The Theoretical Oncological Concern
The most-discussed safety question for Dihexa is mechanistic rather than observed. Dihexa's proposed mechanism — described in papers later retracted (Kawas 2012, PMID 22129598, retracted 2025 per notice PMID 40312092; Benoist 2014, PMID 25187433, retracted 2025 per notice PMID 40312093) — is potentiation of hepatocyte growth factor (HGF) signaling at the c-Met (MET) receptor, driving dendritic spine formation and synaptogenesis via PI3K-AKT.
The HGF/c-Met axis is not a brain-specific pathway. It is a broad growth, proliferation, and angiogenesis signaling system, and it is dysregulated or overexpressed in several cancers. From that biology, researchers draw a theoretical concern: chronically potentiating a proliferative, pro-angiogenic pathway is a plausible mechanism by which an agent could, in principle, support growth of existing malignancy.
The honest framing has three parts:
- This is mechanistic reasoning, not a demonstrated effect. No published study has reported Dihexa accelerating tumor growth, and no human cancer-incidence data exists in either direction.
- The mechanism itself is uncertain. Because the core mechanism papers were retracted, even the premise that Dihexa potently engages HGF/c-Met in a living human brain at community-reported doses is not established fact.
- Community sources treat it as a precautionary pattern. Forum discussion describes users with active or prior malignancy avoiding Dihexa as a precaution. That is a community pattern, not a clinical guideline, and no oncological monitoring protocol exists for it.
The summary a careful reader should take away: the oncological concern is theoretical and pathway-based. It is neither confirmed nor refuted by data, and it sits on top of a mechanism literature that has itself been called into question.
Note on labeling: the effects below come from forums (LongeCity and similar), vendor comment sections, and user write-ups for research-chemical Dihexa used as powder, oral capsules, or transdermal preparations in DMSO. They are anecdotal. They are not from clinical trials, are not incidence rates, and are frequently inconsistent — a meaningful share of community reports describe no noticeable effects at all.
Headache and mental fatigue
The most frequently mentioned anecdotal complaint is headache, sometimes paired with "brain fog" or mental fatigue that runs counter to the cognitive sharpening users were seeking. Community sources variously attribute it to dose, to the DMSO carrier in transdermal use, or to vascular effects, with no way to confirm cause in self-reports.
Irritability and anxiety
Some community reports describe irritability, restlessness, or heightened anxiety, occasionally framed as overstimulation. Reports are mixed: others describe improved mood and motivation, so the pattern reads as individual variation rather than a consistent Dihexa effect.
Nausea and GI upset
Nausea and general gastrointestinal discomfort appear sporadically in anecdotal reports, more often mentioned with oral capsule use than with transdermal application in community write-ups.
Appetite and sleep changes
Self-reported appetite shifts (in both directions) and altered sleep — difficulty falling asleep, lighter sleep, or vivid dreams — are described inconsistently across community sources, with no documented dose relationship.
Blood-pressure changes
Because Dihexa derives from angiotensin IV, some community sources speculate about blood-pressure effects and a minority report feeling changes. There is no human cardiovascular data for Dihexa, so this remains an anecdotal observation tied to a structural rationale rather than a measured effect.
Skin reactions with transdermal-DMSO use
Users applying Dihexa transdermally in DMSO describe local skin reactions — redness, itching, irritation, or a burning sensation at the application site. DMSO is itself a known skin irritant and penetration enhancer, so community sources often attribute these reactions to the carrier as much as to Dihexa.
What Would Warrant Stopping — and Why the Usual Framework Doesn't Apply
For trial-backed peptides, this is where documented discontinuation criteria from clinical protocols would go. For Dihexa, no such criteria exist, because no trial has ever defined them. There is no published guidance on monitoring, no validated stopping rule, and no human safety threshold to reference.
What community sources describe, explicitly as anecdote rather than guidance:
- Reports of pausing when headache, irritability, or GI symptoms persist or intensify.
- Reports of users with active or prior malignancy avoiding the compound outright on the mechanistic grounds described above.
- Reports of discontinuing transdermal preparations when skin reactions to the DMSO carrier become uncomfortable.
None of these are clinical recommendations. They are documented patterns in self-reported community sources, included to show how users have described handling adverse effects in the complete absence of trial-defined safety data. Decisions about a compound with no human safety profile are squarely matters for a licensed physician.
The Research-Integrity Context
Dihexa's safety picture cannot be separated from the state of its evidence base. A Washington State University investigation found that lead author Leen H. Kawas — later CEO of Athira Pharma — altered images in her dissertation and in four or more co-authored papers between 2011 and 2014. The two core mechanism papers were retracted in April 2025 (PMID 22129598, retraction notice PMID 40312092; PMID 25187433, retraction notice PMID 40312093), and a third (PMID 23055539) carries a 2021 Expression of Concern.
The widely repeated claim that Dihexa is "seven orders of magnitude more potent than BDNF" traces to McCoy 2013 (PMID 23055539), the paper now under an Expression of Concern. That figure came from an assay-specific in-vitro spine-density assay (picomolar Dihexa versus nanomolar BDNF) and does not mean Dihexa is millions of times more effective in a living brain. It should be cited only with that caveat.
The one piece of cleaner support is Sun 2021 (PMID 34827486), an independent group's APP/PS1 mouse study reporting improved Morris water maze performance, increased synaptophysin, and reduced IL-1β/TNF-α via PI3K/AKT. It is a small rodent study, not human evidence, but it stands outside the retracted body of work. Read together with the failed fosgonimeton trial, the takeaway is that the human-relevant evidence base for this class is both thin and, where it touched humans, negative.
