This site has attached the same caveat to Epithalon and Thymalin roughly a dozen times: the evidence comes from a single national research lineage, largely in one language, and independent error-correction has not run. That is a statement about the structure of a literature, not a verdict on the compounds — but it has been the single most important thing to know about them.
In 2025 and 2026, that structure changed — not dramatically, and not in the direction most vendor copy suggests. For the first time, a meaningful portion of the AEDG tetrapeptide literature was produced by groups with no connection to Saint Petersburg. One of those papers is a genuine independent test of the core claim. One only looks independent until you read the author list. And the same window produced two corrections.
The lineage lost its principal investigator in January 2024
Epithalon exists because of one research program. Vladimir Khavinson (born 27 November 1946, died 6 January 2024, aged 77) directed the Saint Petersburg Institute of Bioregulation and Gerontology and built the peptide-bioregulator field around the pineal extract Epithalamin and the synthetic tetrapeptide Ala-Glu-Asp-Gly derived from its amino acid composition. His published output across a four-decade career is generally given as roughly 775 papers and close to 200 patents.
That concentration is exactly what the single-lineage caveat describes, and it makes the death of one investigator a structural event for the literature rather than a biographical footnote.
The practical consequence: work carrying Khavinson's name published after January 2024 reflects collaborations and manuscripts already in flight, while the originating institute's continuing output appears largely in Advances in Gerontology, a Russian-language journal, on applied topics — peptide bioregulation in prosthetic stomatitis in elderly patients (2025), cellular aging marker proteins in buccal epithelium (2024). Same venue, same study design that generated the caveat in the first place.
The first genuinely independent test of the telomerase claim
The core Epithalon hypothesis — hTERT induction, telomerase activation, telomere extension — finally got a quantitative workup from a lab with no lineage connection.
Al-Dulaimi, Thomas, Matta and Roberts, Biogerontology 2025;26(5):178 (Centre for Genome Engineering and Maintenance, Brunel University London) treated breast cancer lines 21NT and BT474 alongside normal epithelial and fibroblast cells, then ran qPCR and immunofluorescence across telomere length, hTERT mRNA, telomerase enzyme activity and ALT (alternative lengthening of telomeres) activity. The paper reports concentration-dependent telomere extension in normal cells through hTERT and telomerase upregulation, and telomere extension in the cancer lines through ALT activation instead, with only minor ALT activity in normal cells.
Two things matter. First, it is a rung-1 result — cultured human cell lines, no organism, no ADME. It establishes the interaction is real in a dish, in someone else's dish. That is more than the compound had before, and exactly the kind of finding the evidence hierarchy says to weight modestly and specifically.
Second, the least-quoted finding is the more interesting one. The mechanism observed in the cancer lines was not the mechanism observed in the normal cells. The cancer-caution flag that has hung over telomerase-induction framing since this site first covered Epithalon's pineal and telomerase literature is no longer just a theoretical objection about immortalisation — an independent group looking for it found a distinct telomere-lengthening route running in transformed cells. That is a research finding to sit with, not a selling point.
On 15 November 2025 Biogerontology published a correction (2025;27(1):1) to that paper. The erratum states that the wrong figures appeared in Figures 1, 2 and 3, and replaces them. Captions were unchanged, the text was unchanged, and the scientific conclusions stand.
Figure-replacement corrections are ordinary science and carry no implication of misconduct. But in a paper whose entire quantitative result is carried by three data figures, all three were replaced — so anyone citing the September version is citing figures that no longer represent the record. The version matters.
It was not the only correction in this literature in the same window. In January 2025, Pharmaceuticals published a correction to Khavinson et al. 2021;14:515 on tripeptide neuroprotection in an Alzheimer's mouse model, stating that Figures 5 and 8 were the same image described differently — one labelled male, one labelled female. Corrected figures were supplied and the conclusions were stated as unaffected. Both notices are a reminder that "peer-reviewed and published" is a starting condition, not a finished one.
The paper that looks independent and isn't
Gatta, Dovizio et al., Stem Cell Reviews and Reports 2025;21(6):1822-1834 comes from "G. d'Annunzio" University of Chieti-Pescara in Italy. It reports that in ARPE-19 human retinal pigment epithelial cells injured by high glucose, AEDG restored impaired wound healing while inhibiting hyperglycaemia-induced epithelial-mesenchymal transition and fibrosis gene expression — framed as an antioxidant effect, in an in vitro model of diabetic retinopathy.
An Italian university, a Western journal, a mechanism not obviously borrowed from the pineal story. It reads like independent corroboration. Then check the author list. Trofimov and Khavinson are on it, both affiliated to the Saint Petersburg Institute of Bioregulation and Gerontology. This is lineage extension through international collaboration, which is a normal and useful way for a research program to travel — but it is not the thing the single-lineage caveat asks for. Replication means a group that could have failed to reproduce the result and had no stake in the outcome.
The transferable reading skill: the country of the lead institution tells you nothing; affiliations on the full author list do. Same check as trial registry records — read the record, not the summary.
