Dihexa vs Semax: Mechanism, Evidence, Cost, and Which One Fits

Semax is a heptapeptide, seven amino acids long: Met-Glu-His-Phe-Pro-Gly-Pro. The first four are positions 4 to 7 of adrenocorticotropic hormone (ACTH), the pituitary hormone that tells the adrenal glands to make cortisol. The last three, Pro-Gly-Pro, were added to slow its breakdown. The design keeps the part of ACTH that was studied for attention and learning and drops the part that raises cortisol, so Semax is described in the literature as having no corticotropic activity. It is not a steroid and not a stimulant. The full reference is at what is Semax.

Semax was developed at the Institute of Molecular Genetics of the Russian Academy of Sciences. It is a registered medicine in Russia, sold there as nasal drops, and has been used in Russian neurology since the 1990s for acute ischaemic stroke and chronic cerebrovascular disease. Most of the clinical work was published in Russian journals, several with English abstracts.

• BDNF and TrkB in the hippocampus. A 2006 paper in Brain Research reported that Semax raised BDNF protein and activated its receptor, TrkB, in the rat hippocampus. BDNF is the growth factor most closely tied to synaptic plasticity and forming new memories, which is the basis of the "cognitive" label.
• Neurotrophins where the brain is injured. A 2010 paper in Cellular and Molecular Neurobiology reported increased transcription of BDNF, NGF and their receptors in the ischaemic rat cortex, selectively in the injured tissue.
• Immune and vascular genes. A 2014 genome-wide study in BMC Genomics found the largest effect after stroke in rats was on immune-response genes, with a smaller set of vascular genes.
• Serotonin and dopamine signalling. Animal work describes increased serotonergic signalling. A mouse study reported Semax increasing amphetamine-induced dopamine release, which is why anyone taking a stimulant should tell their physician.

All of the mechanism work is in animals.

There is a real body of Russian clinical work, mostly in stroke and cerebrovascular patients, using the nasal product:

What does not exist: a randomized, placebo-controlled trial published in an English-language peer-reviewed journal, or any trial of Semax for focus or memory in healthy adults. The FDA's 2026 review excluded the Russian studies because complete English translations were not supplied.

Dihexa is a very small modified peptide built from a fragment of angiotensin IV, a breakdown product of angiotensin II (the hormone most people know from blood-pressure medicine). A group at Washington State University led by Joseph Harding and John Wright found that angiotensin IV and close relatives improved memory-task performance in rats, then built a version that would survive in the body. Its chemical name is N-hexanoic-Tyr-Ile-(6)-aminohexanoic amide; "dihexa" refers to the two six-carbon groups on either end, which protect it from enzymes and help it cross into the brain. It is also called PNB-0408, and pharmacies label it dihexa acetate. It is not a stimulant and does not act on dopamine the way attention medicines do. The full reference is at what is Dihexa.

• The angiotensin IV signal. A 2018 systematic review in Neuroscience and Biobehavioral Reviews concluded that angiotensin IV and related fragments improve learning in animals consistently across models. That is the independent foundation Dihexa rests on.
• The HGF and c-Met proposal. The developers proposed that Dihexa makes hepatocyte growth factor a stronger activator of its receptor, c-Met, which drives the growth of new synaptic connections in the hippocampus. The two laboratory papers making that case (2012 and 2014, Journal of Pharmacology and Experimental Therapeutics) were retracted in April 2025 after expressions of concern in 2021. The mechanism may still be right, but its strongest evidence can no longer be cited.
• PI3K/Akt signalling. An independent 2021 paper in Brain Sciences gave Dihexa to APP/PS1 mice, a strain bred to develop Alzheimer's-type plaques and memory deficits, and reported better memory-task performance along with increased PI3K/Akt signalling. It is one mouse study, but it did not come from the developers.

None. There are no randomized trials, no open-label trials and no published case series. The FDA states it has not identified any human exposure data for dihexa acetate by any route. The 2013 paper that introduced the compound, which reported improved water-maze performance in impaired and aged rats, carries a 2021 expression of concern.

Semax is the usual starting point when the complaint is focus, mental fatigue late in the day, or difficulty getting started on tasks. It has the longer record of human use, a clear story around BDNF, and it combines with Selank for people whose poor focus is partly stress. A physician who is cautious about Dihexa's evidence base will often suggest Semax first.

Dihexa is the one built specifically around learning and recall, and some people choose it for exactly that: remembering names, learning a language, holding onto new material. It is a more experimental choice, and it suits people who understand that the human evidence is blank and who have a specific, checkable goal rather than general "brain fog."

When stress, rumination or anxiety sit behind poor concentration, Selank, a sibling from the same Russian research programme, is described around calm rather than drive. Semax + Selank puts both in one vial. If energy or body composition is part of the goal, SYNAPSE adds MOTS-c and tesamorelin; if recovery and inflammation are, ELEVATE adds BPC-157 and KPV.

They act on different systems, and some physicians do prescribe them together. There is no study of the combination, so it means two unknowns at once rather than one. Most people start with one, keep a simple log for four weeks, and add the second only if there is a reason to.