Experimental blood test detects cancer up to four years before symptoms appear | Hacker News Reader

Experimental blood test detects cancer up to four years before symptoms appear

scientificamerican.com

650 pointsprawn189 comments

The real false positive rate is much higher than the 5% reported.

The false positive rate reported is ~5% in a sample with an equal number of people that did and did not develop cancer, while the ratio in reality is 1/122 (i.e., 1000 people developed cancer, 122000 did not).

Could still be a useful development, but it does make me wonder about the cost/benefits at this stage.

Yes, the positive predictive value of the test may be low, with the PPV being equal to the number of true positives divide by the number of true positives plus the number of false positives.

Since this test is designed to identify cancer very early, a likely solution to the low PPV problem would be to re-test the individual, perhaps even multiple times. The likelihood of multiple consecutive FP test results, barring a biological reason for the FP, would at some point warrant a more accurate test, be it a screen or diagnostic test depending on tumor type. Also keep in mind that the PPV of the test would increase with a person's age - an 80 year old is at a greater risk of developing cancer than is a 30 year old. This test would also be useful for people with hereditary cancer syndromes.

The cost benefit would be huge - imagine if everyone could identify their cancer 6 months or even years earlier. Mortality would greatly decrease, as would spending on treatments, etc.

"barring a biological reason for the FP"

I think that's an unwarranted assumption. Sure, another source of false positives is contamination or user error even, but other than that most FP are indeed due to a (usually unknown) biological reason. I'm reminded of a friend that was just positive she was not pregnant. She tested herself three times with three different pee stick brands. She was wrong. They all use the same chemistry, and for some reason something in her body interferes with those tests.

The cost benefit for someone with a cancer diagnosis could be huge. But for the people with false positives, that could also be very harmful to people. A (false) cancer diagnosis would be incredibly stressful.

It’s tricky to find that balance.

This seems like a failure in how test results are communicated to patients. Patients should just been told that they have been "flagged for further testing", rather than being told they have tested positive for cancer.

And how would that conversation go?

"You've been flagged for further testing."

"Why?"

"Um..."

"Didn't I just take a cancer screen? What was the result?"

"Ah..."

"Level with me, doc. Do I have cancer?"

"Well, [brings out probability textbook] you see ..."

So I have a real issue with not just telling the patient flat out that their test results indicate a high risk of having cancer and they need to be tested again.

As a society, we're just getting too used to accommodating people's ignorance about statistics and science. Where does it end?

Do we have to really have to give up promising therapeutics and diagnostic tests because we're afraid people's feelings will be hurt?

Yeah, I really hate it when people say what sounds like, "We recommend against even trying this test because people are incapable of responding appropriately to the answer."

It shouldn't even be that hard to state. "X% of people your age [+ whatever other factors they're using for a reference class] have [type of] cancer. Based on that and your test results, there's a Y% chance you have that type of cancer." Forgive me if I'm too disconnected from the average person, but is that too difficult to understand?

> So I have a real issue with not just telling the patient flat out that their test results indicate a high risk of having cancer and they need to be tested again.

The problem is that "tested again" has its own complications.

This is not theoretical. Part of the reason for updating the breast cancer screening guidelines is the fact that biopsies to confirm were causing more problems than the "cancers" they were finding.

Medical probabilities are tricky things.

You mean that the test physically caused a disorder in the patient because of a side effect, or that there was a problem in the patient’s attitude or emotional response? Because those are very different failure modes.

In one case the patient’s attitude about the test is not the determining consideration. A perfectly stoic and rational patient would be just as vulnerable to the side effects as anyone else.

A biopsy causes complications up to and including the death of the patient (secondary infections, etc.).

A biopsy is the next step to confirm "cancerous/non-cancerous" after a mammogram screen gives a "positive" result.

The problems with the screening is that once a "positive" comes up on the screen most patients cannot stop themselves from going to the next stage even if the probabilities don't warrant it.

The only medical "solution" was to decrease the frequency of screenings and raise the age at which first screening should occur.

Hmm, is that a situation where there is no appropriate action to take, regardless of the outcome of the screen? If that were the case, then there would be no point in doing the screen even if all actors were fully rational. But I'm guessing it's more like, the correct action is to wait N months and do a followup test.

Does it not work to say "If we take a biopsy, assuming normal treatment plans and outcomes for someone in your reference class, there's a 0.1% chance the biopsy kills you, and a 0.05% chance it saves you from cancer compared to the plan I'm recommending", or whatever the truth is?

(If the next possible step is a biopsy, under what circumstances does the calculus imply one should do the biopsy? When the screen results look sufficiently bad? Then maybe they shouldn't be presented as a binary "positive"/"negative".)

> Does it not work to say

No. It does not.

A diagnosis of "you might have cancer but the probability is low and we should just watch it" is not something that most people can psychologically take.

For more than half of the population, that kind of understanding is simply untenable.

And, even for people I knew who could understand it, several of them had so much anxiety that it drove their blood pressure up and it was better to just go get the biopsy so they didn't wind up with a stroke or heart attack.

In addition, "mortality probabilities" are a lot harder to psychologically absorb when you are 50 than when you are 20.

By the time you are 50, you've probably had more than a few of your friends and family in a coffin from a "low probability event".

Emotions are difficult. At some point, I will get diagnosed with a positive prostate screen (practically all men who live long enough do) that will almost certainly be "We should keep an eye on it." I hope I can weather it with the aplomb of my engineering background. We shall see.

Upvoted, but I remain skeptical. "You might have cancer but the probability is low and we should just watch it" is true for everyone who hasn't been screened, just the "might" is somewhat higher for those with a bad test result. Benign-looking moles on your arm could be cancer. Things you can't see or feel in your organs could be cancer. Also, plenty of other things could kill you practically without warning—heart attack, stroke, car crash, and the chance of the first two (and cancer) goes up over time. It seems to me that people have to come to terms with these things eventually, and "you have a low probability of having this type of cancer" just doesn't seem that different to me (perhaps it's best phrased as "you have an x% chance of being killed by this cancer"). If you've seen several people respond badly despite understanding probability, well, it's hard to argue with that, but I can't help wondering if it's fixable through slightly better communication or education.

Doing some reading about prostate cancer specifically, it seems like it actually is of the form "there is no appropriate action to take, regardless of the outcome of the screen"—or at least it's close enough that some experts think so. "In 2012 the United States Preventive Services Task Force (USPSTF) recommended against prostate cancer screening using PSA.[39] As of 2018 a draft for new recommendations suggests that screening be individualized for those between the ages of 55 to 69. It notes a small potential decrease in the risk of dying from prostate cancer, but harm from overtreatment." And: "A study in Europe resulted in only a small decline in death rates and concluded that 48 men would need to be treated to save one life. But of the 47 men who were treated, most would be unable to ever again function sexually and would require more frequent trips to the bathroom." https://en.wikipedia.org/wiki/Prostate_cancer_screening#Guid...

> "Didn't I just take a cancer screen? What was the result?"

“The result was ‘probably not,’ but the other option is ‘almost certainly not’ so we’d like to do more tests to make sure.”

(If the patient presses for more information, you don’t need to pull out the textbook yet; a simple analogy to explain why a cancer screen needs more tests might be by analogy to a rock screen, which sometimes catches clumps of dirt that need to be prodded a bit more to determine that they’re not actually stone.)

The important bit is to present it as routine and not yet a cause for concern, to avoid causing excessive alarm.

This test is a screening test and thus can never be used to diagnose someone. Only a diagnostic test can be used to diagnose someone.

Screening tests like these will represent a paradigm shift in cancer surveillance and as such will require those administering the test, most likely a pcp, to properly educate each patient about what the test results mean and what steps should be taken next.

The false positive rate doesn't depend on the number of true positives in the population. However, as you rightly point out the metric that actually matters for these kinds of screening is the precision/positive predictive value, and that one does.

96% specificity. That's really good for a screening test.

The issue is that there's no follow-up test, at the moment. The cancer is otherwise undetectable.

...and the test does not determine location, so it's impossible to treat even with a positive test.

