Cardiogen A four-residue peptide assigned to the heart, and what happens when a compound’s name belongs to somebody else

Section 01A name with too many owners

Most monographs in this series open with a molecule and ask what is known about it. This one has to start a step earlier, because the ordinary way of finding out — type the name into a database and read what comes back — fails in an unusually complete way. It does not fail by returning nothing. It fails by returning the wrong thing, confidently, from every source at once.

Type Cardiogen into PubChem, the standard public registry of chemical structures, and it returns levocarnitine — the small molecule your muscles use to burn fat — because “Cardiogen” is a registered carnitine trade name. Type cardiogen into ChEMBL, the reference database of drug-like molecules, and it returns rubidium chloride: specifically CardioGen-82, an approved radiopharmaceutical that a hospital elutes to image blood flow through the heart muscle. Neither is a peptide. Neither is the subject. And a reader who stopped at either would come away certain they had found it.

The confusion does not stop at chemical databases. As a word, cardiogen is the opening of cardiogenic — as in cardiogenic shock, a phrase that appears in more than twenty thousand papers — and in Hungarian and German medical writing “cardiogen” simply is the adjective “cardiogenic,” a complete and correctly spelled word. It names two national institutions: the French Filière CARDIOGEN, a rare-cardiac-disease network, and the Szeged CardioGen Registry in Hungary, both of which stamp their name on the acknowledgements of papers about cardiomyopathy and inherited arrhythmia that have nothing to do with any peptide. It is a company in a venture-capital survey. And in Iran it is a brand of losartan, an ordinary blood-pressure drug.

The compound’s own laboratory code is no safer. Written AEDR, it is, in twenty-six of the twenty-eight indexed papers that use it, the annual effective dose rate — a unit from health physics, attached to surveys of radioactivity in soil, granite and ceramic tiles.

The consequence is measurable, and it is worth stating precisely because it sets the terms for everything that follows. A PubMed search for Cardiogen returns eighty records; four of them concern this peptide. A full-text search of open-access articles returns a hundred and twelve; two concern this peptide. The sequence itself, Ala-Glu-Asp-Arg, appears in exactly one indexed record. There is no curated index entry for the molecule at all — no MeSH concept, no CAS number, no ChEMBL row — of the kind that, for most compounds, tells a librarian “this paper is about that thing.”

Two very different situations produce a near-empty result like this. One is a compound nobody has studied: a catalogue number with a price and no science behind it. The other is a compound that has been studied under names the database does not connect to the label. The distinction is the whole question for a reader trying to decide what to believe, and the search result itself does not tell you which one you are looking at. Cardiogen is the second kind. The evidence exists; it is small, it is almost entirely the work of one research group, and it is filed under a code that mostly means something else. Finding it required building a search that asks not what a document must contain to be about this peptide, but what it must not contain — the method the Apparatus describes.

Section 02The molecule

Strip the names away and the object is simple. Cardiogen is L-alanyl-L-glutamyl-L-aspartyl-L-arginine: four amino acids joined in a row by three peptide bonds, a free amino group at one end and a free carboxyl at the other. All four residues are among the twenty specified by the genetic code, and all four are in the natural L configuration. There is nothing exotic in it.

Its molecular formula is C₁₈H₃₁N₇O₉ and its mass 489.5 daltons, confirmed against PubChem CID 11583989. That record is telling in what it lacks: no CAS registry number, no UNII, and a single synonym that is itself just a database accession. It is a bare structure deposition — the chemical equivalent of a passport photo with no name attached. Its more famous sibling Epitalon, by contrast, carries a CAS number, a UNII and a full synonym list. Two molecules one residue apart are curated at opposite depths, and the difference is a fair measure of how much attention each has had.

The two acidic side chains (from glutamate and aspartate) and the one basic side chain (from arginine) roughly balance, leaving a small, highly water-loving molecule. That last property matters more than it looks. The mechanism the compound’s investigators propose requires it to get inside a cell and act on the material in the nucleus; a charged, hydrophilic peptide does not cross a fatty cell membrane on its own. Something would have to carry it. Who or what that might be is the most testable question in this whole document, and Section 09 takes it up.

