Crystagen: what the studies show, status and safety
Summary
Crystagen is a synthetic chain of three amino acids from a St Petersburg institute, traded as an immune peptide. Exactly one paper in the medical literature carries its name, and it reports that the substance activates B cells but renews nothing in the ageing spleen. Every immune number quoted for it sits in the inventors' own patent application, including the one occasion on which people received it. It is approved as a medicine nowhere, and it is banned in sport at all times.
Key findings at a glance
- No trial in people has ever been registered. A registry query on 20 September 2026 returned nothing for Crystagen or Kristagen, as an intervention or as free text. The few hits for the sequence belong to unrelated studies [1].
- One record in the medical literature carries the name. A PubMed search returns a single paper. Europe PMC returns three, two of which are reviews from the same group [2].
- That one paper is partly negative. Comparing four short peptides in the ageing spleen, it reports that Crystagen activates the B-cell arm "however, the peptide doesn't cause cell renewal in spleen as it ages". The helper-cell effect belonged to two other peptides [3].
- The only reviewed paper testing the sequence is about skin. In skin from young rats it raised cell growth alongside three other peptides. In skin from old rats, a marked effect was reported only for a different peptide [4].
- The human numbers come from a patent, not a study. 38 older patients, ten days, and a claim that the immune profile normalised in 82 per cent of them against 56 per cent of a comparison group. No randomisation, no blinding, no registration, no review [5].
- It is a three-part molecule of 359.33 g/mol. Glu-Asp-Pro, formula C14H21N3O8, chemical record CID 145455337 [6].
- The trade name is in none of the reference databases. The American substance register returns nothing for it, the chemical database finds it only under the sequence, and it has no CAS number, no regulator code and no entry in the open knowledge base [6], [7], [8].
- In Russia it is a food supplement taken by mouth. Capsules and sublingual drops carry a fraction of a milligram of the peptide complex a day, and the seller's own page states it is not a medicine. The injectable form sold in the West is licensed nowhere [9], [10].
- Circulating injection schedules run 100 to 200 times higher. Pages aimed at users give daily injections for twelve weeks, against ten days in the patent example. The figures themselves are listed under doses [5], [11].
- Banned in sport around the clock under the class that catches every substance no health authority has approved [12].
What it is
Crystagen is a laboratory-made tripeptide: glutamic acid, aspartic acid and proline, joined in that order. Three amino acids and 359.33 g/mol make it a very small molecule, chemically hard to distinguish from an ordinary fragment of digested protein. The American medical library indexes the sequence as a peptide that stimulates the growth of skin cells [6], [13].
It comes from the St Petersburg Institute of Bioregulation and Gerontology, the group around Vladimir Khavinson. They patented it from 2006 onwards in the United States and in Europe [5], [14]. The group derived the sequence from Thymalin, a mixture extracted from cattle thymus, and later reported finding the tripeptide inside that mixture as a minor component [15], [16].
The words "bioregulator" and "cytogen" come from the same institute and its distributors. They appear in no pharmacopoeia, in no medicines classification and in no regulator's definition. A cytogen, in that house vocabulary, is a synthetic short peptide as opposed to an extracted tissue mixture [17], [18].
One thing is worth stating plainly at the outset. The name Crystagen exists in sales material rather than in chemistry. The chemical database has no entry under it, the American substance register has none, and the open knowledge base has none. Only the sequence Glu-Asp-Pro finds the molecule [6], [7], [8].
Quick facts
| Field | Value | Ref |
|---|---|---|
| Name | Crystagen; Кристаген in Russia; no international non-proprietary name | [9], [10] |
| Other names | Glu-Asp-Pro, EDP, H-Glu-Asp-Pro-OH, T-36 tripeptide | [6], [13] |
| Class | Synthetic tripeptide; "bioregulator" and "cytogen" are the maker's own words | [18] |
| Chain | Glu-Asp-Pro, three amino acids, proline at the end | [5], [6] |
| Formula and mass | C14H21N3O8, 359.33 g/mol for the free acid | [6] |
| Structure key | InChIKey CKOFNWCLWRYUHK-XHNCKOQMSA-N | [6] |
| Open point on the structure | The chemical record describes the proline ring as (2R), where the ordinary L-form would read (2S); unresolved | [6] |
| Chemical record | PubChem CID 145455337, reachable only under the sequence | [6] |
| CAS registry number | None found | [6] |
| Code with the American regulator | None; a control search in the same database works | [7] |
| Half-life | Never published, for any route or species | [2] |
| Registered studies in people | None | [1] |
| Approval status | Approved nowhere, for nothing | [7], [9] |
| Sold as | Capsules and sublingual drops in Russia; 20 mg vials of powder in the West | [9], [10], [19] |
| Originating group | St Petersburg Institute of Bioregulation and Gerontology | [5], [9] |
How it works
Nobody knows. No receptor has been identified for it, no binding has ever been measured, and every mechanism on offer is either a computer calculation or a description of what changed in a dish.
The group's own account runs as follows. The peptide would enter the cell through a carrier protein, reach the nucleus, settle on a particular stretch of DNA and change which genes are read. Each step deserves separate treatment, because the evidence behind them differs sharply.
