> One study in 2020 found that 95% of asymptomatic patients had some type of "abnormal" finding, but just 1.8% of these findings were indeed cancer.
So a bit less than 1.8% of the time in this study
> Prenuvo's recent Polaris Study followed 1,011 patients for at least one year following a whole-body MRI scan. Of these patients, 41 had biopsies. More than half of the 41 were diagnosed with cancer.
That's 2.0%
Note that this doesn't mean that 1.7~2.0% of people have cancer without knowing it. It could be more:
> A negative scan doesn’t mean you’re disease-free. Some cancers and conditions simply aren’t visible yet or aren’t reliably detected on a one-time full-body MRI."
But also perhaps less, in a way:
> "You're finding something that never would have caused you any problem in your life, and in cancer, we call that overdiagnosis," Vickers says.
I pay an extra $60 a year to have my ophthalmologist take a digital image of my retina. It comes back as normal every year, but if something does change we can diff the image against the baseline.
Maybe I don't want to look for cancer right now but if I spend $1,000 every 5 years to take an image for later use... isn't that useful?
Maybe the right answer isn't to do a biopsy, but to monitor the area with follow-up scans? It seems like that addresses much of the harm that a false positive can cause (invasive biopsy leading to complications) while maintaining most of the gains (still very early detection).
> Maybe the right answer isn't to do a biopsy, but to monitor the area with follow-up scans?
Doctors have already thought of this. Several issues with it:
* Monitoring still causes anxiety and mental health issues which come with real effects on patient's quality of life. It's not "harmless".
* Unclear when to monitor and when to treat. It's also really hard to get enough data to characterize these early unspecific findings enough to get confidence on what to do.
* Monitoring via MRI might be just as useful as monitoring via symptoms or any other "passive" methods that do not require a previous scan.
There’s a major difference between having insurance cover something (socialized cost, immediately drives up provider fees for bizarre reasons) and letting the market allow people to buy it themselves (individual cost, the market drives the cost down fast and hard). Notice the pattern with LASIK and GLP1 where lack of insurance coverage has counterintuitively made it cheaper and more accessible.
Let everyone who wants to pay get their scans! But don’t make me pay for you
“ One study in 2020 found that 95% of asymptomatic patients had some type of "abnormal" finding, but just 1.8% of these findings were indeed cancer.”
This has been my experience. And I’ve had oncologists echo exactly this. In the words of one: MRIs find too much.
The CT and the PET/CT are the gold standards for finding cancer, finding recurrences, and staging cancer. The trouble is the radiation dose.
MRI provides very inconclusive results. You’ll see something but it’ll be unclear what it is. And often what you see is not even visible on a CT. Or it’s visible on a PET/CT and is showing metabolic activity indicating its cancer.
MRIs are great for certain things like herniated disks in your back. They suck at cancer.
> MRIs are great for certain things like herniated disks in your back.
I have had a lot of experience with MRIs on both myself (back and knee [1]) and my dogs with herniated discs. The doctors always make it sound like MRIs are great to confirm what's suspected because of other symptoms like pain, but a point in time MRI alone is not that valuable. Everyone's bodies (including animals!) are surprisingly different inside making normal be somewhat unique. I think what would be interesting is if scanning technology like MRIs could be made so inexpensive and easy that everyone had one done 4x/year. That way it's the differential being checked and I'm guessing it would be way more valuable. Normalization such as this could also lower anxiety around findings.
[1] Even when I tore my ACL the MRI came back only as probable.
That’s completely and 100% false. It’s much easier to characterize things on MRI and MRI is indeed phenomenal for cancer! The problem is with screening, not actual staging or follow-up, and whole-body screening in ct and pet/ct is even worse than MRI screening even if you ignore radiation.
1. I had a chest x-ray
2. It showed a small dark patch, and my lungs over-inflate.
3. Erring on the side of caution, doctor ordered a lung biopsy.
4. Lung biopsy is painful, annoying, expensive, and non-zero-risk.
5. Lung biopsy turns up nothing. "Maybe you aspirated some food?" We learn nothing.
Just to point out, cancer isn't the only reason to get these. Aneurisms, hemachromatosis, etc can all be serious. I know someone who got scanned for $500 and they caught hemachromatosis via iron deposits in the liver. Much better than eventual chirrosis and liver failure.
Doctors here are cognitively captured by a system designed to limit cost (and that's mostly a good thing)
But scanning frequently is overwhelmingly good for the patient. The problem is the doctors. Imagine two possibilities.
1. You scan every six months and a doctor reviews your scans but never tells you anything no matter what
2. You scan every six months and a doctor reviews your scans and only tells you results if you have an obviously growing mass that has a probability greater than 95% of being cancerous
Obviously #2 is better for the patient than #1, but #1 is equivalent to never testing if you ignore cost.
