

Astellas Pharma killed the gene therapy it paid $3 billion to acquire after four patient deaths and years of setbacks. Its replacement uses 100-fold lower doses and a smarter delivery system, but the real test is just beginning.
Four dead patients. A clinical hold from the FDA. Years of development gone sideways.
Astellas Pharma just pulled the plug on AT132, a gene therapy for X-linked myotubular myopathy (XLMTM), a rare muscle disorder that primarily affects young boys. The therapy was the crown jewel of Audentes Therapeutics, which Astellas bought in 2019 for $3 billion. That crown jewel turned out to be cubic zirconia.
But Astellas isn't giving up on the disease. It's swapping AT132 for a newer, sleeker replacement called ASP2957. Think of it like trading in a car that keeps catching fire for a model that (hopefully) doesn't.
AT132 was supposed to deliver a working copy of the gene these boys are missing. Simple concept; brutal execution.
In the ASPIRO trial, three patients died from liver failure after receiving a high dose in 2020. The FDA slapped a clinical hold on the program in June 2020. Astellas eventually got permission to resume at a lower dose, but then a fourth patient died in September 2021. The company voluntarily paused screening and dosing again, this time because liver problems showed up even at the lowest tested dose.
The therapy's Achilles heel was its delivery vehicle: an AAV (adeno-associated virus) that wasn't picky about where it went. It was supposed to target muscle tissue, but it kept accumulating in the liver with devastating consequences.
ASP2957, licensed from Kate Therapeutics in 2023, uses a next-generation AAV capsid designed specifically to seek out muscle and avoid the liver. In preclinical studies, it showed high muscle specificity and reduced liver targeting, which is a fancy way of saying it goes where it's supposed to go.
The practical result? Doctors can use a dose roughly 100-fold lower than what AT132 required. That's not a minor tweak; it's a fundamentally different safety profile. Less virus floating around means less chance of the liver problems that killed those patients.

Janux Therapeutics just axed its EGFR-targeted T-cell engager after disappointing Phase 1 results, joining a growing graveyard of solid tumor programs. The company still has a promising prostate cancer asset and an $800 million BMS deal, but the bigger question looms: can anyone crack the solid tumor code?


Join thousands of biotech professionals who start their day with our free, daily briefing.
ASP2957 is currently in a phase 1/2 open-label trial with enrollment underway. For the boys with XLMTM still waiting for a treatment, it represents the next realistic shot.
Astellas didn't stop at gene therapy. The company also terminated two early-stage programs as part of a broader housecleaning:
Both were killed for business reasons, not safety concerns. They simply didn't clear the bar anymore.
And these three cuts are just the visible tip. Astellas has terminated numerous clinical-stage programs as part of its R&D restructuring. That's not trimming the hedges; that's ripping out the garden and replanting.
Astellas isn't alone here. The entire biopharma industry contracted for the first time in 30 years, with the global R&D pipeline dropping 3.9% from about 23,875 drugs at the start of 2025 to 22,940 at the start of 2026.
Roche narrowed its focus to just three core therapeutic areas. Pfizer trimmed from 271 to 257 programs. Novartis cut about 10% of its pipeline. Everyone is playing the same game: stop spreading resources thin across long-shot bets, and concentrate firepower on programs most likely to actually work.
For Astellas specifically, the math is urgent. Its blockbuster prostate cancer drug Xtandi faces patent cliffs after 2027. The company needs new revenue sources, and it can't afford to waste R&D dollars on programs with single-digit odds of success.
CEO Naoki Okamura has laid out a three-phase plan running through 2034. The company is reorganizing around four "Primary Focus" areas: Cell and Gene Therapy, Vision Loss/Ophthalmology, Immuno-Oncology, and a new addition called Targeted Protein Degradation (essentially drugs that tag unwanted proteins for destruction by the cell's own recycling machinery).
R&D spending is set to increase significantly, with the company targeting a 30% R&D spending ratio by FY2027. That's a big number, and it signals that Astellas isn't cutting for austerity; it's cutting for precision. The company wants to be a sniper, not a shotgunner.
The honest answer: sometimes.
Pharma companies love to announce "strategic portfolio rationalization" (translation: we're killing the stuff that isn't working). The question is always whether the remaining programs are genuinely better, or whether the company is just rearranging deck chairs.
In Astellas' case, the AT132-to-ASP2957 swap has real scientific logic behind it. A 100-fold dose reduction isn't marketing spin; it's a meaningful safety improvement rooted in better capsid engineering. The preclinical data suggests Kate Therapeutics solved the targeting problem that doomed AT132.
But preclinical data is just a promissory note. The phase 1/2 trial will tell us whether ASP2957 delivers on that promise, or whether XLMTM continues to be one of biotech's most heartbreaking dead ends.
For the families waiting, Plan B needs to be better than Plan A. The science says it should be. Now Astellas has to prove it in actual patients.
Two clinical-stage biotechs filed for IPOs on the same day, both targeting over $200 million. After a brutal drought that saw only 11 biotech listings in 2025, the twin filings may signal the public markets are finally warming up again.