Skip to content
PeptideAI
Human trialssleeplongevityhormone support

Epitalon (circadian use)

A four-amino-acid pineal peptide taken in short annual courses to pull a flattened, age-related melatonin rhythm back into a normal night-time peak.

Also known as Epithalon, AEDG, Ala-Glu-Asp-Gly, epithalamin tetrapeptide

Human trials — Studied in people, typically early phase or small — promising rather than proven.

For the circadian claim specifically there is real human data — Russian studies in older adults reported restored nocturnal melatonin and improved sleep after peptide courses — but those trials are small, decades old, largely unblinded, published in Russian, and almost all from the group that invented the compound. Independent replication outside Russia is essentially absent. The mechanism claims are far ahead of the evidence, even though the human circadian signal is more than nothing.

How it works

Epitalon is the synthetic distillate of Epithalamin, a bovine pineal extract developed by Vladimir Khavinson's group in St Petersburg. The circadian claim rests on human work in older adults showing that courses of the peptide raised night-time melatonin and sharpened the day-night melatonin gradient, alongside changes in cortisol rhythm. Khavinson's proposed mechanism is that short peptides pass into the nucleus and bind specific promoter regions, changing transcription of pineal and clock-related genes — an interesting hypothesis with real spectroscopic work behind it, but not something independently confirmed outside his institute. The famous telomerase-activation finding comes from cultured human fibroblasts and does not speak to the sleep effect at all.

Targets: Pineal gland / melatonin synthesis, AANAT and pineal clock gene expression (proposed), Telomerase (hTERT) in cultured cells, Gene promoter regions via direct DNA binding (proposed)

Dosing

ProtocolDoseFrequencyRoute
Standard courseEvening dosing is the convention when the target is the melatonin rhythm, though nothing has formally compared morning to evening.5 mg – 10 mgonce daily for 10 to 20 consecutive dayssubcutaneous
Epithalamin-style clinical courseGiven as a short intensive block, repeated once or twice a year.5 mg – 10 mgonce daily for 5 to 10 daysintramuscular
Low-dose intranasalEvening.500 mcg – 1 mgonce daily for 10 to 20 daysintranasal
  • · 5 mg daily for 10 days is the most common single protocol. Some people run 10 mg every other day for 20 days to cover the same total.
  • · Mirrors the Russian geriatric protocols, which used repeated short courses over years rather than continuous dosing.
  • · Intranasal absorption for Epitalon has never been quantified. This route is a community convenience, not a validated equivalent to injection.

Cycling

Epitalon is explicitly a pulsed compound, not a daily one. The Khavinson pattern is a 10 to 20 day course repeated every 4 to 6 months — two or three courses a year. Continuous year-round use has no basis in any of the source literature.

Work out your exact syringe units →

Pharmacology

Half-life
Cleared from plasma within minutes — short peptides like this are chewed up by peptidases almost immediately — yet the intended effect is a slow rhythm shift over a 10 to 20 day course, so plasma half-life is close to irrelevant here.
Onset
Sleep and rhythm changes are typically reported in the second week of a course rather than the first night; the Russian melatonin data were measured after multi-day courses.
Routes
subcutaneous, intramuscular, intranasal
Molecule
Synthetic tetrapeptide
Sequence length
4 amino acids
Molecular weight
390.3 Da

Handling

Diluent
Bacteriostatic water
Typical mix
2 or 3 mL
Vial sizes
10, 20, 50 mg
Lyophilised
Room temperature is tolerated for transit; refrigerate at 2-8 C and freeze for long-term storage.
Reconstituted
Refrigerated, use within about 30 days — which usually covers an entire course anyway.
Light sensitive
Yes — keep it out of the light

Intranasal — usable, with a caveat

Same as the parent record: the trial data used injection, and nasal dosing is a practice-based extrapolation with no absorption study behind it.

Mixing

A 50 mg vial in 5 mL gives 10 mg/mL, so a 5 mg dose is 50 units on an insulin syringe. Because the whole vial is normally used inside a two-week course, reconstitution volume is mostly about convenience.

Side effects

  • commonInjection-site redness or itching— The most frequently reported effect by a wide margin.
  • commonDrowsiness or earlier sleep onset— Often the intended effect rather than a side effect.
  • uncommonHeadache in the first few days of a course— Usually settles without changing the dose.
  • uncommonShifted sleep timing or early waking— A rhythm-shifting compound can move the rhythm in an unwanted direction.
  • rareMild nausea— Reported occasionally, no consistent pattern.

