Cardiogen (Ala-Glu-Asp-Arg): what the research amounts to

Has Cardiogen been tested in humans?

No, and that belongs at the top rather than in a caveat at the bottom. Not thin, not preliminary, not mostly animal. We went looking for a human record and came back with nothing at all.

The searches, with the exact strings: on PubMed the search term "Ala-Glu-Asp-Arg" AND humans[MeSH Terms] returned 0 records, and cardiogen AND (Khavinson OR bioregulator OR tetrapeptide) AND humans[MeSH Terms] returned 0 records. Both ran on 4 August 2026, in the same session as the control semaglutide AND humans[MeSH Terms], which returned 3,085 records. The human filter was working when the two zeros came back, which is the only reason a zero is worth reporting.

The registries agree: on the same day, ClinicalTrials.gov returned no study for AEDR and none for Khavinson as a search term, against 757 studies for semaglutide on the identical interface. The patent that generated this molecule contains no occurrence of the word human, none of volunteer, and none of clinical trial across its full retrieved text.

What that does and does not mean: an empty literature is not a finding of harm and it is not a finding of safety. It means every effect statement anyone can make about this compound is a statement about a rat, a slice of rat heart in a dish, a mouse cell line, or a computer. There is no dose, no duration, no endpoint and no adverse event profile in a person to report, because no such study has been published where we can find it.

The comparison that makes this concrete: a compound with one human paper behind it still tells you something a compound with none cannot. PE 22-28 rests on a single published study and we say so on its own page; this one does not have that single study.

Where the threefold survival number comes from

The figure you meet first: a threefold drop in death rate after a heart attack. It is not invented and it is not a misquote. It is also not from a trial, and it is not from a paper. Every link in the chain below was pulled and read, one document at a time.

Link one, the review. A 2022 review in Cells, from the same research group, states that giving the peptide in experimental myocardial infarction, induced by tying off a coronary artery in rats, "led to a threefold decrease in mortality after the heart attack" compared with control. That sentence carries citation number 123.

Link two, the citation. We pulled the review's full reference list out of the open-access full text. There are 150 entries. Number 123 is not a journal article. It reads: Khavinson VK, Ryzhak GA, Grigoriev EI, Ryadnova IY, Peptide Substance Restoring Myocardium Function, US Patent 7,662,789. Nothing was concealed. The citation sits in full in the reference list of a peer-reviewed review, and opening it is the only step between the number and the document it rests on.

Link three, the primary document. We retrieved the granted patent and read the experiment. It sits in Example 2. The setup: 40 white mongrel rats weighing 180 to 200 g, split at random into two groups of 20. Infarction was induced by tying off the left coronary artery. The treated group received the tetrapeptide in sterile saline at three points after the occlusion, and the control group received the saline alone. Animals were decapitated at 6 and 24 hours. Mortality in the first 24 hours was 45 percent in the control group and 15 percent in the treated group.

The arithmetic nobody prints: 45 divided by 15 is the three. With 20 animals in each arm, 45 percent is 9 deaths and 15 percent is 3. The entire threefold effect is a difference of six rats in a single experiment.

What kind of document a patent is: the experimental section is written by the inventors, filed to support a claim of ownership over the molecule, and examined for patentability. The face of US 7,662,789 names a primary examiner. It names no referees, no journal, and no independent laboratory, because a patent has none of those things. That is not a scandal. It is simply a different category of evidence from a paper, and the distinction disappears the moment a number is lifted out of it.

A small tell that the chain was never walked: the review dates the patent to 16 October 2010. The patent's own front page gives the grant date as 16 February 2010. An eight-month slip is harmless on its own, and it is also what a citation looks like when it has been copied forward rather than opened.

