I hadn’t intended on writing about the profit-driven and non-evidence-based product that is “whole body scanning” again so soon, given that I just wrote about this perennial money-making medical venture last summer in the context of companies offering newer products that add artificial intelligence to the scanning analysis without demonstrating that AI actually improves specificity and/or sensitivity. Of course, the AI is always proprietary, which doesn’t allow independent investigators to determine how and if it works, and the companies promise to do clinical trials, while continuing to charge customers up to thousands of dollars per scan. At the time, I noted that various “whole body scanning” products had been hyped for at least 20 years before and that one of my earliest forays into medical skepticism occurred in the early 2000s when I was faculty at a cancer center in New Jersey and noted radio ads for AmeriScan, Dr. Craig Bittner. The overblown claims being made for the scans, which included breast MRI and various whole body CT scans, being able to “catch cancer early” and save lives led me on a bit of a crusade that I wrote about back in 2005. Ultimately, AmeriScan went out of business, as whole body scans of the type it sold went out of favor, but bad ideas in medicine are evergreen. It was thus inevitable that, as the technology progressed by leaps and bounds over the next two decades, other companies would see profit potential selling old wine (whole body scans) in a new bottle (AI-augmented CT and MRI).

And so it was, with companies like Prenuvo seemingly thinking that whole body MRI + AI = profit. I was wondering whether there was room for the “old fashioned” sorts of whole body scanning. I need not have worried. As I was idly flipping channels Saturday night waiting for our dinner companion to be ready so that we could go out and get some food, I came across NBC World News Tonight and a story entitled Medical professionals weigh in on growing trend of people getting full-body scans to screen for cancer:

On second thought, maybe there isn’t room for old-fashioned MRIs without AI, because I learned as I watched the story that—spoiler alert!—Ezra, the company selling these scans and whose CEO is interviewed in the story is selling an AI-augmented whole body MRI scans that is claimed to reduce screening time to 30 minutes.

The NBC story on whole body scans was practically an infomercial

As this story appeared on my TV, let’s just say that I immediately got a vibe similar to what I used to get watching stories about autism 15-20 years ago, when reporters routinely seemed to think that they had to interview parents convinced that vaccines had made their child autistic whenever they did a story about vaccines or autism. In these stories, reporters would focus on the human interest story about the parents and the child and then include a couple of doctors almost as an afterthought as the token skeptics to tell the audience that the science overwhelmingly shows that vaccination is not associated with an increased risk of autism and autism spectrum disorders (ASDs). Overall, it was false balance in that science had no chance against the sympathetic human interest part of the story about a family dealing with autism. The very best impression left by these stories was that there was a real scientific debate going on over the topic—there wasn’t—while the worst impression coming from this stories was that vaccines really did cause autism. Fortunately, over the years, the press learned, particularly after Andrew Wakefield was disgraced due to losing his medical license in the UK, the retraction of his 1998 Lancet case series that had sparked the 21st antivax movement, and Brian Deer’s devastating exposé on his scientific fraud, and false balance became much less common, although sadly it’s not gone.

This story reminded me a lot of that. Here’s what I mean.

The story was bookended with the story of a woman named Andrea Schaffer. Ms. Schaffer, we’re told, had purchased a whole body MRI scan as a “present to herself” last June. When NBC reporter Vicky Nguyen asked Ms. Schaffer why she had purchased the scan, Ms. Schaffer mentioned that she was a 44-year-old woman whose mother had been diagnosed with breast cancer when she was about her age. We then learned that the scan had found a cancerous lesion in her brain, shown here:

Andrea Schaeffer whole body scan
Yes, that’s a big brain tumor.

Ms. Nguyễn then asked Ms. Schaffer, “Do you think the scan saved your life?”

Ms. Schaffer, of course, replied, “Yes.”

First off, let me just say that I’m very happy that this tumor was found in time to do something about it with surgery and whatever other treatments were offered to Ms. Schaffer. I never want to see patients die of cancer when that can be prevented.

My good wishes to Ms. Schaffer notwithstanding, though, no matter what the rest of the story would go on to say, who might be interviewed, or what studies might be presented, I knew upon seeing how it started that the story was going to be hopelessly slanted. Here was a woman who, concerned about her health given her family history of breast cancer, had purchased a whole body MRI scan, which then found a “cancerous brain tumor.” We were not told what kind of brain tumor—which is actually rather important information given that different types of brain cancer have different prognoses—only that this woman clearly believes that the scan had saved her life. And why shouldn’t she believe that? From the perspective of the patient, surely finding the tumor must have saved her life! True, the tumor was only found about a year ago; so she isn’t that far out from her treatment. But surely she would have died, right? Maybe. Probably, even.

So what’s the problem?