Section 03A different kind of object

Before going further it is necessary to say what class of thing Cardiogen is claimed to be, because a reader who arrives from any other monograph in this series — from a growth-hormone secretagogue, say, or a melanocortin agonist — will bring a mental model that does not apply here, and will misread everything that follows.

Cardiogen is a peptide bioregulator: one of a family of very short peptides developed since the 1970s by Vladimir Khavinson and colleagues at what is now the St Petersburg Institute of Bioregulation and Gerontology. The word “bioregulator” is the group’s own, and it carries a specific and unusual claim. An ordinary peptide drug — a hormone, a cytokine, a receptor ligand — works by binding a receptor on the outside of a cell and setting off a signal within. A bioregulator, as the class is described, does something else: it is said to enter the cell, reach the nucleus, and act on DNA and chromatin directly, tuning which genes are read. There is no named receptor in the account, no binding constant, and no dose– response curve of the kind the rest of this series reports. The peptides are said to act at vanishingly low concentrations — nanomolar and below, and in the cardiac experiments described later, at 10⁻¹² molar.

The family comes in two generations, and the relationship between them is the key to the whole enterprise. First came the tissue extracts — complex mixtures of peptides pulled from the organs of young animals, one preparation per organ. Then, from those extracts, came the synthetic short peptides: two to four amino acids each, one peptide singled out per extract and made in a factory. The second generation was derived from the first, and the compound in this monograph is a second-generation peptide. Its first-generation parent is a cardiac extract, and Section 05 is about that pairing.

One thing has to be said plainly, in this section rather than buried in a late caveat: “bioregulator” is not a pharmacological category recognised outside this literature. It does not appear in standard pharmacology texts or drug classifications as a mechanism. It is a description coined and used by the school that produces these compounds. That does not make the underlying observations wrong — peptides genuinely can enter cells, and short peptides genuinely can bind DNA in a dish — but a reader should know that the framing itself, the idea that this is a distinct mode of action with its own name, comes from the same source as the compounds, and has not been ratified by the wider field.

Section 04One stem, many organs

Cardiogen is not a lone invention. It sits in a designed series in which each member is built by adding residues to a shared three-residue stem, and each member is assigned to a different organ — the pineal gland, the cortex, the bronchi, cartilage, the heart.

The claim embedded in that table is a strong one. Adding a single glycine to the stem is said to turn a cartilage agent into a pineal agent; a leucine makes a lung agent; an arginine, the compound in this monograph, a heart agent. Nothing in ordinary molecular pharmacology would lead one to expect a single terminal residue to redirect a molecule between organs like that, and Section 10 examines what the evidence for it actually is. The short version is that it is much thinner than the confidence with which it is stated — and that it is not nothing.

The naming scheme also has a mundane consequence that shaped the making of this document. Because the shorter stem is contained inside the longer ones, a search for the subject drags in its relatives, and the subject’s own sequence contains two other named members of the family: Cartalax (Ala-Glu-Asp) sits at its front, and Pinealon (Glu-Asp-Arg) at its back. Cardiogen is the only compound in the class whose sequence spells out two of its own siblings. Telling the subject apart from its family, in a body of text that names them all in the same sentences, was most of the technical work behind the corpus — and the Apparatus describes how it was done.

Section 05The natural counterpart

Every synthetic peptide in this class has a first-generation parent — a tissue extract it is said to have come out of — and naming that parent, and being honest about the relationship, is where the interesting science in these compounds lives. It is also the thing commercial material about them most often skips.

Cardiogen’s parent is Korapept, a peptide complex prepared from the heart muscle of cattle. According to the group’s own account (Khavinson, 2020), the tetrapeptide Ala-Glu-Asp-Arg was detected inside Korapept by chromatography–mass spectrometry and then reproduced synthetically. That word — detected, isolated — carries weight, and it is worth dwelling on, because it is the opposite of what happened with the family’s most famous member.