The strangest fact about this compound: the name Cardiogen does not appear in its own patent. Not once in the full text we retrieved. The document only ever writes Ala-Glu-Asp-Arg or alanyl-glutamyl-aspartyl-arginine. The trade name is joined to the sequence only in later review tables: a 2022 paper in the International Journal of Molecular Sciences carries a table row naming Cardiogen as AEDR and classing it a cardioprotector, and a 2021 systematic review in Molecules lists AEDR and Cardiogen together under monofunctional peptides, described as regulating cardiovascular system function. So the commercial name and the primary experiment have never appeared in the same document.

Is Cardiogen an FDA-approved drug?

The peptide is not. There is no US label to read for it: the FDA label database returns nothing for the exact phrases alanyl-glutamyl-aspartyl-arginine or Ala-Glu-Asp-Arg, checked on 4 August 2026. But the question is confusing for a reason that has nothing to do with the peptide, and this is the one page in this family that has to explain it.

There is an approved product with this name. Drugs@FDA holds exactly one application under the brand CardioGen: NDA 019414, sponsored by Bracco Diagnostics, active ingredient rubidium chloride Rb-82, an injectable prescription product, originally approved 29 December 1989. Its own label describes it as a lead-shielded column of accelerator-produced strontium-82, used to produce an injection for positron emission tomography of the heart muscle. It is a radioactive diagnostic agent, not a peptide, not a sequence of amino acids, and related to this molecule in nothing but the name.

It also carries a boxed warning, which is the part that does real damage when it migrates. The current label opens with a boxed warning about high radiation exposure if the wrong eluent is used, and about excess exposure when the eluate testing protocol is not followed. That warning is about handling a radiopharmaceutical generator. Attached to the wrong molecule it is meaningless, and read as reassurance in reverse it is worse than meaningless.

How thoroughly the name is shared: on PubMed the search term cardiogen[Title] returned 6 records on 4 August 2026. Two of them are this peptide. The other four belong to the rubidium generator and to an unrelated cardiomyopathy registry that borrowed the word. A search on the name alone is not a search on the compound.

What we could not settle: registration status in Russia and in Europe. Both official search tools returned the same page for every query we sent them, including for aspirin and for semaglutide, so neither tool was answering the question at all. We report that as unretrieved. Reporting a broken instrument as a zero is how a page ends up stating an absence that was never measured, and it is the mistake we made on the first pass here before a control query exposed it.

What an approval record looks like when there is one: a sponsor, an application number, an indication, and a label with sections you can cite. The bremelanotide approval is a worked example of that shape, and it makes the contrast easy to see.

Is Cardiogen the same thing as Chelohart?

No, and the two are not interchangeable in either direction. Chelohart is documented in a 2015 Russian-language paper as a polypeptide complex extracted from the heart of calves, studied alongside extracts from brain cortex, pineal gland, liver, prostate, thymus and cartilage. Cardiogen is a synthetic chain of four amino acids made in a reactor.

Why the distinction is load-bearing: an organ extract is a mixture whose composition depends on the animal and the process. A tetrapeptide is a defined sequence with a molecular formula, given in the patent as C18H31N7O9 with a molecular weight of 489.48. Evidence gathered on one of those does not transfer to the other, and the names sit close enough together that the transfer happens without anyone deciding to make it.

The paperwork angle: for a defined synthetic sequence there is at least something to verify against, which is what a certificate of analysis is for. The identity and purity sections of a certificate answer a different question from an efficacy claim, and for a compound with no human record the identity question is the only one a document can settle.

What has been measured, and in what?

Everything retrievable, in one table. Species and study type sit in the same row as the result, because that is the pairing that goes missing when a figure travels.

The cell-culture result, read closely: the 2012 paper reports the tetrapeptide raising expression of cytoskeletal proteins and nuclear matrix proteins in cultured mouse embryonic fibroblasts. Those are mouse embryo cells, not heart cells, and a dish is not an animal. The paper's own conclusion ties the result back to previously reported cardioprotective activity, which places it downstream of the patent rather than independent of it. It is a proposed mechanism for a result, not a second observation of the result.

