlled trials discussed below show why comparisons can change our interpretation. [36,38]
We want meaningful outcomes such as better function, less disabling pain, a safe return to activity and fewer reinjuries—not just a change in a laboratory marker. Safety requires active monitoring too. Larger studies, longer follow-up and research across relevant doses and patient groups improve our chance of identifying harms. Observational follow-up can add valuable safety information, but does not by itself establish efficacy or remove confounding. No study can prove that a treatment carries zero risk.
A new injury and persistent pain are not always the same problem
I think it is important to distinguish between a new injury and persistent pain in the context of using any nodality designed at targeting a specific tissue. The hope behind recovery peptides is understandable: if tissue heals faster, perhaps it will hurt less and we can return to activity sooner. With a new hamstring tear, for example, there is a clear tissue injury to assess and rehabilitate. A treatment that could safely improve that healing would be welcome, but we still need clinical studies to show that it actually does.
With persistent pain, the picture can be more complicated. Symptoms may not be easily linked to one specific tissue, and a finding on a scan does not necessarily explain why someone hurts. Research has identified rotator cuff tears in people without shoulder symptoms, meniscal tears in people without knee symptoms, and disc bulges or protrusions in people without back pain. [26,34,35] Conversely, someone can experience substantial pain with little detectable tissue damage.
Even when a tear is present, its size does not necessarily tell us how much pain someone will experience. In a study of 393 people with atraumatic full-thickness rotator cuff tears, measures of tear severity did not correlate with pain severity. [25] These findings do not mean tissue damage is irrelevant; they mean it needs to be interpreted alongside the person’s symptoms, examination and circumstances.
The biopsychosocial model of pain helps explain this wider picture. Tissue injury and nervous-system responses can contribute, alongside factors such as sleep, emotions, expectations and someone’s circumstances or support. Their influence varies between people. This does not make pain imaginary or mean someone is responsible for their symptoms. It means tissue damage is not always the only or main factor driving their pain. [24]
For that reason, focusing solely on “healing” a scan finding with a peptide can miss important parts of the problem. Less pain after treatment would not, by itself, demonstrate tissue repair. Equally, a laboratory finding suggesting repair would not establish that someone feels better or can do more.
Symptoms and function can also improve without every imaging abnormality disappearing. Research in Achilles tendinopathy found that structural restoration was not required for clinical improvement in the group studied. [27] That is encouraging for rehabilitation: recovery does not always depend on making a scan look normal.
There is an important distinction here though. After a fracture, acute tendon rupture, major traumatic tear or surgical repair, tissue integrity and protection still matter. Feeling less pain does not automatically mean the area is ready for unrestricted loading.
A useful treatment therefore needs to demonstrate more than a lower pain score or a change in tissue appearance. We want to know whether people regain strength and function, return safely to activity, and maintain those improvements without unacceptable harms or reinjury.
Why a plausible mechanism is only the starting point
Before recommending a treatment, we need to see meaningful effects in people. MSK care has several familiar examples of treatments whose routine use or recommendations changed when better evidence challenged an appealing explanation.
Prescribed bed rest for uncomplicated low-back pain
Resting a painful back once seemed an obvious way to let it recover. In Malmivaara and colleagues’ 1995 randomised trial, continuing ordinary activity as tolerated produced better recovery than prescribed bed rest for two days. This was acute, nonspecific low-back pain, not a fracture or another condition requiring protection. Current NICE guidance encourages normal activities rather than routine prescribed bed rest. Complete bed rest was the advise for years, and people did still get better, lot’s of those people would have been convinced they got better because of the bed rest despite us now knowing it is less effective than an active approach. During an example such as this you would describe the secondary nagtive effects of bed rest to be other factors such as loss of muscle mass, mobility, confidence etc (especially for prolonged bed rest). Yet again people got better therefore it was viewed as the best way to manage that kind of back pain at the time.[36,37]
Arthroscopic knee washout or “clean-up” for osteoarthritis
Washing out a joint or removing worn tissue sounds as though it should reduce pain. However, Moseley and colleagues’ placebo-controlled trial found that arthroscopic lavage or debridement produced no better outcomes than placebo surgery in the knee-osteoarthritis patients studied. NICE now recommends not offering these procedures for osteoarthritis. This is not a claim that all knee surgery is ineffective, or that surgery for a different diagnosis has been abandoned. [38,15]
I am not writing about these examples to show that BPC-157 cannot work. They show why a convincing explanation and improvement after treatment are not enough to establish its specific benefit.
