Skeptoid #1063: Multi-Cancer Detection: Panacea or Problem?

The Promise of Multi Cancer Detection

0:00All of us want to live long and healthy lives, and nobody looks forward to getting some painful and deadly disease later in life, especially earlier in life. So when we hear that there's a new testing technology that we can buy over the internet without even visiting a doctor, and that promises to catch any of up to 50 deadly cancers earlier than conventional testing ever could, we tend to perk up and listen. And we might be sorely tempted to reach for our credit cards.

0:34And in our extended content for premium members, a fun mnemonic to help you remember the crucial epidemiological terms from today's episode. And if you happen to be an epidemiologist, you probably already know the terms I mean. All of that and more is coming up right now on Skeptoid.

0:56Hey everyone, Brian here. This December, join me for another great Skeptoid adventure. This one to New Orleans, where we'll take a deep dive into the history, pseudo-history, myths and legends of one of America's most storied cities. We'll explore French Quarter ghost, voodoo, and vampire lore with local expert Chris Purley, and walk through the historic St. Lewis Cemetery number one. Did you catch my recent episode on the legendary Honey Island Swamp Monster?

1:28We'll be heading out to tour that very swamp, one of the most pristine and least altered river swamps in the U.S. Perhaps we'll catch a glimpse. We'll also catch live jazz on Bourbon Street, run a Bourbon Street scavenger hunt, tour the historic mansions of the Garden District, and stop in at the Mardi Gras Museum, the Pharmacy Museum, an oddities shop, and much more. And of course, we'll eat real Creole and Cajun cuisine, plus beignets and café au lait.

2:00It's a full weekend of true stories, tall tales, and the city that inspired both. Get the full itinerary and book your spot now at Skeptoid.com slash adventures. This show is supported by Odoo. When you buy business software from lots of vendors, the costs add up and it gets complicated and confusing. Odoo solves this. It's a single company that sells a suite of enterprise apps that handles everything from accounting to inventory to sales.

2:31Odoo is all connected on a single platform in a simple and affordable way. You can save money without missing out on the features you need. Check out Odoo at O-D-O-O dot com. That's O-D-O-O dot com. You're listening to Skeptoid. I'm Brian Dunning from Skeptoid dot com. Multi-cancer detection. Panacea or problem. Welcome to the show that separates fact from fiction, science from pseudoscience, real history from fake history, and helps us all make better life decisions by knowing what's real and what's not.

3:06If you've been on the internet at all, you've probably heard of multi-cancer detection, MCD testing, also called multi-cancer early detection. The basic promise of these tests is one that almost sounds too good to be true. From a single blood test, you can be tested for virtually every cancer we know how to test for, and nobody wants to have cancer if there might be a way to catch it earlier than ever possible before, and especially if it's a rare cancer that there's no chance you would have otherwise been screened for.

3:39The result is that marketers have jumped all over this, and now MCD tests are available from any number of online direct-to-consumer DTC pharmaceutical websites, right alongside Viagra, Propecia, and the latest GLP-1 weight loss drugs. Yours for online delivery without ever having to go to a doctor's office. The concept behind how these MCD tests work is fairly straightforward.

4:09When any cell in your body dies, it sheds tiny fragments of DNA into your bloodstream. This includes cancer cells and tumors. So, in an ideal world, and even then theoretically, your blood carries signatures of any cancers that might be going on in your body. One difference between DNA shed from healthy cells and DNA shed from cancer cells is a set of chemical tags on the DNA called methylation patterns, making it possible to discriminate between the two.

4:39In some cases, different organs can chemically tag their DNA differently, so the particular methylation patterns can even suggest where in the body these DNA fragments came from. All of this data is then fed into machine learning algorithms which, ideally, can then produce a list of all the types of cancers found in your body, and some of the MCD tests on the market today claim early detection of more than 50 different types of cancer. One quick blood test, and a near-miraculous early detection of any of up to 50 cancers that was never before possible.

5:18Where is my credit card? And this is where the big BUTT comes in. The first red flag is that as of the date of this episode, October 2026, not a single MCD test is FDA approved. The most established test, called GALLERY, with an I, and produced by a company called GRAIL, located in Silicon Valley, has applied and expects final approval in early 2027. Nevertheless, GALLERY and a few others are available commercially if you have a prescription.

5:51And this is the second red flag, in my opinion, a larger one. There are about five online DTC websites selling these, most for around $1,000 per test. You can call them or do an online chat. They connect you to some authorized medical professional, possibly a doctor, but more likely a nurse practitioner or a physician assistant, someone who has never examined you, and who knows only as much of your medical history as you choose to share with them verbally.

6:21And people who have used these report that nobody verifies the information you give them, and there's typically no follow-up. These are the McDonald's via DoorDash version of telehealth. Routine screenings for cancer cover only a few types. Do you know how many? It's four. Only four types of cancer are routinely screened for. Breast, cervical, colorectal, and lung. In light of that, it sounds like an MCD covering 50 types is a fantastic idea.