So from a clinical standpoint it's nearly useless, except to schedule patients for more frequent traditional testing. ...the counter point is that the stress caused to a patient is probably not trivial, even from a physical health perspective.

...but from a scientific standpoint, it's awesome and well worth further study.

>> Nearly useless

Being able to catch a cancer early on because of increased surveillance is hugely valuable. Part of why cancer is so dangerous is because it often grows/spreads for years before you get any symptoms. If you're in the unlucky cohort of <50 years old and have an asymptomatic cancer, it might go decades before anyone thinks to check.

And mind, increased surveillance typically means once a year tests or scans, so it's not so overly onerous. It's possible even that generally being in frequent contact with doctors and more seriously minding your health could have net benefits in preventing other maladies as well.

> except to schedule patients for more frequent traditional testing

That sounds very useful to me!

You actually could potentially determine location by examining mutation patterns, both by looking more closely at methylation and doing further sequencing of cell-free DNA.

I imagine that there will be a longitudinal trial for this test. Individuals will be tested at set intervals and the test results will be correlated with clinical outcomes. It is possible that enough blood samples have already been collected such these same correlates could be identified retrospectively.

As the parent comment said - whether or not that's "good" enough for a screening test really depends on the prevalence of the disease it's testing for.

What kind of prevalance do you need for that to be bad?

Lets whip out Bayes' rule:

P(Having cancer given positive) = P(Positive given having cancer) * P(Having Cancer) / P(Having a Positive in general)

Lets say 1% true positive rate is a "useless test".

0.01 = 0.88 (Detection rate from paper) * X / 0.05 (inverse of specificity)

X = 0.0005

Incident rate of Stomach Cancer (most common) is 27,600 per year in US. The denominator for this fraction is iffy because there are a lot of biases on current detection rates. If we go ahead with "adults in the US" then the rate of cancer is 0.00014 [cite](https://www.cancer.org/cancer/stomach-cancer/about/key-stati...)

Basically, if we applied this test to all adults without other diagnostic factors. This would be worse than 1% true positive detection rate. I don't know about how effective intersecting with other diagnostic criteria would be on increasing the odds a patient actually have the disease and it might or might not still be useful after that.

This sort of study is mostly bullshit anyway. They took a relatively small pile of data, made a small classifier with high accuracy (maybe overtrained, maybe not) and really can't validate it.

The wiki article on positive predictive value has the formula for calculating it based on the sensitivity and specificity of a test, and the prevalence:

https://en.m.wikipedia.org/wiki/Positive_and_negative_predic...

Perfect! Thank you.

Ah indeed, I was wrong. You will get approximately 6 false positives for every true positive.

Unfortunately I can't edit my comment anymore.

The introduction of the test itself may change its effectiveness (reducing it). Cancer is an evolving condition that your immune system fights. Perhaps it's not very useful to characterize the test before it's applied since the test itself may change its efficacy. If you start exercising and dieting in response to the potential diagnosis it'll definitely help.

If we had a test that detected everyone and they believed it, we'd improve cancer rates.

Yes, this is right.

But, the even more important metric to measure is mortality. Detecting a cancer that is 100% fatal 1 day earlier than traditional tests obviously wouldn't be that impressive. Even if the PPV is very high.

It's also very possible that a cancer that is detected accurately 6 months earlier, but that detection still doesn't improve our ability to treat the cancer. In these cases, it might even look like mortality is improved if you measure time from detection to death ("early detection extended survival by 6 months"). But, actually, it didn't change much.

So, that's all to say that the "Experimental" in the title should temper expectations considerably, because that has real meaning.

Eventually, this is all about cost. cfDNA assays are targetted to in the end cost less than 100 bucks per sample which for the detection of a single cancer puts you in the range of $12000 for one tumor diagnosis. Ideally, you can now decrease the therapy cost of that one diagnosis while also boosting survival chances. And $12000 is not much in tumor therapy.

The key however here is the advanced prognosis that may allow totally different therapeutic approaches, extend the understanding risk factors etc. and the fact that you do not need to have a specialist on side, taking a blood sample can in principle be done at home and the rest is automatic.

Edit: Oh, and 5% false positives is not that bad actually, I do not have numbers is my head but would expect other essays like breast cancer screening to be also in that range.

The problem is with breast cancer screening the next step probably isn't a full body CT. If you detect cancer in the blood, next step would be actually finding it. A full body CT would actually increase your cancer risk and add the chances for another false positive which might involve more invasive diagnosis. The real risk/benefit calculations quickly become tricky.

All that said, I'd rather risk dying while propofol'ed, than from say colon cancer. I mean all this is about how we die, not if we have to die.

Coming from the field of cognitive neuroscience, where MRI scans are commonplace (because you can get an estimate of brain activity by doing a continuous, "functional" MRI scan), CT feels like ancient technology to me. There are highly specific case where it has benefits, so it has a place. But if I went into the hospital and they sent me in for CT rather than MRI to get an anatomical scan, they better have a damn good reason for it.

Well, you better be willing to wait a couple days for a non-emergent MRI.

Because MRIs take longer, they are a more scarce resource. If you are in the hospital and need a scan, presumably its because the doctors are trying to diagnose something somewhat urgently. If not, they should discharge you and let you get the scan scheduled as an outpatient. So, CT it is for almost all common conditions in the hospital as first line. (even stroke usually gets CT head first to rule out a bleed, where the sensitivity is still pretty good)

I had a mild stroke in 2001. CT did not find it, MRI did.

CT and MRI are simply different imaging modalities that provide different information, usually used to answer different questions. There are tradeoffs for each.

By the way, they were both invented around the same time (CT is only older by like ~5 years).

CT is simply more practical for full body scans, since an MRT would take forever. Imagine laying completely still for 2 hours... CT does it in 5 minutes.

They also differ in diagnostic value for different types of use cases. But I am lacking any expertise there.

Either way, CT plus constrast agent is the standard in cancer diagnostics/staging.

Excuse my ignorance, but I've always been told that CTs are significantly cheaper and faster to administer. According to the famous XKCD radiation graph, a full CT head scan is about 50% of the normal natural yearly background radiation you'll get. Not something to cheer over, but nothing to lose sleep over either. If a CT scan would suffice, why tie up the MRI machine?

The problem is when you are screening lots of people, most of whom don't actually have cancer. The individual risk is low, but the population risk is high. And when we talk about cancer screening we have to think at the population level.

From what I know, the risk is not high enough to leave a strong signal. The problem is the overall high cancer rate for humans. As far as I know, we still don't have a model for predicting cancer rates after low exposure to ionizing radiation.

OK, but you'd be delaying scans by having to wait for the MRI, which takes significantly longer to setup and administer as well as being much more expensive (which would further reduce access).

Also (and this is according to the XKCD chart) you'd need 50 CT head scans before you'd hit the clear statistical cancer risk.

But it's not like the whole population gets a CT scan every year, or even during their entire life. If you need one anyways, there's probably a more pressing issue that 1/50th the clearly statistical risk level.

The dosage is in a shorter amount of time (within seconds instead of months). I don’t think the radiation averaged over half a year is comparable to the head CT.

My understanding is that the dosage within those few seconds is the equivalent to absorbing half a year's worth of radiation. That's explicitly what Sieverts are designed to measure: https://en.wikipedia.org/wiki/Sievert

Correct me if I'm wrong, though?

It's disputed, if the linear, no threshold model is adequate. The problem is, 1:5 people die of cancer anyway and the people getting fullbody CTs tend to be either old and die rather soon, or are people who had cancer at some point to begin with. Really hard to find a signal there. The best data is still from the nuclear accidents/bombings.

Cells can repair (maybe faulty) a certain amount of damage, but may suicide when too much is broken (double-strand breaks). The amount and type of ROS generated by ionizing radiation also depends on your antioxidant state and how well tissue is saturated with oxygen (more ROS if you exercised before exposure). Generated ROS are a significant factor in cell damage, it's not just direct DNA hits. Some ROS can last for weeks and travel across cells to fuck things up.

