A FormBlends network publication
Peptide Reference

How to read a peptide study in ten minutes

Six questions to ask of any study a peptide seller cites: which molecule, which species, what control, what endpoint, who ran it, and has anyone repeated it. Worked examples from TB-500, epitalon, MOTS-c and the tesamorelin label.

By FormBlends editorial teamUpdated September 4, 2026Educational, not medical advice

You do not need a research degree to check a peptide claim. You need the PubMed abstract and six questions. Each question below comes with an example from this site where the answer changed the grade.

1. Is it the same molecule?

Sales pages routinely cite studies of a related compound as if they were studies of the product. Three cases on this site:

  • TB-500 pages cite the phase 1 and phase 2 trials of thymosin beta-4 (PubMed 20536472 and others). Thymosin beta-4 is a 43-amino-acid protein. TB-500 is a 7-amino-acid fragment of it. FDA states it has no human exposure data for the fragment. See TB-500.
  • Epitalon pages cite the 2003 "prolong human life" trial (PubMed 14523363). That trial used epithalamin, a pineal gland extract containing many peptides. Epitalon is a synthetic tetrapeptide designed to mimic it. See epitalon.
  • NAD+ infusion pages cite randomised trials of oral nicotinamide riboside. Different molecule, different route. See NAD+.

Check the first sentence of the Methods for the exact substance and, for peptides, its form (free base, acetate, fragment, analogue).

2. Is it the same species?

The 2025 systematic review of BPC-157 in orthopaedics (PubMed 40756949) screened 544 articles and included 36: 35 preclinical, 1 clinical. That ratio is typical for research peptides. Rat tendon healing is a reasonable place to start and an unreasonable place to stop. The MOTS-c paper most often cited for human relevance (PubMed 33473109) contains human data, but read carefully: the human part measured how much MOTS-c people's own bodies made after exercise; the treatment experiments were in mice. See MOTS-c.

3. What was the control?

"Patients improved" means nothing without knowing what the comparison was. The three BPC-157 human reports have no control group; people with knee pain or interstitial cystitis improve and worsen over time on their own, and the expectation of a novel injection is itself a treatment. The selank anxiety studies (PubMed 18454096 and others) are randomised, but against benzodiazepines with no placebo arm. Anxiety has one of the largest placebo responses in medicine; matching a benzodiazepine without a placebo does not show either drug beat placebo in that sample. See selank.

The hierarchy, roughly: randomised and placebo-controlled beats randomised against an active comparator, which beats a concurrent non-randomised control group, which beats a historical control, which beats no control. Our grading scale draws its C/D line at "any concurrent control."

4. What was the endpoint, and was it the one they set out to measure?

Hormone levels are not outcomes. The CJC-1295 trial (PubMed 16352683) showed growth hormone and IGF-1 rose for days. It did not measure fat, muscle, recovery or anything a buyer wants. The ipamorelin phase 2 trial (PubMed 25331030) had a proper clinical endpoint, time to tolerating a meal after surgery, and missed it (p = 0.15). A page that reports the hormone rise and omits the clinical miss is telling you which half of the literature it read.

Also look for the word "secondary." A significant secondary endpoint in a trial that missed its primary is a hypothesis for the next trial, not a result.

5. Who ran it, and where?

Sponsor-run and single-institute studies are not invalid; most approved drugs were tested by their sponsors. But a literature that consists entirely of one group's work, without any independent replication, is unconfirmed by definition. The semax, selank and epitalon literatures share this shape. The BPC-157 human reports are from one private clinic. Ask whether anyone with no stake in the answer has repeated it.

6. Has anyone repeated it?

Replication is what separates a finding from a result. Bremelanotide has two identical phase 3 trials that agree (PubMed 31599840). Tesamorelin has two phase 3 trials pooled (PubMed 20554713) and independent academic trials that followed (PubMed 25038357, 31611038). Epitalon's telomere effect in cells was repeated by an independent group in 2025; its human retinal study never has been.

A bonus: what a label tells you that a study cannot

Section 12.6 of the Egrifta label reports anti-tesamorelin antibodies in 50% of patients at 26 weeks. That number is only known because a regulator required it to be measured. No research peptide has an equivalent figure, which is exactly FDA's immunogenicity concern in every Category 2 entry: nobody has looked. A label is not just a longer document than a study; it is the record of questions a sceptical reviewer forced someone to answer. When a page cites a label for one product and studies for another, note the asymmetry.

The ten-minute routine

Open the abstract. Read the Methods sentence for molecule, species, n and control. Read the Results sentence for the primary endpoint and the p-value. Check the author affiliations. Search PubMed for the compound name with the "humans" filter and see whether anyone else has published. Then look up the substance in the evidence grade lookup and see whether you reached the same letter we did. If you did not, the about page says how to tell us.

Questions people ask

What is the fastest way to check a peptide claim?

Take the PubMed id from the citation (or search the title), open the abstract, and read the Methods sentence. It will tell you the species, the number of subjects and whether there was a control group. Most peptide claims fail at that sentence.

Does a study being in a Russian or Chinese journal make it less valid?

No. Language and country say nothing about validity. What matters is design, transparency and independent replication. The Russian semax and selank studies are limited by open designs and single-source authorship, not by where they were published.

How do I tell an approved-drug label from a marketing page?

A label on DailyMed has numbered sections (Indications, Dosage, Warnings, Adverse Reactions with placebo comparison rates, Clinical Studies with trial sizes), a revision date and a manufacturer accountable to FDA. A marketing page has none of that, and usually a dosing table in its place.

Sources

  1. Vasireddi N et al. Emerging Use of BPC-157 in Orthopaedic Sports Medicine: A Systematic Review. HSS J 2025. PubMed 40756949 Accessed September 4, 2026.
  2. Ruff D et al. A randomized, placebo-controlled, single and multiple dose study of intravenous thymosin beta4 in healthy volunteers. Ann N Y Acad Sci 2010. PubMed 20536472 Accessed September 4, 2026.
  3. Khavinson VKh, Morozov VG. Peptides of pineal gland and thymus prolong human life. Neuro Endocrinol Lett 2003. PubMed 14523363 Accessed September 4, 2026.
  4. Reynolds JC et al. MOTS-c is an exercise-induced mitochondrial-encoded regulator of age-dependent physical decline and muscle homeostasis. Nat Commun 2021. PubMed 33473109 Accessed September 4, 2026.
  5. Zozulia AA et al. Efficacy and possible mechanisms of action of a new peptide anxiolytic selank in the therapy of generalized anxiety disorders and neurasthenia. Zh Nevrol Psikhiatr Im S S Korsakova 2008. PubMed 18454096 Accessed September 4, 2026.
  6. DailyMed: EGRIFTA SV (tesamorelin) prescribing information, section 12.6 Immunogenicity Accessed September 4, 2026.

Canonical URL: https://formblendspeptides.com/guides/how-to-read-a-peptide-study. Written by the FormBlends editorial team. This page is educational and is not medical advice; see the medical disclaimer.