- Settling on DNA. A docking calculation covering 19 short peptides puts Crystagen on a stretch written "agat". The same calculation gives that identical stretch to the two-part peptide Vilon, so a tripeptide and a dipeptide would recognise one site [20].
- Getting into the cell. A 2023 calculation ranked 26 short peptides at four carriers. Crystagen came 9th of 26 at LAT1 and 23rd of 26 at LAT2, and it is not among the peptides the authors single out as promising [21].
- Waking up immune cells. Peritoneal cells from ageing mice released more signalling molecules when the peptide was added, across 0.5 to 500 ng/ml. This appears only in the patent [5].
- Pushing thymus cells to divide. In a human thymus cell line the effect appeared only in the seventh passage and not in the first or fourth, and it did not rise with concentration [5].
There is a size problem underneath all of this. A chain of three amino acids carries very little information, far too little to pick out one gene from a whole genome. Glutamic acid, aspartic acid and proline also circulate freely in the blood as ordinary amino acids.
Calculated docking scores from one 2023 paper, not measurements: a more negative number means a better computed fit. No transport of this peptide has ever been measured. Source [21].
One boundary is worth drawing against the neighbouring molecules. Thymogen, a two-part peptide from the same series, is a licensed medicine in Russia and the group's own table labels it so. Crystagen carries no such label anywhere [18].
What the studies found
The record is one reviewed paper naming the substance, one reviewed paper testing the sequence, two calculation papers, a handful of reviews from the same institute, and a patent that contains almost everything else [2], [5].
Literature searches and a registry query run on 20 September 2026. The three Europe PMC records include two reviews by the originating group. Sources [1], [2], [3], [4].
The ageing spleen, the only paper that names it
Tissue culture Four short peptides were compared in spleen tissue as it aged: Vilon, Thymogen, Crystagen and one coded R-1. Crystagen activated the B-cell arm of the immune system. The paper then states that it causes no cell renewal in the spleen as the tissue ages [3].
Tissue culture The helper-cell activation in that experiment came from R-1 and Vilon. Thymogen worked through less cell death and more growth, a route Crystagen did not show. The abstract gives no group sizes, no concentrations and no p-values [3].
This is the only record in PubMed that carries the trade name, and its result for this peptide is the weakest of the four tested [2], [3].
Skin cells, where the sequence was actually tested
Animal tissue Five tripeptides were compared in cultured skin from young and old rats: T-32, T-33, T-34, T-36 and T-38. T-36 is Crystagen. In skin from young rats, every peptide except T-34 stimulated cell growth [4].
Animal tissue In skin from old rats, a marked stimulating effect was reported only for T-38, a different peptide. Nothing separate is reported for T-36 in the aged tissue. For a substance sold against ageing, that is an awkward result [4].
This is the only reviewed primary paper in which the sequence appears as a test substance with a stated outcome. Its subject is skin, not immunity, which is also how the American medical library describes the molecule [4], [13].
Immune findings, all of them from the patent
Patent document Thymus and spleen fragments from rats aged 24 months were grown in a dish. On day three the thymus pieces grew 24 per cent larger than controls at 0.5 ng/ml, and the spleen pieces 28 per cent larger at 10 ng/ml. By day seven the spleen effect was weaker [5].
Patent document In a human thymus cell line, growth rose only in the seventh passage: 523 ± 32 in the controls against 705 ± 46, 680 ± 41 and 795 ± 49 at rising concentrations, each below the 0.05 threshold. In the first and fourth passages there was no difference at all [5].
Patent document In irradiated rats given the peptide under the skin for ten days, the growth index reached 55 per cent against 26 per cent in undamaged animals. The value for the irradiated control group, the one that would carry the comparison, is given only as "high", without a figure [5].
None of these experiments has ever been published in a journal. A review from the same group repeats them and cites the patent as its source [5], [17].
The 38 patients, and why they are not a trial
Patent example, not a study 38 older patients aged 62 to 83, all with long exposure to low-dose radiation, received the peptide into the muscle for ten days. A further 32 received conventional treatment only. The share of T cells rose from 43.5 ± 2.3 per cent to 55.7 ± 2.6 per cent [5].
Patent example, not a study The patent reports that immune profiles normalised in 82 per cent of the treated group against 56 per cent of the comparison group. Helper-suppressor ratio moved from 1.1 to 1.4. The CD8 and B-cell figures did not move [5].
These numbers cannot be read as clinical evidence, and the reasons are structural. There was no randomisation, no blinding, no registration, no ethics record, no sample size calculation and no defined main outcome. The comparison group received nothing extra rather than a dummy treatment [1], [5].
The patent also never says how many of the 38 received which of its three amounts. One of those three is 500 times the other two, and it sits outside the range the same patent claims, so it cannot be relied on as printed [5].
One further detail shows how loosely these figures travel. A 2021 review by the same group repeats the 82 and 56 per cent as results of taking Crystagen by mouth. The patent describes injection into the muscle [5], [17].