6:54But one of the reasons it's only four is that these are the cancers where the benefits of early detection exceed the harm. Now, that's a concept that sounds ridiculous, doesn't it? How could finding out that you have cancer be harmful? To understand this, it's important to first understand a few basic concepts in epidemiology.

Sensitivity And Specificity In Screening

7:16So let's take a quick look at those. The first is called sensitivity and specificity, which are scores that we give to a certain type of test to see how good it is at finding whatever it's testing for. Sensitivity is how good a test is at finding a disease when it is actually present. If we take ten people who all have a disease, and the test finds it in seven of them, but misses it and gives false negatives to three of them, then this test has a sensitivity of 70%.

7:49Specificity tells how good the test is at giving the all clear when the disease is not present. If all ten people are known to be free of the disease, and the test correctly clears eight of them but gives false positives to the other two, then this test has a specificity of 80%. A well-designed test often has to perform a balancing act between these two. If we make the test too sensitive, it has more false positives. If we tone it down to make it too specific, it has more false negatives.

8:22The ideal test would have sensitivity of 100% and specificity of 100%, which would give zero false negatives and zero false positives. So all its results would be reliably accurate. For example, some of the gold standards in established medical testing are HIV tests and hepatitis C antibody tests. Both of these have sensitivity and specificity, both above 98%.

8:52But in contrast, the pap smear for women has very high specificity with very few false positives, but it has a sensitivity of only 50% to 60%, missing nearly half of precancerous lesions. Luckily, they grow slowly and the test is repeated every few years to give more chances to catch them. This December, join Skeptoid for Mysteries of the Maya,

9:22a week-long Skeptoid adventure running December 6th through the 13th, 2026. Spots are limited and time is running out, so now is the moment to claim yours. We depart on December 6th for the Yucatan Peninsula and a deep dive into the Maya history of each port. First, Cozumel, the ancient island of swallows and a once sacred pilgrimage site. Then, the Roatan Bay Islands with their famed ancient coral reefs and stories of haunted caves. Lastly, Harvest Key, home to the majestic temple of Nimli Punit.

9:57While on board, we'll host a Skeptoid exclusive mini-conference featuring guest expert Curly Tlapoyawa. As an archaeologist, ethno-historian, and host of the Tales from Atlantis podcast, Curly is a leading voice in reclaiming indigenous history and debunking the pseudo-history and ancient alien myths that often distract from the truth. He'll reveal the sophisticated scientific and mathematical achievements of the Maya and explore the deep roots of Nahuatl philosophy, giving us a powerful, authentic lens into this vibrant history

10:32before we step off the ship to explore the ruins ourselves. Spots are filling fast, and once they're gone, they're gone. Visit Skeptoid.com slash adventures today to secure your place on the mysteries of the Maya before it's too late at Skeptoid.com slash adventures.

10:59This is the danger with imprecise tests. People who receive false negatives go on to suffer from the disease they thought they didn't have. And false positives go in for more invasive testing, which in their case carries only risk with no potential benefits. If a young healthy guy who's asymptomatic asks his doctor for a PSA, the prostate-specific antigen used to catch prostate cancer, many doctors might decline on the principle that any positive result is extremely likely to be false.

11:33And this results in the guy going in for a biopsy, which has side effects such as bleeding, pain, urinary retention, temporary erectile problems, infection, the most serious risk, and of course the anxiety that goes along with a perceived positive cancer diagnosis. In fact, for men younger than 55, the standard of care is for doctors not to give the PSA test at all. What's the harm in having more information is the obvious question that nearly everyone would ask,

12:05quite reasonably, when considering something like an MCD. The harm is that this additional information has a good chance of being wrong and of sending healthy people in for side effect-laden invasive testing that carries only risk and no benefit. But the far more common error is Galleries' 30% sensitivity. 70% of real cancers are missed, meaning a lot of customers will walk away $1,000 poorer

12:35and with no knowledge that they actually do have cancer. Related to these is positive predictive value, PPV, which is the percentage of a test's positive results that turn out to be real. PPVs can vary a lot depending on who is being tested and what their risk factors are. Let's take the example of lung cancer screening via a low dose CT scan. If you give this test to a low-risk group, say children who have never smoked,

13:05there will always be some positive results. The PPV of this test in that case is close to zero, far less than 1%, meaning that of the children who had a positive result, less than 1% actually do have lung cancer. Compare this to the PPV of that test for a high-risk group, say elderly, lifelong heavy smokers with severe symptoms like coughing blood. You'll get lots of positive results.

