Illumina Chases the Grail

Plans for a groundbreaking cancer screen are risky and could take years.

If you want to highlight the combination of extreme ambition and potential futility in a startup, you can do worse than naming it "Grail."

Gene sequencing company Illumina did exactly that with an immodest new spinoff it announced on Sunday and then fleshed out at the JPMorgan health care conference in San Francisco on Monday. With financial backing from Jeff Bezos and Bill Gates, the plan is for Grail to build a test for cancer by sequencing tumor DNA fragments found in blood. This startup's name gives a hint of exactly what investors should think of it -- this could be transformative if it succeeds, but has a high chance of failing to deliver on those high hopes.

As far as risky bets go, this is a solid one for San Diego-based Illumina, which will own the majority of Grail. If it fails, then it's a comparatively modest investment. If it succeeds, then Illumina can expect huge royalty payments from a company that could conceivably be worth more than its parent's current $24 billion market cap.

Excitement over Grail helped Illumina's shares rise on Monday, a day when the Nasdaq Biotech Index touched a 52-week low, though a small beat on 4th quarter earnings estimates may have helped. But investors may be getting ahead of themselves.

Safe Port in a Storm

On an otherwise grim day for health care stocks, investors cheered Illumina's cancer screening moonshot.

Intraday times are displayed in ET.

The potential is certainly huge. Right now, there really isn't a way to screen for cancer in otherwise healthy individuals. Existing technologies aren't as sensitive or specific as doctors would like. Diagnosis often doesn't come until there are symptoms, by which time the cancer is in a later, more dangerous stage. Earlier detection would save many lives. Grail raises the possibility of being able to screen just about everyone every two years or so for cancer.

In Illumina's "base case" for Grail -- where it can screen for tumor DNA in high-risk individuals and find cancer at stage 2 or later -- the market for such testing could be $20 to $40 billion. In the "bull case" -- where the technology could detect cancer at stage 1 and pinpoint a tumor's location in the body -- Illumina predicts Grail would be the "leading cancer company" with a $100 to $200 billion market. Tests would cost between $500 and $1000.  

But there are big risks. Grail will attempt to catch tiny and rare fragments of tumor DNA and use its gene-sequencing prowess to determine the tumor's location. It's the blood-test equivalent of finding a certain needle in a haystack full of fakes. Considering the stakes of a false positive or negative, accuracy is going to be a huge concern. Getting from finding DNA scraps to a definitive and accurate diagnosis is going to be a long, tough road. And there's also the possibility that early-stage detection might lead to costly treatment for slow-progressing or not particularly dangerous cancers. 

Given the amount of work that's left to do, Grail's self-set goal to hit the market in 2019 seems aggressive. 

Illumina has promising early data from lung cancer screenings, and if any firm has the expertise and technology to make this work, it's probably Illumina. It certainly has a superior pedigree to say, Theranos, which has made similarly grand claims about blood screening that are now being questioned.

But grail quests, for ancient cups or equally mythical $200 billion markets, are always longer and harder than people imagine when setting out. 

This column does not necessarily reflect the opinion of Bloomberg LP and its owners.

    To contact the author of this story:
    Max Nisen in New York at

    To contact the editor responsible for this story:
    Mark Gongloff at

    Before it's here, it's on the Bloomberg Terminal.