Do peptides work? The evidence - and who pays to produce it

TL;DR

• A peptide of exactly this kind won FDA approval in September 2025 - SS-31, on randomised placebo-controlled data. The category can clear the bar.
• It got there because it had a sponsor. Rare-disease incentives made the trial economics work, and the evidence followed immediately.
• Everything else stalls at the same place: a pivotal trial pays back only through patent exclusivity, and you can't patent a fifteen-amino-acid fragment of human gastric juice.
• So the literature is mostly preclinical and small. That's cheap evidence, not negative evidence - a study count measures funding, not biology.
• "Not demonstrated" ≠ "disproven." Most of this category is unestablished, which is a real third state.
• The comparator most readers actually face isn't a proven drug. It's rest, physio, and wait - which is why the option exists at all.

Educational content, not medical advice. No doses or protocols appear. Peptide therapy is physician-directed.

---

Two claims circulate about peptides and both are wrong. One says the science is settled. The other says there's no evidence and the category is hype.

The first is marketing. The second mistakes an absence of funded trials for an absence of effect - and that distinction is the whole of this page.

Start with the compound that went the distance

SS-31 (elamipretide). On 19 September 2025 the FDA granted it accelerated approval under the brand name Forzinity, to improve muscle strength in Barth syndrome - the first treatment ever approved for that disorder.

It got there the ordinary way. A defined mechanism: SS-31 binds cardiolipin, a phospholipid in the inner mitochondrial membrane, stabilising the cristae structure that electron transport depends on. A randomised, double-blind, placebo-controlled crossover trial (TAZPOWER). An FDA advisory committee, which voted 10–6 in favour. Then approval. The FDA's own integrated review is public.

Nothing about SS-31 is biologically special compared with its neighbours in this category. What was different is that it had a sponsor. Barth syndrome is an ultra-rare disease, and rare-disease designation carries incentives - market exclusivity, fee waivers, tax credits - that make the economics of a trial work. The moment the money existed, so did the evidence.

That is the whole argument of this page, demonstrated rather than asserted. Full detail: is SS-31 FDA approved? - and the SS-31 compound reference, where it sits in the V-04 Cellular energy and longevity protocol.

Why almost nothing else in this category gets tested

A pivotal trial is one of the most expensive things in commercial science. Hundreds of participants, a year or more of follow-up, multiple sites, independent monitoring, regulatory-grade data capture.

Companies make that bet for exactly one reason: they expect to own the answer. Patent exclusivity is the mechanism - fund the trial, get a protected window, recoup the spend inside it. Remove the window and no rational sponsor proceeds.

Now look at what these molecules are. BPC-157 is a fifteen-amino-acid fragment of a protein found in human gastric juice. TB-500 is a fragment of a naturally occurring protein. GHK-Cu is a tripeptide described in the literature since the 1970s. These are short, natural, long-published sequences. A company that spent nine figures proving BPC-157 repairs tendon would publish the result and hand it to every competitor on the same day.

So the trial doesn't happen. Not because anyone doubts the compound - because the spreadsheet says no. (The patent landscape around BPC-157 is mapped in this 2025 literature-and-patent review, which is worth reading precisely because it treats the two as the same question.)

What gets funded instead is academic and investigator-led work on budgets orders of magnitude smaller. That doesn't produce bad studies. It produces small ones - fewer participants, shorter follow-up, one site, surrogate endpoints.

Which is cheap evidence, not negative evidence. Collapsing those two is the single most common error in coverage of this field, and most of the "peptides aren't proven" genre rests on it.

What the 2026 review actually measured

In September 2026, a UCLA-led team published a review in The American Journal of Sports Medicine covering 565 studies across BPC-157, TB-500, CJC-1295, ipamorelin, MK-677 and GHK-Cu (UCLA Health summary · UCLA Newsroom).

It found more than two-thirds of that work was animal-only, the human studies small, and no adequately powered trials demonstrating clinical benefit for musculoskeletal conditions.

Those findings are accurate and we're not disputing them. But read what the instrument measures. A study count measures research funding. It tells you how much money has been spent asking a question. It does not tell you the answer. Run the same review across any unpatentable molecule in medicine and you get the same shape - which is a fact about the pharmaceutical economy, not about biology.

"Not demonstrated" and "disproven" are different claims

Worth stating precisely, because the two get swapped constantly:

Not shown to work - the adequate trial hasn't been run. A statement about evidence. Shown not to work - an adequate trial was run and the compound failed it. A statement about the compound.

Almost everything in this category sits in the first. The honest word is unestablished - a real third state, distinct from proven and from debunked, and one that most of medicine's off-patent shelf occupies.

What the preclinical work legitimately establishes

Animal and mechanistic work is over-claimed by sellers and dismissed too fast by skeptics. What it genuinely supports:

• That a mechanism exists. A compound that reliably drives angiogenesis or fibroblast migration in living tissue is doing something real. That's a finding, not a slogan.
• That a pathway is plausible. Mechanism tells you where to look, not what you'll find. Plenty of good mechanisms have disappointed. But a coherent mechanistic story plus a consistent preclinical signal is a materially different position from neither.
• That safety is broadly characterised across the exposures studied.

What it cannot give you is effect size in a human, or durability. Those need the trial nobody funded. How to read a study without swallowing or dismissing it: what a study actually measured.

The comparison that actually matters

Most coverage frames this as peptide versus proven medicine. For a lot of readers that isn't the choice in front of them.

With a stubborn tendinopathy, the conventional pathway is load management, physiotherapy, time, and a conversation about injection or surgery if it's still there in a year. That pathway is legitimate. It is also, for a great many people, slow, incomplete, or both.