Chromatin, a finding that points two ways
Patent document Lymphocytes from 95 donors aged 75 to 88 were grown with the peptide at 0.01 µg/ml. Exchanges between sister chromatids rose from 5.9 ± 0.2 to 8.4 ± 0.5 per cell, and the loosening of packed chromosome regions from 1.17 ± 0.05 to 2.32 ± 0.12 [5].
Such exchanges are an established marker of DNA damage and stress. The authors read the rise as a return to the level of young donors, 7.7 ± 0.4, which is one possible reading rather than the only one. No genotoxicity assessment to modern standards exists [5].
The group's own reporting of this experiment is not stable either. The patent describes a loosening of packed chromosome regions; the 2021 review describes the opposite, a tightening, citing the same source [5], [17].
What is still unknown
- Everything about how the body handles it. No half-life, no bioavailability, no distribution, no breakdown and no excretion data exist, for any route or species [2], [21].
- Whether it reaches a cell at all. Uptake has only ever been modelled, and in that model this peptide ranks in the middle to lower part of the field [21].
- Whether anything happens in a living immune system. Every immune figure comes from a dish, an animal or a patent example [5].
- Whether any of it can be reproduced. No laboratory outside the originating group and its co-authors has repeated a single finding [2].
- What is in the material sold. There is no pharmacopoeia monograph and no official reference substance, so a traded vial can be measured against nothing [7], [19].
- Which exact form of the molecule the record describes. The chemical database names the proline ring as (2R), where the ordinary building block of proteins would read (2S). Whether the entry or the expectation is wrong could not be settled, and no seller states a configuration at all [6], [19].
Side effects and safety
There is no safety record. That is not the same as being well tolerated: it means nothing has ever been collected that could be reported [1], [2].
- No adverse event reports exist. The American pharmacovigilance database returned no record for the name on 20 September 2026, and no case report was found in the literature [2], [7].
- The reassurance on offer comes from the inventors. The patent states that the preparation causes no side effects, complications or dependence. It covers 38 people over ten days, was never independently reviewed and appears in no register [5].
- The one laboratory signal points the other way. Exchanges between sister chromatids, a marker of DNA stress, rose from 5.9 ± 0.2 to 8.4 ± 0.5 per cell under the peptide. Nobody has followed this up [5].
- The toxicology is the inventors' own. Mice, rats and guinea pigs showed no toxic reactions up to 5 mg/kg as a single dose, and none over 90 days and six months at daily amounts. No test guideline, no raw data for the higher groups and no tissue findings are given [5].
- The injected form is the least controlled of all. 20 mg vials of freeze-dried powder are sold outside any pharmaceutical quality system, with sterility, endotoxin content and actual peptide content unknown [19].
Data gaps run right through the file. Nothing is known about how the body takes it up or clears it, about immune reactions to the substance itself, about cancer risk or about effects on reproduction. Nothing is known about pregnancy, breastfeeding, children or reduced liver and kidney function [2].
One gap deserves naming for what it is. The substance is sold for its supposed effect on the immune system. If it genuinely modulated immunity, that would matter for anyone with an autoimmune condition, on immune-suppressing treatment or after a transplant. Since no mechanism is established, nobody could say what to watch [5], [18].
The mismatch of amounts is a risk in itself. The longest exposure ever described in a person lasted ten days. The schedules circulating among users run 100 to 200 times higher and for twelve weeks, a range no data covers at all [5], [11].
Doses used in studies
The doses below are simply what the cited sources gave, listed so that the results can be understood. They are not advice on how to use anything, and the human figures come from a patent example, not from a reviewed study.
Every line in the table is a concentration in culture fluid, an amount given to an animal, a range claimed in a patent, or one uncontrolled patent example in people. Not one comes from a clinical study, and the routes and models cannot be compared with one another.
| Study | Model | Amount | Route | Duration | Ref |
|---|---|---|---|---|---|
| Patent example 2, thymus and spleen growth | Explants from rats aged 24 months | 0.5, 10 and 100 ng/ml | Added to the culture fluid | 3 and 7 days | [5] |
| Patent example 3, signalling molecules | Peritoneal cells from mice aged 19 to 20 months | 0.5 to 500 ng/ml | Added to the culture fluid | Not stated | [5] |
| Patent example 4, thymus cell line | Human VTEC2.H/S cells | 2, 20 and 200 ng/ml | Added to the culture fluid | Across 7 passages | [5] |
| Patent example 5, irradiation model | 40 male Wistar rats, 180 to 210 g | 0.5 µg per animal | Under the skin | 10 days | [5] |
| Patent example 6, chromosome markers | Human lymphocytes, donors aged 75 to 88 | 0.01 µg/ml | Added to the culture fluid | Until preparation | [5] |
| Patent example 7, toxicology | Mouse, rat, guinea pig | 1 to 5 mg/kg once; 1 µg/kg, 0.1 mg/kg and 1 mg/kg daily | Into the muscle | Once, 90 days, 6 months | [5] |
| Patent claim 4, a property right rather than a tested amount | Not tied to an experiment | 0.01 to 100 µg/kg of body weight | Parenteral, per claims 3 and 5 | "At least once a day", no length given | [5] |
| Patent example 8 in people, not a reviewed study | 38 patients aged 62 to 83 | 10.0 µg, and a third subgroup printed as 5.0 mg | Into the muscle | 10 days, once daily | [5] |
| Russian retail product, not a study | Adults, by mouth | 0.1 mg of the peptide complex per day | Sublingual drops or capsules | Pack sized for 10 days | [10] |
A concentration in a dish cannot be converted into an amount for a body. Between the smallest and the largest figure in this table lies a factor of many thousands, and no dose-finding study explains any of it [5].