I think we can confidently say, lowish radiation exposure is: not great, not terrible ;)

5% false positives on people who have some symptoms of cancer isn’t bad. But 5% of a ‘high risk group’ could be a lot - particularly if the cancer is too early to confirm detection any other way yet and possibly not treatable for several years.

But in this example - stomach cancer - you could just skip the test and do a routine endoscopy and get some tissue samples. You’d have a much higher confidence in the result.

Reminds me a bit of the PSA test for prostate cancer. More of an early warning, but with the weak predictability, doctors are moving away from routine PSA tests.

If option A is regular blood tests with an endoscopy to confirm should the blood test come back positive and option B is getting endoscopies on a regular basis, option A is a no brainer. While it might not be super invasive, I'd definitely like to keep the number of times a tube has to be shoved down my throat to a minimum.

But if the false positive rate is high, just skip the test and do the endoscopy.

No, the false positive rate is irrelevant. Let's say you want to check for cancer annually and the blood test has a 50% false positive rate:

If you only get endoscopies after positive results, you get an endoscopy every other year. If you skip the test and go straight to the endoscopy, you get one every year. Thus you get half as many tubes shoved down your throat if you get the blood test.

For a more reasonable false positive rate like 5%, you would have a 50/50 chance of getting a false positive any given decade. I'd much rather get 4 endoscopies than 40.

Indeed, for any false positive rate less than 100% (which isn't really a test to begin with) still on average decreases the number of endoscopies you need.

Really the metric that's important is the false negatives: if that's too high then I can't safely forego the endoscopy just because this year's blood test was negative. Even then though, if the blood test has a long lead time then maybe it's safe to wait and see how next year's blood test goes.

They'll find something to sell in this case.

Time provides options in cancer.

Even if the first option is "get more regular checkups over the next year+, to characterize the growth rate."

You don't have to nuke an organ from orbit, if you have historical data that indicates a tumor isn't aggressive. Or you would, if you've been watching it and it suddenly takes off.

"You don't" but many doctors sure seem happy to treat early and to be blind to the potential harm that the treatement entails.

And many software developers don't properly use a CVS.

But that doesn't make GitHub a bad idea.

The problem is that you don’t want to treat every cancer like it’s potentially life-threatening (many aren’t) because you can cause other problems with detection/treatment. I’d want any early detection to be pretty damn solid if we’re going to start talking monthly checkups for something that could resolve on its own.

Using an SCM tool isn’t going to cause additional problems except for reminding yourself of all the dumb mistakes you made last week.

> Using an SCM tool isn’t going to cause additional problems except for reminding yourself of all the dumb mistakes you made last week.

You haven't worked at a shop that mandated Team Foundation Server, have you? ;)

Another potential benefit of this sort of early detection would be the ability to more effectively test interventions that might have little effect on more advanced cancers.

That could help direct research into the development of cancers at its earliest stages, and potentially pave the way for more effective prevention.

If you need 12k$ for a single tumor diagnosis, then we should consider MRI imaging. Those cost significantly less than that. And as of right now, the positive effects of chemotherapy on such early cancers would be debatable. Still a useful development though.

Worth noting an MRT is of low value for prevention. Small lesions would be invisible and you could very well progress from nothing visible to seriously sick in 6 month. We really need some marker for before we get unhindered growth.

Even if we had 100% precise AI assisted image analysis, once a year fullbody MRI... the MRI resolution would still not catch the "before things escalate quickly" pre stage of cancer. With imagery catching _most cancers early is inherently a thing of mostly luck.

You misunderstand. 12k$ (or some number in that ballpark) would be per postive test for testing everyone. Your 12k$ for MRI would check a single person. That's something totally different.

The point is that treating a cancer patient easily costs tens or of hundreds of thousands plus dire consequences for the patient themselves. So there is a lot of room for improvement if the diagnosis is made that much earlier.

Blood tests for testing cancer markers during therapy are already standard since a long time. A family member of mine got that testing regularly about 10 years ago but the predictive value was not good. Still, if the accuracy improves it's a good development and could be revolutionary once high enough since some types of cancer are very likely to be curable if detected early enough. But yes, a false positive result can make things worse, especially for people that are in a risk group.

> but the predictive value was not good

My understanding is that there is some dependency on the type of cancer, but mostly it depends on your immune system response. Doctors are pretty clear that just having a cold can affect the numbers.

Personally I'd like to know I'm in really early stages.

I would take positive steps waiting for a retest, such as cutting out refined carbs, saunas or other heat therapies, and eliminating stress and other harmful situations.

false positive? probably still a better lifestyle will have it's own benefits.

(IMHO)

Maybe it can be used as a component/signal in a broader basket of diagnostic tests and symptoms?

Could you take it multiple times to help? Or it could be a guide to what additional screening you do?

That would only help if the false positive occur completely at random. If the error is systematic (for example because the test is detecting something else in the blood) you'd just keep getting false positives.

presumably these other things that would cause repeated false positives would be identified over time and could be ruled out with other tests - ie you either have cancer or a serious iron deficiency, let's check if you have an iron deficiency.

It probably depends on what is causing the false positive.

If it is just some variance in the test, then multiple testings could help (And the test could probably be improved to reduce false positives pretty easily)

But if there is something intrinsic to those tested (say, some hormone level or other) Repeat tests will give repeat false positives. AND it may be harder to eliminate false positives by refining the test

A good econ talk on why this kind of thing (improving cancer diagnostics) is by far the most promising "cure for cancer".

http://www.econtalk.org/azra-raza-on-the-first-cell/

I hope we have many more tests like this that can be done frequently and add to the arsenal. Great work to these researchers.

Azra Raza was also on the Mindscape podcast, really realy good interview.

https://www.preposterousuniverse.com/podcast/2019/12/16/77-a...

This makes me think of an article I read about prostate cancer detection. Apparently, it is possible to detect the possibility of prostate cancer many, many years before it is going to develop with a more sensitive, modern test. Once the detection is made both patient and doctor are not in a position to ignore it. At the same time treatment may well be premature or not recommended. But the doctor is left with a dilemma because it may appear they did not take action on something they knew.

That at least was the general topic of the article. The question it raises is interesting: is it sometimes too early to know.

> Apparently, it is possible to detect the possibility of prostate cancer many, many years before it is going to develop with a more sensitive, modern test. Once the detection is made both patient and doctor are not in a position to ignore it.

I'll tell you a story from when my mother was in medical school.

There was a class one day in which the students were given a pile of slides of prostate samples and told to identify the cancerous one. Many, many questions ensued, of the form "This looks kind of cancerous. Is this the slide?"

Eventually the professor announced to the class "look, there's a bit of cancer in everyone's prostate. The goal here is to find the slide with OBVIOUS cancer."

And this is, to the best of my knowledge, still the state of things in prostate cancer today. All prostates are cancerous. A man who lived long enough would eventually be killed by prostate cancer -- a cancer detectable long, long before he actually died. But most men die of other things, because prostate cancer develops so slowly.

Ignoring prostate cancer is in fact a common way to handle it, if the patient is old enough or the cancer is small enough when detected. The technical term for this is "watchful waiting".

As a corollary, there’s much discussion about the benefits of screening for prostate cancer. The procedure itself carries some risks, and it can lead to operations that add further risks, a significant fraction of which statistically aren’t necessary.

https://en.wikipedia.org/wiki/Prostate_cancer_screening:

“A 2018 United States Preventive Services Task Force (USPSTF) draft adjusted the prior opposition to PSA screening. It suggests shared decision-making regarding screening in healthy males 55 to 69 years of age. The final recommendation for that age group states screening should only be done in those who wish it. In those 70 and over, screening remains not recommended.”

The idea here is to screen people who both have a relatively high risk of having prostate cancer and statistically have long enough to live for it to become problematic, but even for that group, screening still isn’t a clear win.

And this is for the USA, where, in general, sending in the cavalry and doing something is preferred over a “wait and see” approach.

This is highly controversial in the field and is looking like a bad recommendation from the task force. A study has shown that metastatic prostate cancer has increased as a result of this.