Do not use if

  • Active or recent malignancy - telomerase activation is the exact biological argument against use here, and Khavinson's own oncology claims run in the opposite direction without the trial data to settle it.
  • Pregnancy and breastfeeding - no data at all.
  • Anyone on a strictly timed shift-work or clinical sleep schedule should be aware this compound is intended to move circadian timing.

Combining it

  • synergydsip — Different timescales — Epitalon works on the rhythm over weeks, DSIP works on tonight.
  • synergypinealon-circadian — The standard Khavinson pairing: a pineal bioregulator with a neuronal one, run in the same short course.
  • synergythymalin — Epitalon plus Thymalin is the classic Khavinson combination used in the long-term Russian geriatric cohorts.
  • cautionmk-677 — Both alter sleep architecture and neuroendocrine rhythm; running them together makes it impossible to attribute any effect.

What to monitor

  • · Track sleep timing and sleep onset latency across a course — the effect you are looking for is when you get sleepy, not just how long you sleep.
  • · If you have access to it, salivary or urinary melatonin (6-sulphatoxymelatonin) before and after a course is the only direct readout of the claimed mechanism.
  • · No routine bloodwork is established.

Legal status

Not approved in the US or EU; sold as a research chemical. Epithalamin, the parent pineal extract, has had registered medical status in Russia.

References

  • Korkushko et al. 2007, Epithalamin and melatonin rhythm in elderly subjects (trial)
  • Khavinson & Morozov, long-term geroprotective studies of Epithalamin/Epitalon (review)
  • Anisimov et al., epithalamin and lifespan in rodent models (preclinical)
  • Khavinson et al., AEDG peptide telomerase activation in human fibroblast culture (preclinical)

Mechanism in depth

There are two mechanistic stories here and they should be kept strictly separate, because vendors deliberately blur them. The circadian story is the one with human data. Korkushko, Khavinson, Shatilo and colleagues ran courses of pineal peptide preparation in elderly subjects and in old monkeys and reported that the flattened, blunted melatonin curve of ageing was pushed back toward a normal shape - higher nocturnal melatonin, a sharper day-night gradient. Their 2004 paper in Bulletin of Experimental Biology and Medicine and the 2007 Advances in Gerontology paper are the primary sources. The proposed proximate mechanism is an effect on pineal function itself, plausibly through AANAT, the rate-limiting enzyme in melatonin synthesis, and through clock gene expression in pinealocytes. None of that intermediate biochemistry has been demonstrated directly in humans. The epigenetic story is the one with no human data. Khavinson's model is that very short peptides pass through the cell membrane, enter the nucleus and interact directly with specific sequences in gene promoters, changing transcription in a tissue-specific way. This is not pure hand-waving - Fedoreyeva, Kireev, Khavinson and Vanyushin published fluorescence work in Biochemistry (Moscow) showing short labelled peptides entering the nucleus of HeLa cells, and there is a body of spectroscopy and modelling work on peptide-DNA interaction from the same group. What is missing is independent replication outside St Petersburg and any demonstration that this nuclear entry is what produces the melatonin effect. A tetrapeptide binding a promoter with meaningful sequence specificity is a large claim: four residues is not enough information content to specify a unique genomic site, and the group's own answer to this is that the peptides act on chromatin structure rather than as sequence-specific transcription factors in the classical sense. The telomerase result is a third thing entirely and it is the one most often misused. Khavinson, Bondarev and Butyugov reported in 2003 and 2004 that Epitalon induced telomerase activity and telomere elongation in cultured human somatic cells, allowing them to exceed the Hayflick limit. That is a cell-culture finding. It says nothing about sleep, nothing about melatonin, and nothing about what happens in a person. It is also the single best reason to think carefully before using this compound if you have any oncological history. The timescale tells you what kind of drug this is. Plasma exposure is minutes, the course is ten to twenty days, the effect is reported in week two, and the protocol repeats twice a year. Nothing about that is consistent with receptor occupancy. Whatever Epitalon does, it does by triggering a change that then persists on its own.