Two lines need care. The 5.0 mg printed for one subgroup is 500 times the other two amounts in the same experiment. It also falls outside the patent's own claimed range, so it cannot be taken at face value [5]. The 0.01 to 100 µg/kg line is wording in a patent claim; it was never tested as a dose [5].
The schedules that circulate among users sit far outside all of this. Pages aimed at buyers of 20 mg vials give 1,000 micrograms daily in week one, rising to 2,000 for weeks three to twelve [11].
That is 100 to 200 times the only amount ever given to a person, and 10 to 20 times the top of the patent's claimed range. It comes from no study, because no study exists [5], [11].
Development and approval status
There is no development programme. Nothing was filed with a medicines regulator, no phase of testing ever began, and the substance appears in no regulator's substance file [1], [7].
A patent in 2006, and almost nothing since
- 2006Patent filed by the St Petersburg groupPriority filing in Russia; the application carries eight examples, including the one in 38 people, ref [5]
- 2011Patents granted in Europe and the United StatesEP granted 22 June, US granted 15 November; a property right, not an approval, refs [5], [14]
- 2012The sequence is tested in a reviewed journal, on skinFive tripeptides in rat skin culture; in old tissue a different peptide carried the effect, ref [4]
- 2014The only paper naming the substance appearsAgeing spleen: B cells activated, no cell renewal, ref [3]
- 2016 and 2023Two calculation papersDNA docking, then carrier docking placing the peptide mid-field, refs [20], [21]
- 2021Reviews repeat the patent figuresThe same group reports the patent numbers as results of oral use, ref [17]
- 20 September 2026Registers checkedNo approval in any market, no registered trial, no regulator substance code, refs [1], [7]
- 2006 the St Petersburg group files the patent family claiming an immune and anti-ageing peptide [5].
- 2011 the European and American patents are granted, five months apart [5], [14].
- 2012 the sequence is tested in a reviewed journal for the first and only time, in rat skin [4].
- 2014 the single paper carrying the trade name is published, with a partly negative result [3].
- 2016 and 2023 two calculation papers place the peptide among short peptides docked to DNA and to carriers [20], [21].
- 2021 reviews from the same group repeat the patent findings, in one case attributing them to use by mouth [17].
| Market | Status | Note | Ref |
|---|---|---|---|
| United States | Not approved | No substance code, no label, no adverse event file; never put forward for pharmacy compounding | [7], [22] |
| European Union | Not approved | No assessment report and no trial; the agency's own search could not be read on the day of checking | [1], [23] |
| Germany | Not approved | An unlicensed injectable preparation may not be placed on the market | [23] |
| United Kingdom | Not approved | The register returned only a page shell, so the finding rests on every other register | [23] |
| Australia | Not approved | Absent from the Poisons Standard, which is not permission | [24] |
| Canada | Not approved | The medicines database returns nothing, while a control query works | [25] |
| Switzerland | Not approved | No authorisation identified, and no register query was possible | [23] |
| Russia | Not approved | Capsules and sublingual drops are sold as a food supplement, which is not a medicine licence | [9], [10] |
Three rows need care. The Australian entry means the substance was never scheduled, not that it may be supplied. The European, British and Swiss entries rest on absence everywhere else, because those registers could not be read on the day [23], [24].
The Russian row is the one most often misread. A food supplement is cleared as a food and passes no test of whether it works. The institute's own page gives a food registration number, the state medicines register could not be queried, and no injectable form is licensed there either [9], [10], [23].
There is one further sales market worth noting. A Ukrainian trade page advertises the product for a long list of conditions, from immune deficiency to recovery after heart attack and stroke. None of that rests on a study, and such claims would be unlawful in German advertising law [26].
All registers were checked on 20 September 2026. The wider legal picture is under are peptides legal and FDA-approved peptides.
Anti-doping
Crystagen is prohibited at all times, in and out of competition. It is not named on the 2026 list. Class S0 covers "any pharmacological substance which is not addressed by any of the subsequent sections of the List and with no current approval by any governmental regulatory health authority for human therapeutic use" [12].
The classification needs no interpretation here, because there is no approval anywhere to argue about. The Russian food supplement status is not an approval for therapeutic use. Substances in S0 are specified substances, which affects how a sanction is measured but not the ban itself [12].
German anti-doping law reads differently. Its annex lists banned substances one by one and names no tripeptide of this kind, so possession is not an offence there. That is a gap in a closed list, not a permission [27].
No validated test for detecting it in a sample has been published, and detecting a three-part chain of ordinary amino acids would be demanding [2]. The full picture is at peptides banned in sport.
