Peptide basics

Peptides with no registered human trial: 25 of the 68 we track, named

This page counts registry entries. It does not evaluate whether any compound works, is safe, or is appropriate for anyone, and no absence described here should be read as proof that a compound does nothing.

The count, and the rule that produced it

On 1 August 2026 we queried ClinicalTrials.gov for each of the 68 compounds in the market directory and then re-checked every result by hand-equivalent rule: a study is counted only if the compound's name appears in one of the study's registered intervention names, or in an intervention's other-names field. A keyword match in a title or a condition does not count.¹

That rule matters more than the number. ClinicalTrials.gov's intervention search is deliberately loose, and short names collide badly: a raw query for MGF returns 102 studies, almost all of them about African swine fever virus genes or unrelated oncology drugs. Publishing 102 would have been a fabricated figure laundered through a real API. Under the intervention-name rule, MGF returns zero.¹

By that rule, 25 of 68 compounds have no registered human trial. All but one of them also have no FDA-approved product. The exception is sermorelin, which has both an approved application and an empty registry row, for reasons of chronology rather than science.¹⁸

The 25, with what exists instead

Each line gives the compound, the number of PubMed records returned for its name on the same date, and the average monthly English Wikipedia readership over complete months in the first half of 2026 where an article exists. The publication figure is a raw term count, not a curated bibliography — see the note beneath the list.¹⁵⁷

Cognitive, mood and sleep chemistry

• Semax — 230 PubMed records, 14,065 readers a month.
• Selank — 135 PubMed records, 6,791 readers a month.
• Dihexa — 18 PubMed records, 4,044 readers a month.
• Noopept — 106 PubMed records, 224 readers a month.
• DSIP (delta sleep-inducing peptide) — 543 PubMed records, 2,664 readers a month.¹⁵⁷

Growth hormone secretagogues

• Sermorelin — 331 PubMed records, 8,297 readers a month. Has an approved FDA application; see the separate page on why.
• GHRP-6 — 784 PubMed records, 1,552 readers a month.
• GHRP-2 — 251 PubMed records, 73 readers a month.
• Hexarelin — 312 PubMed records, 30 readers a month.¹⁵⁷⁸

Tissue repair and growth factors

• IGF-1 LR3 — 44 PubMed records, 2,390 readers a month.
• IGF-1 DES — 723 PubMed records, no English Wikipedia article.
• MGF (mechano growth factor) — 2,001 PubMed records, 597 readers a month.
• Follistatin 344 — 21 PubMed records, no English Wikipedia article.
• KPV — 52 PubMed records. No readership figure: no Wikipedia article about the peptide could be matched with confidence.¹⁵⁷

Longevity and bioregulator chemistry

• Epithalon — 126 PubMed records, 5,080 readers a month.
• Pinealon — 22 PubMed records, 2,562 readers a month.
• FOXO4-DRI — 19 PubMed records, 1,079 readers a month.
• Humanin — 558 PubMed records, 671 readers a month.
• Thymalin — 293 PubMed records, no English Wikipedia article.¹⁵⁷

Metabolic and body-composition chemistry

• SLU-PP-332 — 10 PubMed records, 4,136 readers a month.
• AOD-9604 — 23 PubMed records, 26 readers a month.
• Adipotide — 7 PubMed records, 58 readers a month.
• 5-Amino-1MQ — no PubMed records for the name, no English Wikipedia article.¹⁵⁷

Melanocortin and cosmetic chemistry

• Palmitoyl pentapeptide-4 — 6 PubMed records, 603 readers a month.
• Acetyl octapeptide-3 — 2 PubMed records, no English Wikipedia article.¹⁵⁷

What the empty row actually says

Trial registration in the United States is a legal obligation attached to a defined class of studies. The 2016 final rule implementing Title VIII of the Food and Drug Administration Amendments Act sets out which trials of drug, biological and device products must be registered and have results submitted, and it took effect on 18 January 2017. Registration is therefore an artefact of a regulatory regime with edges, not a universal census of human research.²

So an empty row is consistent with several very different worlds. It is consistent with nobody ever having given the compound to a person in a study. It is also consistent with studies that ran before registration was required, studies outside the rule's scope, studies registered under a sponsor code the name-match rule cannot see, and studies registered somewhere other than ClinicalTrials.gov.¹²

That last possibility is not hypothetical for this list. Semax, Selank, Epithalon, Pinealon, Noopept and Thymalin are Russian-origin compounds with Russian-language literatures. The WHO Registry Network's list of primary registries names eighteen registries, in Australia and New Zealand, Brazil, China, Korea, the European Union, India, Cuba, Germany, Iran, the United Kingdom, Japan, Lebanon, Thailand, Africa, Peru and Sri Lanka. None of them is in Russia.³

What a regulator said about the same gap

The registry gap is not only our reading. In its published safety assessments of bulk drug substances nominated for use in compounding, the FDA repeatedly gives the same reason for placing peptides in the significant-safety-risk category: for dihexa acetate and for KPV it states it has not identified any human exposure data for any route of administration, and for emideltide (DSIP), epitalon, Semax and Selank acetate it states it lacks sufficient information to know whether the drug would cause harm if administered to humans.⁴

That is a regulator, working from its own review rather than from our count, describing the same hole in the same compounds. It is corroboration of the gap. It is not an endorsement of anything, and it is emphatically not a finding that these compounds are inert.⁴

Why the gap is worth publishing

Nineteen of them have an English Wikipedia article, and between them those articles draw about 55,000 readers a month. Semax alone draws 14,065. Attention and evidence are separate quantities, and on this list they are not merely uncorrelated — several of the most-read compounds have the thinnest registry record on the site.¹⁷

The statistical point underneath this is old and well stated: absence of evidence is not evidence of absence. A study that finds nothing and a body of research that never happened produce the same silence, and only one of them is informative. Applied here, it cuts both ways. It forbids concluding that Epithalon does nothing, and it equally forbids treating 126 papers as though they had settled a question no registered trial has asked.⁶

The useful posture is narrower than either. For these 25 compounds, the public human-trial record on ClinicalTrials.gov is empty. Anyone making a claim about what one of them does in people is drawing on something else, and the honest next question is what.¹