

ARPA-H just handed $160 million to seven teams tasked with turning custom gene editing from a $4 million luxury into a scalable platform. The targets: rare childhood diseases. The timeline: patients dosed within three years.
A single baby changed everything.
Sometime in the last couple of years, doctors at the Children's Hospital of Philadelphia used a custom-built CRISPR therapy to treat an infant named KJ, who had a devastating metabolic disorder called CPS1 deficiency. It was a one-patient, one-therapy, one-shot deal. And it worked well enough to spark a much bigger question: what if we could do this for thousands of rare diseases, not just one?
Now the federal government is putting serious money behind that question. ARPA-H, the Pentagon-style research agency but for health, just launched a $160 million, five-year program called THRIVE (Treating Hereditary Rare Diseases with In Vivo Precision Genetic Medicines). Its mission is audacious: build reusable gene-editing platforms that can crank out personalized cures for rare childhood diseases the way a software company ships updates.
Seven teams got the call. And the implications reach far beyond rare disease.
To understand why THRIVE matters, you need to understand the ugly economics of gene therapy right now.
The most expensive approved treatment in the U.S. is Lenmeldy, a gene therapy for a rare brain disease. Price tag: $4.25 million per patient. It's not alone in the stratosphere. Hemgenix for hemophilia B runs $3.5 million. Elevidys for Duchenne muscular dystrophy costs $3.2 million. Even Casgevy, the landmark CRISPR-based sickle cell therapy, lists at $2.2 million.
These prices aren't just greed (though there's a debate to be had). They reflect a brutal math problem. Each therapy requires its own manufacturing run, its own quality checks, its own regulatory package. Think of it like commissioning a custom oil painting every time someone needs a new poster. The artistry is incredible, but the process doesn't scale.
For truly personalized therapies (ones designed for a single patient's specific mutation), the economics get even worse. There's no batch production. No economies of scale. Every dose is a bespoke product built from scratch.

Corcept Therapeutics just got FDA approval for the first drug that blocks cortisol from shielding tumors against chemotherapy. The approval came nearly four months early, and it could reshape treatment for the thousands of ovarian cancer patients who've run out of options.


Join thousands of biotech professionals who start their day with our free, daily briefing.
THRIVE's bet is that this doesn't have to be the model.
ARPA-H is explicitly trying to kill the "artisanal editor" approach. Instead of building each therapy from the ground up, THRIVE wants teams to create platforms: standardized toolkits where the gene-editing machinery, delivery system, and safety testing are shared across many diseases. Only the targeting instructions change from patient to patient.
The analogy ARPA-H seems to be reaching for is software. Imagine the editing platform as an operating system and each disease-specific therapy as an app. You don't rebuild the OS every time you want a new app. You just write new code that runs on the same infrastructure.
The milestones are aggressive. By year one, teams must show editing platforms that can generate multiple drug products sharing the same safety and distribution profiles. By year three, they need to be dosing actual patients in first-in-human trials. By year five, those trials should be expanding to cover additional diseases while proving the whole thing can work in real-world clinics, not just ivory-tower hospitals.
Program manager Mimi Lee, M.D., Ph.D. has described a vision where these therapies reach patients through regional treatment centers and virtual clinics, turning gene editing from a destination procedure into something closer to routine care.
The seven teams selected read like a rare-disease Avengers roster:
Each team covers a different organ system, but they all share one requirement: build a platform, not just a product.
There's a quiet subplot here that might matter more than the money.
THRIVE is launching in tandem with what insiders describe as a forthcoming FDA regulatory pathway for individualized gene-editing therapies. The teams behind KJ's treatment have confirmed active discussions with both FDA and ARPA-H about how to approve gene-editing platforms rather than individual products, one painful application at a time.
This is a big deal. Today, getting a gene therapy approved means navigating the full regulatory gauntlet for each product. If the FDA creates a pathway where a validated platform can add new disease targets under an "umbrella" approval structure, the cost and time to reach patients could drop dramatically.
THRIVE teams are required to pursue exactly this kind of umbrella approach; clinical and regulatory innovation isn't optional, it's mandatory.
ARPA-H isn't working alone. It has a companion program called GIVE (Genetic Medicines and Individualized Manufacturing for Everyone) focused on building compact, automated manufacturing units that could sit inside hospitals. Think of THRIVE as the recipe book and GIVE as the kitchen.
Meanwhile, NIH continues to be the heavyweight funder of CRISPR research, pouring roughly $2.4 billion into CRISPR-related projects in fiscal year 2025 alone. That's the basic science engine. ARPA-H's $160 million is smaller by comparison, but it's pointed at a different problem: not discovery, but deployment.
The timing is notable, too. Federal research budgets face political headwinds, with proposals to cut NIH funding by as much as 40%. In that environment, ARPA-H positioning gene editing as strategic national technology (complete with "first-shoring" language about keeping manufacturing in the U.S.) looks like a savvy survival move.
Right now, personalized gene editing is like early commercial aviation: technically possible, breathtakingly expensive, and available to almost nobody. THRIVE is a bet that the right platforms, paired with smarter regulation and distributed manufacturing, can turn it into something more like commercial broadband: widespread, affordable, and life-changing.
Five years and $160 million will tell us if that bet pays off. For the thousands of families watching their children battle rare genetic diseases, the stakes couldn't be higher.
AstraZeneca and Ionis' gene-silencing drug eplontersen did everything right biologically in a massive ATTR cardiomyopathy trial, yet still failed to keep patients alive or out of the hospital. The reason why could reshape how the entire industry thinks about combination therapy.