Therini Bio.

There is a useful trick in biology, which is that the thing keeping you alive is frequently the same thing trying to kill you, and the whole game is figuring out which version is in the room. Fibrin is a good example. When you cut yourself, fibrin is the mesh that shows up, cross-links, and stops you from bleeding out. You want fibrin. You would not last an afternoon without it. But fibrin also has a second, less advertised job: when it leaks out of a damaged blood vessel into tissue where it does not belong - say, into the brain, past a blood-brain barrier that has started to fail - it stops behaving like a bandage and starts behaving like an alarm that will not turn off. It binds a receptor on the brain's immune cells and tells them, essentially, to attack. And they do, chronically, for years.

Therini Bio, a clinical-stage company founded in 2019, is built entirely around that second job. Its bet is that a meaningful amount of neurodegeneration - the slow damage in diseases like Alzheimer's - is not only about amyloid plaques and tau tangles, the targets the field has chased for decades. Some of it, Therini argues, leaks in from the bloodstream. Fix the vascular inflammation, and you may slow the disease from a direction almost nobody else is coming from.

The obvious objection writes itself: if fibrin is causing brain inflammation, get rid of the fibrin. But you cannot get rid of fibrin, because getting rid of fibrin means you cannot clot, and not clotting is a much faster way to die than Alzheimer's. This is the design constraint that makes the whole thing hard and, if it works, valuable. Therini's answer is precision. Fibrin has a specific region - an "inflammatory epitope" - that is only exposed when the protein is doing its inflammatory job. Its lead antibody, THN391, is engineered to grab exactly that region and nothing else.

The result, at least in principle, is a drug that walks into the brain, blocks the inflammatory switch on fibrin, and leaves the clotting machinery entirely alone. It is the difference between disarming a system and removing it. And the early human data suggests the engineering held: in a Phase 1a study in healthy volunteers, THN391 was well-tolerated, produced no serious adverse events, did not trigger anti-drug antibodies, and - critically - left coagulation and fibrinolysis untouched. In a first-in-class drug, that kind of uneventful safety readout is not boring. It is the entire point.

None of this arrived overnight. The intellectual engine is Dr. Katerina Akassoglou, who directs the Center for Neurovascular Brain Immunology at UCSF and the Gladstone Institutes, and who spent roughly two decades asking a question the mainstream mostly ignored: what if the blood-brain barrier is where neurodegeneration starts? Therini Bio is, in a real sense, that question turned into a company - a lab's patient, unfashionable work capitalized into a pipeline. It is a reminder that deep foundational research is not a detour from a product; sometimes it is the product, waiting twenty years for its funding round.

Here is the part that is genuinely clever from a strategy standpoint. Therini is not developing THN391 for one disease. It is developing it for two - Alzheimer's disease and diabetic macular edema, a retinal condition that causes vision loss. On paper those are wildly different: one is the brain, one is the eye. But Therini's framing is that they are the same problem wearing different anatomy. Both involve vascular dysfunction and fibrin-driven inflammation. If your drug is defined by the mechanism it fixes rather than the market it lands in, then the brain and the retina are two doors into the same building.

That logic extends to the pipeline's second molecule, THN622, a bispecific antibody that fuses fibrin-targeting with VEGF inhibition - the established standard-of-care mechanism in retinal disease. Two mechanisms, one injection. It is the elegant move that shows up again and again in medicine and in product design: rather than adding your thing next to the incumbent's thing, you merge them into a single molecule and let the physician give one shot instead of two.

For a company of roughly fifteen people, Therini has raised a striking amount - about $145 million in total, including a seed round of $17 million and a combined Series A of $75 million, the latter assembled from a $36 million round in 2023 and a $39 million extension in May 2025. The investor list is the tell. Eli Lilly, Sanofi Ventures, and Merck's MRL Ventures Fund are all in. Those are the venture arms of three pharma giants who compete with each other, and when competitors co-invest in the same small biotech, they are usually not chasing a fad. They are buying an option on a mechanism they cannot afford to be wrong about. Add the Dementia Discovery Fund, SV Health Investors, Dolby Family Ventures, and 2025 newcomers Angelini Ventures and Apollo Health Ventures, and you have a syndicate that reads like a hedge against the possibility that the fibrin hypothesis is right.

What people can actually do with all this is, for now, participate: the near-term users of Therini Bio are the patients in its clinical trials, the clinicians tracking a new mechanism, and the investors and partners underwriting it. The long-term promise - the reason any of this matters - is a class of medicines that treats neurodegeneration and vision loss at a root that current drugs largely skip. Whether THN391 delivers on that is what the Phase 1b trials, now beginning, are built to answer. Until then, Therini Bio remains one of the more legible bets in neuroscience: a company named after a protein, going after that protein, in two organs at once, and being careful not to break the one job that protein is supposed to keep.