So, what do the human BPC-157 studies tell us?
Human research is extremely limited. A large preclinical literature is not a large clinical evidence base, even when a long reference list makes it look that way. A 2025 systematic review included 35 preclinical studies and only one clinical study. Its search ended in June 2024, so it is a useful snapshot rather than a complete account of subsequent publications. The small human studies below need to be judged on their own design and limitations. [7]
The small knee-pain study
Lee and Padgett’s 2021 retrospective study reviewed 17 patients and reached 16 by telephone. Twelve had received BPC-157 alone and four a combination with thymosin beta-4. Eleven of the 12 receiving BPC-157 alone reported improvement; across both groups, 14 of 16 reported relief. [8]
That is an interesting observation, but there was no randomised comparison group. Follow-up varied and relied on recall, and the researchers did not use specific tools to assess function, quality of life, stiffness or daily activities. The study did not demonstrate structural healing.
It therefore cannot establish a reliable treatment success rate, show superiority to rehabilitation or prove that cartilage or a tear had repaired. The authors’ suggestions about regeneration go beyond what these data measured. This is a reason to run a better trial, rather than a basis for promising the same outcome to patients.
The two-person safety pilot
Lee and Burgess’s 2025 intravenous pilot involved two adults, both of whom had previously received intravenous BPC-157. They received infusions on two consecutive days, with blood tests, vital signs and side-effect questions assessed over three days. No side effects were reported, and the tested biomarkers showed no measurable adverse changes. [9]
That is a narrow, short-term tolerability observation. Two previously exposed participants cannot establish long-term safety, detect uncommon complications, or tell us about other preparations, injection routes or people with different medical conditions. The study did not test the effect on any injury or tissue change so can be used to support or deny the claim of healing.
So, do they work and are they safe?
The honest answer is that we do not yet have a reliable answer for the MSK healing benefits being claimed. A lack of convincing evidence does not prove experimental recovery peptides are ineffective, but it means we cannot confidently say they work for those outcomes. The mechanisms sound promising, and there are preliminary findings, including the small human studies discussed above. Robust human research has not yet established the healing benefits often claimed. [7–10]
To add to this, people obtaining products through unregulated or unverified routes face another uncertainty: we may not know whether the identity, strength, purity or sterility matches the label, or whether contaminants are present. Spending more does not guarantee an authorised, quality-assured medicine. UK rules do permit some unlicensed medicines to be supplied for special clinical needs, but that is different from having a marketing authorisation; it does not mean every supplier or product has the same standards. [28,39]
What about cancer? BPC-157’s experimental effects on blood-vessel signalling raise a question worth investigating; they do not demonstrate cancer promotion in people. FDA reported that it had not identified carcinogenicity studies for BPC-157 free base or acetate. The available human data cannot establish or exclude an increased cancer risk. That is uncertainty—not evidence that “peptides cause cancer”, and not reassurance that BPC-157 is cancer-safe. [2,10]
Finally, we do not have adequate long-term human safety data for these experimental recovery uses. We cannot reliably characterise delayed or uncommon adverse effects, drug interactions, contraindications, or suitability for different patient groups, including people with coexisting conditions. Established medicines are not risk-free, but their benefits, harms and appropriate patient groups are generally much better characterised. Small, short-term studies cannot give us that same confidence about recovery peptides. [9,10,17,22]
Peptides to heal rotator cuff injuries
Shoulder pain can make sleep, dressing, work and exercise difficult, so the appeal of a treatment advertised to repair the rotator cuff is easy to understand. However, “rotator cuff injury” can describe different clinical situations: tendon-related pain, a partial tear, a full-thickness tear or recovery after surgery. Evidence for one cannot automatically be applied to all the others.
There is currently no published, convincing controlled human evidence showing that BPC-157 or TB-500 heals rotator cuff tears or reliably improves recovery after repair.
There is an important emerging study. ClinicalTrials.gov record NCT07803250 describes a planned randomised Phase 1 study of BPC-157 after rotator cuff repair, with an estimated 30 participants. On 5 October 2026, it was listed as not yet recruiting, with an estimated January 2027 start and no posted results. [11] Its existence is encouraging as a research development
For a patient, the questions that matter include whether treatment improves strength and function, changes repair integrity, reduces re-tears or shortens a safe return to activity. A convincing explanation involving collagen or blood vessels cannot answer those questions.
