26 September 2023 Blog: On Whole Body MRI Scans
A patient asked me my thoughts on a recent New York Times post on Prenuvo, a whole body MRI. Here is the post:
“For $2,499, Prenuvo will try to predict your future. The celebrity-endorsed company offers a roughly hour-long session of magnetic resonance imaging, or MRI, that scans your entire body, searching for early signs of cancer, aneurysms, liver diseases and even multiple sclerosis. It’s part of a crop of companies that claim they can revolutionize preventive health care with full-body scans — but experts warn this might not be the right approach.
In recent months, images of celebrities and influencers posing in branded scrubs in front of a glossy, cylindrical MRI machine have popped up on social media. Kim Kardashian wrote in the caption of a post she shared last month that Prenuvo “has really saved some of my friends’ lives.” In May, the television host Maria Menounos said that a Prenuvo scan had alerted her to a mass that turned out to be Stage 2 pancreatic cancer.
Many celebrities talk about their health on social media. But the ones that are documenting their body scans — complete with nearly identical photo ops — have taken the celebrity health endorsement to new heights in terms of cost. And considerable harm can come from screening, experts said.
What do experts say about the scans? Read more at the link in our bio.”
INTRODUCTION and GENERAL CONCEPTS of SCREENING
There is so much to unpack here that it is hard to know where to begin. But let’s start at the beginning “Prenuvo will try to predict your future.” It’s never a good sign when the first sentence is factually incorrect. What a whole body MRI is trying to do is illustrate your present. There have been no studies showing that whole body scanning can materially impact one’s health. In fact, MRI and CT scanning were developed for diagnostic purposes for patients with either known disease, serious complaints or trauma (the so-called ‘pan-scan’ is a mainstay of Emergency Medicine and Trauma Surgery). Only low-dose CT scans for lung cancer screening among at risk patients (i.e. prior or current smokers with at least 20 pack years of smoking – or 1 pack per day for 20 years) have undergone careful evaluation and study with two large randomized controlled trials showing that CT screening reduces lung cancer mortality rates in these at risk patients.
Moving along, the suggestion that the Prenuvo MRI will detect early signs of cancer, aneurysms, liver disease and even multiple sclerosis is a curious constellation of outcomes. Multiple sclerosis has typical clinical signs but is best confirmed using an MRI protocol both with and without contrast. Prenuvo does not use contrast stating ‘we believe the combination of sequences that we take performs as well as contrast-enhanced MRI for tumor detection’ but no mention of MS detection. Similarly, aneurysms are best detected with magnetic resonance angiography (MRA) which is similar to an MRI but focuses on vessels, rather than organs. So oddly, 4 of the conditions they choose to highlight as detectable are not best identified using Prenuvo.
Pancreatic cancer (to which I would add ovarian cancer and kidney cancer) is really an outcome where Prenuvo could make a difference. Typically diagnosed at later stages, the annual incidence rate of pancreatic cancer is 12.9 cases per 100,000 person-years with a death rate of 11.0 per 100,000 person-years. Compare that to colon cancer which has an annual incidence rate of 36.6 cases per 100,000 person-years and a death rate of 13.1 per 100,000 person-years. I did find it odd that they used the experience of “television host Maria Menounos” to highlight early detection at Stage 2 of disease. Diagnosing pancreatic cancer at Stage 2 only has a 5 year survival rate ranging from 13-39% as opposed to 84% for Stage 1A. A far better example would have been finding localized pancreatic cancer at Stage 1. Once again, an MRI is not the imaging modality of choice for pancreatic cancer detection in a high risk population – endoscopic ultrasound or a magnetic resonance cholangiopancreatography is preferred.
LEAD TIME BIAS
Those who follow my posts know that I won’t let a chance to introduce an epidemiologic concept go to waste – and this blog post is no different. One critique of Prenuvo comes from lead time bias. Lead time is the period of time between the detection of a medical condition by screening and when it ordinarily would have been diagnosed because a patient experiences symptoms and seeks medical care. A disease for which early treatment is no more effective than late treatment can make early detection look more ‘effective’, when in fact all the patient experiences is more disease time. This is illustrated in the figure below:
Specific screening for breast (mammography), lung (low dose CT scan among appropriate risk patients, detailed above) and colorectal cancers (colonoscopy) are known to be effective because randomized controlled trials have shown that the mortality rates of those screened are lower than a comparable group of unscreened individuals.
LENGTH TIME BIAS
A second epidemiologic Achilles heel of Prenuvo comes from length time bias. Screening works best when a medical condition develops slowly (this is part of the reason that colonoscopy is such a good screening tool and can be done once every 10 years for an average risk patient with no polyps). Screening tests like Prenuvo are more likely to find slow-growing tumors because they are present for a longer period of time before they cause symptoms. This is illustrated in the figure below:
Because screening tends to find tumors with an inherently better prognosis, mortality rates may appear to be better but this has nothing to do with the screening process itself. It simply reflects what the Prenuvo is good at detecting.
FALSE POSITIVE RATE
One of the critiques I have specifically left out (until now) are concerns about false positive results arising from the Prenuvo screening. This actually may be one aspect where Prenuvo performs fairly well as it is an expected cost of any screening test. In short, a false-positive screening test results in an abnormal result in a person without disease. These can then lead to the inconvenience, expense and potential risks associated with obtaining follow-up procedures. About 10% of mammograms lead to a false positive result, and in a study of ovarian cancer screening (CA-125 lab testing and ultrasound) returned an 8.4% false positive rate and one third of those cases underwent surgery. In that study, five times more women without ovarian cancer underwent surgery than did women with ovarian cancer.
My assumption would be that the Prenuvo screening would have a similar false positive rate of between 8-10%. But where clinicians get into trouble with higher false positive rates is when multiple tests are run (the so-called ‘covering all the bases’ phenomenon) occur at once wherein among several dozen tests, it is not unusual to have one or more return as ‘abnormal.’ So having the Prenuvo testing by itself could be a reasonable approach to reduce a cumulative false positive rate. It is important to note, however, that practice variation can contribute to the false-positive rate. This is a phenomenon we see with mammography wherein similar large screening programs in the US returned false positive rates nearly twice as high as in the UK – but with similar cancer detection rates. I would not imagine that the true false positive rate of Prenuvo screening will ever be known or publicized but an 8-10% rate remains a fair assumption.
SUMMARY: IS IT WORTH IT?
I suppose that the real question is for $2499, is it worth it? I’m not really sure how to answer that, because there are no guidelines on how often the Prenuvo should be performed. Is it annual? Every five years? Every ten? But to provide some framework, we know that colonoscopies ($1500, done every 5 to 10 years), mammograms ($200 performed annually), and Low Dose CTs of the Lungs among smokers ($500, done annually) do reduce mortality rate. Prenuvo has placed itself strategically with newer detection technologies such a Galleri which is a blood-based test seeking to detect early stage cancers ($1000, recommended annually). A vigorous evaluation of its efficacy would be reasonable, but as a cash-pay service there is little likelihood that Prenuvo will invest in research when they can put those dollars into marketing and endorsements.