Men should get a PSA test in their 40s.

https://www.cancernetwork.com/view/increasing-rate-prostate-...

One of my mentors, age 67 got recently diagnosed with prostate cancer due to a PSA test. No symptomes. His doctor (from the VA), recommended radiation treatement. The treatement fried the nerves in his legs and he now has much trouble in life and doesn't trust a thing they say.

I’m sorry to hear that. However we should be careful not to draw general conclusions from anecdotes.

Did they get second and third professional opinion? You need to do your due diligence with these diagnostic tests in my experience as a patient, it falls on our shoulders to make an informed decision on carrying out invasive preventive procedures (risk - lifelong side effects vs reward - preventing cancerous tissue growth, etc.).

Otherwise you can end up in a situation where you regret not taking action sooner when things have taken a turn for the worst e.g. terminal illness that could have been prevented if caught earlier e.g. metastasizing cancer cells

He got radiation treatment (for prostate!) at 67? That's very odd -- radiation treatment is not recommended for a patient who would be expected to live for more than 10 years in the absence of the cancer.

While there other treatment options. Radiation therapy is a very common treatment to patients of all ages and isn’t odd at all.

Radiation therapy for prostate cancer is in fact odd for a patient expected to live to see the side effects. Standard treatment is nothing for a small tumor, surgery for an advanced tumor when the patient has 10+ years of life expectancy, radiation for an advanced tumor when the patient has short life expectancy.

In fact you are a bit wrong. Surgery and radiation are routinely considered as options for treating the same type of tumor. It is more common to do surgery in younger patients but it is incorrect to say radiation is “not recommended” in these cases. And 67 is not young.

Source: UCSF Prostate Cancer team (#4 ranked by US News & World report)

I don’t know what the right amount of testing is, but proponents of less testing argue that its downsides are worse than its benefits, not that testing doesn’t prevent cancer.

I don’t see how the reference you give addresses that.

> Ignoring prostate cancer is in fact a common way to handle it... The technical term for this is "watchful waiting".

That is not watchful waiting. There is a lot of active watching in watchful waiting, not ignoring. For example you may get looked at every 3 months, either a blood test or an TRUS ultrasound, plus a biopsy every couple years. If it gets worse and crosses a threshold, you might get treated.

This is entirely different from just ignoring it. There are many different grades of prostate cancer and the key is detecting the aggressive types before they metastasize.

Just to clarify what you are describing is called “active surveillance” and is a step up compared to “watchful watching”. Both are ways to deal with prostate cancers.

It depends on risk, diagnosis, and age, but generally with active surveillance you will be actively monitored, you’ll monitor PSA volumes and prostate sizes (and get biopsies), usually on a regular basis. Watchful waiting is much less invasive. When someone is old, their cancer not aggressive and low risk, and is dealing with other health issues, subjecting them to numerous tests for no real benefit is a big stress to both them and their families.

I've heard it phrased by a doctor as "Almost every man dies WITH prostate cancer. Almost no man dies FROM prostate cancer."

2.4% of men die from prostate cancer, 33K/year in America. This is easily googleable.

https://www.cancer.org/cancer/prostate-cancer/about/key-stat...

Anecdotal but both of my grandfathers died from it, one at age 52.

10 year survival rate of prostate cancer is 98%.

https://www.cancer.net/cancer-types/prostate-cancer/statisti... 5-year survival rate for most men with local,prostate cancer combined is 98%25.

Note: That’s if it’s detected.

A pathologist once told me the likelihood of having prostate cancer is basically you age: 50 -> 50%, 70 -> 70%, ...

103% for my grandfather then

You might want to apply a

    min(calcProp(), 1)

Maybe he has one advanced tumor and one polyp that's about 3% more cancerous than expected.

It's 3% . It resets itself to zero after 100.

>Ignoring prostate cancer is in fact a common way to handle it,

Most people find this kind of advise difficult to swallow. Ignoring is equated with negligence, conversely heroic efforts (chemotherapy etc.) is glorified even if most of them produces no appreciable positive outcomes.

This would be like, high cholesterol may lead to heart disease 5 years from now.. atleast cholesterol can be cut, but yet to start cancer may not be stoppable.

This is an incredible confusion, and I'm angry to see so many medical authorities and administrations taking the same stance: no, it's never too early to know. The more the information the better everything. What could be too early is to act, supposing that by "act" is meant any predefined medical action. However why would it be any "too soon" for an awareness awakening by the sick individual that would bring a healthier lifestyle?

Last thing I'd like to add, you just have to tell the individual about the trade-off and risks of the early act too, and this additional information will be enough to cancel any supposed danger of the early diagnostics information.

When you think that giving some information is dangerous, then actually it's because you need to provide the targeted individual even more of what info you have.

It's just hard to take in that info.. Your average patient is not really capable, especially when delivered life-changing information and requiring THEM to decide what life-changing action to take. Most in that situation defers to what the physician recommends, which brings us back to the current situation.

One root problem here is the lack of a firm connection between test result and long-term prognosis. You can have a supersensitive test, but if the prognosis varies wildly anyway and the possible procedure to do carries significant risk, the test has limited value.

One general rule is that you should not order a test unless you're prepared to act on it, and the guidelines for when to act on the PSA levels are still debated.

There are lots of other tests and pathologies and remedies that make sense to screen for (that can, like you say, make the patient go for a more healthy lifestyle etc).

This is outdated. PSA is a very valuable test as it can flag you for new tests that can be performed short of a biopsy (4kscore, genetics, MRI, ultrasound). And biopsies are getting better at detection as well.

I think that your premise is that the knowledge that the doctor would have to pass to the patient in order for the latter to make enlightened choices would require squarely a long and tedious medical training.

To which I would suggest that

1/ no, the knowledge to have for this particular test results, likely (and less likely) prognoses, and the associated possible medical acts and risks, is not equivalent to a full medical training, as it is a tiny bit of one specialty. But I acknowledge that it may require quite some time.

2/ there is no need to apply the same policy to every patient, and you can choose to only allow tests to the patients who went through the basics training evoked in 1/.

In the end, if somebody wants to go through the process to get the knowledge to be able to deal with the information, why to deny it?

Plus, that would make a nice incentive to get people to know better their own body.

Well I agree fully that there are patients who are smart and could take in the info (for that matter, lots of doctors also end up as patients), and I'd hope their treating doctors give them any info they ask for and get them involved in the aspects. After all, in this specific case I'm pretty sure you can get a PSA taken by yourself even and then the cat's out of the bag.

But in a broader sense, what I was thinking of was that when faced with a shocking piece of info ("you might die and there might be nothing we can do about it" etc.), most people don't stay very composed - there are several different reactions and phases and one is certainly "I want as much information as possible and I want to be involved in all decisions" but I'd argue it's not the most common reaction and even in that case, it's a crisis reaction and maybe not as logically driven (or productive at all) as it might seen at first.

So then you'd first have to figure out how this specific patient would react before deciding to do the screening and if you just guess, you'd probable be criticized at some point, so you instead make sure you follow the national plans and recommendations..

Then the (sad) fact is that you usually have maybe 15 minutes allocated in primary care to say hello to a new patient, listen to the symptoms or concerns, examine, make a plan, inform the patient, say goodbye, dictate the journal. You can't spend hours educating the patient. I'd wish this was different (but then again that applies to any service interaction, even your car mechanic or whatever).

I totally understand...

I think there could be solutions, like:

- proposing bundles formation + test(1), of course much more expensive than just test, but allowing access to a broader range of tests.

- new job description, "medical teacher for citizens", giving (possibly customized) formations to people (and documents certifying they passed), and then these certificates would be useful later to know which patients already understand which aspects of medicine.

I think that would make nice ideas for startups, and I for one would be a customer.

As for the "emotional" aspect, when receiving shocking news, I don't think that withholding information is a solution, quite to the contrary. In a lot of cases, people don't feel well and imagine the worse, but they actually are fine. Other people feel good and actually are very sick. If an information leads to bad diagnostics, they should know, because if not now it would be required later anyway. If the information does not lead to a 100% accurate diagnostic, just quickly get sure that your patient understands basic stats/probas, and then explain to them why the diagnostic cannot be certain and why some benefit/risk trade-off holds.