Ms. Nguyễn, in fairness, did follow up the brief snippet of her interview with Ms. Schaffer with a quote from authorities trying to tell us what the problem is. However, watching such a story constructed in this manner, by the time audience hears from those authorities, it’s too late. Ms. Schaffer’s belief had been conflated with actual cause-and-effect, leading to the message being indelibly imprinted that a whole body MRI scan had saved her life.

From there, went on to note that these scans have proliferated, promoted by celebrities like Kim Kardashian and Maria Menounos, the latter of whom had a stage II pancreatic neuroendocrine tumor discovered (PNET) on a Prenuvo scan. It is important to note here that PNETs are much less aggressive and deadly than the more common run-of-the-mill pancreatic cancer, although in fairness I also need to point out that it was a PNET that ultimately claimed Steve Jobs’ life. You might read stories about Maria Menounos and see this story about Andrea Schaffer and then ask again, “What’s the problem?” Didn’t these scans save these women’s lives?” The question of screening asymptomatic people for disease, be it cancer or whatever, is, unfortunately, a much more complex question than that, and there are, in fact, harms from such screening that have to be balanced against the rare cases like those of Ms. Menounos or Schaffer, something this story did in such a perfunctory fashion that it almost might have been a brief infomercial for Ezra scans. Yes, Ms. Nguyen quoted the American College of Radiology’s statement that these scans are expensive, not covered by insurance, and can result in unnecessary medical tests and procedures, all of which is totally true, but sounds a lot less convincing coming hot on the heels of an interview with someone like that of Andrea Schaffer.

Any consideration of the potential harms of such scans was then undermined even further by what followed next, an interview with the founder and CEO of Ezra, Amit Gal, described on the Ezra website as a “software engineer and entrepreneur with a 15-year track record for building successful technology companies, and bringing them to a worldwide stage.” Mr. Gal, predictably for one of these stories, enthusiastically hyped how his company “has helped our members find numerous types of cancer,” which is all well and good but rather useless information without knowing which cancers, how many, and the denominator of how patients were screened. In other words, how many is “many of our members” and how many members underwent the scans to detect those cancers? It would also be helpful to know how many required additional workup and tests, particularly additional invasive tests, to rule out disease and were ultimately found not to have anything requiring treatment, as well as the complication rates for those additional tests.

The overall impression given was that early detection is always good. Of course, doctors serious about developing screening tests for asymptomatic populations know that this is not always the case—and often not the case—and that there are harms involved with any screening test. None of this stopped Mr. Gal from noting that his test “might find cancer, or any of 500 other diseases” and that “the vast majority of Ezra scans” don’t find anything but do give you “peace of mind.” That made me wonder: What’s the false negative rate of the scans? Every test has false negatives? In other words, how “peaceful” is the “peace of mind” provided by these scans? After all, MRIs are great at detecting some cancers and diseases, not so great at detecting others. Of course, Ezra had an answer for that, as the next shot showed Dr. Dan Sodickson, Chief of Innovation at NYU and Ezra’s chief scientist, saying that the Ezra scans are best if you get one every year. (Of course they are. Ezra needs a steady, predictable income stream, after all.) Sadly, Dr. Sodickson didn’t even present anything resembling a scientific rationale for saying this. Then, when Ms. Nguyen asked him whether it’s realistic to expect someone to pay several thousand dollars a year for these scans for “peace of mind,” his response was telling: “I think it is, if it provides that confidence about health.

After that avalanche of promotion, the story finally brought out the token skeptic, Dr. Matthew Davenport, a professor of radiology at the University of Michigan, who tried valiantly to say the things that needed to be said about a test like Ezra’s. For example, when Ms. Nguyen asked him if he thinks that the result of these scans will be more tests and medical procedures that will cause harm to patients, Dr. Davenport, to his credit, answered that “unquestionably” that will be the result, citing more testing, biopsies, procedures, and operations as potential harms and concluding that you should check with your doctor before shelling out $1,350 for the most basic Ezra scan (or $2,500 for the “advanced” scan). (I also couldn’t help but note that Ezra offers financing options with no-interest monthly payments, so that you can go into debt to get a scan.)

Of course, this being a story with false balance, Dr. Davenport’s valiant attempt to inject some evidence and science into the discussion was immediately undercut by how the story was bookended, with a shot of Ms. Schaffer walking next to a pond, with a voiceover saying that she plans on getting another full body scan this year, after which she said, “You can’t put a price on it. This saved my life.”