The tissue results, read closely: the two Russian-language papers in Advances in Gerontology worked with organotypic explants, small pieces of organ kept alive outside the animal. The 2009 paper reports that the tetrapeptide stimulated proliferation in myocardium from both 3-month and 24-month rats and lowered p53 expression. The 2006 paper is the tissue-specificity one: heart, lung, prostate and pancreas explants, each exposed to the peptide matched to it. Both are small, both are old, and both are published in a journal with no digital object identifier, so only the abstracts are verifiable from outside.

Where that 2006 paper ends up: its own closing sentence proposes that these peptides could be used clinically to stimulate repair in the matching tissue during aging. That proposal rests on explants in dishes and on nothing else. It is worth seeing the leap happen inside the literature, because it is the same leap that later happens outside it.

What none of this is: a measurement of anything in the powder anyone can buy. The vial form and the paperwork are separate questions from the biology. What a freeze-dried peptide is is worth understanding on its own terms, and it settles nothing about whether the sequence inside does what a patent says it did in rats.

Why does a heart peptide have a tumor result?

Because two researchers ran that experiment and published it in 2009. It is the odd result in this small literature and it is genuinely specific to this compound, so leaving it out would be its own kind of dishonesty.

What the paper reports: cardiogen was given to senescent rats carrying transplanted M-1 sarcoma. Growth of the tumor was inhibited in a dose-dependent way. Apoptosis of tumor cells was higher in every experimental group than in control. The authors attribute the effect to hemorrhagic necrosis acting through the tumor's own blood vessels, and state explicitly that the proliferative measurements do not support a direct cytostatic action on the tumor.

What it is not: a cancer finding in the sense a reader will reach for. One transplanted tumor line, in one aged rodent strain, in one paper, with no replication we could locate. A vascular mechanism that shrinks a graft in a rat is a long way from anything in a person, and the shortest path from this paragraph to a wrong conclusion is to drop the words rat and transplanted.

Why it belongs on the page anyway: a compound marketed for the heart with a published tumor result is exactly the finding that gets rediscovered, stripped of its species, and repeated. Better to meet it here with the rats attached. The same problem in a much larger form sits on the cardarine record, where the animal carcinogenicity data is the whole story and is routinely told without it.

What do the transporter scores actually measure?

Software, not biology. A 2023 paper docked 26 short peptides against the binding sites of amino acid and peptide transporters, using published protein structures and a modeling package. The output is a score: a computed estimate of how well a molecule fits a site. Lower is better in this scale, and a positive number means poor fit.

This compound's own rows: against LAT1 the tetrapeptide ranks 3rd of 26 with a score of -35.35. Against LAT2 it ranks 7th of 26 at -27.54. Against PEPT1 it ranks 3rd of 26 at -27.50. Against PEPT2 it ranks 24th of 26 at +13.15, near the bottom of the list. Those four numbers were read out of the source tables, not out of the abstract.

Why they are here: not because they show anything works. The docking study ran 26 sequences at once, and the 2006 explant work ran four compounds side by side, which is how this small literature is built: a finding arrives attached to a group. These four scores are attached to this sequence and to no other, which makes them at least a number somebody would have to argue with.

Why they prove nothing about a person: no cells were involved, no animals, and no measurement of the peptide entering anything. Docking is a hypothesis generator. Treating a rank order from a computer as evidence of absorption is the same error as treating a patent's rat experiment as a clinical result, one step further removed.

Where the record runs out

The open items, stated as gaps rather than as findings:

• Any human data. No study, no trial, no registry entry, no case report that we could retrieve. This is the first and largest gap and everything else is downstream of it.
• Replication of the rat experiment. The threefold mortality result appears in the inventors' own patent and in reviews written by the same group. We found no independent repetition of it.
• Regulatory status outside the United States. The Russian and European search tools we used were returning identical pages regardless of the query, so we have no answer, only a failed measurement.
• What is in a given vial. The patent describes a synthesis and a purity analysis for the substance the inventors made. That says nothing about any other material, and a purity figure answers a narrower question than its percentage suggests: what an HPLC purity number leaves out applies here with unusual force, because there is no label to check it against.
• One provenance claim we deliberately do not repeat. The 2022 review states that the tetrapeptide was detected in the polypeptide complex of the heart by chromato-mass spectrometry. No citation is attached to that sentence. The patent cited nearest to it contains no mass spectrometry of any kind, and a PubMed search pairing the sequence with mass spectrometry or chromatography returned nothing on 4 August 2026. So the claim is absent from this page, on purpose.