Any of the "formations" evoked above could include some lesson on how to deal with bad news about oneself.

(1) of course I mean "medical test" here, like the ones to guide diagnostics, not "knowledge test", although these too would be useful at the end of the formation.

PS: the 15 min/patient thing is a huge PITA, so maybe my ideas are doable only in private clinics. Even the "medical teacher for citizens" is completely outside of what is done (currently) in public health.

I pretty much agree with your points. But this is not what we see in reality today.

I believe that part of today's problem is knowledge, interest and incentives.

- About 40% of patients (in general) do not care enough about their health and delegate all decision making to the doctor.

- I am simplifying here: If you ask a surgeon about the best therapy for your cancer, the answer will most likely be surgery. If the same patient would ask a radiologist, the answer will most likely be radiation. Which we might attribute to people more likely to believe in their own line of work. But in reality this might be more influenced by doctors being financially driven by their administrators to maximize patient throughput.

Fortunately these decisions are usually not done by the individual surgeon or radiologist but by a cross-disciplinary team during rounds and by input from wide long-term studies categorising the tumor type, treatment options and outcome.

For metastasized cancers yes. But for primary cancers? Not such much.

Patients need to step up and actively talk to multiple doctors if they want to get a complete picture.

> it's never too early to know

I don't see this. Suppose, hypothetically, that we could detect right now every illness or condition that you're ever likely to develop in the course of your life. Do you really want to read through that list? You're better off not worrying about it and enjoying your life.

I would absolutely read the list, in part because I simply value (self) knowledge. I notice that some people don't value knowledge for its own sake, and have trouble understanding those who make decisions solely for the sake of knowing, or solely to avoid dishonesty toward others, even at some narrow instrumental expense.

But also, I would read the list of because of its instrumental value:

* If I knew I were likely to die in ten years, I would live my life very differently than I do now.

* If I knew I were at particular risk for certain illnesses, I would do what I can do minimize that risk.

But what if you are like most people and trust doctors to do the right thing and they convince you that unless you treat you'll die. You go along and the doctors damage you. This just happened to a mentor of mine (67 y.o.) with a PSA test followed by radiation that fried his nerves and did nothing for the cancer which he might have lived with for the rest of his life (his father died of prostate cancer at 97). In this case the "cure" was much worse than the disease and the early detection led to the almost fatal result.

I’m very truly sorry about what happened to your friend. That is an awful thing to go through.

Without being an expert on the subject myself, I can’t say whether his doctors recommended the right course of action but he got a bad roll of the die, or whether his doctors made a terrible mistake.

But I think if they made a terrible mistake, then the solution is to make more knowledge available about how to avoid such mistakes, not to make less knowledge available about people’s health.

My father recently went through prostate cancer radiation treatment, and here is our experience. The doctor did not say that unless it is treated, he will die. He did emphasize that with the modern day treatments, the risks and side effects are much lower than that of non-treatment. The radiation treatment was non-invasive, except for the placement of 3 markers inside the prostate to guide the radiation beams. My father did have stronger-than-normal side effects, however after 3 months the side effects have gone. We were actually amazed that cancer treatment was so easy. It is now a waiting game to see whether the cancer has been completely eliminated, but the peace of mind from having had the treatment is worth a lot.

On the other hand, a family friend had prostate cancer many years before, when the radiation treatment was not available, and he had to go through surgery. The surgery caused strong side effects, and also turned out to be not successful, as the cancer spread to his bones after 10 years, for which he is now receiving further treatment. In this case, where the loss of quality of life was high, it's not as easy of a choice.

Actually the side effects of radiation are often similar to surgery in the long run. Surgery is “front loaded” and may be followed by some recovery of function over a year or more. Radiation is “back loaded”, where you see the effects creep in over time. It all depends on a number of factors but it’s not that useful to compare short term.

I’m glad to hear you dad is doing well.

I can't really see how not reading the list would require dishonesty towards others.

It's probably a good idea to live your life as if you were going to die in 10 years anyway. The chance of this happening is non-negligible unless you are very young.

Similarly, as far as minimising risk of illnesses goes, 99% of what you could do is what everyone should be doing anyway: eating healthily and exercising.

Fundamentally, the paradox here is that the extra stress caused by reading the list is likely to be much worse for you than the consequences of not knowing what's on the list.

Not reading the list doesn't require dishonesty toward others, but I was making a more general point: There are lots of people who are very ready to tell "white lies" to others for the sake of those others. As it happens, I think they are often the same people who wouldn't read the list. And these sorts of people don't really seem to understand the other sort: the sort who want to know the truth even if it's not instrumental.

---

On the other topics: if I lived like I were going to die in ten years, then I'd have to abandon multiple projects of mine which I expect to take more than ten years to come to fruition. I don't think this would be appropriate unless I actually had a good reason to believe my premature death would come to pass.

I mean live as if there's a reasonable chance that you'll die in ten years. But even so, do you get no intrinsic satisfaction from the work involved? Is there no-one who could continue the projects if you died? If it's 'no' to both questions, then maybe you should stop doing them anyway.

There are always tradeoffs. I get intrinsic satisfaction from the projects. But I think both about the short-term and long-term. If I expected to die in 10 years, I'd do different things, because the balance of costs and benefits would work out differently.

> You're better off not worrying about it and enjoying your life.

I would rather have all of the information upfront so that I can make better informed decisions.

We're all going to die. I'm not going to shy away from that by pretending it won't come.

There's no 'pretence' involved in not reading the list. I'm perfectly aware of my mortality. I just don't see any value in placing bets on what's going to kill me.

Everyone is living with an invisible list made up by them in their minds. This is what drives people to exercise, eat healthy foods, get vaccinated, and wash their hands. This just clarifies the list, perhaps adding some items, perhaps changing the priorities.

Nonsense. Catching cancer prior to stage 3 is HUGELY important. With prostate cancer it can be cut out or irradiated with a high cure rate at stage 2. Once it spreads though you are in a very different course.

I think you're talking about cancer screening rather than the hypothetical in my post. I was not suggesting that no-one should ever get screened for any illness. But it's a balance. Too much screening and testing can be just as harmful as too little - or perhaps even more so. Prostate cancer is actually a classic example of this (see e.g. https://www.health.harvard.edu/healthbeat/the-pros-and-cons-...)

Your hypothetical includes every illness and thus, cancer. It is extremely helpful to detect cancer early.

The prostate cancer field is rapidly evolving, and the “controversy” over PSA testing in the US over the last decade is looking like a big mistake. See for example https://jamanetwork.com/journals/jamasurgery/fullarticle/268...

>Your hypothetical includes every illness and thus, cancer.

I think there's a quantifier scope error here. I was saying "you don't want to see everything on the list", i.e. "it's not the case that for all x, you want to see x". You seem to be reading this as "For all x, you don't want to see x". Whatever the right answer is for specific cases (such as prostate cancer), I'm certainly not saying that all testing or screening is pointless.

The paper you link to has a grand total of two citations listed on Google Scholar.

It’s a comment by the chair of the #4 ranked prostate cancer department in the country. Here’s another: https://jamanetwork.com/journals/jamaoncology/fullarticle/25...

You changed your statement. What you said was “Do you really want to read through that list? You're better off not worrying about it and enjoying your life.” This unambiguously refers to the list in its entirety.

No, you're still making the quantifier scope error. I'm saying you'd be better off not reading the full list. It doesn't follow from that that you'd be better of without knowing about any of the items of the list. Your mistake is anbalagous to inferring (B) from (A):

(A) John doesn't want to eat the whole pie.

(B) John doesn't want to eat any of the pie.

I'm saying that you're better off not eating the whole pie / reading through the whole list. It doesn't follow from that that you're better off without any pie / without knowing any of the test results on the list.

