RESEARCH PEPTIDE FUNDAMENTALS / CROSS-COMPOUND
Three peptides, three evidence bases, three legal positions
Side by side, the three compounds on this desk share almost nothing except the word peptide — and the comparison is what makes the identity problem visible.
The short version
These three compounds get discussed in the same breath and belong in three different categories. Tirzepatide is an approved prescription medicine with large controlled trials behind it. Ipamorelin is an unapproved compound whose one published trial in people did not work. BPC-157 is an unapproved compound that has barely been tested in people at all.
Their receptors are unrelated, their evidence is at completely different stages, and their legal standing runs from fully approved to specifically excluded from pharmacy compounding. The only thing they truly share is that all three are short chains of amino acids.
Comparing them is useful because it separates two things that usually get merged: how good the evidence is, and how legal the compound is. Those are different axes, and a preparation of unknown content undermines both of them at once.
Three regulatory positions in one table
The table sets the three side by side on the dimensions that actually differ. Every quantitative entry carries its source.
| BPC-157 | Ipamorelin | Tirzepatide | |
|---|---|---|---|
| Class | cytoprotective pentadecapeptide | selective growth hormone secretagogue | dual GIP/GLP-1 receptor agonist |
| Target | VEGFR2 and repair machinery [4] | ghrelin receptor GHS-R1a [11] | GIPR and GLP-1R [14] |
| Largest human dataset | three small pilot reports [2] | one Phase 2 trial, 114 adults [8] | phase 3 trials of 2,539 and 1,879 adults [16] [17] |
| Did it work in people? | never formally tested for efficacy | no — primary endpoint missed [8] | yes, repeatedly and in head-to-head trials [13] [16] [17] |
| Regulatory position | not approved anywhere; excluded from 503A compounding pending review | not approved anywhere; removed from the interim 503A Category 2 list in 2024 | FDA-approved from May 2022, prescription-only [14] |
| Prohibited in sport? | yes, at all times | yes, at all times | not specifically listed |
| What an unverified vial costs the reader | everything — no human baseline exists to compare against | everything — the reported effects were never trial endpoints | the trial numbers, which belong to characterised material, not to the name |
The last row is the one this desk cares about most, and it is the row that would be missing from an ordinary comparison chart.
Evidence maturity and approval are not the same axis
It is tempting to read approval as a score for evidence. It is closer to a record of who filed what, and when.
Ipamorelin illustrates the distinction. Its pharmacology is described with unusual precision: a founding characterisation across rat pituitary cells, anaesthetised rats and conscious swine established both potent growth hormone release and the absence of an ACTH or cortisol rise at doses more than 200-fold above the growth-hormone half-maximal dose [11], and human pharmacokinetics were mapped with a terminal half-life of about two hours and a single growth hormone pulse peaking around forty minutes after dosing [9]. That is more mechanistic clarity than many approved drugs enjoyed at launch. It is unapproved because the trial that mattered missed its endpoint [8], not because the mechanism is murky.
BPC-157 sits at the other corner. Its animal literature is broad and largely consistent — ulcer-formation inhibition of 45.7–65.6 per cent at higher doses in rats [5], a coherent VEGFR2-linked angiogenic mechanism [4] — while its human literature is three pilot reports [2]. It is unapproved because almost nobody has run the human trials, and a 2025 review says so explicitly [2].
Tirzepatide is the case where the two axes happen to line up: strong controlled evidence in large populations [13] [16] [17], and approval to match [14]. That alignment is common enough to be mistaken for a rule. The other two pages on this desk are the counterexamples.
What none of these three literatures can say
For all their differences, the three bodies of evidence share a blind spot of exactly the same shape.
None of them contains an assay of material outside the study system. Trials characterise what they administer, because a result about an uncharacterised substance is not a result. That characterisation stays inside the study; it does not extend to anything bearing the same name elsewhere. No paper in this reference list examined a preparation obtained outside a controlled research setting, and no paper claims to.
None of them establishes that a research-use-only designation carries a purity standard, because that designation is a statement about permitted use rather than about content.
And none of them can transfer a trial number onto an unverified preparation. The weight changes in the tirzepatide trials [16] describe what happened to characterised drug given in recorded amounts to enrolled participants who were followed for 72 weeks. Detach any one of those conditions and the number stops describing anything.
Reviewers occasionally say this in the margins — the 2025 BPC-157 review lists non-regulated availability alongside limited human data as reasons for caution [2] — but no journal publishes a paper whose subject is the contents of unlabelled vials. The silence is structural, and it is not evidence of safety.
The interpretation problem, stated plainly
Suppose someone reports that a compound helped. To read that report as evidence about the molecule, four things have to hold: the preparation contained the named molecule, it contained a meaningful amount of it, it contained nothing else that could explain the effect, and the effect would not have occurred anyway.
For an approved medicine taken through a prescription, the first three are underwritten by a manufacturing system and the fourth is what the placebo arm exists to test. For an unverified preparation, the first three are unknown and the fourth is untested. That is not a weaker version of evidence. It is a different kind of statement — a report about an experience, not about a compound.
The most defensible position available to a site like this one is therefore also the simplest. The published literature cannot say where anything came from, and any effect attributed to a preparation of unknown content is uninterpretable. This desk states that plainly rather than hedging it, and then declines the follow-up question about material entirely, because the literature has no answer to it and neither does this site.