It’s a new test that is said to detect multiple forms of cancer early. What’s not to like?
The Galleri Multi-Cancer Early Detection (MCED) test is a relatively new cancer screening test that its manufacturers say can detect a cancer signal from 50 types of cancer through a simple blood test. The test, according to the manufacturer, Grail, can detect types of cancers that currently have no recommended screening tests. It can also alert you to hard-to-detect, aggressive and often fatal types of cancer like pancreatic, ovarian and esophageal cancers, says Grail, a unit of the gene-sequencing company Illumina.
It costs $949, and health insurance doesn’t cover it.
The Wall Street Journal headlined its story last October: “This $1,000 test finds signs of cancer in your blood. The Galleri test is designed to detect more than 50 cancers.” Only then does the headline inject a note of caution: “Doctors are split on whether it is worth the risks for patients.”
Some systems are advertising it, like the Mercy Center for Performance Medicine in St. Louis, Mo., and of course the Galleri marketing is full of happy customers with testimonials — either an early cancer was detected, or none was found, giving peace of mind. Tony Robbins, the life coach, has a testimonial on his site. Commentary on the Wall Street Journal’s site explores the topic and comes out in favor. Longevity doctors offer it frequently.
Many medical professionals, though, are much more reluctant to say this is a good idea. “I Am Afraid of Early Cancer Detection” is the title of a three-part Substack series by internist Adam Cifu. A French scientist discussed her reservations in Medscape. British scientists writing in The Lancet about Galleri’s testing in the United Kingdom wrote about “the adoption of poorly evaluated interventions that might be of little or no benefit, harm people, and waste resources that could be better used elsewhere.” A number of rival tests are reportedly in development.
I wrote recently about the whole-body “preventive” MRI and CT scan phenomenon, which are marketed as early detection strategies but that earned not a single approving comment from the medical and healthcare professionals I interviewed.
False positives
So what about Galleri? I asked Al Lewis, CEO of the Quizzify employee health education company and author of a textbook on population health management and vendor math. He wrote: “I’ve looked into it pretty carefully and even if price were no object, it wouldn’t be worth considering. And price is an object. Not just the price of Galleri itself, but doing this for employees commits you to the prizes of all the followup as well.
“Most of the positives will be false, too. You’re looking for some 1-in-100,000 incidence rate cancers with Galleri. Assume the test is 99.5% accurate. So 500 of those 100,000 will test positive, but only 1 will have cancer. Nonetheless, the other 499 will have to endure future expensive testing and stress.
“It also misses some cancers too.”
Galleri got in hot water last year when it incorrectly informed people they had positive tests. “The company, Grail, said in an emailed statement on Sunday that a vendor it works with had sent hundreds of letters with incorrect test results because of a ‘software configuration issue’ that has since been resolved,” The New York Times reported. Yet its marketing has numerous stories of people who think their early cancers were caught by the Galleri test.
Currently, testing for common cancers like breast, colorectal, cervical and prostate cancers are covered by Medicare and other insurers, and recommended by the U.S. Preventive Services Task Force at certain ages. Grail is reportedly seeking legislation to let Medicare cover Grail and similar multi-cancer detection tests if they are approved by the Food and Drug Administration.
Similar skepticism
Dr. Blake Long, a pediatric cardiologist who is chief medical officer of Alva10, a consulting group working to bring diagnostics to the forefront of precision medicine, was similarly skeptical about the Galleri test.
“The upside is the potential ability to detect cancer at an earlier stage,” he said in a Zoom interview. “The downsides are the false positive rate. And then the evaluation of patients that they have to undergo because they have suspected cancer when they actually do not.
“Second is overdiagnosis, where you detect potential cancer, but that cancer was not going to really cause problems for you, or would have been detected using standard detection at an early stage. So catching it early really doesn’t change the outcome for the patient, and potentially exposes them to treatments earlier with toxicities when they may not have really needed them. The third downside is cost. The tests are roughly $1,000. The question becomes ‘Who do you test? Starting at what age and how often?’ If you were to think about it, it’s starting at age 50 every year. That’s a lot of tests and a lot of money.”
