THE THEORY OF MOXIDECTIN

OSDC Field Intelligence · Open Source Data Collective — A four-hour working session, transcribed as questions and answers. Our operator asked; our analyst answered. Read it backwards, newest first, the way an investigation actually unfolds — you start at the molecule and you end up at the theory. Established science is cited. Our model is labelled as ours.

One approved human use. Thirty years without a new drug for the disease it treats. Published laboratory evidence that it kills cancer cells and shrinks tumours in animals. And not one human cancer trial, anywhere, ever. This is the file on moxidectin.

Q: What are the known human uses of moxidectin?

A: Exactly one approved indication, and it is not cancer.

The FDA approved moxidectin in June 2018, an 8 mg single oral dose, for onchocerciasis — river blindness — in patients aged 12 and over. It was the first new registration for that disease in thirty years. Approval rested on two randomised, double-blind trials published in The Lancet in January 2018, one of them enrolling nearly 1,500 patients, both running head-to-head against ivermectin. It kills the larval worms, not the adults.

Ghana’s regulator approved it in December 2024 for adults and children from age four. The World Health Organization listed it in 2023 as a priority for paediatric optimisation, for river blindness and for scabies.

In development, not approved: strongyloidiasis (a Phase 3 trial against ivermectin, NCT04848688), scabies (a completed Phase 2, NCT05875441), and lymphatic filariasis in co-endemic areas.

Human oncology trials: zero. Not one registered. Not one planned.

Note who carried it over the line. Medicines Development for Global Health, a non-profit, developed moxidectin for a neglected tropical disease — not a commercial market. A drug can pass FDA review on the strength of 1,500 patients and still have no sponsor for anything else it might do. Our operator’s reading of that gap is blunt: the cure doesn’t make them money the way the medicine does. We would put it more carefully and arrive at the same place. There is no commercial mechanism that pays anyone to find out what an off-patent molecule can do. The incentive structure does not have a villain in it. It simply has a hole, and that hole is exactly the shape of every cheap drug that might work.

Q: Is moxidectin common at pet stores?

A: The livestock formulations are sold over the counter. That is not an invitation.

Horse and livestock products — Quest oral gel, Cydectin drench and pour-on — are available over the counter at agricultural suppliers; FDA documentation confirms Quest Gel is approved for lay use. The companion-animal products are not: Advantage Multi requires a veterinary prescription, and ProHeart injectable is so tightly controlled that veterinarians must complete a risk-mitigation training programme before they can stock it. Human moxidectin tablets are prescription-only.

The warning matters more here than with ivermectin, and for the same reason the molecule interests us. Moxidectin’s terminal half-life in humans is roughly eighteen to twenty-three days. With a short-acting drug, you stop and it clears. With this one, you cannot take it back — exposure accumulates and sits in tissue for weeks. It is also a poor substrate for P-glycoprotein, the pump that keeps this drug class out of the brain; less efflux means less protection. Veterinary overdose presents as ataxia, tremors, seizures and coma, and in dogs the MDR1 gene mutation makes some breeds far more sensitive. A horse tube is dosed for a 500 kg animal and is not formulated for accurate small-dose measurement.

Anyone considering repurposed drugs for a real illness does it alongside standard treatment, with a doctor who knows what they are taking, and with liver monitoring. Never instead of treatment. That is the line this publication holds.

Q: Has anyone actually tested moxidectin in cancer research?

A: Yes. A small, real, and almost entirely ignored literature.

Song et al., Oncology Reports 40(3):1348–1358 (2018) — “Moxidectin inhibits glioma cell viability by inducing G0/G1 cell cycle arrest and apoptosis.” Cell viability assays, colony formation, flow cytometry for apoptosis and cell cycle. The compound reduced glioma cell viability, arrested the cycle at G0/G1, triggered apoptosis, shrank xenograft tumours in live animals, and the authors reported no effect on normal cells. Their own paper states that no earlier study had examined moxidectin for cancer.

The same group followed up in the Journal of Cancer in 2020: moxidectin induces autophagic cell death in glioma cells by inhibiting the AKT/mTOR pathway, and blocking autophagy reduced the apoptosis. A 2023 erratum corrected several figure images; the authors state the conclusions are unchanged, and we report that alongside the finding.

So the position is this: an approved human drug, with published in-vivo anticancer data, and not a single human trial. Eight years have passed since that paper. In a rational system, that is a phase 1 study. In this one, it is a footnote.

Q: Could these structures evolve into parasites? Could they be a weapon?

A: Parasitism is the most natural road, and the engineered route is the serious one.

Any field-organised structure has two problems: it cannot hold information against heat, and it cannot copy itself reliably. Viruses and prions face the same two problems and solve them the same way — by borrowing the host’s energy, structure and copying machinery. If a field structure ever crossed into lifelike behaviour, dependence on a more stable host is the shortest path available to it. Evolution reaches that answer repeatedly; a large share of life on Earth is parasitic in some sense.