What usually goes wrong

The most common failure is running it continuously. Epitalon is a pulsed compound in every source protocol - 10 to 20 days, twice or three times a year. People buy a 50 mg vial, feel something, and keep going for months. There is no evidence base for that, the entire Russian literature is built on short repeated courses, and the honest answer about what continuous daily AEDG does to a human is that nobody knows. The second failure is expecting a sleeping pill. Epitalon does not make you sleepy on night one. If it works at all it works by shifting when you get sleepy, and it does that over two weeks. People who take one 5 mg injection, notice nothing, and double the dose have misunderstood what the compound is. The third failure is the attribution problem. Almost nobody runs Epitalon alone. It arrives inside a stack with Thymalin, Pinealon, melatonin, a GH secretagogue and a new sleep schedule, all started in the same week. If you want to know whether Epitalon does anything for you, run it alone with a measured aMT6s before and after. Otherwise you are buying a story. The fourth is the oncology question, and it is the only one that could actually hurt you. The compound's most cited property is telomerase activation. That is a mechanism you do not want running unopposed if you have an occult malignancy, and the counter-argument - that Khavinson's group reported reduced tumour incidence in rodents - is from the same institute that produced the telomerase paper and has never been independently replicated. If you have any personal or strong family cancer history, this compound is a genuinely poor risk-reward proposition and there is no sleep benefit large enough to justify it. The smaller ones: reconstituting a 50 mg vial into 2 mL and then discovering a 5 mg dose is 20 units of a very concentrated solution, which is fine but wastes the vial if you only wanted one course; and assuming that a nasal spray delivers anything at all, when the intranasal route for AEDG has never been quantified.

Bloodwork worth running

MarkerWhenWhy it matters
Urinary 6-sulphatoxymelatonin (aMT6s), first morning void, or a salivary dim-light melatonin onset seriesBaseline in the week before a course, then again in the week after the course ends. Repeat under the same conditions - same wake time, same light exposure, same collection window - or the comparison is worthless.This is the only direct readout of the mechanism Epitalon is actually sold on. If the compound does what the Russian data claim, aMT6s in the overnight or first-morning urine goes up and the day-night gradient widens. If it does not move, nothing else in the marketing matters.Act if: No safety threshold. This is an efficacy readout: a nocturnal aMT6s that has not moved after a full 10 to 20 day course is a reason to stop paying for the compound.
Morning (08:00) serum cortisol and, if available, a diurnal salivary cortisol curveBaseline and post-course, alongside the melatonin sampling.The Russian geriatric work reported changes in cortisol rhythm alongside the melatonin changes, and cortisol and melatonin rhythms are coupled. A flattened cortisol curve is also one of the better objective markers of the circadian disruption people take this for.Act if: Morning cortisol below roughly 140 nmol/L (5 mcg/dL) on repeat testing warrants a proper adrenal workup, not another course.
IGF-1Baseline before any stack that includes a secretagogue or growth hormone peptide.Not an Epitalon effect. Included because IGF-1 is the marker most often being tracked in parallel by the same people running Khavinson protocols, and because a rising IGF-1 from a stacked compound will be misattributed to Epitalon otherwise. Know your baseline so you can tell the compounds apart.Act if: None specific to Epitalon.
Age-appropriate cancer screening, plus PSA in men over 50 and a documented breast and cervical screening status in womenBefore your first course, and stay current with routine screening thereafter.This is the one that matters. The headline preclinical finding for Epitalon is telomerase activation and telomere elongation in human somatic cells. Telomerase reactivation is a hallmark of the great majority of human cancers. Nobody has shown that Epitalon causes cancer, and Khavinson's own oncology claims run the other way, but a compound whose most famous property is switching telomerase on deserves a clean screening baseline before you take it and not after.Act if: Any active or recent malignancy, or any unexplained abnormal screening result, means do not run the course until it is resolved.
Full blood count and comprehensive metabolic panelBaseline, then annually if you are running two or three courses a year.Baseline hygiene rather than an Epitalon-specific concern. Short peptide courses have not been associated with haematological or hepatic changes, and having a clean baseline means an unrelated abnormality later does not get blamed on the peptide.Act if: None specific to Epitalon.

Pharmacokinetics

Metabolism
Sequential exopeptidase cleavage to Ala, Glu, Asp and Gly. There is no phase I or phase II metabolism to speak of - the molecule is simply taken apart into ordinary amino acids.
Elimination
As free amino acids into the general metabolic pool. No renal or biliary excretion of intact peptide.

Receptor targets

  • No identified receptor — None established

    Epitalon is not a ligand in the ordinary sense. There is no cloned receptor, no binding assay and no reported Kd for AEDG at anything. The proposed mechanism is intracellular and transcriptional, which is why this section is short.

  • Pineal gland melatonin output — Not a binding interaction

    The only endpoint with human data. Courses of pineal peptide preparation raised nocturnal melatonin and restored the day-night gradient in elderly subjects and old monkeys (Korkushko et al., 2004 and 2007). Whether AEDG specifically or the crude extract drove this varies by study.

