September 3, 2026
Biotech Correspondent

Today, a crushing Phase 3 failure for Ultragenyx in Angelman syndrome, a new accelerator from Nobel laureate David Baker to use AI to map nature’s design rules, and more.

The sniffly 5-year-old in my bed is coloring a remarkable green-blue rainbow at present. Visionary, truly.

rare disease

Ultragenyx Angelman drug fails pivotal Phase 3 trial

Ultragenyx said yesterday that its closely watched Angelman syndrome drug failed a Phase 3 trial — a major setback to both families and investors who had hoped the therapy could become the first of many breakthrough treatments for patients with intellectual disabilities, STAT’s Jason Mast writes.

Despite striking results in earlier studies, GTX-102, designed to unsilence the paternal copy of the UBE3A gene in neurons, showed no benefit over a sham treatment in improving cognition, skills, or behavior in children with the disease.

The failure is also a major blow to Ultragenyx, which had been counting on the drug as a path toward profitability. It now plans significant cost cuts. That said, Ionis and Oak Hill Bio are running Phase 3 trials of similar medicines, with Ionis expecting randomized data next year.

Read more.


artificial intelligence

Seattle scientists launch $95 million AI biology effort

The Allen Institute, University of Washington, and Fred Hutch Cancer Center are launching a $95 million effort to use AI to design biology that doesn’t exist in nature. Called AI BioDesign and led scientifically by Nobel laureate David Baker and genome scientist Jay Shendure, the initiative will use AI models to propose new DNA sequences and biological molecules.

The idea, then, is to test those designs at scale in the lab, and feed the results back into the models — a cycle intended to uncover fundamental rules of biology and make biological engineering more predictable. They hope this could prompt new treatments for cancer and neurodegenerative disease — but might even help spur development of enzymes that break down plastics, or even even biological computers. The project plans to openly release its models, datasets, assays for researchers elsewhere.

“For the first time, the speed of AI is beginning to match the experimental power of synthetic biology,” Baker said in a statement. “That changes the question from ‘what has nature already made?’ to ‘what else is posisble, and how can we test it?’”



stem cells

FDA's success at regulating unproven stem cells

Despite health secretary Robert F. Kennedy Jr.’s enthusiasm for unproven stem cell treatments, the FDA has not backed away from policing the sprawling industry of clinics selling unapproved cell therapies.

In his Lab Dish column for STAT, UC Davis stem cell researcher Paul Knoepfler notes two recent wins: a new FDA warning letter to the company R3 Stem Cell and similar firms over marketing unapproved birth tissue cell and exosome products — which includes concerns over manufacturing and sterility. Regulators also won a permanent injunction in court that bars Cell Surgical Network from marketing stromal vascular fraction, an adipose-derived cell preparation, without FDA approval.

The actions are notable given Kennedy’s past claims that the FDA has suppressed stem cell treatments — and fears have swirled that HHS under his oversight might interfere with enforcement. Instead, CBER has continued issuing warning letters and defending its authority over unproven cell products.

“Now more than ever, the FDA deserves credit when it’s doing things well,” Knoepfler writes, “and in these two recent cases, a hat tip is in order.”

Read more.


M&A

Why hasn't Abivax been swept up by the M&A tide?

Biotech M&A is booming, but Abivax — long considered one of the sector’s most obvious acquisition targets — remains conspicuously single. The French biotech reported strong Phase 3 data for its ulcerative colitis drug obefazimod only eight weeks ago, STAT’s Adam Feuerstein notes in his weekly Biotech Scorecard column. The results eased concerns about a potential risk for cancer — and some investors say a deal could reasonably emerge between now and the J.P. Morgan Healthcare Conference in January.

Abivax shares are down 17% this year even though the XBI has climbed 34%. Investors hoping for a roughly $20 billion deal are getting impatient.

“Without visibility into the actual sale process, there is no way to pinpoint exactly what’s happening between Abivax and its Big Pharma suitors,” Adam writes. “We can guess, of course, starting with the most facile explanation I’ve heard: Abivax is French and the French don’t work in August.”

Read more.


More around STAT

More reads

  • FDA pilot offers generative AI medical devices a path to patients before they are authorized, STAT

  • Owkin signs AI drug discovery deal with Boehringer for cancer, immunology, Reuters



Thanks for reading! Until tomorrow,