How this page is sourced

What was read directly: the granted patent in full, including the experimental examples and the analytical section; the full text of the 2022 review and its complete 150-entry reference list; the docking tables of the 2023 modeling paper, read from the tables rather than from the abstract; the two trade-name table rows in the 2022 and 2021 reviews; and the PubMed abstracts of the five journal papers that are not open access. The application record for the similarly named radiopharmaceutical came from the FDA approval database, and its current label from the National Library of Medicine label archive.

What we could not read: the full text of the three Advances in Gerontology papers. That journal has no digital object identifier and is not in the open full-text archive, so those three entries rest on their PubMed abstracts and are cited by PubMed identifier alone. They are Russian-language papers with English abstracts, and this page says so wherever it uses them.

One correction made during the build: our own working notes recorded the patent's assignee as one company. The patent's front page names a different one, Access Bioscience CJSC of Saint Petersburg. The document wins, and the note was wrong. It is recorded here rather than quietly fixed, because a page that traces other people's citation errors has no business hiding its own.

What is deliberately absent: any figure for how many people use this, any price, any seller, any claim about what is commonly said about it, and the patent's expiry date, which we could find only as a computed field on a third-party site and never confirmed against the issuing office.

Twelve sources: one patent, nine journal papers with a digital object identifier or a PubMed identifier, and two regulator records.

The standard: the full sourcing and citation-verification standard for this site, including who checks a page before it ships, lives on the methodology page.

Last reviewed: 4 August 2026.

1. 1

   Khavinson VK, Ryzhak GA, Grigoriev EI, Ryadnova IY. Peptide substance restoring myocardium function. US Patent 7,662,789 B2, granted 16 February 2010. PCT filed 7 December 2004, priority application RU 2004118699 of 22 June 2004, assignee Access Bioscience CJSC, Saint Petersburg. Full grant text retrieved and read 4 August 2026; Example 2 carries the rat infarction experiment, and the analytical section gives the formula C18H31N7O9 and molecular weight 489.48. A patent is an office document, not a peer-reviewed paper, and no DOI or PubMed identifier exists for it. Europe PMC patent record US7662789.

2. 2

   Khavinson V, Linkova N, Dyatlova A, Kantemirova R, Kozlov K. Senescence-Associated Secretory Phenotype of Cardiovascular System Cells and Inflammaging: Perspectives of Peptide Regulation. Cells. 2022;12(1):106. Narrative review; its reference 123 is the patent above. doi:10.3390/cells12010106. PMID 36611900.

3. 3

   Khavinson VKh, Lin'kova NS, Polyakova VO, Kvetnoy IM, Benberin VV, D'yakonov MM, Titkov YS. Tetrapeptide H-Ala-Glu-Asp-Arg-OH stimulates expression of cytoskeletal and nuclear matrix proteins. Bull Exp Biol Med. 2012;153(4):559-562. In vitro, cultured mouse embryonic fibroblasts. doi:10.1007/s10517-012-1766-9. PMID 22977870.

4. 4

   Levdik NV, Knyazkin IV. Tumor-modifying effect of cardiogen peptide on M-1 sarcoma in senescent rats. Bull Exp Biol Med. 2009;148(3):433-436. Animal study, senescent rats with transplanted M-1 sarcoma. doi:10.1007/s10517-010-0730-9. PMID 20396706.