  • Telomerase (hTERT) in cultured human fibroblasts — Not characterised

    Induced telomerase activity and telomere elongation in vitro, permitting cells to divide past the normal limit (Khavinson et al., Bull Exp Biol Med 2003 and 2004). In vitro only. Relevant here mainly as a contraindication signal, not as a benefit.

  • Gene promoter regions and chromatin (proposed direct interaction) — Not characterised

    Short labelled peptides have been shown entering cell nuclei (Fedoreyeva et al., Biochemistry (Moscow) 2011), and the Khavinson group reports sequence-dependent effects on transcription. Independent confirmation outside the originating institute is essentially absent.

  • AANAT and pineal clock gene expression — Not characterised

    The most commonly proposed intermediate step between the peptide and the melatonin result. Not directly demonstrated in humans. Treat as hypothesis.

Trials

  • Korkushko, Khavinson, Shatilo and Magdich - effect of peptide preparation Epithalamin on the circadian rhythm of pineal melatonin-producing function in elderly people (Bulletin of Experimental Biology and Medicine 2004) Investigator-initiated clinical study, not phase-labelled · 2004

    Circadian melatonin production in elderly subjects. Reported restoration of nocturnal melatonin output toward a younger profile after a course of pineal peptide preparation. This is the primary human source for the entire circadian claim.

  • Korkushko, Lapin, Goncharova, Khavinson, Shatilo et al. - normalising effect of pineal gland peptides on the daily melatonin rhythm in old monkeys and elderly people (Advances in Gerontology 2007) Investigator-initiated, not phase-labelled · 2007

    Daily melatonin rhythm in aged rhesus monkeys and elderly humans. The cross-species version of the same finding, and the study most often cited as the Epitalon circadian evidence.

  • Labunets, Butenko, Magdich, Korkushko, Khavinson and Shatilo - effect of epithalamin on the circadian relationship between thymic endocrine function and melatonin production in elderly people (Bulletin of Experimental Biology and Medicine 2004) Investigator-initiated, not phase-labelled · 2004

    Coupling of thymic endocrine output and melatonin rhythm in elderly subjects after epithalamin. Relevant because it is the study underpinning the standard Epitalon plus Thymalin pairing.

  • Korkushko, Khavinson, Shatilo and Antonyk-Sheglova - peptide geroprotector from the pituitary gland inhibits rapid aging of elderly people (Bulletin of Experimental Biology and Medicine 2011) Investigator-initiated long-term follow-up, not phase-labelled · 2011

    Rate-of-ageing measures in elderly subjects given repeated peptide courses. Included for completeness as part of the same programme; note this one concerns a pituitary peptide preparation, not Epitalon specifically, and it is routinely miscited as Epitalon evidence.

What to expect, and when

Days 1 to 3: nothing detectable in most reports. Occasional early-course headache and injection-site redness, which settle. Days 4 to 7: the first thing people notice, when they notice anything, is a shift in sleepiness timing rather than sleep quality - getting tired earlier in the evening than they used to. That is exactly the direction a restored melatonin peak would produce, and it is also exactly what expectation produces, which is why the urinary aMT6s measurement matters. Days 8 to 20: this is where the reported effect consolidates and where the Russian melatonin measurements were made. Sleep onset latency shortens, sleep timing moves earlier. If nothing has happened by day 20, the course has failed. After the course: the claim, and the reason the protocol is pulsed, is that the change persists for months without further dosing. That is the strongest and least tested part of the whole story. Nobody has followed aMT6s in a Western cohort for six months after a course. Next course: 4 to 6 months later, two or three courses a year. Running them closer together has no basis in any source protocol. A note on direction: this compound is intended to move circadian timing, and it can move it the wrong way. Early waking or a phase advance you did not want is a real reported outcome and it is a reason to stop the course, not to push through it.