So the actual reason we don't have effect frequent scans combined with effective diagnostic techniques is cost, and doctors cope with this reality by saying clearly wrong things about "over diagnosis". It's a local minimum of the payer/provider dynamic that has nothing to do with scans per se.
Your case #2 doesn't have nearly enough information to say whether it's obviously better for a population of patients. There are a lot of other variables you would need to know:
- The accuracy of detecting a mass
- The true distribution of masses in the population
- The likelihood that of falsely detecting a mass in the same place twice (you seem to implicitly assume that false detections are uncorrelated with each other)
- The likelihood that a real mass is cancerous (you stipulate that this is 95% in your scenario, but you don't say what other factors are used to determine this - as opposed to just knowing that there's a mass that grew.)
- The positive effect of treatment in the case of true-positives.
- The negative effect of treatment or further diagnostics in the case of false-positive.
Saying that doctors are lying about over diagnosis to cope with the fact that diagnostic techniques are too expensive is absurd. They have to actually make decisions in the real world, where your two neat little categories can't be known even if they hypothetically exist.
One important point is that many people die WITH cancer but not OF cancer. So even for the 1.8%, only a fraction of those people were going to die of the disease (or even suffer significant symptoms) - the rest were just going to die of natural causes anyway.
But now you've found it you pretty much have to remove it, which has significant quality of life implications.
I don't see the point of testing constantly. It's just creating stress and probably most of the time, the tumor might be benign or it might be small and go away on its own.
And anyway, you have to die of something so for me cancer would just be a sign that time's up.
There are two points in MRIs that limit or (better) regulate their use:
* financial cost vs level of care. A full body MRI costs a huge sum but it is the most detailed non-invasive diagnosis we have for any disease that can lead to earlier therapy. Used as a screening method, does it worth to save one patient in tens/hundreds tests performed? You answer, but public health authorities, health insurances and medical societies are negative.
* MRI shows some minor findings that would never cause symptoms and better not be known to the patient due to the stress they bring and cost of ongoing follow up (eg in the brain small meningiomas or angiomas). This might bring more harm than good and limit their net value.
> One study in 2020 found that 95% of asymptomatic patients had some type of "abnormal" finding, but just 1.8% of these findings were indeed cancer
So that would be 1.71% of people getting a full body MRI catching cancer early.
That seems like an astoundingly good return to me. What does late cancer treatment and lost lives cost? Many times these MRI’s.
If everyone did this, would there be any significant number of late stage cancer discoveries anymore?
Obviously, reducing the cost of false positives is important. But if 1.7% of us have cancer - wow. My guess is that percentage can be tuned up and down by demographics. So maybe there is still an argument against everyone doing it.
Can someone ELI5 why false positives on a MRI are so bad?
From a pure Bayesian PoV, you're better off with a noisy additional observation. At worst it doesn't get much weight.
At a pragmatic level, can't you say, hey here's something thats probably nothing, let's scan it again in 6 months? Why does an MRI necessarily lead to invasive follow ups?
I get that ideally we'd have a crystal ball with 0 type I / type II errors but short of that, why is a noisy predictor bad?
Frankly, this sounds like some people aren't so comfortable with the sheer cost of the machine than their absolute utility. CT and MRI scan machines are something that said to cost like $1m/yr/unit that's ~$500 uninsured/$100 insured per run in Japan that China don't publish data on numbers or distributions of. That says "military grade expensive" written all over.
Arguments against proactive MRI scanning always seem to have a whiff of status quo bias to them. Yes, right now MRIs are expensive and false positives are common, but if regular scans were widespread, it's likely this result in innovations that would drive down costs, improve accuracy, as well as producing a much larger corpus of data with which to guide diagnosis and reduce false positives.
To use a software analogy, if your downtime detection system kept producing false negatives, would your solution to be just turn it off? You'd get some better night's sleep, but you'd pay for it when the system really went down and you had no idea.
> if your downtime detection system kept producing false negatives, would your solution to be just turn it off
It's more like if your CI build fails in 95% of runs, and in only 2% of runs it indicates a real bug, do you have a bad CI which is next to useless for detecting bugs? I'd say yes, that's exactly what you have. No developer is going to pay any attention to this CI, and if you tell the developers they must ensure tests pass on such CI, they would rebel.
Can one solution be always doing two scans, N months apart, before drawing any conclusions (excluding things that can be reliably detected from a single scan)? Initial scan could affect N (if you find something potentially aggressive, you can schedule the second scan sooner). And then do a follow up every M years.
That should exclude benign or very slowing growing things
Doctors absolutely hate the idea of people being checked for diseases. Every time someone comes out with a plan to detect cancer or prevent HIV, they start screeching.