Peptides to heal muscular tears
Claims about peptides for hamstring tears, calf tears and quadriceps injuries commonly draw on animal muscle-repair research.
The proposed benefits include supporting cell survival, vascular responses and the organisation of repair tissue. But the clinical leap is substantial: an improvement in an animal tissue sample is not a measured reduction in days lost from sport.
For muscular tears, we would want trials assessing:
Time to a clearly defined, safe return to sport.
Recovery of strength and sport-specific performance.
Reinjury over meaningful follow-up.
Adverse effects and the need for further treatment.
Whether any benefit remains when both groups receive good rehabilitation.
We did not identify robust controlled human evidence establishing these benefits for BPC-157 or TB-500.
There is human evidence that rehabilitation decisions matter. Bayer and colleagues’ randomised study found a shorter return-to-sport interval when rehabilitation began earlier rather than later after acute muscle injury. [13] This does not mean every muscle injury should receive the same timetable. It shows the value of testing treatments against outcomes that matter to patients.
Peptides for joint pain and cartilage repair
Joint pain is a symptom, rather than a single diagnosis. Osteoarthritis, inflammatory disease and injuries to surrounding structures require different decisions. The small BPC-157 knee series cannot establish that injections rebuild cartilage, repair a meniscus or prevent joint replacement. Those claims require direct investigation.
The GLP-1 finding that is relevant to knee pain
There is credible evidence for a specific peptide medicine in a specific population. STEP 9 randomised 407 people with obesity and knee osteoarthritis to semaglutide or placebo, alongside lifestyle advice. At 68 weeks, average weight change was −13.7% versus −3.2%, and WOMAC pain scores improved by 41.7 versus 27.5 points. [14]
This is evidence of benefit for weight and knee symptoms in the population studied. It does not demonstrate cartilage regeneration, establish the same benefit in people without obesity, or validate unrelated recovery peptides.
The trial was funded by Novo Nordisk. It remains much stronger evidence than an uncontrolled testimonial, while its design does not fully separate weight-mediated benefits from other possible mechanisms.
For osteoarthritis, NICE recommends tailored therapeutic exercise and, where relevant, weight management, with other treatment decisions based on individual needs. [15] The question is what improves a person’s life, not simply what is described as regenerative.
Why genuine testimonials cannot establish effectiveness
A person who says “I felt better after taking peptides” may be describing their experience entirely accurately. We do not need to dismiss that experience to question the explanation.
Several possibilities can coexist:
Symptoms may have improved as part of their usual course.
The person may also have changed training, rehabilitation or other treatment.
They may have started treatment during a particularly bad period, followed by a return towards their usual symptom level.
Expectations and the treatment experience may have influenced symptoms.
The compound might have had an effect.
An uncontrolled story cannot tell us how much each contributed. These are alternative explanations, not proof that the person’s improvement was a placebo response. This is the whole reason we perform a randomised comparison, with the randomised aspect useful to avoid bias of the researcher and the participants. Ultimately it helps answer the main question: what would probably have happened without the active treatment?
That question is especially important when recovery fluctuates significantly from person to person, the complexity of pain and tissue damaged previously mentioned and when several treatments are used together.
Concerns about safety and product quality
Unknown long-term effects
Small, short studies cannot reliably exclude uncommon or delayed adverse effects. They also provide limited information about interactions, repeated exposure and use in people with other illnesses. For BPC-157 and TB-500, the uncertainty is part of the risk assessment. “No harm was noticed in a tiny study” and “long-term safety has been established” are very different claims.
Product identity and manufacturing
A research finding only applies to the material and conditions studied. Commercial products raise separate questions about identity, concentration, degradation products, storage and consistency. A purity certificate alone does not establish clinical effectiveness or demonstrate every property required of an injectable medicine. In particular, chemical purity and sterility are different attributes.
Injection-related harm
Injection introduces risks beyond those of the molecule itself, including infection, bleeding and injury to nearby structures. A joint infection can be serious. Such procedural risks still exist even if someone believes the active ingredient is harmless. After any injection, a hot, swollen, increasingly painful joint, particularly with fever or feeling unwell needs urgent medical assessment.
Growth signalling and cancer concerns
Because some proposed mechanisms involve blood vessels and growth-related pathways, questions about unwanted effects on abnormal tissue are reasonable. However, claiming that BPC-157 has been proven to cause cancer in humans would go beyond the available evidence. Equally, the lack of convincing human cancer data cannot establish that long-term exposure is risk-free. The appropriate conclusion is uncertainty requiring proper investigation, rather than either reassurance or alarm based on a theoretical mechanism alone.
