P21

P21, more precisely written P021 in the primary literature, is one of the few peptides in the consumer market whose preclinical file was built by an academic laboratory rather than by a vendor. It emerged from work by Khalid Iqbal's group at the New York State Institute for Basic Research in Developmental Disabilities, who set out to identify the active neurotrophic principle inside Cerebrolysin, the porcine brain-derived peptide preparation used clinically in parts of Europe and Asia for stroke and dementia. Cerebrolysin is an undefined mixture, which makes it difficult to study and impossible to optimise. The goal was to find the smallest fragment that reproduced its neurogenic effect.

The answer they converged on maps onto a fragment of ciliary neurotrophic factor. The resulting compound, Ac-DGGLAG-NH2, was then adamantylated to improve stability and passage across the blood-brain barrier, producing a small peptidergic molecule that is orally active in rodents. That last property matters: most neurotrophic peptides fail not because the biology is wrong but because nothing reaches the brain.

The rodent literature is the reason P21 attracts attention. In a 2014 Neurobiology of Aging study, chronic administration to aged rats restored hippocampal neurogenesis and improved performance on memory tasks toward the level of young animals [1]. In the same year, a Neurobiology of Disease paper reported a disease-modifying effect after chronic oral treatment in the 3xTg triple transgenic mouse model of Alzheimer's disease, with reductions in amyloid and tau pathology alongside behavioural improvement [2]. A 2017 Scientific Reports study moved into the Ts65Dn mouse model of Down syndrome, where early treatment rescued developmental delay and Alzheimer-like memory deficits [3]. A further 2017 paper in Alzheimer's Research and Therapy focused on structural endpoints, reporting prevention of dendritic and synaptic deficits along with preserved cognition [4]. The programme has continued to broaden: a 2024 paper in the Journal of Neurodevelopmental Disorders tested the same CNTF-mimetic peptide in models of CDKL5 deficiency disorder and reported rescue of neuronal and behavioural phenotypes [5]. A 2022 review in Biomolecules sets out the group's overall therapeutic case [6].

Mechanistically, the reported chain runs from increased neurogenesis in the dentate gyrus, through elevated BDNF and downstream synaptic protein expression, to increased dendritic spine density and restored synaptic plasticity, with secondary reductions in tau hyperphosphorylation. That is an internally consistent story and it is reported across several distinct disease models rather than a single one, which is a genuine strength. Multiple models failing in the same direction is much harder to explain by artefact than a single striking result.

What is entirely missing is human data. There is no published phase I trial of P021. There is no registered clinical study. There is no human pharmacokinetic profile, no tolerability data, no immunogenicity assessment, and no dosing information that has any basis outside rodent milligram-per-kilogram conversions. Anyone quoting a human microgram or milligram protocol for P21 is extrapolating from animal work without pharmacokinetic support, and the history of neurology drug development is a long list of compounds with beautiful rodent data that produced nothing in people. Cerebrolysin itself, the parent preparation, has had a mixed and much-debated clinical record over decades, which should temper expectations for a derivative.

Safety is therefore an open question rather than a reassuring one. Rodent studies at the doses used report no overt toxicity across months of oral dosing, and no organ pathology has been highlighted. But P21 is designed to increase neurogenesis and drive synaptic remodelling. Chronic pharmacological stimulation of proliferation in a stem cell niche is exactly the kind of intervention where long-term human data would matter most, and none exists. There is also no information on interactions with antidepressants, cholinesterase inhibitors, or anything else a person with cognitive complaints is likely to be taking.

Practically, P21 sold on peptide sites is a research chemical. It is not a licensed medicine and it is not a lawful supplement ingredient in the United States or the European Union. Purity and identity depend entirely on the vendor, and the peptide is complex enough that misidentification or partial synthesis is plausible without third-party analysis.

Our position: P21 has one of the better preclinical files of any peptide in this category, produced by an academic group without a peptide business to promote, and published in indexed journals across several disease models. That earns it a solid tier 4 rather than the tier 5 assigned to compounds with only vendor literature behind them. It does not earn it any human claim. The single event that would change the assessment is a registered phase I trial; until that appears, P21 is a promising preclinical compound being used by people as if it were a drug.