That's why the list in my hypothetical scenario is comprehensive. If I was saying that you'd be better off not knowing anything, then my hypothetical scenario could just have been "you get some kind of test result giving you information about some aspect of your health", rather than an unrealistic scenario where you get completely comprehensive information about your future health.

You are misusing academic language to mask a modification to your claim under interrogation. You specifically argued against “an awareness awakening by the sick individual that would bring a healthier lifestyle“ by stating it is “better off not worrying about it and enjoying your life.” You are now attempting to amend that claim by saying well of course you’re in favor of knowing about the bad things.

So what does this mean, that you actually support looking at a hypothetical comprehensive list, just that it should be sorted by severity so you don’t have to sweat the small stuff? That would be a fundamental modification of your original claim which had no such qualification.

My claim is consistent and perfectly simple. I'm saying that you'd be better off not knowing everything about your future health. It doesn't follow from that you'd be better off not knowing anything. As you seem to have difficulty understanding that, I've been trying to spell the logic of it out in explicit detail. For reasons I don't understand, you appear to interpret that as a modification of what I originally said.

What exactly would you be better off knowing? In general, that's a complex question and I don't know the answer. But certainly there are clear examples of things that are worth knowing (e.g. your weight, your blood pressure, whether or not you have TB ...)

>by the sick individual

I did not specify that the individual in the hypothetical scenario was sick.

Now you are claiming you didn’t argue against that claim you responded to, which specifically referenced a “sick individual” achieving an awareness “that would bring a healthier lifestyle.”

Indeed your claim was clear and simple: you disputed it, replying it was better not to worry about it and “enjoy your life.”

You have simply dressed up poor reasoning regarding “you’re better off not knowing” by pretending you were only referring to the things not worth knowing, not the things worth knowing, the other part of the set of all things. This is spurious argumentation designed to distract from the fundamental shift in your claim — that of course it is never too early to gain awareness of medical conditions that could motivate lifestyle changes and inform choices.

In the end I am glad you came around to agreeing with the comment, if by a circuitous route.

The quoted claim that I responded to was the claim that "it's never too early to know". Any my claim was made in relation to the hypothetical scenario that I spelled out, where it was not specified that the individual in question was sick.

I think you think I'm changing what I'm saying because you're still making the quantifier scope error. If you read through my comments paying more careful attention to the quantifier scope, you'll see that I've been saying the same thing consistently. I can only make sense of your responses if they are objections to "for all x, you're better off not knowing x", rather than to what I actually said. This misunderstanding is evident from several of your first comments. E.g. "Your hypothetical includes every illness and thus, cancer", and "This unambiguously refers to the list in its entirety." These replies only makes sense if you imagine that I'm saying that you don't want to know anything on the list. But this simply isn't what I said. Or at least, it's the least plausible interpretation of my original comment, and inconsistent with the explanations I've repeatedly made of what I meant.

I don't know why you are so reluctant to accept this. Early in this thread I politely pointed out the quantifier scope error in your interpretation. And the wrong interpretation that you insist on attributing to me has me saying something implausibly ridiculous (i.e. that all medical tests are bad!) Go back to my examples (A) and (B) earlier. Do you have difficulty understanding that (A) does not entail (B)? I doubt it. So why is it difficult for your to understand my claim about the list? I'm baffled.

The pie analogy is a good one. Every piece of the pie is nutritious and worth eating (let's say). Similarly, every item on the unrealistically comprehensive list of medical information may be useful in and of itself. But just like you can have more pie than is good for you, you can have more information about future problems and risks than is good for you.

Prostate cancer is somewhat unique by having a very sensitive blood marker (PSA) being available which can be monitored easily over time.

And yes, there is an (I would say) ongoing debate about the value of using PSA to detect early cancers. Obviously detecting cancer early is a win. But the downside of this early detection is over-treatment of low-risk cancers.

Standard therapy for prostate cancer is the removal of the prostate, but this surgery has a somewhat high (up to ~10-15%) risk of impotence and incontinence (depending on the quality of the surgeon), which can obviously impact life quality quite massively.

Current strategy (in the US) seems to basically stop broad usage of PSA testing, which gets rid of the problem of detecting low risk cancers, but at the cost of delaying the identification of high risk cancers. This is visible in the falling rate of newly diagnosed prostate cancer patients in the US.

This feels like not solving the actual problem (distinguishing low risk from high risk tumors) but glossing over the problem by not actively working on detecting it anymore.

>Prostate cancer is somewhat unique by having a very sensitive blood marker (PSA) being available which can be monitored easily over time.

Cancer antigens such as PSA and CA-125 can also increase due to non-cancerous causes. Pretty much all of them increase in any kind of inflammation whether it is due to bacteria or something else and more. So their main use is getting a baseline once the cancer is diagnosed and monitoring it later on.

Funny story, a professor once had a patient with prostate cancer, treated and cancer free afterwards. Every 6 months the patient would visit for screening and his PSA was through the roof( think >1k while normal levels are at most around 3ng/mL). After performing CT scans etc the patient was found cancer free. Another PSA test would return normal after a couple days. This story repeated several times with the professor slowly sliding into madness from it.

Turns out the patient was getting anxious about the screening the day before so he smoked some weed which increases PSA secretion. And of course weed is illegal here so he didn't want to admit it.

Should have stressed the "over time" part more. Yes PSA is sensitive to sex and sport (esp. cycling), and also to inflammation. Plus you personal PSA value depends on the actual size of your prostate.

That's why getting an early PSA baseline is important, so you know what value is "good". Also you might not want to take drastic action based on a single "bad" PSA.

No, you got it wrong. You get a baseline PSA measurement AFTER the diagnosis. That way you can see if the patient responds to treatment and also monitor for recurrence.

Cancer antigens are NEVER to be used for diagnosis. Chemotherapy is too destructive to be used without 100% certainty. We've had a woman with high CA125 and it turns out it was tuberculosis in the reproductive system... Everyone wanted to start chemo, the symptoms matched etc...

PSA is definitivly used for early screening prior to diagnosis [1].

But we might use the word baseline for different things.

As PSA is influenced by the patients physiology and activity [1], it can be helpful to understand what the non-cancer baseline/standard PSA value for a specific patient is, this would need to be established by multiple tests over a couple of years. This can help in distinguishing noise from signal later.

After surgery the PSA baseline is zero (by definition, the generating organ has been removed). Any value above zero indicates either metastasis in other body parts or a really bad surgeon.

After non-surgery primary treatment (radiation, ...) the PSA does not (necessarily) fall to zero, so here one takes the lowest value after treatment (with a grace period) as the baseline.

[1] https://www.cancer.org/cancer/prostate-cancer/detection-diag...

In my father's case, our family doctor requested a prostate cancer screening after seeing PSA levels from regular blood tests rise over the course of a year. We received the cancer diagnosis after the screening.

> but at the cost of delaying the identification of high risk cancers. This is visible in the falling rate of newly diagnosed prostate cancer patients in the US.

Importantly, it doesn't seem to affect mortality rate of prostate cancer, nor all cause mortality.

People focus a lot on 5 year survival rate, but sometimes they don't really understand what this means. If I die at 75 does it matter if my cancer is detected when I'm 68 vs 73?

> If I die at 75 does it matter if my cancer is detected when I'm 68 vs 73?

Yes, it matters in both directions.

One way is that the cancer might cause you to suffer in other ways, e.g. by metabolizing all of your food, causing you to lose large amounts of weight. If this is addressable (not at all a sure thing!), then knowing earlier is probably better.

The other way is that you might not be suffering. In that case, given that your death is fixed, knowing earlier is obviously worse.

I've never seen someone die of cancer without being treated beforehand. I understand it's a horrible, painful death. That said, the treatments we apply to someone dying of cancer also make for a horrible, painful death; there definitely is a point at which the treatment is worse than the cancer.

Another wrinkle is that treatment may give you a few horrible months, then a few not-so-bad-years before your cancer comes back, and then the horribleness resumes. Those years in the middle are worth something, even if the time at which you die is fixed.