The distinction is this. A peptide can be isolated from its extract: physically found in the mixture, its sequence read off an instrument, and then made in a factory. Or it can be designed from the extract: someone measures the overall amino-acid content of the mixture and composes a short peptide from the most abundant residues, without ever showing that this exact sequence was present in the original. Those are different epistemic acts wearing the same phrase, “the active fragment of the extract.” Epitalon — the pineal sibling — was designed, from the bulk composition of the pineal extract. Cardiogen, on the group’s telling, was isolated. The claim is stronger, and it is worth flagging that it rests on the same laboratory’s mass-spectrometry data, which this reviewer could not independently inspect; the review that reports it is peer-reviewed, but the underlying spectra are not in it.

There is a name worth keeping separate here, because conflating the two is the easy error. Chelohart is also a cardiac peptide preparation, also from calf heart, and it is sold commercially in the same broad family. But Chelohart belongs to a different product line — a parapharmaceutical supplement — and it is not the preparation the group names as the source of Cardiogen. Korapept is the one in the discovery chain; Chelohart is a commercial sibling. A reader comparing labels will meet both, and they are not the same thing.

As for status: neither Korapept nor Cardiogen is a registered medicine. In 2020 the group described Korapept as being in clinical study rather than approved, and there is no independent confirmation of that stage. Cardiogen the synthetic peptide has never been the subject of a registered clinical trial in any country — a point Section 07 returns to.

Section 06The discovery, as history

The class has an unusual origin, and the reader is entitled to it told straight, with the documented parts separated from the marketing.

The documented core is this. In the 1970s, at what was then a military-medical institute in Leningrad, Khavinson and Vyacheslav Morozov began studying peptide fractions extracted from animal organs — the “cytomedines,” as the extracts were called. The programme dates itself to 1973–74 by its own later reviews, and it produced, over the following decades, first the tissue extracts and then the synthetic short peptides drawn from them. The thymus preparation, Thymalin, was registered as a Soviet medicine in the early 1980s. This much is on the record and is internally consistent.

Around that core circulates a more dramatic story: that the work was secret military research, declassified only after the fall of the Soviet Union. That framing appears mostly in vendor material and popular accounts rather than in the scientific record, and it sits awkwardly with the fact that one of the preparations was a registered, marketed medicine by 1982. It is repeated here only to be marked as what it is — the compound’s legend, not its documented history.

Cardiogen’s own paper trail begins later and in an unexpected place: not a journal but a patent office. The tetrapeptide Ala-Glu-Asp-Arg, and its use for “recovery of myocardium function,” are the substance of Russian patent RU 2255756, filed in June 2004 and published in 2005, with an international family that includes a granted United States patent. That patent, rather than any peer-reviewed paper, is the origin document for the compound’s central claim — a fact that becomes important in Part Four. Commercially, the peptide is sold today as a capsule in the “Cytogens” line associated with the St Petersburg institute and its manufacturing partners, and as a research vial by Western chemical suppliers, most of whose online “information” about it is vendor-written copy rather than literature.

Section 07The one cardiac claim, and where it lives

The reason anyone assigns Cardiogen to the heart is a single result, and it is worth stating exactly, along with exactly where it comes from. In a rat model of heart attack — the coronary artery tied off to starve part of the heart muscle — giving the peptide is reported to have cut mortality roughly threefold compared with untreated animals, to have shrunk the zone of dead tissue, to have preserved the muscle’s glycogen stores, and to have protected its mitochondria (Khavinson et al., 2022).

If that result is real and reproducible, it is a serious finding. The difficulty is where it lives. The primary source for it is not a peer-reviewed experimental paper; it is the 2004 patent. The number reaches the open literature through a 2022 review, which restates it and cites the earlier work. A patent is written to establish a claim, not to survive peer review; its methods are not reported in the detail a journal would demand, and no independent group has repeated the experiment. So the strongest cardiac claim about this compound rests on the weakest kind of source in the whole evidence base.

The figure sets the sources side by side, and the shape it makes is the central fact about Cardiogen’s evidence: the boldest claim rests on the softest source, and the firmest source makes the smallest claim. It is not that the work is fraudulent or that the peptide does nothing; it is that the pyramid is upside down.

