The drug was designed by an algorithm. In cell tests against live Ebola virus, it outperformed remdesivir by a factor of seven, the company says. That is the claim; the work ahead is proving it holds in animals.
Shares in Decoy Therapeutics closed 21.09% higher at $3.10 after the company published the results, on dollar volume of about $559 million.
For holders, the core question is whether the lab data survives animal testing — and whether government funding will pay for that next stage.
Decoy's AI platform produced the candidate against filoviruses — the viral family that includes Ebola and Marburg — in what the company calls a single optimization cycle. In cell-based assays, the compound hit a potency reading of 1.56 μM against wild-type Ebola Zaire. That measures the concentration needed to block half the virus's activity in a dish; lower means the drug works at a smaller dose. The same compound hit 1.04 μM against Marburg virus.
The company plans to seek approval through the FDA's Animal Rule, a pathway that lets drugs targeting lethal pathogens win regulatory clearance through animal studies rather than human efficacy trials. Decoy says it is pursuing non-dilutive government grants and biodefense partnerships to fund the work — meaning it does not plan to issue new shares to pay for it.
There is a second potential payoff. A successful filovirus approval would make Decoy eligible for a Priority Review Voucher — a certificate that speeds the FDA's review of any later drug application and can be sold to another pharmaceutical company. The company says recent voucher transactions have ranged between $100 million and $180 million.
The program remains at the preclinical stage. No animal data has been published, no government grant has been announced, and the company has not given a timeline. The distance from a cell assay to an approved drug is long, and what sits at the start of that road is an AI-designed molecule that worked in a dish.