Peptides: Promising Biology, Uneven Evidence

The word "peptide" tells you almost nothing about how much is actually known -- it spans an FDA-approved medicine and a compound that has never been tested in a human being.

6–45 min read

The Question

"Peptides" has become shorthand in wellness and longevity circles for an entire category of injectable, research-forward compounds -- exciting, a bit underground, evidence-optional. That framing collapses something that shouldn't be collapsed: a peptide is just a short chain of amino acids, and that chemical description covers hormones the body already makes, decades-old approved medicines, drugs in active human trials, and compounds whose entire evidence base is a handful of rodent studies. This piece sorts those apart, one tier at a time, using named examples rather than the category as a whole.

The Biology

Chemically, a peptide is a chain of amino acids linked by peptide bonds, conventionally shorter than what gets called a protein (a rough, fuzzy cutoff around 50 amino acids). Many peptides are signaling molecules -- hormones, neurotransmitters, growth factors -- which means the body is already full of naturally occurring peptides (insulin, glucagon, oxytocin, growth-hormone-releasing hormone) doing exactly the kind of signaling work that makes "a peptide could plausibly do X" an easy sentence to say about almost anything.

That plausibility is also the trap. A molecule that resembles a natural signaling peptide, or mimics one, is a reasonable hypothesis generator -- not evidence of a clinical effect at a given dose in a given person. The distance between "this peptide binds a real receptor in a dish" and "this peptide changes a health outcome in a person" is where nearly all of this category's genuine uncertainty lives.

What We Know

Some peptides are unambiguously established medicines with regulatory approval and large trial programs behind them. Insulin itself is a peptide hormone. GLP-1 receptor agonists such as semaglutide and liraglutide are peptide-based drugs approved for diabetes and obesity on the strength of multiple large randomized trials. Tesamorelin (brand name Egrifta) is an FDA-approved growth-hormone-releasing hormone analogue, approved specifically for HIV-associated lipodystrophy after a 2007 randomized trial (Falutz et al., n=412) found a 15.2% reduction in visceral adipose tissue at 26 weeks versus a 5.0% increase with placebo, later confirmed by a 2014 randomized trial (Stanley et al.) showing reductions in visceral and liver fat alongside a transient early rise in fasting glucose.

What these approvals share is the pattern that defines "established" for the rest of this piece: a specific molecule, a specific dose, a specific patient population, and multiple controlled human trials measuring hard clinical or imaging endpoints -- not a proxy, and not an animal result extrapolated forward.

What We're Learning

Elamipretide (also called SS-31), a mitochondria-targeting peptide, reached a genuine phase 3 randomized trial -- MMPOWER-3, published in Neurology in 2023 (n=218) -- in primary mitochondrial myopathy. It did not improve its primary endpoints, six-minute walk distance or a fatigue-symptom scale, despite being well tolerated. That result is a useful case study in its own right: a molecule with real mechanistic plausibility and a real disease target can still fail to clear the human-trial bar, and a clean trial failure is genuine evidence, not just a disappointment.

GHK-Cu, a naturally occurring copper-binding tripeptide, has a substantial mechanistic and preclinical literature -- gene-expression, wound-healing and cell-signaling studies reviewed in depth by Pickart and Margolina in 2018. Worth noting plainly: that review's senior author has a documented commercial interest in copper-peptide skincare products, and the evidence for topical use remains considerably stronger than for any systemic or injectable use. NORTEAVA covers GHK-Cu's biology in more depth in its own dedicated entry, linked below.

MOTS-c, a peptide encoded by mitochondrial DNA with genuinely interesting metabolic signaling in animal models, is still substantially at the animal- and cell-model stage; human trial data is minimal as of this writing.

Where the Evidence Gets Thin

BPC-157, a synthetic peptide derived from a fragment of a protein found in gastric juice, has an extensive rodent literature on gut healing, tendon repair and angiogenesis -- and, as of this writing, no published randomized controlled trial in humans that NORTEAVA has been able to verify. Its regulatory treatment has shifted more than once and remains unsettled; rather than assert a specific current legal status that may already be out of date by the time you're reading this, we'd point you to the FDA's own bulk drug substances page as the primary source to check. What is stable is the FDA's stated reasoning across this whole class of unapproved peptides: insufficient characterization and a lack of human safety data -- a more basic problem than "we don't yet know if it works."

TB-500, a synthetic fragment modeled on thymosin beta-4, sits in a similar place: real, published biology around actin regulation and tissue repair in animal and cell models, and essentially no controlled human trial evidence of the kind the approved peptides above were built on.

Epitalon (also spelled epithalon), a synthetic tetrapeptide modeled on a natural pineal peptide, has a genuinely interesting finding behind it -- laboratory work reporting increased telomerase activity and telomere length in human cell lines. But "activates telomerase in a dish" and "extends healthy lifespan in a person" are separated by exactly the gap this piece has already described twice, and claims about its use in aging humans rest overwhelmingly on older reports from Russian research groups that are difficult for an outside reviewer to independently verify.

What We Know

"Peptide" is a chemical-structure description, not an evidence tier -- the category includes insulin, FDA-approved drugs like tesamorelin and semaglutide, investigational compounds in real human trials, and compounds with no published human trials at all.
Tesamorelin's FDA approval rests on multiple randomized trials with hard imaging endpoints (visceral and liver fat by CT/MRI) -- the evidence bar the rest of the category is measured against.
A real phase 3 human trial (MMPOWER-3, elamipretide) can fail its primary endpoints despite a plausible mechanism and a real disease target -- trial failure is informative, not just disappointing.

What We Don't

Whether BPC-157 or TB-500 do in humans what they clearly do in rodent tissue-repair models -- neither has a published controlled human trial as of this writing.
Whether Epitalon's cell-line telomerase finding translates into any measured human aging or longevity outcome in a trial NORTEAVA has been able to verify.
The current, exact regulatory status of several research-chemical peptides, which has shifted more than once and is worth checking against a primary source, such as the FDA's own bulk drug substances page, rather than any single article -- including this one.

What People Are Doing Anyway

Online peptide communities routinely treat this entire list -- from tesamorelin to BPC-157 to Epitalon -- as one continuous menu differing mainly by price and where to source it, with dosing protocols passed around that have no connection to any of the trials, or absence of trials, described above. NORTEAVA doesn't publish a protocol for anything in the hypothesis-or-thinner tier, for the same reason a science desk doesn't publish a dosing recipe for a drug that hasn't been through human trials: it would misrepresent how much is actually known.

The Norteava View

The category-level lesson here matters more than any single compound's status, because the compound-specific facts will keep shifting: "peptide" describes a molecular shape, not a claim about evidence. Grouping insulin, semaglutide, tesamorelin, elamipretide, GHK-Cu, MOTS-c, BPC-157, TB-500 and Epitalon into one bucket and then reasoning about that bucket as though it has one evidence level is a category error, and it's the single most common failure mode in how this subject gets discussed online. The useful question is never "are peptides good" -- it's "which peptide, tested how, in whom, with what result," and the answer is different for every name on this list.

What We'd Like to See Next

Any published, adequately powered human RCT for BPC-157 or TB-500, given how large their preclinical literatures already are. Continued follow-up on elamipretide's genotype-specific post-hoc signal from MMPOWER-3, since a clean overall miss doesn't rule out a real effect in a defined subgroup. And any rigorous, independently funded human trial data on GHK-Cu for systemic or injectable use, which right now essentially doesn't exist.

Selected Research

Related Research