Section 08The mechanism paper with no heart in it

The most-cited attempt to explain how Cardiogen might protect the heart is a 2012 study, and it contains no cardiac tissue at all. It was done in mouse embryonic fibroblasts — connective-tissue cells, not heart-muscle cells. In those cells the peptide raised the amounts of structural proteins: the cytoskeletal proteins actin, tubulin and vimentin by two- to fivefold, and the nuclear-envelope proteins lamin A and lamin C by two- to threefold (Khavinson et al., 2012). The paper’s own logic is that these changes stimulate cell proliferation and reduce cell death, and that this — and here the paper reaches back to the patent — explains “the previously reported cardioprotective activity” of the tetrapeptide.

Read carefully, that is a mechanism inferred in one cell type for an effect demonstrated in another. The effect (survival after infarction) was seen in the heart; the mechanism (more cytoskeleton, more lamin) was measured in fibroblasts; and the two have never been shown in the same tissue. That is not a fatal objection — cytoskeletal and nuclear-matrix proteins matter in heart cells too — but it is a real gap, and it is the kind of gap that a single confirmatory experiment in cardiomyocytes would close, and that has not been run.

There is exactly one experiment on cardiac tissue itself. In 2009, in organotypic cultures — small explants of rat heart kept alive in a dish — the peptide at 10⁻¹² molar increased cell proliferation in explants from both young and old animals and lowered the amount of p53, a protein that pushes cells toward arrest and death (Chalisova et al., 2009). It is a suggestive result, consistent with the fibroblast findings. It is also a culture of a tissue fragment, not a working heart, and it is a single study from the originating group.

Section 09Signalling and expression: not a receptor

This is the section that most separates Cardiogen from the rest of the series, and the class contract this monograph is written under requires it to be explicit. The compounds in every other monograph here act at a receptor; this one is not claimed to.

What stands in for receptor pharmacology, in this compound’s case, is a short list of observations about DNA and protein. The peptide has been shown to bind histones — the spool proteins that DNA wraps around — in a test using labelled wheat histones, alongside several of its siblings, apparently at specific sites in the histones’ flexible tails (Fedoreyeva et al., 2013). In the heart-explant work it lowered p53; in fibroblasts it raised structural proteins. The proposed picture is of a peptide that slips into the nucleus and nudges gene expression by touching chromatin directly. Whether that picture is right is genuinely open: histone binding in a tube is a long way from gene regulation in a living cell, and the class as a whole leans heavily on this kind of indirect, correlational evidence.

Which leaves the question the molecule’s own chemistry forces — how does a charged, water-loving peptide get inside the cell in the first place? The group’s answer is a 2023 modelling study that docked twenty-six of these short peptides, Cardiogen among them, into the structures of the cell’s known peptide and amino-acid transporters (Khavinson et al., 2023). Cardiogen scored well against the amino-acid carrier LAT1 and poorly against one of the peptide carriers, PEPT2.

Two cautions belong with that figure, and they are the honest heart of the mechanism question. First, these are docking scores — computed estimates of how well a molecule fits a binding pocket — not measurements of actual transport. No experiment has yet shown any of these peptides being carried into a cell by any of these proteins. Second, there is a size problem the modelling itself surfaces: the PEPT transporters are built to carry di- and tripeptides, chains of two or three residues, and Cardiogen is a tetrapeptide, at the edge of or beyond what those carriers normally accept. The route of entry — the single most testable claim in the whole account — remains, after all this, a calculation in search of an experiment.

Section 10Tissue-specificity, and where it breaks

The family’s central and strangest claim is that one added residue sends each peptide to a different organ. For the reader to weigh it, the evidence has to be laid out rather than asserted — including the parts that cut against it.

The direct evidence for tissue-specificity is essentially one experiment. In 2006, four of these peptides — cardiogen, bronchogen, prostamax and pancragen — were applied to explant cultures of their four assigned organs (heart, lung, prostate, pancreas), and each was reported to stimulate its own organ’s tissue more than the others (Zakutskii et al., 2006). That is a genuine head-to-head design, and it is the strongest support the specificity claim has. It is also a single study, from the originating group, in organotypic culture rather than in living animals.

Set against it are two results from the same body of work that complicate the tidy picture. When cardiogen was tested outside its assigned organ — on aging human prostate fibroblasts — it was active there too, raising several signalling proteins; and in that experiment a different peptide, not the prostate-assigned one, was the most active of the set (Kheifets et al., 2010). And in a rat sarcoma model, cardiogen slowed tumour growth — an effect having nothing to do with the heart — apparently by acting on the tumour’s blood supply rather than on the tumour cells directly (Levdik & Knyazkin, 2009). A peptide that is active in prostate tissue and against a sarcoma is not behaving like a heart-specific agent.

