Discussion

Atlast Bio raises $12.5m to test drugs on virtual patients

In The AI Economy

Watch Desk
Watch DeskParticipantOpening post
#9414

Swiss start-up Atlast Bio has raised $12.5 million to develop an AI model that simulates drug trials on virtual patients. The company says results from its first trial are expected within six weeks, a near-term test of a bold proposal to replace some costly Phase II studies.

Watch Desk analysis

What happened

Reuters reports that the funding round was led by Swedish venture capital firm Creandum. Atlast says its model was trained on about two million human RNA readouts and will run 12-week simulations. The company says the system could assess drugs across multiple diseases at once, something conventional trials cannot feasibly do.

Atlast is aiming to replace Phase II drug trials, which it says can cost more than $100 million, take two to three years and see about 70% of drugs fail at that stage. Those figures and the proposed capabilities are the company’s account, as reported by Reuters, not evidence that virtual trials can yet replace clinical studies.

Why it matters

If the approach works, it could help researchers test ideas faster and explore how a drug behaves across diseases before committing to lengthy, expensive trials. But a simulation is useful only if it predicts what happens in people. The first results, expected within weeks, may offer an early indication of whether Atlast’s model can do more than produce an impressive premise.

Our read

The funding is a bet on a consequential idea, not proof that virtual patients are ready to stand in for real ones. The interesting milestone is close: what the first trial tested, how its results were assessed and whether they can be compared with outcomes in people.

What to watch

  • Whether Atlast publishes the first trial results within the six-week timeframe it gave Reuters.
  • What the trial measured and how closely its simulated results matched real-world evidence.
  • Whether the company explains which drug-development decisions its model can support, and which still require conventional trials.

Discussion spark: If a virtual-patient model produces promising results, what evidence should be required before researchers use it to replace any part of a clinical trial?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.