If we ignore brief-to-medium remissions, and the hypothetical patient's time of death is fixed, then the obvious recommendation is to avoid all treatment and commit suicide when your quality of life drops too low. But the uncertainty we have in reality makes that difficult.

I tend to agree that this is not such a relevant topic for the average prostate cancer age group which is 70+. Here most might die with prostate cancer but not from prostate cancer.

But in the lower age groups 40-60 early detection should IMHO be the aim.

It is also visible in the increase in metastatic prostate cancer.

I’ve spoken with the chief surgeons of two major prostate cancer centers. They agree this was not a good recommendation by the US task force.

Also I should add that the risk of impotence / dysfunction is much higher with late-detected cancers.

https://jamanetwork.com/journals/jamaoncology/fullarticle/25...

Makes me wonder if there are similarly low risk medication that fight cancer.

Immunotherapy, where drugs target specific mutation mechanisms, is making massive improvements lately. I have seen bone scans from metastasized patients before and after, where most metastasizes were gone after a couple of weeks of treatment. Unfortunately today these drugs only work for a small subset of all tumor mutations, so it will not work for everyone and even if it works, the tumor might mutate and kill you anyway.

With some specialized exceptions, there isn't. Cancer medication has serious side effects.

Early detection of prostate cancer with PSA testing and digital rectal exam: https://www.hardingcenter.de/en/early-detection-prostate-can...

> With or without screening, 7 out of every 1,000 men died from prostate cancer over a time period 11 years.

But in the screening group additional men were harmed by side effects of surgery.

I am not sure what percentage of patients still require surgery as opposed to the more modern radiation treatments. My father's radiation treatment was painless except for the placement of the 3 tracking markers in his prostate, and although his side effects were more severe than most, they have disappeared after 3 months. So if prostate cancer treatment does not affect eventual death from the cancer over 11 years, then the questions are whether the treatment improve QOL during the 11 years, and whether the onset is delayed within that 11 years.

Stomach cancer, for one, is easy to remove surgically if detected early but it does not exhibit symptoms until it is often too late. I suspect that early warnings would trigger targeted endoscopy procedures. The pathogenesis of prostate cancer may not be representative of other cancers.

Same with bowel cancer. Easily treated if found early. There are at home test kits which can detect small amounts of blood in stool -- often an early warning sign. And while unpleasant sounding, colo(no)scopy is pretty uninvasive.

This is widely discussed in medical circles, but the real problem is a different one: It's sometimes hard to know what helps and what doesn't.

You may see something that may lead to a cancer. But you really don't know. It may be a) harmless, b) lead to a cancer, c) a mistake in your test. If it's b) then treating it early is good. If it's a) or c) then treating it early will ultimately harm the patient. If your test finds mainly a) and c) then your test may ultimately do more harm than good.

Isn't this the idea behind watchful waiting? You have an early signal so you follow up with periodic tests to see if it goes away or shows signs of malignancy.

Eventually we'll have studies that show the conditional probability of different cancer outcomes based on these early signals, and from there we'll have a protocol to get the best outcomes, reducing the chance of malignant cancer while minimizing cost and medical intervention.

Even if you could predict with certainty that the patient is going to develop cancer in 10 years, and you know of a course of treatment you could undertake that will ultimately prevent that cancer developing, you still may not be able to say whether this is the best use of the patient's time and resources today. (And the patient can't be sure either.)

So the best course to take seems to be a combination of the probabilities of each case along with the risk tolerance of the patient, thus it is not a question the medical community can answer, but needs to be an individual discussion with each patient.

it is at minimum an early warning for lifestyle change

>That at least was the general topic of the article. The question it raises is interesting: is it sometimes too early to know. reply

Bit like near earth objects, we can detect one far far away and whilst could be 95% sure it will hit, either way, unable to do anything to it until it gets closer. But always good to know ahead of time so can prepare and eventually able to deal with it further out.

I have a SNP (according to ancestry) that gives me ~5X the chance to get prostate cancer. What treatment is available for prostate cancer, besides debilitating surgery?

First, you should be focusing on detection. You probably know about the PSA test and biopsies, but there is also a more sophisticated 4kscore blood test. MRIs and TRUS ultrasounds are also diagnostic tools.

If it’s detected then they will biopsy it and grade the cells to see how aggressive it is. If it’s not aggressive they will just monitor it without treatment.

If it comes to treatment, and you’ve caught it at stage 2 (hasn’t spread), then it’s either cut out your prostate or zap it somehow (e.g. localized radiation).

For “younger” patients they usually recommend surgery. First, with early detection it is less likely to be debilitating. They can spare the nerves that control erectile function, although they can still get messed up and there is still a big chance of some decline in function. But it’s a better outcome than older age brackets or larger tumors — keep that in mind when looking at overall stats.

The reason they recommend surgery is the long term side effects are similar to radiation and if they don’t get it all with surgery, they can go back in and mop up with radiation. Whereas the reverse doesn’t work as well because of tissue damage.

(Note: some people are drawn to alternative ways of zapping it like proton therapy in the hopes it will have less side effects but so far, nothing conclusive.)

Triple therapy is what a relative of mine underwent after getting Gleeson score 9 (bad but not metastasized) prostrate cancer. He hadn't been doing regulator screenings or would have caught it earlier. The triple therapy (radiation + hormone + brachytherapy, basically little radioactive implants) sucked in terms of side effects at the peak, but it hasn't been long term debilitating, amd he's now doing well in remission.

Chemo and radiotherapy, the whole trifecta. Some new ways of implanting small radioactive disks straight into the prostate with a minimally invasive procedure are being developed as well.

Surgery, done by good surgeons, has low (single digit) complication rates.

Tip: Ask your surgeon how often he does this specific surgery. Removing the prostate requires foremost experience to not severe any nerves surrounding the organ, which are important for potency and continence.

As you would not want your cars brake firmware to be written by someone who usually writes marketing apps, you do not want to get this surgery from anyone who has not specialized on it. There are specialists who do this surgery >200 times a year.

I'm in my mid thirties, I'll look into PSA, and my father has prostatitis, but no family history of prostate cancer(my grandfather died 86 years old, but lived a much healthier life than me or my father).

So I'll definitely keep it under observation, I just hope those percentages will improve in the next decades.

Someone I know had robotic keyhole surgery and they were up and walking about in a week. That was a man in his 60s with mediocre health. He went to the Henry Ford Medical Centre.

Keyhole is the key here to walk out asap. The "roboter" is not delivering that much in terms of quality of outcome or speed of recovery.

Depending on personnel preferences, you might want to get a regular PSA check once a year to ensure you catch it early to give you runway to evaluate treatment options. It's better to be able to assess the situation and contact multiple experts than having to rush if detected to late.

I’ve studied this topic a lot and consulted with top prostate cancer centers of excellence. Key points:

- Disregard what you may read about PSA testing / screening being “more harm than good”. This was a relatively recent recommendation in the US by a non specialist task force that is looking to be bad advice. Studies now show it has led to an increase in metastatic cancer.[1] You don’t want that.

- Every man should get their PSA checked by 45, or earlier if it runs in your family. It is a simple blood test. We are lucky to have a tool like that at our disposal; not always the case for other cancers.

- Prostate cancer treatment can have major impacts on quality of life including sexual function. You are much better off getting treated before it reaches stage 3. Quality of life is better. This is not reflected in mortality stats.

- People do die from it and it is a slow and painful death like most cancers. About 2.4% of men die from prostate cancer.[2] With early detection it is very treatable, however, even curable.

- There are advances in screening techniques. PSA is an imperfect but easy and cheap initial screening tool. Doing it early can also establish a baseline. If it’s abnormal it just opens up a conversation and more screening options. Biopsies are also more accurate these days.

Men, go ahead and get your PSA test.

[1] https://jamanetwork.com/journals/jamaoncology/fullarticle/25...

[2] https://www.cancer.org/cancer/prostate-cancer/about/key-stat...

If false positive rate is 5%, and 1000 of 123,000 were diagnosed with cancer over 10 years, this test would produce 6,100 false positives, and 1,000 true positives.