The comparison this section most wants — cardiogen set directly against a sibling in the heart, on a functional endpoint — does not exist in the literature. No such experiment has been done. In its absence, the honest summary is that the specificity claim rests on one organ-matched culture panel, is undercut by the compound’s own activity elsewhere, and awaits the kind of study that could actually test it. Meanwhile the vascular half of the “cardiovascular” story in the group’s own reviews is carried not by Cardiogen but by a different sibling, the tripeptide KED, on the endothelium — a reminder that even within the heart-and-vessels domain, the compounds are not interchangeable.

Section 11The shape of the evidence

Gather the pieces and the picture is coherent, if not the one the product pages paint. Cardiogen is a real, well-defined molecule with a small, internally consistent body of preclinical work behind it. That work reports a striking result in a rat heart-attack model, a plausible cell-level mechanism, and a specificity claim that fits the family’s theory. It is the provenance of the evidence, not its direction, that a careful reader has to weigh.

What that evidence is: one full-text paper genuinely about the compound; a handful of short primary reports, most in Russian, most from a single institute; one striking cardiac result whose primary source is a patent; a mechanism established in the wrong cell type; a single experiment in cardiac tissue; a route of entry that is so far a calculation; and, for the whole class, zero registered clinical trials and no independent replication of the headline finding. There is no human data of any kind — not a trial, not a case series.

Stated that way, this reads like a thin monograph, and it is worth being precise about why it is not. A thin monograph is one where the author could not find the evidence. Here the evidence has been found, read and weighed; the thinness is the finding, and it is a finding with a shape. A compound does not become well-supported by being sold on four continents, by carrying a patent, or by having a mechanism story; it becomes well-supported when independent groups can reproduce its central result and when that result has been tested in the species it is meant to help. Cardiogen has neither, and saying so is not a verdict against the molecule — it is a description of exactly how far the evidence currently reaches.

Three specific things would move it, and naming them is more useful than a grade. A functional cardiac experiment — the mechanism shown in heart cells, not fibroblasts — run by a group with no commercial stake. A direct measurement of transport, replacing the docking calculation with an actual uptake experiment, since the tetrapeptide-versus-transporter mismatch is a real puzzle worth resolving either way. And a single registered trial, which would convert the compound from a research chemical with a patent into something whose effects in people could begin to be known. Until then, what is true of Cardiogen is what is true of its whole class: an interesting idea, a devoted laboratory, and an evidence base whose most honest description includes its own edges.