13:35And if, say, 30% of them actually did have cancer, three-quarters of the positive tests would be correct. So in this case, the same type of test has a PPV of around 75%. A test's PPV, its positive predictive value, varies a lot based on the risk factor of the test subject. Herein lies the fundamental problem of pulling up a DTC website and getting your own MCD. You probably have almost no idea what your risk factor is for any of these 50 cancers you're getting screened for.

14:11The PPV for an MCD test for each of these cancers, for you, is essentially unknown. Probably better for these tests' target market, people 50 and older, and probably worse for people younger than that, for whom very few doctors would have prescribed it.

Critiques Of Published Trial Data

14:30More problems arise when we look at the published literature, which mainly focus on gallery. The clinical trials published so far are nearly all funded by Grail, the producer of gallery. A critique in the British Medical Journal found, quote, that the test sensitivity for stage one cancers was only 16.8%, which they described as strikingly low. And yet, quote, many of the authors declared fees, patents, or stock holdings with the company, end quote.

15:02When the test was given to patients already known to have stage one to three cancers, the sensitivity was 44%, meaning 56% of the people with cancer received a false negative result. Another Grail-funded study called Pathfinder, and published in 2023 in The Lancet, found the gallery had a PPV of only 38%, also terrible, and a sensitivity of 29%, even worse,

15:32and yet concluded that this, quote, supports the feasibility of MCED screening for cancer, end quote. Another Grail-sponsored study published in Nature Medicine in 2026 found that the test had an overall sensitivity of just 31%. The BMJ further found that the UK's National Health Service had committed to buy one million gallery tests if it showed a PPV of at least 30%, an eyebrow-raising coincidence.

16:04In a nutshell, there have been many similar published criticisms of gallery and of Grail's practices, but that says nothing about the potential of MCD testing in general. Summarizing the published literature, we can make a few general statements. The biggest criticism that is nearly universal is that MCD tests have not yet proven to be beneficial. Although the potential for harm is there, as discussed previously, there doesn't seem to have been much of it,

16:38so it's less a matter of the test causing harm than it is a case of it simply not having been shown to improve outcomes for patients. In the future, when the tests improve to the point of reaching their potential, almost nobody thinks they will be useful for anything close to 50 cancers. Most say it would be just a handful of the most aggressive and fastest-growing cancers that shed a lot of DNA, cancers like esophageal and pancreatic.

17:09The catch is that these are fast-moving diseases, for which MCD has an abysmal sensitivity of only 17%.

Future Potential Versus Current Reality

17:21Promising but unproven is a good summary. MCD is a real technology, and it does work via the mechanism described. The most realistic eventual goal for MCDs is as supplemental testing for a small number of cancers alongside existing proven screening tests, not as a miracle, check yourself at home and find all your cancers before the doctors do, the way they are marketed. Today, that is late 2026, it is not ready for prime time for consumers seeking to purchase it directly.

17:57This is likely to improve in the future, not guaranteed, but likely, and when it does, that will be a better time to ask your doctor whether you should try it. Whether such testing will ever be recommended for consumers to purchase directly, or whether it will ever be useful for more than a small number of cancers, well, those are questions we can't answer today without a crystal ball. We'll close with a quick recommendation. Next time you see your primary care doctor, ask whether you should get any of the routine cancer screenings.

18:32No crystal ball is needed to know whether those are beneficial. Special thanks to Dr. Erica Anderson for a tech edit and a number of important corrections and clarifications. We continue with the fun mnemonic for remembering sensitivity and specificity in the ad-free and extended premium feed. To access it, become a supporter at skeptoid.com slash go premium.

19:04A great big skeptoid shout out to our premium supporters, including Bill Campbell, Pepsi and Cat Dad, JK Falsknight, and Brian and Emma Dorland. Skeptoid is expanding and diversifying our teacher's toolkit and is looking for an experienced educator with strong skills in K-12 curriculum development to build the program. This would be a part-time, one-year contract position. If this sounds as exciting to you as it does to us, check out the full position announcement at skeptoid.org slash jobs.

19:42The position is only open until filled, so don't wait. That's skeptoid.org slash jobs. It's movie night, so please enjoy our 40-minute movie, Principles of Curiosity, that teaches the basics of scientific skepticism and critical thinking in a far-ranging journey that takes you from the depths of Death Valley to the highest points in space. It's free on YouTube at PrinciplesOfCuriosity.com and on streaming services. Skeptoid is a production of Skeptoid Media.

20:15Our executive director and tinfoil hat counter is Kathy Reitmeyer. Marketing guru and Illuminati liaison is Jake Young. Director of production and Martian outreach coordinator is Will McCandless. Theme music is by Lee Sanders. Researched and written by me, Brian Dunning. Listen to Skeptoid for free on Apple Podcasts, Spotify, Amazon Music, or iHeart.

20:41You're listening to Skeptoid, a listener-supported program. I'm Brian Dunning from Skeptoid.com. From PRX. 1. 3. 4. 5. 4. 5. 5. 6. 6. 7. 6. 8.