Someone eight months into a recovery that has stalled is not being reckless when they ask what else exists. They are responding to a real gap - and that gap is the reason this category exists at all. Conventional medicine is superb at treating disease once it presents. It has much less for the person who isn't ill but isn't recovering.

So the honest framing isn't proven versus unproven. It's a weighed decision under uncertainty, made with someone qualified - against a realistic account of what the alternative actually offers.

Where caution is genuinely warranted

Two things deserve to be taken seriously on their own terms.

MK-677. A trial was halted over congestive-heart-failure risk. That is a different category of fact from thin evidence - an early stop for harm means a monitoring committee with unblinded data decided continuing wasn't defensible. Notably, MK-677 is among the better-studied compounds here; the signal exists because someone looked. Full treatment: MK-677 (ibutamoren), honestly assessed.

Product is a separate question from compound. The literature on a molecule tells you nothing about what's in a given vial. These are independent problems and conflating them muddles both. What addresses the second isn't a study count - it's knowing what you've been supplied and having someone accountable for it.

What supervision actually changes

The thinner the evidence, the stronger the case for doing this properly - not because something is likely to go wrong, but because monitoring is how you find out early if it does, and how you find out whether it's working at all.

In practice: baseline bloodwork before anything starts, a clinician who will tell you when a compound isn't worth it, verified product rather than an unlabelled vial, and follow-up markers that turn an experiment into information. That's the difference between a considered decision and a guess.

Where the compounds in this review sit in that model: BPC-157 and TB-500 in the V-01 GLOW recovery protocol; CJC-1295 and ipamorelin in the V-G1 Build GH-axis protocol; GHK-Cu in V-06 Radiance.

What would move the evidence

Concretely: investigator-led and foundation-funded trials at adequate size, pre-registered so endpoints are fixed before the data lands; registries that capture real-world outcomes instead of letting them evaporate; published null results; independent replication. None of that needs a scientific breakthrough. It needs funding and coordination - a different problem, and a more tractable one.

Frequently asked questions

Is there scientific evidence that peptides work? Yes, at varying strengths. One - SS-31 - holds FDA approval on randomised placebo-controlled data. Several approved peptide medicines sit outside this review entirely (semaglutide, tirzepatide, tesamorelin, bremelanotide). For the research-stage compounds, there's substantial evidence of biological activity and much thinner evidence of clinical effect size in humans - a gap driven mainly by which trials got funded.

Does the 2026 review mean these compounds don't work? No. It measured the composition of the published literature - mostly preclinical, human work small. It did not test these compounds and find them ineffective.

Why would nobody fund a trial if a compound genuinely helps? Because funding a trial and benefiting from it are different things. Without exclusivity, a sponsor who proves a compound works has paid to inform its competitors. SS-31 is the control condition: a rare-disease designation supplied the incentive, and the evidence followed.

So what should I actually do? Start with an assessment rather than a compound. Establish what the problem is, what conventional care genuinely offers you, and where something investigational might reasonably fit - with baseline bloodwork, monitoring, and a clinician willing to say no.

Related reading

• Is SS-31 FDA approved? - the approval, in detail
• SS-31 and MOTS-c: the mitochondrial peptides
• What a study actually measured - reading the evidence yourself
• Does BPC-157 heal injuries? · TB-500 · CJC-1295 and ipamorelin explained · Do copper peptides work? · What 1,700 adverse event reports do and don't tell you
• MK-677 (ibutamoren), honestly assessed - where caution is warranted
• Peptide therapy monitoring panels - what supervision looks like in practice
• Part of The Complete Guide to Peptides

The next step is an assessment, not a purchase. Vivre Labs recommends starting with baseline bloodwork, supplies verified product, and will tell you plainly when something isn't worth doing. That's what turns a decision under uncertainty into a reasonable one. Book a consult - Telegram · see the goal-based protocols.

---

Educational content, not medical advice. This page describes the state of published research and does not recommend or discourage any compound for any individual. SS-31 (elamipretide) is FDA-approved for Barth syndrome only; use for any other purpose is off-label. Other compounds named here are not approved for the uses discussed, and several are prohibited in sport. Evidence changes - verify against current sources. Claims for the Malaysian market are subject to MAB requirements. Position stated as of 15 September 2026.

References

• FDA accelerated approval of elamipretide (Forzinity), 19 September 2025, Barth syndrome, based on the randomised, double-blind, placebo-controlled crossover TAZPOWER trial; Cardiovascular and Renal Drugs Advisory Committee voted 10–6 in favour. Accelerated approval carries a confirmatory-evidence obligation. FDA integrated review, NDA 215244
• UCLA Health / The American Journal of Sports Medicine - review of 565 studies across BPC-157, TB-500, CJC-1295, ipamorelin, MK-677 and GHK-Cu, published 8 September 2026. (uclahealth.org · newsroom.ucla.edu) Safety findings included a trial of MK-677 halted over congestive-heart-failure risk.
• BPC-157 literature and patent review - *Multifunctionality and Possible Medical Application of the BPC 157 Peptide: Literature and Patent Review* (PMC).
• BPC-157 musculoskeletal narrative review - *Regeneration or Risk?* (PMC).
• Trial economics: the dependence of commercial trial sponsorship on patent exclusivity is well described in the health-economics literature. Pivotal-trial costs vary widely by therapeutic area and methodology; the structural point - sponsors fund trials they expect to recoup through exclusivity - is not disputed.
• WADA Prohibited List (S2) - TB-500 and growth-hormone secretagogues.