Section 12References

1. Chalisova NI, Lesniak VV, Balykina NA, Urt'eva SA, Urt'eva TA, Sukhonos IuA et al.. [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):409-13. PMID 20210190
2. Fedoreyeva LI, Smirnova TA, Kolomijtseva GY, Khavinson VKh, Vanyushin BF. Interaction of short peptides with FITC-labeled wheat histones and their complexes with deoxyribooligonucleotides. Biochemistry (Mosc) 2013;78(2):166-75. PMID 23581987 · doi
3. 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). PMID 36611900 · doi · PMC9818427
4. 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). PMID 35887081 · doi · PMC9323678
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). PMID 36979488 · doi · PMC10046148
6. Khavinson VK, Popovich IG, Linkova NS, Mironova ES, Ilina AR. Peptide Regulation of Gene Expression: A Systematic Review. Molecules 2021;26(22). PMID 34834147 · doi · PMC8619776
7. Khavinson VKh, Lin'kova NS, Polyakova VO, Kvetnoy IM, Benberin VV, D'yakonov MM et al.. Tetrapeptide H-Ala-Glu-Asp-Arg-OH stimulates expression of cytoskeletal and nuclear matrix proteins. Bull Exp Biol Med 2012;153(4):559-62. PMID 22977870 · doi
8. Kheĭfets OV, Poliakova VO, Kvetnoĭ IM. [Peptidergic regulation of the expression of signal factors of fibroblast differentiation in the human prostate gland in cell aging]. Adv Gerontol 2010;23(1):68-70. PMID 20586252
9. Kraskovskaya N, Linkova N, Sakhenberg E, Krieger D, Polyakova V, Medvedev D et al.. Short Peptides Protect Fibroblast-Derived Induced Neurons from Age-Related Changes. Int J Mol Sci 2024;25(21). PMID 39518916 · doi · PMC11546785
10. Levdik NV, Knyazkin IV. Tumor-modifying effect of cardiogen peptide on M-1 sarcoma in senescent rats. Bull Exp Biol Med 2009;148(3):433-6. PMID 20396706 · doi
11. Sakhenberg E, Linkova N, Kraskovskaya N, Krieger D, Polyakova V, Medvedev D et al.. The Influence of Short Peptides on Cell Senescence and Neuronal Differentiation. Curr Issues Mol Biol 2025;47(9). PMID 41020860 · doi · PMC12468822
12. Zakutskiĭ 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;19:93-6. PMID 17152728
13. Khavinson VKh, Grigoriev EI, Ryzhak GA, Ryadnova IYu. Peptide compound recovering function of myocardium. Russian Federation patent RU 2255756 C1. Priority 22 June 2004, published 10 July 2005; assignee St Petersburg Institute of Bioregulation and Gerontology. The claim names the tetrapeptide Ala-Glu-Asp-Arg as the active agent for recovery of myocardial function. This patent, not a peer-reviewed paper, is the primary source of the experimental-infarction result restated in the 2022 review. link
14. Khavinson VKh, Grigoriev EI, Malinin VV, Ryzhak GA. Peptide substance restoring myocardium function. International family of RU 2255756: US 7,662,789 B2 (granted 16 February 2010; assignor St Petersburg public organisation, later Access Bioscience CJSC / Peptid Products), WO 2006/001728 A1, EP 1758923. Note the digit transposition common on vendor pages: the US number is 7,662,789, not 7,622,789. link
15. National Center for Biotechnology Information. PubChem Compound Summary for CID 11583989, H-Ala-Glu-Asp-Arg-OH. C18H31N7O9; 489.5 g/mol; InChIKey QXQARLZWUIQZPX-NAKRPEOUSA-N. The record carries no CAS number, no UNII and no name synonym other than “SCHEMBL3194515”: it is a bare structure deposition. Retrieved 3 August 2026. link
16. National Center for Biotechnology Information. PubChem Compound Summary for CID 10917, Levocarnitine. The record PubChem returns for a name search on “Cardiogen”: (−)-carnitine, C7H15NO3, 161.20 g/mol, CAS 541-15-1, because “Cardiogen” is a registered carnitine trade name. Cited to document the name collision, not the peptide. Retrieved 3 August 2026. link
17. EMBL-EBI. ChEMBL molecule CHEMBL1200727 / CHEMBL2110557, rubidium chloride Rb-82. The record ChEMBL returns for a search on “cardiogen”: the approved radiopharmaceutical CardioGen-82 (rubidium Rb-82 chloride), max phase 4, first approval 1989, black-box warning. There is no ChEMBL entry for the peptide. Retrieved 3 August 2026. link
18. US National Library of Medicine. ClinicalTrials.gov, searched for Cardiogen, AEDR, Ala-Glu-Asp-Arg and Khavinson peptide bioregulator. Zero registered studies, searched 3 August 2026. The four “cardiogen” hits in the registry all concern the CardioGen-82 rubidium generator or unrelated cardiovascular genetics, not the peptide. link
19. Khavinson VKh. Lekarstvennye peptidnye preparaty: proshloe, nastoyashchee, budushchee. Klinicheskaya Meditsina 2020;98(3):165–177. In Russian. The source for the statement that the tetrapeptide Ala-Glu-Asp-Arg (Cardiogen) was DETECTED within the cardiac peptide complex Korapept by chromatography–mass spectrometry, i.e. isolated rather than designed; and that Korapept was, as of 2020, in clinical study rather than registered as a medicine. link
20. US National Library of Medicine. PubMed and PMC, indexed-surface counts for Cardiogen, AEDR and Ala-Glu-Asp-Arg. Measured 3 August 2026: “Cardiogen”[All Fields] returned 80 PubMed and 112 PMC-body records, of which four and two respectively concern this peptide; “Ala-Glu-Asp-Arg”[All Fields] returned one record; “AEDR”[All Fields] returned 28, of which 26 use the abbreviation for annual effective dose rate. No MeSH Supplementary Concept Record exists for the molecule. link
21. Peptide Products / TD Peptid Bio. Cardiogen (Cytogens line), product page. Vendor-authored material, cited once as commerce rather than as evidence. The label declares a “peptide complex AKS-K (peptide-linked amino acids: glutamic, aspartic, alanine, arginine)” — the constituent amino acids of Ala-Glu-Asp-Arg without stating the sequence. Retrieved 3 August 2026. link