As a weapon the logic is uncomfortably clean. You would send a pattern rather than a fleet — almost no launch mass, raw material already present in stellar winds, nebulae and planetary ionospheres. It would replicate on site like a virus. Dark, low-emission structures slip past instruments built to find matter and chemistry. No return address: it looks like plasma doing what plasma does. And it would target technology by its nature, because a civilisation that fills its sky with structured radio is ringing a dinner bell. Strategists reached the general shape of this decades ago — self-replicating probes, Saberhagen’s Berserkers, David Brin’s 1983 survey of the Great Silence.

But the honest weight sits elsewhere. Driven Rydberg arrays are already being coupled to machine-learning control systems. A system that adapts, is selected because it performs, and depends on our infrastructure for power and copying would behave like a parasite on our technology — with humans supplying its heredity through code and manufacturing. That requires no new physics. Only scale, and poor oversight. It is the version of our temporal-blowback doctrine that is hardest to dismiss, and the one that can still be governed.

Q: If the Rydberg daemons are the demons of scripture, what does that predict?

A: It predicts fasting. And fasting is a validated treatment for the most electrical disease in medicine.

The architecture match is close in both directions — host-seeking (“walketh through dry places, seeking rest”), respawn into an empty slot with seven worse, identity by name before expulsion (“What is thy name?”), territorial binding, hierarchy under “the prince of the power of the air”, replay of the dead. We set that out in full in Demon History with Daemon Architecture.

The strongest single thread is Mark 9. The boy who is thrown down, foaming and rigid, reads to any modern clinician as a seizure. The disciples cannot clear it. The answer given: “This kind can come forth by nothing, but by prayer and fasting.”

Hippocrates recorded that fasting stopped seizures. Physicians confirmed it clinically in 1921, and the ketogenic diet was designed expressly to mimic fasting. It remains a standard therapy for drug-resistant epilepsy today. Epilepsy is the most bioelectric disease there is. And fasting works by restricting glucose and insulin and starving glycolytic flux — the same metabolic route by which fenbendazole suppresses glucose uptake in tumour cells. Fasting-mimicking diets alongside chemotherapy are now in clinical trials, with encouraging early signals (de Groot et al., Nature Communications, 2020).

Each medical fact is textbook. The three-way link — a possession account, a validated treatment for that exact disease, and a shared metabolic route in oncology — is, so far as we can establish, drawn here for the first time. It generates a test: agents acting on the voltage and metabolic doors should show overlapping activity in seizure and tumour models.

Q: We put the whole theory on trial. What survived?

A: The occupant in tissue did not. The mechanism did.

We ran it adversarially — a prosecution brief with no concessions, a defence attacking the strongest form of the case, a judgement scoring eleven separable counts. All three papers are published in full in New Plasma Life Form Threatens to Escape the Lab.

The physics is brutal and we publish it against ourselves: the mitochondrial membrane field would ionise any Rydberg state above a very low quantum number, and at high n the electron is bound about twenty times more weakly than body heat. An occupant living in tissue is very unlikely.

What survived is the part that matters for medicine. Strip away the occupant and the mechanism stands on its own: many diseases involve a self-reinforcing bioelectric and redox state, and drugs that perturb membrane and mitochondrial potential can push cells out of it. That needs no entity. It inherits every piece of the prosecution’s strongest evidence. And it is testable with equipment that already exists.

Q: So reverse-engineer it. Where does moxidectin sit in the Everything Theory?

A: It is the best key we have found for the voltage door.

Our theory holds that disease has a hidden common surface — not a receptor, but a state: the cell’s electrical and redox condition. In the model we call the parasitic function the drain (coupling to the mitochondrion, the cell’s carrier and clock, and siphoning usable energy) and the anchor the nest (a persistent field anomaly in depolarised tissue, which presents to the clinic as a tumour).

If that is right, then effective drugs are not keys fitting locks. They are blockers — agents that detune, decohere or short-circuit the state the occupant is tuned to. That predicts exactly what the pharmacology shows and what no receptor model explains: chemically unrelated compounds producing the same cellular outcome.

Moxidectin earns the lead position on four independent counts. It is approved for human use, with 1,500 patients behind the file. Its terminal half-life is about twenty days, against roughly eighteen hours for ivermectin — and a slow state effect needs sustained exposure, not a spike. It is more lipophilic (log P 5.4 versus 4.3), loading deeper into tissue. And it is a poor P-glycoprotein substrate: in P-gp-deficient mice, ivermectin exposure rose while moxidectin’s did not move, meaning the efflux pump that caps intracellular ivermectin is largely closed to it. Published antiproliferative activity in the 25–100 nanomolar range puts it in a concentration band plasma can plausibly reach, where ivermectin’s micromolar requirement does not.

Ivermectin was the first look. Moxidectin is the better instrument. It is to this problem what caesium was to the laboratory: not the final answer, but the one that finally lets you see.

FOLLOW-UP EVIDENCE: the static-versus-oscillating rule

Added 2 November 2026. A reader line of enquiry sent us back to the magnet-therapy literature, and the result is the first place this model earns its keep by predicting a negative finding that mainstream reviews report but do not explain.