Combining products makes uncertainty harder to manage
Using several peptides together makes it harder to identify which substance produced a benefit or an adverse effect. It also introduces combinations that may never have been studied adequately.
More compounds do not automatically create a more effective recovery treatment.
There is also an opportunity cost: an expensive experimental intervention can consume attention and resources while an important diagnosis, rehabilitation adjustment or specialist opinion is delayed.
UK regulation
Clinical evidence and legal supply are separate questions
UK rules allow some unlicensed medicines to be supplied for an individual patient’s special clinical need under defined conditions. That does not amount to general approval or prove effectiveness. MHRA guidance also prohibits advertising unlicensed medicines. [28,29]
A product being offered by a clinic, prescribed somewhere or discussed in a US compounding decision should not be interpreted as proof that it heals injuries.
FDA’s July 2026 advisory process concerned substances proposed for compounding; it was not an approval of an injury-healing indication. [30] Regulatory access and clinical evidence should be described separately, especially as rules can change.
Why we cannot currently recommend recovery peptides
Our position is based on the gap between the claims and the evidence.
We cannot presently give patients reliable answers about the size of any benefit, who is likely to benefit, an evidence-based treatment regimen, longer-term safety, or how these products compare with good usual care.
We also cannot equate pain improvement with tissue regeneration, or assume that a laboratory mechanism will translate into a meaningful benefit for the person in front of us.
That is a substantial gap when recommending an intervention involves exposing someone to risk and asking them to spend money.
We would change our position if high-quality evidence justified it. Useful research would include clearly defined injuries, verified products, appropriate comparison groups, adequate sample sizes, meaningful functional outcomes, assessment of repair where relevant, and sufficient follow-up for adverse effects and reinjury.
The right conclusion today is not enough evidence to recommend routine use, rather than certainty that these compounds could never help.
Clinical takeaways
For someone considering peptides, our first priority would be to understand the diagnosis, the main barriers to recovery and whether the current plan is appropriate.
That may mean adjusting rehabilitation, investigating a new or changing problem, addressing relevant health factors or obtaining a medical or surgical opinion. Repeating an ineffective plan indefinitely is not good care either.
For clinicians, discussions should be open and non-judgemental. Ask what the person hopes to achieve, what they have already tried and whether they are already using a product. Explain the evidence without dismissing their frustration.
The practical messages are:
Treat peptides as individual substances. GLP-1 evidence cannot validate an unrelated recovery product.
Distinguish mechanism from outcome. A change in cells, blood vessels or hormones does not establish faster or safer recovery.
Separate pain relief from tissue repair. Both matter, but neither proves the other.
Respect uncertainty about safety. Small studies and reassuring testimonials cannot establish long-term safety.
Keep care focused on meaningful progress. Function, confidence, capacity, participation and appropriate tissue protection matter more than a treatment’s regenerative label.
At The Injury & Performance Clinic, we understand the attraction of a treatment that could make recovery easier. We would welcome one that demonstrated meaningful benefits with a well-established safety profile. For the experimental recovery peptides discussed here, that evidence is not yet strong enough.
Detailed literature review and references
The notes below explain what the main studies contribute and what they cannot establish. This is a targeted narrative review, with some clinical opinion, not a a registered systematic review. Searches covered peer-reviewed research, trial registration and official guidance available on 5 October 2026.
Established peptide medicines and the GLP-1 comparison
[1] Wilding JPH and colleagues, 2021. STEP 1 trial. New England Journal of Medicine. DOI: 10.1056/NEJMoa2032183. Read the study. Large placebo-controlled trial supporting semaglutide for weight reduction in the population studied. Its findings illustrate the standard of evidence achievable for a peptide medicine, without supporting tendon or muscle repair claims for other molecules.
[14] Bliddal H and colleagues, 2024. STEP 9 trial. New England Journal of Medicine, 391:1573–1583. DOI: 10.1056/NEJMoa2403664. Read the study. Randomised evidence for improvements in weight and knee osteoarthritis symptoms among people with obesity. Structural cartilage regeneration was not established.
BPC-157 mechanisms and animal research
[2] Hsieh MJ and colleagues, 2017. BPC-157 and VEGFR2 activation. Journal of Molecular Medicine. Read the study. Experimental vascular assays support a proposed angiogenic mechanism. They do not demonstrate clinical efficacy in injured patients.