Compared with related peptides
| Peptide | Relation to Crystagen | What the checks behind this page found |
|---|---|---|
| Crystagen | — | Glu-Asp-Pro, 359.33 g/mol; one paper carries the name, no registered trial, approved nowhere [1], [3], [6] |
| Cortagen | The same sequence with one extra amino acid in front, assigned to the brain cortex | Ala-Glu-Asp-Pro; most literature hits for Glu-Asp-Pro are about this peptide instead [28] |
| Thymalin | The cattle thymus extract the sequence was derived from | A mixture rather than a single molecule; the group reports Crystagen inside it as a minor component [15], [16] |
| Thymogen | The two-part peptide of the same series, and the only licensed medicine among them | Glu-Trp; labelled "immunoprotector, medicinal product" in the group's own table, while Crystagen carries no such label [18] |
| Vilon | A two-part peptide of the same series, tested beside Crystagen | Lys-Glu; in the ageing spleen it activated helper cells where Crystagen did not [3] |
| Chonluten | A sibling tripeptide of the same laboratory | Glu-Asp-Gly; tested in the same rat skin experiment, where it was the only peptide without an effect in young tissue [4] |
The comparison is not about strength, because no head-to-head measure of effect exists. It shows where this peptide sits among its neighbours: same laboratory, same naming pattern, and the thinnest record of the series.
Two rows do real work. Cortagen explains why a literature search appears to return a substantial body of research when almost none of it concerns this molecule. Thymogen explains why "the Russians approved these peptides" is misleading: one of the series is a licensed medicine there, and it is not this one [18], [28].
Vilon and Chonluten were each tested alongside Crystagen in one of the two reviewed experiments. In both, the interesting result belonged to another peptide [3], [4].
Common misconceptions
- "Crystagen and Cortagen are the same peptide." They are not. Cortagen is Ala-Glu-Asp-Pro, one amino acid longer, and assigned to the brain cortex. Crystagen's sequence sits inside it, which is why searches confuse the two [28].
- "It has a US and a European patent, so it has been checked." A patent examines novelty and inventiveness, not effect, safety or the right to sell. No medicines regulator took part in granting either one [5], [14].
- "38 patients were treated successfully, so there is clinical evidence." That description sits inside the inventors' patent application. No randomisation, no blinding, no registration, no review, and no statement of who received which amount [1], [5].
- "It is registered in Russia." It is registered there as a food, in capsules and drops, and the seller's page says it is not a medicine. A food registration involves no test of whether a product works, and no injectable form is licensed [9], [10].
- "The capsules and the injection are the same product." The Russian supplement is taken by mouth in a fraction of a milligram a day. The Western vial holds 20 mg of powder for injection, and the schedules built around it are orders of magnitude larger. No data links the two [10], [11].
- "Thymalin and Thymogen results show that Crystagen works." Thymalin is an extracted mixture and Thymogen a different, two-part peptide that is a licensed medicine in Russia. Their results belong to them [16], [18].
- "Studies show it strengthens immunity." The one paper carrying the name reports B-cell activation and no cell renewal in the ageing spleen. Every quantified immune result is in the patent [3], [5].
- "The rise in chromosome exchanges proves rejuvenation." Those exchanges are a standard marker of DNA damage. Reading the rise as rejuvenation is the authors' interpretation, and no genotoxicity assessment exists [5].
- "It is not on the doping list, so it is allowed." It is not named, but class S0 catches every substance no health authority has approved. That is exactly its situation [12].
- "Crystalgen makes it." Crystalgen Inc. is an American laboratory-supplies company founded in 1994 and has nothing to do with the peptide. The spelling is one letter apart [29].
Frequently asked questions
What is Crystagen?
It is a synthetic chain of three amino acids, glutamic acid, aspartic acid and proline, written Glu-Asp-Pro or EDP and coded T-36. It weighs 359.33 g/mol, which makes it a very small molecule. It comes from a St Petersburg institute that calls such peptides bioregulators and cytogens, house words with no standing in any official classification.
Has Crystagen ever been tested in people?
Not in any registered or reviewed study. A registry search on 20 September 2026 returned nothing under either spelling of the name. The one occasion on which people received it is described as an example inside the inventors' patent application, covering 38 older patients over ten days, with no randomisation, no blinding and no independent review.
What did the only published paper naming Crystagen find?
It found a partial result. In a comparison of four short peptides in the ageing spleen, Crystagen activated the B-cell arm of the immune system but produced no cell renewal as the tissue aged. The helper-cell effect in that same experiment belonged to two other peptides, not to this one.
Is Crystagen the same thing as Cortagen?
No, and the mix-up is easy to make. Cortagen is a four-part peptide, Ala-Glu-Asp-Pro, with one extra alanine at the front, and it is assigned to the brain cortex. Crystagen has three parts. Searches for the sequence Glu-Asp-Pro return mostly Cortagen papers, which say nothing about Crystagen.
Is Crystagen approved as a medicine anywhere?
No, in no country. Register checks on 20 September 2026 found no approval in the United States, the European Union, Germany, the United Kingdom, Australia, Canada, Switzerland or Russia. In Russia it is sold as a capsule and sublingual food supplement, a status that involves no test of whether the product works.
