Kamuvudine K-9: the modified HIV drug that reversed vision loss and paralysis in an MS model
The experimental drug kamuvudine K-9, derived from HIV therapy, did not just stop but reversed paralysis and vision loss in mice with an MS-like disease. Here is the Science Translational Medicine study, the inflammasome mechanism, the three-million-patient data — and an honest answer to what this means for people.
Every time I publish a piece about a new drug, the same message lands in my inbox: "Is this finally the one?" So I want to start this article about kamuvudine K-9 the way I would want to read it myself: with what is proven, what is plausible, and what nobody knows yet.
In late September 2026 the US National Institutes of Health announced a finding that travelled around the world: the experimental drug kamuvudine-9 (K-9) did not merely halt but reversed vision loss and paralysis in animals with a disease that mimics multiple sclerosis [1](#ref-1). The study was published in Science Translational Medicine by researchers at the University of Virginia, led by ophthalmologist Jayakrishna Ambati [2](#ref-2), [3](#ref-3).
The reason this is news, and not just another headline, is simple: every MS drug we have today does the same job — it reduces future attacks. None of them gives back what the disease has already taken. In the animal model, K-9 did exactly that second thing [3](#ref-3).
> Most MS therapies aim to prevent the next attack. What is remarkable about K-9 is that it did not only prevent further neurological deterioration — the animals recovered function they had already lost.
What kamuvudine K-9 actually is
K-9 belongs to a group of compounds called kamuvudines — derivatives of nucleoside reverse transcriptase inhibitors, or NRTIs. These are antiviral drugs that have been used for decades to treat HIV and hepatitis B [1](#ref-1).
NRTIs have a side property Ambati's team noticed earlier: besides blocking the viral enzyme, they also block the inflammasome — a built-in alarm system of brain and spinal cord cells that, once activated, sets off explosive cycles of cell death [1](#ref-1), [4](#ref-4). The problem is that long-term NRTI use carries neurological and musculoskeletal side effects. So the team chemically stripped the antiviral part of the molecule and kept the part that switches off the inflammasome. The result is K-9, described as a non-toxic derivative [1](#ref-1), [6](#ref-6).
Why the inflammasome, not just the immune system
For decades, neurodegeneration in MS was explained simply: the adaptive immune system's B and T cells mistakenly attack healthy tissue, and that is that. Standard therapies therefore keep those cells in check — and they genuinely do prevent relapses, but they do not repair damage already done, and they often suppress immunity broadly [1](#ref-1).
Newer research suggests rampant adaptive immunity may not be what delivers the killing blow to cells. Instead it may be the trigger that switches on the inflammasome, a key part of the central nervous system's innate immunity. If that is right, then shutting down the inflammasome could protect nerves even where classic therapies have nothing left to prevent [1](#ref-1), [5](#ref-5).
In the paper, K-9 was shown to disrupt the NLRP3-NEK7 and NLRP3-NLRC4 interactions — which the authors read as blocking two inflammasomes at once [6](#ref-6).
What exactly happened in the experiment
The researchers induced an MS-like disease in mice (the EAE model) and — this matters — waited until symptoms were already visible before starting treatment. The animals then received K-9 for two weeks [7](#ref-7).
The result: treated animals were protected from further neurological decline and recovered from earlier movement problems and vision loss. In the same experiments, animals given an approved MS drug recovered nowhere near as much function [3](#ref-3), [7](#ref-7).
The drug preserved nerve fibres and the myelin insulation that MS destroys. It also halted the rise in neurofilament light chain (NfL), a blood biomarker of nerve damage that is increasingly used to track the disease [3](#ref-3). I have written separately about what doctors look for in scans and findings in the article on [MRI in multiple sclerosis](/blog/mri-multipla-skleroza-sta-lekari-vide).
The real-world data point: three million patients
What lifts this story above a typical "mouse" headline is the parallel part of the work. The team analysed health insurance data covering more than three million people and found that NRTI use — in people treated for HIV, those on preventive therapy, and those with hepatitis B — was associated with a 41 per cent lower risk of developing MS and a 36 per cent lower annual relapse rate among people who already had it [3](#ref-3), [8](#ref-8).
That is not proof that NRTIs treat MS — it is an epidemiological association, the same kind you see whenever a large database shows one group of patients falling ill less often. But when that association lines up with a clear laboratory mechanism and with recovery in an animal model, you get what the director of the National Eye Institute called "a very compelling translational pathway" [1](#ref-1).
What this means — and what it does not
I need to be completely plain here, because I know how news like this reads when you live with MS.
K-9 is not a drug for people. It has not been tested as an MS treatment in humans, it is not approved, and it cannot be obtained. The results come from an animal model, and the history of MS research is full of drugs that worked spectacularly in mice and never made it through human trials [7](#ref-7). The EAE model mimics the inflammatory part of the disease, but not its decades-long complexity in a person.
Just as importantly: nobody reading this should change, stop or "top up" their therapy because of this news. The drugs we use today — which I covered in the [DMT efficacy ranking](/blog/najefikasniji-lekovi-multipla-skleroza-rang-lista-efikasnosti) — remain the only proven way to keep the disease under control.
Why this is still big news
First, it changes the question. Until now, almost all drug research revolved around preventing the next attack. K-9 asks a different question: can function already lost be regained? If that principle is confirmed in people, it is a new pillar of treatment, not just another drug in the existing column [3](#ref-3).
Second, it comes with a safety head start. K-9 is derived from drugs whose safety in people has been known for decades, and the molecule itself was designed to avoid their toxic parts [1](#ref-1). That shortens the road to first-in-human trials compared with an entirely new chemical.
Third, it fits a broader trend. For years I have been following research that targets nerve protection and repair rather than immunity alone — from [remyelination and the ATDR molecule](/blog/retsat-q247r-jak-mijelin-remijelinizacija-atdr-multipla-skleroza-otkrice) to [stem cell transplantation](/blog/transplantacija-maticnih-celija-ms-vodic). K-9 is another serious move in that direction, this time backed by the National Institutes of Health.
What to watch next
The next step is clinical trials in people — first safety studies, then the ones that must show whether anything seen in the mice repeats in people with MS. The senior author co-founded Inflammasome Therapeutics and is named on patents covering this use, which is normal for the road from bench to bedside, but also one more reason to read the results with a healthy measure of caution [6](#ref-6).
When and if a human trial begins, I will follow it here. Until then, the honest answer to the message at the top of this article is: this is one of the most interesting stories of the year — and it is not a medicine yet. One does not cancel the other.
This article is informational and does not replace an examination or medical advice. Do not change or stop a prescribed therapy on the basis of results obtained in animal models.