See what I mean? NBC News and Ms. Nguyen can defend themselves by saying that they did quote the American College of Radiology and did interview radiologist who has a different view of these scans. Unfortunately, the way the story was bookended by a compelling anecdote of an appealing woman who was fortunate that a whole body scan spotted her brain cancer in time for her to have it surgically removed and, even more pointedly, got the final word, that final word being that you “can’t put a price on” the scan and that the scan “saved my life” gave a message that was anything but balanced. Let’s put it this way. Even if the story had not been bookended with the story of Andrea Schaffer and concluded with her saying that the test had definitely saved her life, placing her story instead somewhere in the middle, this story would still have been false balance, anecdotes against science.

Anecdotes, not much science

Whenever I see a story like this, I immediately become curious about the company, its product, and what, exactly, the product is claimed to be able to do and then compare all that with current scientific recommendations. The Ezra website immediately struck me as being long on anecdotes and hype but but very short on science. Sure, it says, “Our Scientific Advisors participate in authoring and publishing research papers with our Ezra team to maintain the highest scientific rigor within our work.” Yet, there’s no link to the article shown, and when I looked it up I didn’t see how it was evidence for the utility of scanning asymptomatic individuals at average risk of cancer because it was an article about identifying areas suspicious for cancer in the prostate using MRI. In fairness, it’s likely a perfectly fine article on one of the techniques developed by Ezra, but it is not evidence that their products do more good than harm. Also in fairness, I found other such articles by searching PubMed for “Ezra” and “MRI,” but none of them address the elephant in the room, the question of whether scanning asymptomatic people for cancer improves survival or even just the question of whether it does more harm than good.None of that stops Ezra from claiming on its website landing page without evidence, “Our annual scan catches potential cancer earlier, making it easier to beat.” On another page, the company states, “Screening is our best defense against cancer. And we’ve perfected it.” Is it, though? And have they, though?

Is it, though?
And have they, though?

Unsurprisingly and typically, in lieu of studies Ezra does have a fair number of anecdotes like that of Ms. Schaffer, although, oddly enough, I could not find her story on their website. That led me to go back and see if it was an Ezra scan that she had actually gotten. I’m glad I noticed that her story wasn’t on the Ezra website, as I did some additional Googling and fairly quickly found Ms. Schaffer’s website, Not Just Broccoli. It turns out that she is a “nutritional oncologist” who offers “terrain testing” that looks at the “The Terrain 10”:

  • Circulation
  • Inflammation
  • Oxidation
  • Microbiome
  • Blood Sugar
  • Toxins & Detoxification
  • Hormones
  • Immunity
  • Stress & Biorhythms
  • Epigenetics

“Toxins and detoxification”? Uh-oh. Ms. Schaffer also notes:

As we take a deep dive to find out why your body is cancering, we individualize your nutrition plan based on your health history, biology, environmental exposures and DNA. In order to have a complete picture of which areas of your terrain need work, we must do extensive testing. 

 The types of testing we recommend are genetic testing, extensive blood marker panels which include cancer drivers and inflammation markers, as well as home glucose and ketone testing. 

Why your body is “cancering”? She turned “cancer” into a verb?

She also offers food sensitivity testing, which is generally a bogus test. She is also treating her cancer with unproven, non-science-based dietary interventions:

In February 2023, I did a preventative full body MRI at a company called Prenuvo, and it found that I had a 2.3 by 2.4cm mass in the right temporal lobe of my brain. About the size of a walnut. I had no idea, I was asymptomatic. I have been on a journey through the medical system since, culminating with surgery at Barrow Neurological in Phoenix, Arizona on April 5th, 2023. After surgery, I was released from the hospital and returned to mostly normal function remarkably fast. I credit that to the talents of my surgeon, Dr. Smith, and all the work I did preparing for surgery. As soon as I found out about the tumor, as my training instructs, I immediately went on a therapeutic ketogenic diet (also known as therapeutic carbohydrate reduction- very different from internet keto), and worked to mitigate any other contributing factors I could identify.  Two weeks after surgery, once pathology came in, we found out that it was a malignant grade 3 oligodendroglioma. While this is shocking news, my surgeon luckily was able to remove the full tumor and we are utilizing my diet, while monitoring labs (and using them to know where I need to work on my health), to avoid recurrence. The goal is to avoid chemo and radiation in my case, due to my cancer being of the brain. 

And on her Instagram page:

Fortunately for Ms. Schaffer, nothing heals like surgical steel—as we surgeons like to say—and a skilled neurosurgeon was able to completely remove her malignant brain tumor. Again, despite my pointing out her “alternative” and “integrative” nutrition practice, I really am happy for her. I also note that, while she was one of the lucky ones for whom a whole-body MRI did more good than harm, her anecdote doesn’t let companies like Prenuvo and Ezra off the hook for making grandiose claims for their scans and recommending them yearly, as Ezra apparently does. It’s the companies whom I blame, not Ms. Schaffer, as well as NBC News, which left out a lot of relevant information in its story and constructed it so that it came across to me as a two and a half minute commercial for Ezra. This struck me as odd, given that it was just in October last year that NBC News medical correspondent Dr. Kavita Patel did a decent story cautioning against the hype for these tests, in which she brought up a good point that “full body MRIs” often use fewer “slices” and less detailed images than targeted MRI imaging a specific organ or area of the body.