Does a US patent mean Crystagen has been approved or proven?
No. A granted patent is a property right over an invention. It is examined for novelty and inventiveness, not for whether the substance works, whether it is safe, or whether it may be sold. Neither the American nor the European medicines regulator took any part in granting it.
How much Crystagen has ever been given to a person?
Once, and not in a study. The only occasion ever described is a patent example: one injection into the muscle a day for ten days, in 38 older patients. The amount is listed in the doses table on this page. The injection schedules that circulate online run 100 to 200 times higher and over twelve times as long, and come from no study.
Is Crystagen banned in sport?
Yes, at all times, in and out of competition. The name is not printed on the 2026 prohibited list. Class S0 covers any pharmacological substance that no government health authority has approved for human use, and that describes Crystagen exactly.
What are the side effects of Crystagen?
Nobody knows, because no study has ever collected them. There is no reviewed toxicology, no adverse event report in the American database and no case report in the literature. The inventors' own cell work did show a rise in a marker of DNA stress, which has never been followed up.
Why does this page list study concentrations for Crystagen but no dose?
Because the figures that exist are amounts added to fluid in a dish, doses given to rats and mice, a range claimed in a patent, and one uncontrolled patent example. None came from a clinical study, and none can be converted into an amount for a person.
Sources
- ClinicalTrials.gov, API v2, queried 20 September 2026. query.intr and query.term for "Crystagen" and "Kristagen" each returned a total count of zero. "Glu-Asp-Pro" returns two unrelated studies (NCT05214872, NCT05400733) and "EDP peptide" returns one unrelated cancer study (NCT03095079). https://clinicaltrials.gov/api/v2/studies
- PubMed via NCBI E-utilities and Europe PMC REST search, queried 20 September 2026. "crystagen" returns one PubMed record and three Europe PMC records; "kristagen" returns none in either. The MeSH supplementary concept for the sequence is attached to one PubMed record. No pharmacokinetic study, no case report and no doping-control detection method was found for the substance.
- Chervyakova NA, Linkova NS, Chalisova NI, Kontsevaya EA, Trofimova SV, Khavinson VKh (2014). Molecular aspects of immunoprotective activity of peptides in spleen during the ageing process. Advances in Gerontology 27(1):224-228. PMID 28976144. In Russian with English abstract, read 20 September 2026: "Crystagen also activates B-cells of the immune system; however, the peptide doesn't cause cell renewal in spleen as it ages." The full text is not openly accessible; the abstract gives no group sizes, concentrations or p-values.
- Voicekhovskaya MA, Chalisova NI, Kontsevaya EA, Ryzhak GA (2012). Effect of bioregulatory tripeptides on the culture of skin cells from young and old rats. Bulletin of Experimental Biology and Medicine 152(3):357-359. PMID 22803085. DOI 10.1007/s10517-012-1527-9. Abstract read: T-32 (Glu-Asp-Ala), T-33 (Glu-Asp-Arg), T-34 (Glu-Asp-Gly), T-36 (Glu-Asp-Pro) and T-38 (Lys-Glu-Asp) compared in skin explants from rats aged 3 weeks and 18 months. In young tissue all peptides except T-34 stimulated growth; in old tissue a marked effect is reported for T-38 only.
- Khavinson VK, Grigoriev EI, Malinin VV, Ryzhak GA. Peptide substance revealing an immunogeroprotective effect, pharmaceutical composition on its base and the method of its application. US 8,057,810 B2; priority RU 2006118488 of 30 May 2006, granted 15 November 2011. Full text read 20 September 2026: claim 1 "A peptide of the formula H-Glu-Asp-Pro-OH", claim 4 with the range 0.01 to 100 µg/kg, claims 3 and 5 restricting the composition to parenteral use, examples 1 to 8 and tables 1 to 4, including the synthesis and 97.19 per cent purity, the organ cultures, the thymus cell line across seven passages, the irradiation model, the lymphocyte cytogenetics, the toxicology, and example 8 in 38 patients aged 62 to 83 with the statement that the preparation "does not exert any side effects, complications or drug addiction". https://patents.google.com/patent/US8057810B2/en
- PubChem, compound CID 145455337, queried 20 September 2026 for formula C14H21N3O8, mass 359.33 g/mol, InChIKey CKOFNWCLWRYUHK-XHNCKOQMSA-N, SMILES and IUPAC name. The compound is reachable through the name "Glu-Asp-Pro" but not through "crystagen", which returns no hit. No CAS registry number was obtainable; repeated synonym queries failed with server-busy responses, so the absence of a CAS number is recorded as not established rather than as a confirmed nil. The stored IUPAC name describes the proline ring as (2R), where an N-acylated L-proline would be expected to read (2S); this discrepancy could not be resolved. https://pubchem.ncbi.nlm.nih.gov/compound/145455337
- United States register checks, 20 September 2026. FDA Global Substance Registration System: exact name searches for "CRYSTAGEN" and "GLUTAMYL-ASPARTYL-PROLINE" each returned zero records, while the control search for "EPITALON" returned two, so the substance has no regulator code. openFDA drug label and adverse event endpoints: not found for either name, against 18 control hits for semaglutide. DailyMed web services v2: no entries, database state 18 September 2026, against a control hit for Ozempic. https://gsrs.ncats.nih.gov
- Wikidata, wbsearchentities in English for "Crystagen" and "Kristagen", queried 20 September 2026: no entity exists under either spelling. https://www.wikidata.org/w/api.php
- St Petersburg Institute of Bioregulation and Gerontology, section "Цитогены", retrieved 20 September 2026. The institute's own page lists Кристаген® with a state food registration certificate number and assigns it to immune function, alongside sister products of the same series. Manufacturer page, cited as evidence of how the product is presented, never as a scientific source. https://gerontology.ru
- Russian trade page for the sublingual product Кристаген лингвал, retrieved 20 September 2026. Declares the contents as peptide complex AC-6 made of proline, glutamic acid and aspartic acid, gives an average of 0.1 mg of the complex per daily portion, recommends 5 to 6 drops three to four times a day before meals, sizes the pack for ten days, and states in its own words that the preparation is not a medicinal product. Trade record, cited only as evidence of what is sold. Not linked.