To the authors' credit, their abstract states that more mechanistic investigation is needed to confirm benefits and safety, and proposes developing ophthalmic forms to improve retinal delivery. That last clause is the familiar pattern: when the next step is a new formulation, delivery is the limitation being diagnosed.
What the second independent group was actually doing
The other 2025 paper from outside the lineage is Ullah et al., Life Sciences 2025;362:123381 (Gyeongsang National University, South Korea), and it is not a longevity study at all. Working on bovine in vitro embryo production, the group used AEDG to activate telomerase in cumulus-oocyte complexes and reported improved oocyte maturation rate, and improved blastocyst hatching and implantation potential in post-thaw embryos, with supporting ROS and mitochondrial membrane potential assays.
Read that as what it is: a laboratory tool being used to optimise a culture system, in the same category as IGF-1 LR3's documented role as a serum-free cell-culture supplement. It says nothing about a healthspan endpoint in an organism. It does say a second unaffiliated group found the telomerase axis responsive enough to build a method around.
Thymalin: no 2025-2026 file to report
The contrast is stark. A PubMed search returns no 2025 or 2026 primary research on Thymalin. The most recent indexed work is a pair of 2024 rat mandibular bone-defect studies from Ukrainian groups, and a 2023 IJMS paper from Khavinson's own group on the KE and EW dipeptides in the composition of the Thymalin drug.
That title is the most useful documentation sentence in the whole Thymalin literature. The originating group's own modern mechanism work is on defined dipeptides said to be present in the preparation, not on the preparation itself — because Thymalin is a heterogeneous thymic fraction, not a molecule. As covered in the immune class primer, that is why a purity percentage on a Thymalin certificate has no clean referent, the same structural problem as HMG.
Searched against the ClinicalTrials.gov v2 API on 4 September 2026, the terms Epitalon, Epithalon, Thymalin, AEDG and peptide bioregulator return zero records between them. Not zero results — zero registrations. There is not an intention on file for either compound, which places both below rung 3 entirely.
What the 2026 reviews say
Mavrych, Shypilova and Bolgova, Frontiers in Aging 2026;7:1790247 (7 April 2026) surveyed peptides in gerontology across PubMed, Scopus and FDA/WADA databases through January 2026, covering nine compounds including epitalon, GHK-Cu, BPC-157, TB-500 and Semax alongside approved drugs. Its methods selected 20 primary sources across all nine — itself informative about the size of the underlying evidence base — and it separates the shelf cleanly: approved agents have robust safety data from large trials; non-approved peptides show promising preclinical and limited clinical evidence but lack long-term safety data and systematic validation.
Araj, Brzezik, Mądra-Gackowska and Szeleszczuk, IJMS 2025;26(6):2691 (Medical University of Warsaw) surveyed 25 years of Epitalon work and flagged a gap the marketing literature never mentions: despite the considerable volume of research on biological and pharmacodynamic characteristics, the quantity of physico-chemical and structural investigation of this peptide remains quite limited. Twenty-five years of bioactivity papers, and a thin file on the molecule as a chemical object. The biology file and the analytical file are separate files, and only one has been worked.
Scorecard: what 2025-2026 did and did not do
Nothing this year changed the documentation picture. AEDG still carries no tryptophan or tyrosine, so UV quantitation at 280 nm is unavailable and it depends on other quantitation routes; its C-terminal Asp-Gly is still a named aspartimide and deamidation motif, making its degradation liabilities predictable from the letters.
Did: produce the first quantitative test of the telomerase mechanism by a lab with no lineage connection; produce a second unaffiliated group using the peptide as a culture tool; produce a Western review that names the structural-characterisation gap out loud; produce two corrections that tightened the record.
Did not: produce a single human study, a single trial registration, any pharmacokinetic data, any Thymalin work at all, or any result that moves either compound off the preclinical rung.
The caveat is smaller than it was in June. It has not been retired.
FAQ
Does the Brunel study validate Epithalon? It validates that the interaction is measurable in cultured human cells by a group with no stake in the result — which is meaningful, and is rung 1. It is not evidence about an organism, and the paper's cancer-cell ALT finding is a reason for more caution about telomerase-induction framing, not less.
Two corrections in one literature in one year — is that a red flag? No. Figure corrections are routine and both notices state the conclusions are unaffected. The practical implication is narrower and duller: cite the corrected versions, and treat "published" as a status that can be updated rather than a permanent seal.
Why cover Thymalin's absence rather than skipping it? Because an empty file is a finding. The honest statement is that no new primary research on Thymalin was indexed in 2025 or 2026, and that the originating group's own recent mechanism work targets defined dipeptides rather than the heterogeneous preparation. Filling that silence with extrapolation is how vendor copy gets written.
Compound pages are in the library, and the documentation standards behind these points are in quality.
This article is educational and for the laboratory research community. Trulogic Labs products are sold for laboratory and research use only and are not for human consumption.