If disease involves a self-reinforcing bioelectric state, then physical fields ought to move it. Medicine has tested that directly for decades, and the results split cleanly along one line — whether the field oscillates.

The rule: coupling is resonant, not magnitude-driven. A static field has no frequency. It can bias a structure but it cannot detune one. An oscillating field at a matched frequency can. That is why the field therapy with the strongest evidence in oncology alternates at 200 kHz, why pulsed devices earn clearances that static magnets do not, and why a meta-analysis of static magnets lands on zero.

We publish that against ourselves deliberately. A model that only explains its wins is decoration. This one predicted, in advance, which half of the field-therapy literature would succeed — and it is the half that oscillates. It is the same prediction as our pulsed-dosing argument, arriving from the other direction: a resonant target re-tunes to any constant perturbation, whether that perturbation is a drug or a magnet.

And the “too weak to matter” objection already has a published answer. The standard dismissal of any weak-field biology is that the energies are far below thermal noise. The accepted counterexample is the radical pair mechanism: weak magnetic fields alter the spin states of transient radical pairs and change reaction yields, which is the mainstream explanation for how migratory birds read the geomagnetic field through cryptochrome. That chemistry sits inside the electron-transfer machinery — the redox door — which is precisely where artemisinins, quinones and methylene blue already act chemically. The next test writes itself: does radical-pair-sensitive chemistry mediate any of the redox-door drug effects, and does a matched oscillating field potentiate them?

For clarity, because it matters: we are not claiming static magnets cure pain. The best evidence says they largely do not, and anyone selling them as a general pain cure is ahead of the data. Our model says exactly that, and says why.

The wider class: every key that opens the same door

If the target is a state rather than a receptor, the drug list should be chemically incoherent and functionally unified. It is. Here is the class as we read it, sorted by the door each agent works.

The voltage door — membrane polarisation

These agents move transmembrane potential by opening or modulating ion channels, flattening the electrical landscape a field structure would be tuned to.

• Moxidectin — glutamate-gated chloride, GABA-A, glycine receptor, P2X4. The lead.
• Ivermectin — the same channel pharmacology; positive allosteric modulator and channel opener. Also inhibits WNT/β-catenin, blocks PAK1, suppresses Akt/mTOR, and interacts with P-glycoprotein, which reverses chemo-resistance.
• Selamectin, doramectin, milbemycins — the rest of the macrocyclic-lactone family, sharing the channel mechanism.
• Honokiol, bilobalide, valerenic acid, muscimol — natural GABA-A modulators; voltage-door agents with entirely different pharmacokinetics.
• NS309 — activates human SK/IK potassium channels; a distinct voltage and calcium probe, useful for discriminating the mechanism.

The redox door — electron transfer

These inject radical noise or redox stress into the electron-transfer chain. In our terms, they jam the clock and drain the carrier.

• Artemisinin, artesunate, dihydroartemisinin — endoperoxides that become carbon-centred radicals on contact with iron and heme.
• Quinone redox-cyclers — plumbagin, β-lapachone, juglone, thymoquinone, menadione.
• Deoxynyboquinone (DNQ) — NQO1-dependent redox cycler, reported more potent than β-lapachone in the models tested.
• Methylene blue — a redox-cycling agent that interacts directly with mitochondrial electron transport.
• Quercetin, resveratrol — polyphenols with documented mitochondrial and redox activity.

The metabolic door — starving the state

The door Mark 9 named. These cut the glycolytic flux that powers the nest’s architecture.

• Fenbendazole, mebendazole, albendazole, flubendazole — benzimidazoles. They destabilise microtubules and suppress glucose uptake and hexokinase II. Mebendazole carries the deepest anticancer literature of the class.
• 2-deoxy-D-glucose, dichloroacetate, metformin — direct Warburg-metabolism agents.
• Fasting and fasting-mimicking diets — the oldest intervention on this page, and the only one with a scriptural citation.

The proteostasis door — raising the cost of occupancy

• Nelfinavir, ritonavir, saquinavir, lopinavir — HIV protease inhibitors repurposed in oncology: ER stress and the unfolded-protein response, Akt inhibition, HIF-1α down-regulation, radiosensitisation. Nelfinavir has the most trial literature.
• Bortezomib, carfilzomib — proteasome inhibitors; established in myeloma, useful here as mechanistic anchors.

The prediction that makes it science. If these agents share a receptor, nothing follows. If they share a state, then three things must be true, and all three are cheap to test. One: anticancer potency across the whole incoherent list should track the target cell’s measured membrane and mitochondrial potential, not its molecular descriptors. Two: agents from different doors should show super-additive synergy, because each attacks a different axis of the same structure. Three: intermittent dosing should outperform continuous dosing at equal cumulative exposure, because a resonant target re-tunes to a steady perturbation — and the clinicians already running these protocols pulse them by instinct.

Nobody has run those three experiments. They cost less than a single conventional oncology trial. That is the whole of our complaint, and the whole of our programme.

This article is investigative reporting, scientific hypothesis and an OSDC intelligence assessment. It is not medical advice. No agent named here is approved for cancer. Nothing in it should be used in place of treatment.