A vaccine that protected every vaccinated monkey in 2011 against a strain of Ebola now killing hundreds in Central and East Africa sits unused because the system that funds vaccines for wealthy nations has no mechanism for diseases that hit poor countries first.
In 2011, three crab-eating macaques at a University of Texas Medical Branch lab were vaccinated against the Bundibugyo strain of Ebola, a lesser-known cousin of the virus that killed more than 11,000 people in West Africa between 2013 and 2016. All three survived. Roughly two-thirds of their unvaccinated companions died. Fifteen years later, the candidate vaccine that produced that result has never been injected into a human arm, never opened an investigational new drug application, and never attracted a pharmaceutical partner. It is, in the words of the scientist who developed it, "sitting on the shelf."
The shelf is not empty by accident. Thomas Geisbert, a UTMB immunologist, built the rVSV platform (a harmless livestock virus repurposed as a delivery vehicle for the vaccine) and has watched Ebola vaccines move from bench to bedside exactly once. Merck's Ervebo, the first licensed Ebola vaccine, was rushed through trials during the 2013 to 2016 epidemic and deployed in ring vaccination in the Democratic Republic of the Congo. Ervebo was built partly on Geisbert's earlier work and helped earn him a spot on Time's 2014 "Ebola Fighters" Person of the Year, as recounted in the WIRED feature on his shelved candidate. What Ervebo was not built to do was stop Bundibugyo. It targets the Zaire strain, not the one now spreading in the DRC and Uganda.
That gap is the story. The same WIRED feature frames Geisbert's candidate as the most promising vaccine against the current Bundibugyo outbreak, attributing that assessment to the World Health Organization. WHO's Alert and Response page for Ebola disease (Bundibugyo virus), which draws on surveillance data from the DRC's COUSP, confirms active outbreak response. The "most promising" label itself is not independently confirmed in the readability-extracted WHO content and should be read as WIRED-attributed until WHO publishes a direct statement.
The science is not what is stuck. A 2009 study from Geisbert's group showed the rVSV platform protecting nonhuman primates against Marburg virus and three species of Ebola. A 2023 post-exposure study led by Courtney Woolsey at UTMB, in which Geisbert is a coauthor, found that most vaccinated macaques were protected from Bundibugyo even after exposure. The 2023 trial's post-exposure window was roughly 20 minutes, an unrealistically short interval for real-world use, but the ring vaccination mechanism it supports is the same one that helped contain the West African epidemic. WHO has separately noted the success of a similar rVSV-based vaccine against the Sudan ebolavirus strain in a 2025 ring vaccination trial, evidence that the platform works for filoviruses other than Zaire.
What is stuck is funding. CEPI, the Coalition for Epidemic Preparedness Innovations, has offered up to $3.2 million to prepare the manufacturing material needed for the first steps toward human trials of Geisbert's candidate, according to Rachael Bonawitz, CEPI's filovirus disease programme lead, as reported in WIRED. CEPI has also pointed to "extensive safety data and prior regulatory experience" from the licensed rVSV-based Zaire vaccine as a potential expedited pathway. Moderna and the University of Oxford have separate Bundibugyo candidates in earlier-stage development. None of these efforts has a buyer.
That is what the market produces. Vaccines are developed by companies that recover costs through sales. A Bundibugyo vaccine has no high-income market. The customers are the same low-income countries now reporting cases, and the use case is sporadic outbreaks that may or may not arrive in any given decade. Geisbert and Woolsey both describe a candidate that survived a 2011 primate challenge and produced a 2023 post-exposure result, and is still waiting for an entity with a balance sheet to take it through Phase 1. The 15-year shelf is the predictable output of a system designed to fund products for paying customers, applied to a product whose customers cannot pay.
There is a second, smaller bottleneck the WIRED piece surfaces: scientists have not yet obtained a live Bundibugyo virus sample to test the candidate against, because of stretched resources in the DRC and the logistical complexity of shipping refrigerated blood. Even with full funding, the candidate has not been tested in vivo against the virus now circulating. "Most promising" remains preclinical optimism, not proven protection. The current WHO epidemiological update confirms active surveillance. The precise case and death counts circulating in coverage, "hundreds infected" and "around 200 dead," are WIRED's rounded framing and should be confirmed against a WHO Disease Outbreak News bulletin before being repeated as hard numbers.
A system that took this seriously would look familiar from COVID-19: advance purchase commitments that let manufacturers plan Phase 2 and 3 capacity before an outbreak starts, sustained Phase 1 and 2 funding pools for neglected-strain candidates, and platform trial infrastructure that can pivot from one filovirus to the next when surveillance flags a new cluster. CEPI's $3.2 million is closer to a feasibility study budget than to the cost of taking a candidate through to licensure, and the agency has acknowledged as much. Geisbert's vaccine is not stuck because the science is hard. It is stuck because the only entities that fund Phase 1 trials for filovirus vaccines want a paying market, and the customers for a Bundibugyo vaccine are exactly the countries that cannot pay.