“But what if the person would have died anyway without noticing they had cancer? Think of the shareholders. They would have paid for treatment for nothing”
> The tests range from several hundred to several thousand dollars, depending on which sections of the body are scanned, and are not covered by insurance.
Even ignoring the overdiagnosing problem (I don't understand how they can determine from MRI when a cancer is a cancer; there are also benign growth and often when they are a certain size, people notice them, but how would MRI help here? Too small areas could be classified as malign; any further procedures can be dangerous - see that Dawson Creek actor recently, the cost of clinical intervention did not help), I think that medicine is increasingly becoming a "only affordable for those who have money". You can see this with regards to gene therapy too - if we ignore the success ratio, many of these therapies are impossible to acquire for Average Joe. Granted, the prices will go down for various reasons (we saw this with Moore's law and many other inventions too), but at the end of the day I feel we are stepping closer and closer to a very unfair society model - more and more superrich, but prices also go up immensely for average people. That model is not sustainable; people will be angry since this is not fair.
I much prefer tests with low false positive rates.
I recently had such a cancer-related test. A cousin had a BRCA2 mutation and I was concerned I could have it also. Insurance would not pay for the testing, but one can get a panel of such genetic tests for just $250 now, so I went ahead. And it was negative. This is reassuring not just to me, but also to my children, and (somewhat) my sibling (the relevant parent is no longer alive).
Had this test been positive, the chance of pancreatic cancer would have gone way up, so frequent scans (I think annual MRI and ultrasound?) would have been justified.
Extremely anecdotal: my dad’s lung cancer was found incidentally - shoulder pain → shoulder MRI → radiologist noticed lung nodules. The shoulder pain was unrelated, but the scan itself was clinically indicated and there were prior scans to compare against, which made it immediately suspicious and triggered follow-up imaging.
He was also in a higher-risk group (age + history), which in hindsight probably made the incidental finding meaningful rather than noise.
So cases like his make me think less “scan everyone periodically” and more “imaging can be very valuable once risk is non-trivial or there’s a baseline to compare to”. For the general population you’d mostly generate false positives, but for higher-risk groups (older, smokers, prior findings) it seems much more defensible.
40 comments
[ 3.3 ms ] story [ 51.8 ms ] thread> One study in 2020 found that 95% of asymptomatic patients had some type of "abnormal" finding, but just 1.8% of these findings were indeed cancer.
So a bit less than 1.8% of the time in this study
> Prenuvo's recent Polaris Study followed 1,011 patients for at least one year following a whole-body MRI scan. Of these patients, 41 had biopsies. More than half of the 41 were diagnosed with cancer.
That's 2.0%
Note that this doesn't mean that 1.7~2.0% of people have cancer without knowing it. It could be more:
> A negative scan doesn’t mean you’re disease-free. Some cancers and conditions simply aren’t visible yet or aren’t reliably detected on a one-time full-body MRI."
But also perhaps less, in a way:
> "You're finding something that never would have caused you any problem in your life, and in cancer, we call that overdiagnosis," Vickers says.
Maybe I don't want to look for cancer right now but if I spend $1,000 every 5 years to take an image for later use... isn't that useful?
2. i can afford the money for the chance of early detection. Many cancers are symptomatic only in the latter stages. It does not hurt to check.
Doctors have already thought of this. Several issues with it:
* Monitoring still causes anxiety and mental health issues which come with real effects on patient's quality of life. It's not "harmless".
* Unclear when to monitor and when to treat. It's also really hard to get enough data to characterize these early unspecific findings enough to get confidence on what to do.
* Monitoring via MRI might be just as useful as monitoring via symptoms or any other "passive" methods that do not require a previous scan.
Let everyone who wants to pay get their scans! But don’t make me pay for you
This has been my experience. And I’ve had oncologists echo exactly this. In the words of one: MRIs find too much.
The CT and the PET/CT are the gold standards for finding cancer, finding recurrences, and staging cancer. The trouble is the radiation dose.
MRI provides very inconclusive results. You’ll see something but it’ll be unclear what it is. And often what you see is not even visible on a CT. Or it’s visible on a PET/CT and is showing metabolic activity indicating its cancer.
MRIs are great for certain things like herniated disks in your back. They suck at cancer.
I have had a lot of experience with MRIs on both myself (back and knee [1]) and my dogs with herniated discs. The doctors always make it sound like MRIs are great to confirm what's suspected because of other symptoms like pain, but a point in time MRI alone is not that valuable. Everyone's bodies (including animals!) are surprisingly different inside making normal be somewhat unique. I think what would be interesting is if scanning technology like MRIs could be made so inexpensive and easy that everyone had one done 4x/year. That way it's the differential being checked and I'm guessing it would be way more valuable. Normalization such as this could also lower anxiety around findings.