5. 5

   Khavinson VK, Linkova NS, Rudskoy AI, Petukhov MG. Feasibility of Transport of 26 Biologically Active Ultrashort Peptides via LAT and PEPT Family Transporters. Biomolecules. 2023;13(3):552. In silico docking only; the four ranks quoted here were read from the LAT1, LAT2, PEPT1 and PEPT2 tables. doi:10.3390/biom13030552. PMID 36979488.

6. 6

   Khavinson V, Linkova N, Kozhevnikova E, Dyatlova A, Petukhov M. Transport of Biologically Active Ultrashort Peptides Using POT and LAT Carriers. Int J Mol Sci. 2022;23(14):7733. Cited here only for the table row that joins the trade name to the sequence. doi:10.3390/ijms23147733. PMID 35887081.

7. 7

   Khavinson VK, Popovich IG, Linkova NS, Mironova ES, Ilina AR. Peptide Regulation of Gene Expression: A Systematic Review. Molecules. 2021;26(22):7053. The second independent source joining the trade name to the sequence. doi:10.3390/molecules26227053. PMID 34834147.

8. 8

   Chalisova NI, Lesniak VV, Balykina NA, Urt'eva SA, Urt'eva TA, Sukhonos IuA, Zhekalov AN. [The effect of the amino acids and cardiogen on the development of myocard tissue culture from young and old rats]. Adv Gerontol. 2009;22(3):409-413. Russian-language paper with an English abstract; ex vivo organotypic culture of rat myocardium. No DOI exists for this journal, so it is cited by PubMed identifier and only the abstract is verifiable. PMID 20210190.

9. 9

   Zakutskii AN, Chalisova NI, Ryzhak GA, Aniskina AI, Filippov SV, Zeziulin PN. [The tissue-specific effect of synthetic peptides-biologic regulators in organotypic tissues culture in young and old rats]. Adv Gerontol. 2006;19:93-96. Russian-language paper with an English abstract; ex vivo organotypic culture, rats aged 3 weeks and 18 months. Cited by PubMed identifier, abstract only. PMID 17152728.

10. 10

    Ryzhak AP, Chalisova NI, Lin'kova NS, Khalimov RI, Ryzhak GA, Zhekalov AN. [Polypeptides influence on tissue cell cultures regeneration of various age rats]. Adv Gerontol. 2015;28(1):97-103. Russian-language paper with an English abstract; the source for Chelohart being an extract of calf heart rather than a synthetic peptide. Cited by PubMed identifier, abstract only. PMID 26390619.

11. 11

    U.S. Food and Drug Administration. Drugs@FDA record for NDA 019414, CARDIOGEN-82, sponsor Bracco Diagnostics Inc., active ingredient rubidium chloride Rb-82, injectable, prescription, original approval 29 December 1989. Retrieved 4 August 2026. This is the radiopharmaceutical, not the peptide. accessdata.fda.gov, NDA 019414.

12. 12

    U.S. National Library of Medicine, DailyMed. CARDIOGEN-82 (rubidium chloride Rb-82) injection, solution, Bracco Diagnostics Inc. Label set id ee95aa18-9f2f-40eb-9b4c-583bea6f36bf, retrieved 4 August 2026. Cited only for the boxed warning on high radiation exposure with an incorrect eluent and on failure to follow the eluate testing protocol. dailymed.nlm.nih.gov, CardioGen-82 label.

Related pages

Around this page: five neighboring pages on thin evidence, animal data, and the documents that are supposed to settle things.

• PE 22-28 explained: another compound whose entire published record is small enough to count, and what we do with a page like that.
• Cardarine and the carcinogenicity data: what happens when the animal record is the whole record, told with the animals left in.
• What the bremelanotide approval covers: the shape of a real regulatory record, for comparison with a name collision.
• How to read a COA: identity, purity and content, and which of the three a certificate can actually settle.
• What a purity number leaves out: why a high percentage answers a narrower question than it appears to.

Where this one sits: the same tracing job is being done for the rest of this peptide family, one primary document at a time, and each finished page lands first in The Decadewise briefing.