He pointed out that the test is not approved by the Food and Drug Administration; were it to be approved, he said, it might be covered by Medicare and commercial insurance. But as it stands now, it’s testing only people who can pay out of pocket, which excludes a lot of people.
How does the science work?
“It measures DNA in the bloodstream,” he said, “commonly termed cell-free DNA or circulating tumor DNA — CFDNA or CTDNA. If one of those tumor cells dies, the DNA that’s inside that will get released into the bloodstream in very, very tiny quantities. The technology has the ability to actually detect that in very small amounts and pull it out and say ‘it’s not your DNA.’
“Then they determine if it’s cancer or not. There’s some very sophisticated testing and algorithms and they look at not only the DNA, but things about the DNA, not just the sequence, but what’s called the methylation, which is a covering which turns genes on and off — sort of the switches for genes — and also proteins that are attached to it. And so all those things together help the test to say ‘what I’m seeing is coming from a tumor, not from normal tissue.'”
But does it work consistently? He said different companies have different technology to assess what they see.
“But one key questions is, wow, these are very tiny amounts of tumors circulating, because you don’t have symptoms. That’s kind of the idea — one of the challenges is detecting the tumor at an early stage.
“So it turns out the tests are not very sensitive at picking up tumors at an early stage. In their current format, they are better at picking up later stage. And that intuitively makes sense, because an early-stage tumor may not be shedding nearly as much DNA into the bloodstream as a late-stage tumor, meaning it’s already metastatic, it’s in multiple places in the body and may be larger. So while we want the test to detect early stage, it needs to be very sensitive, and the tumor may not be shedding enough at an early stage to easily detect that. We suspect that will get better over time. But that’s currently where we are.
“The second piece is that some tumors don’t shed a lot of DNA and other tumors shed a lot, regardless of whether they’re early stage or late stage. So one of the challenges is, depending on what kind of tumor cancer you’re talking about how good these tests are, varies both early stage and late stage.”
The current widely available Cologuard test, a stool test for colorectal cancer, has the same issue, he said.
“It’s detecting little pieces of DNA along with blood. Same problem, the tumor has to be a little bit farther along, or to shed enough for Cologuard to pick it up.” (While Cologuard is a stool based test, not a blood based test, he noted, it’s somewhat the same principle.)
“So it’s not really good as good at picking up really early tumors, or what we call pre-cancer lesions or adenomas in the colon.”
Maybe a targeted approach
So false negatives — not detecting early stage disease, because it can’t pick it up — and false positives — it picks up something but it’s actually not cancer — are serious issues, he said. Screening tests in general, Cologuard and mammography, have false positive rates of 10% to 15%, he said. Data on false positive and false negative tests for Galleri are limited, he said, because the studies aren’t very big.
But looking at the numbers, he said, “If you tested 10,000 patients, roughly 140 would have a positive test, 50 would actually have cancer and 90 would have to get a big workup.”
Instead of using such a test on a wide swath of the population, he said, perhaps a targeted approach would make sense.
“We’re seeing an increasing incidence of early onset cancer, particularly colorectal cancer. And there, there’s a lot of research being done about why that might be — is it diet, environmental exposure? There’s some evidence to suggest that obesity is a big driver of early onset incidence. It’s not a cause yet, it’s just a correlation. But it’s a thought.
“So then the question becomes, the screening trigger for age has usually been much higher, but it is now starting to move younger, because we are trying to catch this. But would use of a Multi Cancer Early Detection test be more useful if you did it in a risk-stratified population? Meaning you have a risk factor like obesity, and therefore you would more likely be a positive test and really have cancer with this, than an average patient who doesn’t have many risk factors, and is a little bit older. So we’re trying to think about it as a test that may have more value for outcomes and utility by using it in select populations, rather than the general population. That needs to be evaluated and thought through, and there’s a lot of data that will need to be collected to demonstrate that, but that’s where we think the utility will be.”
