Multi-Cancer Blood Test: Does Grail Galleri Work to Spot Cancer Early?

Is a $950 test that searches for upwards of 50 types of cancer in your bloodstream worth it? That depends on who you ask.

Grail, the Silicon Valley healthcare company developing blood tests to pick up early cancer, presented its largest study to date on Saturday at the American Society of Clinical Oncology meeting (ASCO) in Chicago.

The study, tracking more than 140,000 older people in the UK, assessed whether people who took Grail’s $950 Galleri test each year for three years could catch cancer before it hit the later stages (Stage 3 and Stage 4).

The study was unsuccessful at its stated goal of meaningfully curbing more late-stage cancers. Still, the company says these results are a clear signal that their test is working.

“When you add Galleri to single cancer, standard-of-care screening, we found four times more cancer in the population,” Grail President and incoming CEO Joshua Ofman told Business Insider, shortly before the results were released.

The fresh data from ASCO shows the sticking point: while an additional 937 were picked up by Grail’s Galleri, the test is still not the best at detecting early-stage cancer, the kind that is generally easier to treat and cure. While Grail’s testing did pick up some new cancer cases, it missed a lot more, and the majority of those it did find — 54% — were in Stages 3 and 4.

Grail’s test is not yet approved by the US Food and Drug Administration, but it is available across the country with a doctor’s prescription, or through telehealth services like Hims & Hers.

Business Insider has zeroed in on five key data points to keep in mind as you weigh whether the test is worth your time and your money:

1: If Grail says you have cancer, you probably do.

The test is designed with high specificity, so that it doesn’t scare people into thinking they have cancer when they don’t. More than 70,000 people took this test, and the odds of any individual without cancer getting a false positive test were very low, at less than 0.5%.

But there were at least a few false alarms in the crowd. 864 people were given a cancer “signal” who were not diagnosed with cancer afterwards. That’s hundreds of people who worried about cancer alongside their doctors, and ultimately couldn’t find anything wrong.

Grail’s founding CEO Jeff Huber said that may not mean the test failed. Previous research has shown that about a third of Grail’s “false positive” tests, when followed over longer periods of time, do become cancer diagnoses that doctors simply couldn’t identify right away.

“When you get a Grail test result, and it says you have cancer, you can be highly, highly, highly confident you have cancer,” Huber said. “If Grail says you have cancer, unfortunately, you should freak out and do something about it as quickly as you can.”

Dr. Scott Ramsey, who directs the Institute for Cancer Outcomes Research at Fred Hutch Cancer Center, and who was not involved in this study, agreed with Huber.

“If I did Grail and I had a positive test, I would be worried, I would be looking for cancer,” he said.

2: Grail missed 2,114 cancers

Getting a negative Grail test does not mean you’re in the clear.

In the study, 2,114 people with “negative” Grail test results were diagnosed with cancer in the following 12 months, through traditional oncology methods like regular screenings and biopsies.

This means Grail’s test is not yet a viable safety net — it’s missing twice as many cancers as it is finding, with a specificity around 30%.

“If it’s a negative test, you should not feel safe because they are not very sensitive,” Ramsey said.

3: The test caught most cancers before they were terminal.

People who took Grail had fewer Stage 4 diagnoses, and more cancer cases were picked up in Stage 3. Grail says that’s a good thing, because it means doctors can catch cancer before it progresses to the metastatic, and often terminal stage. Ideally, that should mean Grail helps make cancers easier to treat, and patients can live longer lives. But that’s all speculation for now.

“This is where it gets so complicated,” Ramsey said. “In theory, and intuitively, detecting cancer in an early, curable stage is what you want to do because it should work. But the devil ends up being in the details because most of these cancers that Grail is screening for — the ones that aren’t screened for now — we don’t know the behavior of those cancers.”

In general Stage 1 and Stage 2 cancers are the easiest to treat, then things get a little more challenging in Stage 3, depending on the cancer, and in Stage 4, survival is generally very low. But that’s not universally true.

“Stage 4 to Stage 3? For some of these cancers, that might be a good thing, but for others it may make no difference at all,” Ramsey said.

4: The test appears best at identifying a few specific early-stage cancers

After three years of running Grail, Stage 1 and 2 cancer detection improved by 16%.

For some cancers, like lung cancer and bladder cancer, there wasn’t a huge surge in detection after three annual tests. But for some that cannot be routinely screened for, the test really made a dent in early detection:

• Ovarian cancers: Grail doubled the number of Stage 1 and 2 cases detected in the study
• Myeloma/plasma cell neoplasm: There was a 118.2% increase in Stage 1 and 2 diagnoses
• Esophagus cancers: And a 92.3% increase in Stage 1 and 2 diagnosed

“Those are horrible cancers that kill you when they’re detected late,” Huber, the former Grail executive, said.

5: Should you do it? Maybe as an add-on

Ultimately, none of these new figures tells cancer researchers what they really want to know: can Grail testing save lives?

In the meantime, patients are left to decide on their own if the testing is worth adding to their annual medical routine.

When friends ask Ramsey if they should do Grail, he tells them to keep two things in mind: “I say they’re very promising tests that I think someday will be part of screening. But right now, I don’t have any data to tell them that getting this test will reduce their risk of dying from cancer. And that’s the whole reason you do screening.”

Huber, a true believer in this technology, says he gets a Grail test every year. But he also gets regular full-body MRIs and colonoscopies. He believes the Galleri test should be part of comprehensive clinical care. But it will be some time before we can tell whether that extra vigilance saves lives.