When Xaira Therapeutics debuted in 2024 with a gobsmacking $1 billion in funding, a natural expectation would have been for a flurry of further activity to follow. Instead, the closely watched AI biotech went quiet.
Now, Xaira has divulged the first details of the pipeline it’s been building, alongside a clearer picture of its generative AI antibody design platform, dubbed X-Design.
“X-Design is central,” Xaira CEO Marc Tessier-Lavigne, Ph.D., told Fierce in an interview. “It's been a major focus for us to develop a model that can design antibodies that are drug-ready from the start, and to then use that to fuel a pipeline of therapeutics.”
X-Design stands alongside other models developed by the company, X-Cell and X-Patient, which together are meant to establish Xaira as a full-stack drug-making machine.
“We're tackling all three stages of drug discovery: the process of identifying targets; making the drugs themselves; and also identifying the patients most likely to respond,” Tessier-Lavigne said. “We've been taking our time.”
“For decades, some of the proteins that drive disease have been off-limits to antibody drugs because traditional discovery approaches failed on them,” David Baker, Ph.D., Nobel Laureate protein designer and co-founder of Xaira, said in a statement. “X-Design shows that AI is beginning to design those antibodies from scratch. This changes what's possible in drug discovery.”
In a blog post describing X-Design, Xaira has also for the first time shared information on two of its pipeline prospects, XA-1 and XA-4.
Though the targets of each are not being disclosed, XA-1 is a cancer candidate that was created from scratch in just seven weeks, the CEO said. XA-4 is meant to drug a tricky G-protein coupled receptor that was otherwise “seemingly intractable” to target.
Both drugs are for targets that Xaira has “great conviction” in, Tessier-Lavigne explained, “but where prior efforts to make antibodies have failed.”
In a past conversation with Fierce, Jeff Jonker—then Xaira’s chief operating officer and now an advisor to the company—said the biotech was working in immunology and inflammatory diseases.
Xaira plans to advance both candidates into human studies, but the chief exec declined to share any details on timing. Other preclinical drug candidates, aptly named XA-2 and XA-3, also populate Xaira’s budding pipeline.
The AI biotech may ultimately decide to partner some of its assets with larger biopharma players, and is also actively in discussions about teaming up to turn its antibody design platform against targets of interest to other companies.
Tessier-Lavigne, who formerly served as president of Stanford University and chief scientific officer at Genentech, remained tight-lipped about details of those conversations as well as about the current status of Xaira’s bank account. One billion dollars is a lot of cash to burn, but AI models are famously expensive to build and train.
“We wanted to capitalize the company so as not to have to go back to the capital markets until we had molecules in the clinic,” the CEO told Fierce. But, he caveated, he is open to seeking more money sooner “if there's appropriate investor interest or the opportunity to accelerate or extend our programs.”
Though the field of AI antibody design is relatively new, competition is rapidly heating up. BigHat Biosciences, Absci, Generate:Biomedicines and Roche subsidiary Chugai, to name a few, are all in the game of using AI to design or optimize antibodies. Tessier-Lavigne thinks X-Design’s focus on properties that turn a molecule into a medicine stands out from the crowd.
“Most AI design work stops at finding hits, molecules that bind,” he explained. “What sets X-Design apart is that it designs for drug readiness from the start.”
While generating new antibodies is nice, and doing it quickly is even better, Tessier-Lavigne is also keeping Xaira squarely focused on the ultimate goal: helping patients, especially those who currently have no options.
“Despite a huge amount of activity in the industry, we still remain focused on a relatively narrow slice of disease space,” he said. “These AI models—for targets, for patients—are going to help unlock that potential as well.”
The role of AI in drug development is one of the hot topics being discussed at Fierce Biotech's high-impact event in London next month. Find out more information and register by clicking on the banner below:
