Research Peptide Fundamentals research peptides: what has actually been checked

The short version

Checked Peptides is a reading desk with an unusual job: it keeps a written record of what peptide research has not shown.

A peptide is a short chain of amino acids, the same building blocks that make up proteins, only much smaller. Some peptides are approved medicines. Some are sold as supplements. Some circulate as research chemicals, meaning material sold for laboratory use with no regulator checking what is in the vial.

Almost every page written about these compounds tells a reader what a study found. Far fewer say what the study did not look at, how many people were in it, how long it ran, or whether an unrelated laboratory has ever repeated it. Those absences are the whole story, because they are what separates a promising idea from a settled fact.

Four compounds sit on this desk: BPC-157, retatrutide, NAD+ and PT-141. Each of them is missing something different, and each missing piece is named here in plain language, next to the study that reveals it.

How a gap gets written down here

A gap only counts on this desk if it can be stated precisely. "More research is needed" is not a finding; it is a way of avoiding one. So every claim summarised on these pages is asked the same five questions, and the answers are what the pages report.

1. Which species? A result in a rat is a result in a rat. It is a reason to run a human study, not a substitute for one.
2. How many participants? A trial of 1,267 people and a report on two people are both "human data", and they carry nothing like the same weight.
3. How long did it run? A compound studied for eight weeks has an eight-week safety record, whatever anyone hopes about year five.
4. Which endpoint was measured? Body weight, blood sugar and a blood level are surrogate endpoints: stand-ins that are easy to measure and are assumed to track something that matters. Heart attacks, fractures and deaths are outcomes. Trials of these four compounds have measured far more of the former than the latter.
5. Has anyone independent repeated it? A literature produced largely by one research group is a different kind of evidence from one confirmed across unrelated laboratories.

Applied honestly, those five questions sort the four compounds on this desk onto very different rungs. They also explain why two compounds can both be described as "studied" and mean almost opposite things.

What research peptides are, in plain terms

Peptides work by fitting into receptors, which are best pictured as locks sitting on the surface of a cell. A molecule that fits the lock and turns it is an agonist. Most of what a peptide does, and most of what can go wrong with it, follows from which locks it turns and where in the body those locks happen to be.

The four compounds here turn very different locks. Retatrutide turns three gut-hormone receptors at once ^[9]. PT-141 turns melanocortin receptors clustered in the brain rather than in blood vessels ^[22]. BPC-157 is linked most consistently to a receptor called VEGFR2 that governs the growth of new blood vessels ^[4]. NAD+ is not a peptide at all in the strict sense; it is a coenzyme, a small helper molecule that cells use to move electrons around and to fuel a set of repair and signalling enzymes ^[16]. It sits on this desk because it is sold, injected and argued about in exactly the same market.

Two more terms recur throughout the site. Half-life is how long it takes the body to clear half a dose, and it is the number that dosing schedules are built from. Preclinical means the work was done in cells or animals; Phase 1 asks whether a compound is tolerated in a small number of people, Phase 2 looks for a signal of effect, and Phase 3 is the large, controlled test that regulators judge. Knowing which of those words applies to a claim is usually enough to know how much of it to believe.

Four peptides, four different things missing

The reason these four sit together is that their gaps are not the same gap. Lining them up makes the differences visible.

• BPC-157 has a species problem. The famous results, a fully cut rat Achilles tendon healing faster ^[6] and rat stomach ulcers shrinking ^[5], are animal results. A 2025 review states plainly that only three small pilot studies have examined BPC-157 in humans and that rigorous large-scale trials are lacking ^[2]. Its pharmacokinetics have been characterised only in rats and beagle dogs ^[3].
• Retatrutide has an endpoint and time problem. The human trials are real and substantial: 338 adults over 48 weeks for obesity ^[11], 281 adults over 36 weeks for type 2 diabetes ^[12]. But what they measured was weight, HbA1c and liver fat ^[10]. No published trial has yet reported whether that translates into fewer cardiovascular events, or whether the weight stays off after the injections stop.
• NAD+ has a compartment problem. Oral precursors raise NAD+ in blood dependably and dose-dependently ^[14][17]. A 2025 review of the human evidence nonetheless concludes that clinical efficacy has been limited and that data on tissue-specific NAD+ remain sparse ^[13]. The blood number is not in dispute; what it means for the tissues is.
• PT-141 has a population problem. It is approved, which puts it above the others on evidence, but approved for one narrow group: premenopausal women with low sexual desire, studied in two Phase 3 trials totalling 1,267 participants ^[20]. Most of the people discussing it online are outside that group entirely, and there is no trial for them.

Read as a set, the comparison page makes the point this desk exists to make: "the evidence is limited" is almost never a useful sentence, because the limits are in different places and matter in different ways.