It’s true, too. Targeted MRI scans generally take much longer, more detailed scans of limited parts of the anatomy based on the clinical indication, thus producing more detailed images. For example, a targeted pancreas scan takes around 30-45 minutes to do. In comparison, Ezra’s “Whole Body Flash” scan (the least expensive scan at $1,350) takes only 30 minutes to scan the whole body (except lungs, which do not visualize well on MRI because they are filled with air). Its images can’t possibly be as detailed, AI or not. Moreover, Ezra does not use gadolinium contrast, which is used for most MRI scans in order to outline the blood vessels more effectively and provide more detail. In fairness, I understand the desire not to use gadolinium, particularly when there’s a concern about potential cumulative gadolinium toxicity with repeated MRI scans done over many years (although the risk appears to be low), but there’s a price to be paid in terms of less resolution and less ability to pick up abnormalities. I also can’t help but note that the more expensive ($1,950) Full Body Scan takes 60 minutes, and the most expensive ($2,500) Full Body Plus Scan adds a screening CT scan of the lungs.

But what about the actual evidence, rather than corporate hype? First of all, a recent review article on whole body MRI (WB-MRI) states these recommendations succinctly in its conclusion:

The use of WB-MRI for cancer screening is recommended by current guidelines for subjects with cancer predisposition syndromes, including Li–Fraumeni syndrome, hereditary pheochromocytoma–paraganglioma syndromes and constitutional mismatch repair deficiency.

I note that these are all hereditary cancer predisposition syndromes in which affected patients have a very high lifetime risk of developing a number of different cancers. But what about patients at average or low risk of cancer, which is the general population targeted by ads for whole body MRI scans? The review notes:

…all the reports provide the number of subjects in which a malignant tumour was suspected, with cancer detection rates from 0% up to 10%. This wide range is probably a consequence of the small sample sizes of many studies, and the above-mentioned technical heterogeneity. Taken together, findings suspicious for malignant cancers were reported in nearly 2.0% (119 of 5809) of the screened subjects. Unfortunately, the number of studies in which follow-up and verification of findings was performed is lower, comprising 3287 screened asymptomatic subjects, in whom there was a 1.5% overall rate of histologically confirmed malignant cancers. This rate of malignant tumours detected with WB-MRI in asymptomatic subjects of the general population should not be ignored, justifying further studies. However, critics highlight the high rate of indeterminate incidental and false‐positive findings, which can lead to unnecessary additional examinations and treatments, with potential negative psychological impact [7475]. Therefore, the clinical utility of WB-MRI for cancer screening in the general population remains a matter of debate.

Last summer, the Dana Farber Cancer Institute blog published an article bluntly stating:

There is no evidence that full-body MRI scans are beneficial for people of average risk. If there were clinical evidence that they were effective, such as a clinical trial showing that many people benefitted from these scans, then they would be incorporated into standardized screening protocols. The United States Preventive Services Task Force (USPSTF) recommendations and the National Comprehensive Cancer Network (NCCN) provide excellent, meticulous analyses of screening technologies, and neither entity currently recommends full-body MRI for average-risk screening. 

Regarding some harms:

If a full-body MRI finds a suspicious signal on the pancreas, the next step is to do an endoscopic ultrasound. A doctor will put a little camera down your esophagus and into your stomach to look at the pancreas. If they see something suspicious, they’ll take a little piece, a biopsy, for investigation.  

These procedures introduce a risk of complications, even when performed by experts with low complication rates. In the average risk population, the risk of pancreatic cancer is low enough that many of those biopsies would be false alarms. It’s a lot of potential harm, possibly even the risk of death, that isn’t necessary for most of those people. 

That is exactly what I’ve been saying all along, going back over 20 years and going back to the days of my blogging about AmeriScan. I can also understand someone like Mr. Gal, who has no background in medical screening, being well-meaning and thinking he’s saving people by offering them these AI-enhanced scans. However, radiologists really should know better. Most do, but not all. They’re the ones who sell products like whole body MRI scans, even though there’s no evidence of benefit for people at average risk, they aren’t covered by health insurance, and they have the potential for causing more harm than good, exceptional cases like Andrea Schaffer notwithstanding. Even the presence of Siddhartha Mukherjee, M.D., D.Phil., author of the excellent book about the history of cancer Emperor of All Maladies, on Ezra’s list of scientific advisors does not change that.



Posted by David Gorski