[3] Chang CH and colleagues, 2011. Tendon outgrowth, cell survival and migration. Journal of Applied Physiology, 110:774–780. DOI: 10.1152/japplphysiol.00945.2010. Read the study. Laboratory tendon-cell research involving FAK–paxillin signalling. Useful for mechanism, with no patient recovery or safety outcome.
[4] Chang CH and colleagues, 2014. Growth hormone receptor expression in tendon fibroblasts. Molecules, 19:19066–19077. DOI: 10.3390/molecules191119066. Read the study. Shows a cellular response under experimental conditions. Increased receptor expression does not prove greater tendon strength in humans.
[5] Staresinic M and colleagues, 2006. BPC-157 after quadriceps transection in rats. Journal of Orthopaedic Research, 24:1109–1117. DOI: 10.1002/jor.20089. Read the study. Positive findings in a surgically created injury model support further investigation. Translation to human sporting strains remains unresolved.
[6] Novinscak T and colleagues, 2008. BPC-157 after muscle crush injury in rats. Read the study. Another preclinical injury model. It does not establish an effective clinical regimen, return-to-sport benefit or long-term human safety.
[7] Vasireddi N and colleagues, 2025. Emerging Use of BPC-157 in Orthopaedic Sports Medicine. HSS Journal, 21:485–495. DOI: 10.1177/15563316251355551. Read the systematic review. A useful evidence map showing the dominance of preclinical research. Its June 2024 search cutoff needs to be recognised when discussing newer publications.
BPC-157 human evidence and trial development
[8] Lee E and Padgett B, 2021. Intra-Articular Injection of BPC 157 for Multiple Types of Knee Pain. Alternative Therapies in Health and Medicine, 27:8–13. Read the study. Small retrospective series with no control group. The reported symptom response cannot establish cartilage repair, prevention of surgery or the effectiveness of combination treatment.
[9] Lee E and Burgess K, 2025. Safety of Intravenous Infusion of BPC157 in Humans. Alternative Therapies in Health and Medicine, 31:20–24. Read the pilot. Two previously exposed adults and brief monitoring. It addresses a narrow tolerability question, not clinical effectiveness or a dependable long-term risk estimate.
[10] FDA, 2026. BPC-157 compounding advisory briefing. Read the assessment. Regulatory review documenting limited human studies, gaps in pharmacokinetics and safety, and older abstract-only evidence. A regulatory evidence assessment, not a treatment trial.
[11] ClinicalTrials.gov NCT07803250. Read the registry. Planned Phase 1 BPC-157 study following rotator cuff repair. The live registry was checked directly on 5 October 2026: not yet recruiting; estimated enrolment 30; no results. Registration must not be reported as a positive trial.
Safety and nutritional peptides
[17] FDA, 2026. TB-500 compounding advisory briefing. Read the assessment. Regulatory assessment documenting limited evidence for the substance reviewed, relevant to the uncertainty surrounding its clinical use and safety.
[22] FDA. Safety concerns for substances proposed for compounding. Read the safety information. Compound-specific concerns include immune responses, impurities and limited human data. This source does not quantify the likelihood of every possible adverse effect.
[23] Praet SFE and colleagues, 2019. Collagen peptides alongside Achilles rehabilitation. Nutrients, 11:76. DOI: 10.3390/nu11010076. Read the trial. Small crossover pilot studying a nutritional product with exercise. Encouraging but insufficient for broad healing claims or extrapolation to injectable peptides.
Pain and rehabilitation
[24] IASP, 2020. Revised pain definition and accompanying notes. Read the terminology. Authoritative framework distinguishing pain from nociception and acknowledging biological, psychological and social influences. It supports person-centred assessment, not dismissal of tissue injury.
[25] Dunn WR and colleagues, 2014. Rotator cuff tear severity and pain in 393 patients. Read the study. Cross-sectional evidence that structural severity did not track pain in this atraumatic full-thickness tear cohort. It cannot establish causes of an individual’s symptoms or be generalised to every traumatic injury.
[26] Tempelhof S and colleagues, 1999. Rotator cuff tears in asymptomatic shoulders. Journal of Shoulder and Elbow Surgery, 8:296–299. Read the study. Ultrasound study of 411 volunteers illustrating that a tear can coexist with an absence of symptoms. An imaging finding still needs clinical context.