Stacking and comparisons

Epitalon is a course, not a daily habit, and that shapes every stack. With Thymalin: the original Khavinson pairing and the one with actual human data behind it. Labunets and colleagues published on the thymic-pineal coupling specifically, and the long-term Russian geriatric cohorts used both. If you are running the Khavinson protocol as designed, this is the combination it was designed as. With Pinealon: the other half of the standard pineal course. Epitalon for the gland, Pinealon for the neurons, both run over the same 10 to 20 days. Be clear-eyed that Pinealon is the weaker half - it has no human circadian data at all - so if either one is doing something, attribution is impossible. With DSIP: genuinely complementary and the most practical pairing on this list. Epitalon is trying to move where your melatonin peak sits over the course of two weeks. DSIP is trying to make tonight better. Run DSIP nightly through the Epitalon course if you want, then keep DSIP as an as-needed tool once the course ends. With low-dose melatonin: the arguments cut both ways. If Epitalon works by restoring endogenous melatonin, adding exogenous melatonin muddies the readout completely and you will never know which one did anything. If you are going to measure aMT6s, stop exogenous melatonin for at least a week before baseline and do not take it during the course. If you are not measuring anything, the combination is harmless but pointless. With MK-677 or any GH secretagogue: run them at different times of year. Both alter sleep architecture and neuroendocrine rhythm, both are being taken for overlapping reasons, and running them together guarantees you learn nothing about either. With bright morning light and a fixed wake time: this is the actual synergy nobody sells. Morning light exposure is the strongest circadian entrainment signal available to a human, it is free, and it has vastly more evidence than any peptide in this class. If you are running an Epitalon course without also fixing your light exposure and wake time, you are optimising the wrong variable.

Against exogenous melatonin: melatonin is oral, costs almost nothing, has hundreds of controlled trials, and works immediately for phase shifting at doses of 0.3 to 0.5 mg taken several hours before target bedtime. Epitalon proposes to restore your own melatonin production instead. That is a more elegant idea and a far weaker evidence base. For a fifty-year-old whose only problem is that their melatonin amplitude has flattened, low-dose melatonin is the rational first move and Epitalon is the experiment you run afterwards. Against DSIP: different timescale, different target. DSIP acts tonight and has more polysomnography behind it. Epitalon acts over weeks and has the melatonin data. Neither substitutes for the other. Against Pinealon: Epitalon has human circadian trials, however flawed. Pinealon has none. If you are choosing one of the two Khavinson peptides for a sleep goal, it is Epitalon, and it is not close. Against other telomere or longevity products: the telomerase result is real in cell culture and irrelevant to sleep. If longevity rather than sleep is the actual goal, be honest about that, because the risk calculation around telomerase activation is completely different when the benefit you are buying is speculative. Against morning bright light plus a fixed wake time: this comparison is uncomfortable and it should be made. Timed light exposure has stronger, more replicated and more mechanistically clear effects on human circadian phase than any peptide in this category, and it is free. Epitalon is worth considering after that has been implemented properly, not instead of it. Within the Khavinson bioregulator family: Epitalon is the flagship and the only one with a circadian human readout. The rest of the family - Pinealon, Cortexin, Thymalin and the cytogens - are largely riding on its reputation.

Rough cost

$12–$40/month. Epitalon is priced per course, not per month. A 50 mg vial covers a full 10-day course at 5 mg daily and typically sells in the range of a mid-double-digit to low-triple-digit dollar figure on the research-chemical market. Two to three courses a year amortises to roughly 12 to 40 dollars a month. If you buy 10 mg vials instead of 50 mg you will pay substantially more per milligram. These are indicative retail observations, not sourced or surveyed pricing.

Genuinely uncertain

  • Much of the human circadian evidence used Epithalamin, the crude bovine pineal extract, rather than synthetic AEDG. These are not the same compound and the literature routinely conflates them. Which studies used which is often unclear from the abstracts.
  • Participant numbers, blinding and randomisation status for the Korkushko studies were not extracted in this session. They are generally described as small and largely unblinded. Verify individually before treating them as strong evidence.
  • The Korkushko 2004 and 2007 papers are published in Russian-language journals and I verified their existence, titles, authors and citation details through PubMed but did not read the full texts. Effect sizes are not characterised here.
  • No independent replication of the melatonin finding outside the originating institute could be identified.
  • No pharmacokinetic data of any kind exists for AEDG in humans - no tmax, no bioavailability, no clearance, no brain penetration measurement.
  • Whether oral or intranasal AEDG delivers any intact peptide at all is unknown and there is a reasonable argument that oral delivery does not.
  • Evening versus morning dosing has never been formally compared. The convention of evening dosing is inference from the melatonin target, not evidence.
  • The proposed mechanism of sequence-specific promoter binding by a tetrapeptide is not straightforwardly plausible on information-theoretic grounds, and the group's own framing has shifted toward chromatin-level effects. This is unresolved.
  • The relationship between the telomerase finding and any real-world cancer risk is completely uncharacterised in humans. The contraindication in this record is precautionary reasoning, not an observed harm.
  • Whether the reported effect persists for months after a course, which is the entire justification for pulsed dosing, has never been tested outside the originating group.
  • The 390.3 Da molecular weight was confirmed by calculation from the AEDG sequence rather than from a cited analytical measurement.
  • Cost figures are indicative retail observations, not sourced or surveyed data.

Papers