Loading today’s stories
Science5 distinct publishers3 min readPublished
The protection ran through the same serotonin receptor that carries the hallucinations, though a non-hallucinogenic compound worked too, and a randomised trial in about 80 patients is due to start in November.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
The mechanism is precise, not vague: protection ran through 5-HT2A, the same serotonin receptor responsible for many of psilocybin's psychedelic effects, and blocking that receptor reversed the neuroprotection [8]. What is unusual sits downstream. In human neurons, cisplatin depleted mitochondria and ATP from the fibres that extend toward the skin; psilocybin did not restore the mitochondria's overall energy-producing capacity, but preserved their ability to travel and held ATP at normal levels inside those projections [7]. Amit compares the microtubules the mitochondria ride along to a railroad that chemotherapy damages along with the cargo [30].
The separability claim rests on TBG, a compound built to activate 5-HT2A without hallucinations, which also protected mice [9]. Amit's own caveat is the one to keep: the psychedelic effects may be doing neuroprotective work nobody has characterised yet, so it is not clear the hallucination should be engineered out [10].
The design earns some trust. Mice with abdominal tumours received two injections, each equivalent to a 25-milligram human dose, a week apart, then daily cisplatin for a week, against saline controls [22]. On filament testing, the psilocybin animals responded like a third group of tumour-bearing mice given neither cisplatin nor psilocybin, while the cisplatin-only animals were much more sensitive to touch [23]. That third arm is what makes the number readable: the comparison is to a never-poisoned baseline, not to a partial improvement. And the protection held in tumour-bearing mice with no measurable change in tumour growth [6], addressing the standing concern that a psychoactive compound given during chemotherapy could accelerate the cancer itself.
Now the denominator, which the coverage does not settle. A UCLA neuropathy researcher not involved in the study puts CIPN at a third to a half of people on certain chemotherapies [13]; the authors write that platinum-based regimens affect up to 60% [14]; Dougherty told a briefing about 75% are affected to some extent [15]; Amit said roughly two in three develop neuropathic pain requiring dose adjustment or treatment breaks [16]. Those are different endpoints measured on different populations, and the spread between the lowest and highest figure is 42 percentage points [17]. It matters because the planned randomised trial enrols about 80 people across breast, colorectal and head and neck cancer [20], roughly 40 per arm on even randomisation [21]. A control-arm event rate of 33% versus 75% is the difference between a readable result and a shrug.
The thing this does not tell you is whether preventing filament hypersensitivity in a mouse corresponds to sparing a patient years of numb feet, a symptom duloxetine currently addresses only moderately and only after the fact [19]. Human tissue entered the study as donor sensory neurons and patient skin biopsies [11], not as patients. Results from the trial are expected in about five years [20].
Ranked by verification strength, evidence, and original report placement.
Two doses of psilocybin, given nine days and two days before the first chemotherapy session, completely prevented mice from developing hypersensitivity, and also reduced cold sensitivity and preserved nerve endings in the skin.
Two doses of the drug given prior to chemotherapy produced protection that persisted through as many as six rounds of the cancer treatment.
In experiments with human neurons, cisplatin depleted mitochondria and ATP from the wires extending from nerves; psilocybin did not restore the mitochondria's overall energy-producing capacity but preserved their ability to travel and maintained normal ATP production specifically within the nerve projections.
A mouse study on psilocybin and chemotherapy-induced peripheral neuropathy was published Thursday, Sept. 3, in the journal Science.
The Science paper is titled "Psilocybin prevents chemotherapy-induced peripheral neuropathy through mitochondrial trafficking preservation" and was co-led by Moran Amit, professor of Head and Neck Surgery, and Patrick Dougherty, professor of Pain Medicine, at The University of Texas MD Anderson Cancer Center.
The effect depended on the serotonin receptor 5-HT2A, which is also responsible for many of psilocybin's psychedelic effects, and blocking the 5-HT2A pathway reversed the neuroprotective effects.
Follow any of these and your For You feed starts watching them — no settings page required.
science
Psilocybin Heads Into Anorexia Trials, Where the Bar for Progress Is a 5% Death Rate1 distinct publisher
science
Graz group measures the geometry a single-molecule collision has to hit1 distinct publisher
science
Psilocybin's 'oneness' now has a brain-network correlate, and it was found by sorting people on self-report1 distinct publisher
science
Two cases a year: a Louisiana amoeba death tells northern states little about their own lakes1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
1 article · September 3, 2026
1 article · September 3, 2026
1 article · September 3, 2026
1 article · September 3, 2026
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Strong for a preclinical result, and that is the ceiling
The underlying work is unusually well-buttressed for an animal study: three chemotherapy agents, six cycles, tumour-bearing animals, a receptor antagonist that abolishes the effect, and — per GEN — donor neurons and patient skin biopsies showing the same trafficking network in human tissue. What none of it contains is a treated patient. Live Science's outside expert also notes the mitochondrial account of nerve injury predates this paper; what is new is the protectant, not the pathology.
Nothing in use; one trial not yet open
Scientific American states the flat fact the other accounts imply: the results stop at mice and cell models. The nearest thing to uptake is a trial that had not started when this was published, sized at about 80 patients, with a read-out New Scientist puts five years out. The only human signal anyone offers is second-hand and anecdotal — patients in psilocybin trials run for other reasons who mentioned improved nerve symptoms.
Mildly overstated, mostly by borrowed superlatives
Three of the five headlines carry the mouse qualifier or a hedge, which is better discipline than this genre usually manages. The inflation comes from elsewhere: "unparalleled" and "first-in-class" are the investigators' words, repeated as findings, and the disease burden is quietly enlarged as it travels — the briefing's 75 per cent in New Scientist against one-third to one-half from the unaffiliated researcher in Live Science. Only Live Science tells the reader how far this is from a patient.
One institution briefing on its own programme
Every number and nearly every quotation in this story originates in a pre-release press conference held by the institution that produced the paper, will run the trial and is publicly building a psychedelics research programme — Amit says as much in GEN, where he notes MD Anderson is actively exploring the multiple facets of psychedelics and that he leads other psilocybin trials. That is not a reason to disbelieve the result; it is a reason to notice that the framing, the superlatives and the largest incidence figures all come from the party with the most to gain, and that two outlets sought no outside voice at all.
Confident about what was shown, not about what it means
The factual core is stable: five accounts agree on the design, the receptor, the trafficking mechanism and the mouse outcome, and two of them quote researchers with no stake. Our confidence drops on the things a reader actually wants — how many patients are at risk, whether cisplatin protection generalises, and whether the psychedelic experience is load-carrying — each of which rests on a single source or is openly unresolved.
1 article · September 3, 2026