[27] de Jonge S and colleagues, 2015. Achilles tendon structure and symptom change. American Journal of Sports Medicine. DOI: 10.1177/0363546515605077. Read the prospective study. Shows that structural restoration is not required for symptom improvement in the cohort studied. It does not mean tendon structure is irrelevant to every clinical decision.
[12] Rotator cuff tendinopathy clinical practice guideline, 2025. Journal of Orthopaedic and Sports Physical Therapy. DOI: 10.2519/jospt.2025.13182. Read the guideline. Supports active rehabilitation for tendinopathy. Its scope should not be substituted for surgical advice after a major traumatic tear.
[13] Bayer ML and colleagues, 2017. Early versus Delayed Rehabilitation after Acute Muscle Injury. New England Journal of Medicine, 377:1300–1301. DOI: 10.1056/NEJMc1708134. Read the trial. Human comparative evidence supporting timely rehabilitation in the injuries studied, rather than a universal timetable for every tear.
[15] NICE NG226, 2022. Osteoarthritis in over 16s. Read the recommendations. Clinical guidance supporting tailored exercise and relevant weight management alongside individualised care.
Regulation
[28] MHRA. Supply unlicensed medicinal products. Read the guidance. Explains the UK special-clinical-need framework; it does not constitute product approval.
[29] MHRA. Advertise your medicines. Read the guidance. Sets out advertising restrictions, including for unlicensed medicines.
[30] FDA. July 2026 Pharmacy Compounding Advisory Committee. Read the meeting materials. Establishes the scope of the compounding discussion. Advisory consideration is distinct from approval of an injury treatment.
Additional sources for treatment evaluation and the revised discussion
[32] Finucane ML, Alhakami A, Slovic P and Johnson SM, 2000. The affect heuristic in judgments of risks and benefits. Journal of Behavioral Decision Making, 13:1–17. General experimental psychology research on affect and perceived benefits and risks; it does not test GLP-1 medicines, BPC-157 or confidence transferring between them. Read the paper.
[33] Kreymann B and colleagues, 1987. Glucagon-like peptide-1 7-36: a physiological incretin in man. Lancet, 2:1300–1304. DOI: 10.1016/S0140-6736(87)91194-9. Human physiological study supporting the explanation of endogenous GLP-1. It does not establish the safety of experimental recovery products. Read the study.
[34] Englund M and colleagues, 2008. Incidental meniscal findings on knee MRI in middle-aged and elderly persons. New England Journal of Medicine, 359:1108–1115. DOI: 10.1056/NEJMoa0800777. Population study of 991 adults aged 50–90; meniscal findings were common without symptoms. It does not mean every acute traumatic tear is incidental. Read the study.
[35] Jensen MC and colleagues, 1994. Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of Medicine, 331:69–73. DOI: 10.1056/NEJM199407143310201. MRI study of 98 asymptomatic people showing that disc bulges and protrusions can occur without pain. Imaging still requires clinical interpretation. Read the study.
[36] Malmivaara A and colleagues, 1995. The treatment of acute low back pain—bed rest, exercises, or ordinary activity? New England Journal of Medicine, 332:351–355. DOI: 10.1056/NEJM199502093320602. Randomised trial of 186 people with acute nonspecific low-back pain; ordinary activity as tolerated outperformed prescribed two-day bed rest. Not a study of fractures or every cause of back pain. Read the trial.
[37] NICE NG59. Low back pain and sciatica in over 16s: assessment and management. Current recommendations encourage continuation of normal activities, with care adapted to the individual. Read the recommendations.
[38] Moseley JB and colleagues, 2002. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. New England Journal of Medicine, 347:81–88. DOI: 10.1056/NEJMoa013259. Randomised placebo-controlled trial of 180 patients; lavage and debridement were not superior to placebo surgery over two years. It does not evaluate all knee operations or diagnoses. Read the trial.
[39] MHRA, 2024. Illegal medicines worth more than £30 million seized in the UK in 2023. Official advice explaining risks of unlicensed and illegally supplied medicines and the distinction from MHRA-assessed products. It does not demonstrate that every peptide supplier sells contaminated products. Read the safety advice.
[40] MHRA. GLP-1 medicines for weight loss and diabetes: what you need to know. Official UK guidance distinguishing semaglutide brands and licensed uses, including Ozempic for diabetes and Wegovy for weight management in eligible patients, and explaining recognised risks. This is regulatory guidance, not evidence that GLP-1 success increases confidence in recovery peptides. Read the guidance.