- User-facing dosing page for 20 mg vials of the substance, retrieved 20 September 2026. Sets out a twelve-week course of daily injections rising to 2,000 µg a day, while stating on the same page that the substance is not approved by the American or European regulator for human administration and that no rigorous large-scale clinical trials have been completed. Trade record, cited only as evidence that such schedules circulate. Not linked.
- World Anti-Doping Agency. The 2026 Prohibited List, in force 1 January 2026, class S0 Non-Approved Substances: "Any pharmacological substance which is not addressed by any of the subsequent sections of the List and with no current approval by any governmental regulatory health authority for human therapeutic use … is prohibited at all times." Full text checked 20 September 2026: no match for "crystagen", "kristagen", "glu-asp-pro" or the Russian spelling; BPC-157 appears in the same text as a control. https://www.wada-ama.org/en/prohibited-list
- United States National Library of Medicine, MeSH supplementary concept "glutamyl-aspartyl-proline", retrieved through E-utilities 20 September 2026. Entry terms Glu-Asp-Pro and T-36 tripeptide; heading mapped to Oligopeptides; described as "a synthetic peptide that stimulates proliferation of skin cells". The trade name is not an entry term, and a MeSH search for it returns nothing.
- Khavinson VK, Grigoriev EI, Malinin VV, Ryzhak GA. Peptide substance revealing an immunogeroprotective effect, pharmaceutical composition on its base and the method of its application. EP 2032595 B1, granted 22 June 2011, applicant OOO "SIA Peptides". Claim 1, grant date, designated states and family status checked 20 September 2026. The family also covers RU 2301074 C1, EA 010575 B1, KR 101382947 B1, UA 91135 C2 and WO 2007/139435 A1, whose full texts were not read. https://patents.google.com/patent/EP2032595B1/en
- Khavinson VK, Zhurkovich IK, Ryzhak GA, Mironova ES, Kovrov NG (2021). Identification of short peptides: optimization of target therapeutic properties of the thymus medicine. Molecular Medicine 19(3):32-37. DOI 10.29296/24999490-2021-03-05. Abstract read 20 September 2026: "the short peptides KE and EDP were found to be a part (minor components) of the thymus polypeptide complex (thymalin)". No independent confirmation of this finding was located.
- Khavinson VKh (2020). Medicinal peptide preparations: past, present, future. Clinical Medicine 98(3):165-177. DOI 10.30629/0023-2149-2020-98-3-165-177. Read in Russian 20 September 2026: the tripeptide Glu-Asp-Pro was detected in Thymalin by liquid chromatography with mass spectrometry and given the name Кристаген.
- Khavinson VK, Linkova NS, Trofimova SV, Kuznik BI, et al. (2021). The Use of Thymalin for Immunocorrection and Molecular Aspects of Biological Activity. Biology Bulletin Reviews. PMCID PMC8365293. DOI 10.1134/S2079086421040046. Full text read 20 September 2026: it names EDP (crystagen) as a component identified in Thymalin, repeats the 82 and 56 per cent figures as results of oral administration although the patent describes injection into the muscle, and states that the peptide promotes tightening of chromatin in the lymphocytes of elderly people, the opposite direction to the patent, citing the same 2006 source.
- Khavinson V, Linkova N, Kozhevnikova E, Dyatlova A, Petukhov M (2022). Transport of Biologically Active Ultrashort Peptides Using POT and LAT Carriers. International Journal of Molecular Sciences 23(14):7733. PMID 35887081. PMCID PMC9323678. DOI 10.3390/ijms23147733. Full text read: the cytogen table carries the row "Crystagen (EDP) −20.22 Immunoprotector", whose cited evidence is reference 174, the US patent above rather than a study, and the row "Thymogen (EW) −24.72 Immunoprotector, medicinal product".
- Western research-chemical listings for the substance in 20 mg vials, retrieved 20 September 2026 from sellers in the United States, the United Kingdom and Canada. They give the sequence as Glu-Asp-Pro, state purity above 99 per cent without independent testing, and carry wording restricting the material to research use and excluding human or animal consumption. Trade records, cited only as evidence of what is sold and how it is labelled. Not linked.