[1] Even when I tore my ACL the MRI came back only as probable.
MRIs are fine for certain kinds of cancer like liver cancer.
That can also be reframed as 1.71% of asymptomatic people having cancer, which is a really good argument for better screening.
But scanning frequently is overwhelmingly good for the patient. The problem is the doctors. Imagine two possibilities. 1. You scan every six months and a doctor reviews your scans but never tells you anything no matter what 2. You scan every six months and a doctor reviews your scans and only tells you results if you have an obviously growing mass that has a probability greater than 95% of being cancerous
Obviously #2 is better for the patient than #1, but #1 is equivalent to never testing if you ignore cost.
So the actual reason we don't have effect frequent scans combined with effective diagnostic techniques is cost, and doctors cope with this reality by saying clearly wrong things about "over diagnosis". It's a local minimum of the payer/provider dynamic that has nothing to do with scans per se.
But now you've found it you pretty much have to remove it, which has significant quality of life implications.
And anyway, you have to die of something so for me cancer would just be a sign that time's up.
* financial cost vs level of care. A full body MRI costs a huge sum but it is the most detailed non-invasive diagnosis we have for any disease that can lead to earlier therapy. Used as a screening method, does it worth to save one patient in tens/hundreds tests performed? You answer, but public health authorities, health insurances and medical societies are negative.
* MRI shows some minor findings that would never cause symptoms and better not be known to the patient due to the stress they bring and cost of ongoing follow up (eg in the brain small meningiomas or angiomas). This might bring more harm than good and limit their net value.
So that would be 1.71% of people getting a full body MRI catching cancer early.
That seems like an astoundingly good return to me. What does late cancer treatment and lost lives cost? Many times these MRI’s.
If everyone did this, would there be any significant number of late stage cancer discoveries anymore?
Obviously, reducing the cost of false positives is important. But if 1.7% of us have cancer - wow. My guess is that percentage can be tuned up and down by demographics. So maybe there is still an argument against everyone doing it.
From a pure Bayesian PoV, you're better off with a noisy additional observation. At worst it doesn't get much weight.
At a pragmatic level, can't you say, hey here's something thats probably nothing, let's scan it again in 6 months? Why does an MRI necessarily lead to invasive follow ups?
I get that ideally we'd have a crystal ball with 0 type I / type II errors but short of that, why is a noisy predictor bad?
To use a software analogy, if your downtime detection system kept producing false negatives, would your solution to be just turn it off? You'd get some better night's sleep, but you'd pay for it when the system really went down and you had no idea.
It's more like if your CI build fails in 95% of runs, and in only 2% of runs it indicates a real bug, do you have a bad CI which is next to useless for detecting bugs? I'd say yes, that's exactly what you have. No developer is going to pay any attention to this CI, and if you tell the developers they must ensure tests pass on such CI, they would rebel.
That should exclude benign or very slowing growing things
“But what if the person would have died anyway without noticing they had cancer? Think of the shareholders. They would have paid for treatment for nothing”
Even ignoring the overdiagnosing problem (I don't understand how they can determine from MRI when a cancer is a cancer; there are also benign growth and often when they are a certain size, people notice them, but how would MRI help here? Too small areas could be classified as malign; any further procedures can be dangerous - see that Dawson Creek actor recently, the cost of clinical intervention did not help), I think that medicine is increasingly becoming a "only affordable for those who have money". You can see this with regards to gene therapy too - if we ignore the success ratio, many of these therapies are impossible to acquire for Average Joe. Granted, the prices will go down for various reasons (we saw this with Moore's law and many other inventions too), but at the end of the day I feel we are stepping closer and closer to a very unfair society model - more and more superrich, but prices also go up immensely for average people. That model is not sustainable; people will be angry since this is not fair.
I recently had such a cancer-related test. A cousin had a BRCA2 mutation and I was concerned I could have it also. Insurance would not pay for the testing, but one can get a panel of such genetic tests for just $250 now, so I went ahead. And it was negative. This is reassuring not just to me, but also to my children, and (somewhat) my sibling (the relevant parent is no longer alive).
Had this test been positive, the chance of pancreatic cancer would have gone way up, so frequent scans (I think annual MRI and ultrasound?) would have been justified.
He was also in a higher-risk group (age + history), which in hindsight probably made the incidental finding meaningful rather than noise.
So cases like his make me think less “scan everyone periodically” and more “imaging can be very valuable once risk is non-trivial or there’s a baseline to compare to”. For the general population you’d mostly generate false positives, but for higher-risk groups (older, smokers, prior findings) it seems much more defensible.