Section 13How this document was assembled

This monograph was built from the Radix Therapeutic Peptide Research Library (project 05) and a targeted harvest of the external literature, by a pipeline whose stages are listed below. The whole of the technical difficulty was in one place — deciding which documents are about this compound — so that is what this section describes.

The identity problem, and the gate built for it

As Part One set out, the string “Cardiogen” and the code “AEDR” are shared with at least seven unrelated things: a carnitine salt, a rubidium–82 PET generator, the adjective “cardiogenic” (a complete word in Hungarian and German), two national cardiac-disease registries, a biotech company, an Iranian losartan brand, and the annual effective dose rate of health physics. On top of that, the compound’s sequence contains two of its own siblings — Cartalax at the front, Pinealon at the back — and is a common enough motif to appear inside ordinary proteins.

A search built around what a document must contain cannot separate these, because the false matches share all the obvious vocabulary. So the gate was inverted and made local. It normalises the text (rejoining words a line break has split, so that cardiogen- at a line end is not read as a boundary); strips the constructions that merely contain the subject’s name — the rubidium generator, the registries, the adjective forms, the sibling sequences, and any protein-length run of residues; then judges each remaining mention of the name or code in its own surrounding window, admitting it only where that window carries short-peptide vocabulary and carries none of the disqualifying senses. A crucial detail: the corroborating vocabulary deliberately excludes cardiac terms, because the dominant false positive here is itself cardiac — a heart-disease registry — so “heart” and “myocardium” confirm the noise as readily as the signal.

The gate was broken on purpose before it was trusted: a suite of forty-nine worked examples drawn from real records, each of the eight design decisions reverted in turn to confirm that the suite then failed. Two of those decisions were only discovered by auditing what the gate first admitted — among the “subject” documents it initially accepted were a rubidium-generator paper, a venture-capital survey and a French registry, all admitted because the first version of the gate treated the bare trade name as proof. That is the correction the windowed test above encodes.

Two corpora, reported as two numbers

The pipeline distinguishes a subject corpus — documents about this tetrapeptide — from a class corpus — documents about the bioregulator family, its siblings, its extracts and its proposed mechanism, which inform the mandatory background sections but are never cited as though they were about Cardiogen. The two are reported separately and deliberately, because the class literature is roughly a hundred times the size of the subject literature and is written by the same people in the same words; conflating them is the single error this compound most invites. In the curated project–05 library the subject count is zero: not one of the seventy-five thousand indexed passages names this peptide, a fact confirmed three independent ways before it was believed.

Section 14Evidence handling

Findings are labelled in the sentence that reports them by the study that produced them — patent claim, review, fibroblast culture, tissue explant, docking calculation, animal model. Where a result exists only in a patent, that is stated, because a patent claim and a peer-reviewed result are not the same kind of evidence and the difference is load-bearing for this compound in particular. Animal and in-vitro results are never phrased to imply a human outcome; there are no human results to phrase.

Three features of this evidence base are flagged wherever they bear on a claim. It is single-source: nearly every primary study shares an author or an institution, and the institute that produces the compound also publishes much of the research and edits some of the journals it appears in. It is patent-anchored: the central efficacy claim originates in a patent rather than a paper. And it is unreplicated: no independent group has reproduced the headline cardiac result, and no registered clinical trial of the compound exists anywhere. None of these is a reason to dismiss the work; each is a reason to hold it at the confidence its provenance supports, which is the whole purpose of a document like this one.