- Khavinson VKh, Lin'kova NS, Tarnovskaya SI (2016). Short Peptides Regulate Gene Expression. Bulletin of Experimental Biology and Medicine 162(2):288-292. PMID 27909961. DOI 10.1007/s10517-016-3596-7. Molecular docking of 19 short peptides to DNA: "Peptides KE and EDP bind agat sequence". No structural determination and no binding experiment accompanies it.
- Khavinson VK, Linkova NS, Rudskoy AI, Petukhov MG (2023). Feasibility of Transport of 26 Biologically Active Ultrashort Peptides via LAT and PEPT Family Transporters. Biomolecules 13(3):552. PMID 36979488. PMCID PMC10046148. DOI 10.3390/biom13030552. Full text and tables 2, 3, 5 and 7 read: EDP scores −32.60 at LAT1 (rank 9 of 26), −18.75 at LAT2 (rank 23 of 26), −17.29 at PEPT1 (rank 18 of 26) and −5.40 at PEPT2 (rank 9 of 26), all by calculation. EDP is not among the peptides the authors highlight as favourable. No transport measurement and no pharmacokinetic data appear here or anywhere else searched.
- United States Food and Drug Administration (14 May 2026). Bulk Drug Substances Nominated for Use in Compounding Under Section 503A of the Federal Food, Drug, and Cosmetic Act. Checked in full text 20 September 2026: neither the trade name nor Glu-Asp-Pro nor EDP appears in any of the three categories, while the dipeptide alanyl-L-glutamine appears in category 1 as a control that short peptides are indeed listed there. https://www.fda.gov/media/94155/download
- Registers that could not be read on 20 September 2026, documented here rather than treated as negative findings: the European Medicines Agency full-text search returned HTTP 401; the British medicines register returned only the JavaScript shell of its search page; the Swiss overview page returned HTTP 404; the Russian state medicines register returned page scaffolding without result rows, including for the control query with a genuinely registered medicine; the Russian consumer-protection register of supplement certificates was unreachable; and the Australian goods register could not be queried programmatically. The findings for those markets therefore rest on the absence of any entry in every register that could be read.
- Commonwealth of Australia. Therapeutic Goods (Poisons Standard—June 2026) Instrument 2026, version of 27 May 2026, full text checked 20 September 2026: no entry for the trade name, for Glu-Asp-Pro or for "glutamyl", and no general heading covering peptides. Controls found in the same text: alanylglutamine in schedule 4, pralmorelin and the GHRP entries.
- Health Canada, Drug Product Database API, brand-name queries for "crystagen" and "kristagen" on 20 September 2026: empty results, while the control query for Ozempic returns DIN 02471469. https://health-products.canada.ca/api/drug/drugproduct/
- Ukrainian trade page for the product, retrieved 20 September 2026. Advertises it for immune deficiency, rehabilitation after heart attack and stroke and cancer prevention, names a peptide complex as the contents and gives no registration number. Trade record, cited only as evidence of how the product is advertised. Not linked.
- Germany, annex to section 2(3) of the Anti-Doping Act, read in full 20 September 2026: no entry for the trade name, for Glu-Asp-Pro or for "glutamyl". The peptide hormone section lists erythropoiesis-stimulating agents, chorionic gonadotrophin and luteinising hormone, corticotrophins, growth hormone with its fragments and releasing factors, and growth factors including IGF-1, MGF and TB-500, all by name, with no catch-all position covering a thymus tripeptide.
- Kazakova TB, Barabanova SV, Khavinson VKh, Glushikhina MS, Parkhomenko EP, Malinin VV, Korneva EA (2002). In vitro effect of short peptides on expression of interleukin-2 gene in splenocytes. Bulletin of Experimental Biology and Medicine 133(6):614-616. PMID 12447482. DOI 10.1023/a:1020210615148. Abstract read: it names "Cortagen (Ala-Glu-Asp-Pro)", confirming that the four-part sequence belongs to Cortagen and not to Crystagen. A companion paper, PMID 12420072, uses the same naming.
- Crystalgen Inc., company website, retrieved 20 September 2026. An American manufacturer and distributor of laboratory consumables founded in 1994 and based in New York, with no connection to the peptide. Recorded only because the two names differ by one letter. https://crystalgen.com
Cite this page
The searches and register checks behind this page were run on 20 September 2026. For a substance with almost no scientific record, the date on which those searches came back nearly empty is part of the finding. Three things could still move it: a Russian-language report outside the international databases, an independent replication of any single result, or a change in the anti-doping classification.
myPeptides Research & Editing. (2026). Crystagen: what the studies show, status and safety. Version 1.0, 20 September 2026. myPeptides Peptide Register. Retrieved from https://mypep.app/peptides/crystagen
How pages in this register are compiled and graded is described under methodology; the full register is at peptides.
| Version | Date | Change |
|---|---|---|
| 1.0 | 2026-09-20 | Initial publication |
Last verified: 20 September 2026. Next review: on any registered trial, on any independent replication of the spleen or skin findings, or on any change to the anti-doping classification.