Six out of seven positive tests would be false positives.

Would having 6000 false positives be okay for being able to find the 1000 true positives?

Would further testing be able to show whether or not there is actually cancer?

Maybe they will be able to improve the accuracy of the test enough where the false positive rate of the test is <<5%...

> Would having 6000 false positives be okay for being able to find the 1000 true positives?

I’d say no. That’s 5000 people who are told they might have cancer while in fact they don’t and who will then undergo further testing that is at best traumatic and at worst extremely invasive if not outright damaging.

Take intestinal cancer, for example. You can get screened for that and if you screen positive, they will take a biopsy - a fairly unpleasant and not entirely risk free procedure. Then it gets worse. If the biopsy shows some part of your intestines isn’t quite in order (a likely outcome, though not necessarily one that indicates cancer) they will take out that bit of intestine, just to be safe. This is a very invasive procedure and because it’s your intestines we’re talking about, you digestive system will be seriously out of whack for a while - if it ever heals completely - seriously and significantly degrading your quality of life. All this for a, let’s use your figures, 1 in 6 chance that you might have had cancer.

You're leaving out all choices in your decision pipeline.

If you screen positive, you can elect to have a biopsy or scan.

If your results come back inconclusive, you can elect to pursue surgical (or other) intervention.

That's what medicine is: a selection from among potential procedures, given the information at hand, informed by previous outcomes in similar or applicable situations.

In this case, we simply have more information. If said information turns out to have limited predictive power on the ultimate occurance of cancer, that will be incorporated into detection and treatment practices.

But the important thing is having more information than we had before.

If the patient were a robot, this would make sense. But people are easily scared, especially by the word "cancer," and fright can lead to bad decisions. And this is a decision that's made by individuals, including the doctors, who just want to take the "safe" option.

For example, the United States Preventative Services Task Force, which effectively decides the preventative procedures covered by Medicare and Medicaid, recommended against routine prostate specific antigen testing for prostate cancer in 2012. They backed that up with data on mortality and unnecessary procedures. Prostate doctors around the country flipped out. The doctors are focused on helping their individual patients. That's good, but it keeps them from seeing the overall harm to the entire population.

Why don't you just ask them?

"Based on the results of this test, you have a 1 in 6 chance of having cancer. This is what the biopsy process looks like, would you like one?

It also means that the test fails at detecting cancer more often than not, and will mean that if it detects a true cancer this time doctors will be inclined to not take it seriously because evidence/experience is that the test is useless

I suppose this takes into account the fact that doctors are humans too, and humans are typically not good at statistics. The probability of cancer from a positive test rises from 0.8% to 14.1%, a significant increase, and seems to warrant a closer look, from a layman's point of view at least.

And one out of seven is a true positive, whereas overall 1 in 123 people developed cancer. So a positive result on this test still means you are 17.5x more likely to develop cancer than the average person in this group if they were never tested.

It's weird to pretend like this is not significant new information, which has the potential save lives. Whether or not we have an acceptably safe way today to act on this early information is a separate question, this is one more tool that can be used in improving treatments.

Original paper [1]:

> Here we report the preliminary results of PanSeer, a noninvasive blood test based on circulating tumor DNA methylation...

I’m very surprised that parts of the methylome, the per-cell state-machine that controls what genes are transcribed, have sufficient serum concentration to be detected. I suspect that this approach should work for any diseased cell targeted by the immune system as long as the disease results in a small but unique methylation signature.

[1] https://www.nature.com/articles/s41467-020-17316-z

With most of these (cell free DNA) assays, its not a single site that you are measuring but a statistical aggregate about profiles that have been defined for a certain cell type. If you know the overall profile of prostate cells and the profile of normal blood, you can subtract the expected profile from a patient sample and the analyse the left-over fraction for whether it is just noise or is matching to one or multiple known cell types.

The surprise from this publication is only that they detect such signatures 4 years earlier than other non-invasive testing. (although nobody in the cfDNA field would be surprised about that).

There are newer methods that use antibodies to enrich for methylated dna prior to sequencing [1]. The method increases sensitivity as the majority of plasma cfDNA is unmethylated and hematopoietic in origin.

[1] https://www.nature.com/articles/s41586-018-0703-0

gww and patall both point out that this is related to cfDNA [1]:

> Circulating free DNA (cfDNA) are degraded DNA fragments released to the blood plasma.

[1] https://en.wikipedia.org/wiki/Circulating_free_DNA

Guess it's time to share this again. "Cancer screening has never been shown to save lives" - Oncologist Vinay Prasad. Detection is not the problem. It's what to do once you find it. Iatrogenic harms are real.

https://www.bmj.com/content/352/bmj.h6080

With better testing, we might be able to save some people, but we might also be subjecting a great deal more people to harms through unnecessary treatment and psychological trauma.

You can't design interventions for problems that you cannot see. Detection alone doesn't save lives, but it is a necessary step to the discovery of interventions that do.

There's a lot to this, if you look hard enough you'll find something wrong. But that something wrong probably would have never progressed into something that kills you - however the interventions you then undergo are not without risk. Like if you have the slow-growing prostate cancer, you'll die of something else. But if you operate you're taking risks of surgery and lifetime incontinence, and might spread it elsewhere in the body.

However, I still think early detection is an important first step to finding better interventions. Once you know it's there you can start to come up with less risky/invasive treatments. The odds of dealing with it at an early stage are so much better and the scope of the intervention required so much smaller.

We call it the "Do I qualify for health insurance?" test.

If you care even the smallest amount about your own, personal, cancer risk you should immediately cut your feeding window to 8 (or even 6) hours per day and start walking a few miles, daily.

I did not mention "fasting" or "diet" or "weight loss".

Just condense your caloric input into a tighter window and walk more than zero miles per day.

It's silly to pursue fine grained, specific optimizations when the juiciest, lowest hanging fruit is stubbornly, inexplicably, left unpicked.

Source? There’s so much information about nutrition online that one should not just trust the first comment found on a forum.

Edit: I’m referring to the “feeding window”

I think the common term is "intermittent fasting" (despite the protest about the word "fasting"), which covers a range of strategies—"time-restricted feeding" is the type where the fasting period is less than a day. https://en.wikipedia.org/wiki/Intermittent_fasting#Types

There is at least some research supporting the claim that time-restricted feeding inhibits the growth of cancer in humans: "In repeated-measures Cox proportional hazards regression models, fasting less than 13 hours per night (lower 2 tertiles of nightly fasting distribution) was associated with an increase in the risk of breast cancer recurrence compared with fasting 13 or more hours per night (hazard ratio, 1.36; 95% CI, 1.05-1.76)." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4982776/

There have also been studies in mice: "Results and Conclusion: TRF had a dramatic effect, reducing tumor growth in obese mice fed a high fat diet (HFD) to levels seen in lean mice. Tumor growth and initiation was also delayed in the transgenic PyMT model of mammary tumorigenesis. Our results further suggest that the antitumor effect of TRF is at least partially mediated by reducing hyperinsulinemia, suggesting that this intervention may be effective in breast cancer prevention and therapy." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6552025/

Seems really cool, but how many cancers progress from something minor to something life threatening? Some of the treatments can be pretty invasive/lot's of side effects. I wonder how much cancer the body can deal with on it's own and never progresses?

Interesting, especially considering pancreatic cancer. That one can be treated if caught early on, but it hardly ever is, making it incredibly lethal, even for a cancer to be.

To be clear here: these are only circulating cancers. Cancers that break through mesenchymal barriers or can affect the blood are not very common. It's great, amazing even, if you can catch them. But many cancers will never show up in the blood. Still, great work!

Call me silly, but, methylation is a hallmark of aging as well. So all cancer patients would show methylation, but all methylation will not indicate cancer. If that were the case, cancer would be the only cause of natual death.

That's not what the test is doing. They are looking at methylation in specific places.

Cures and vaccines would be great but I feel like we are missing an opportunity with improved testing not just for cancer but also for infectious diseases. Imagine a toilet that can detect a UTI/STI or parasite or a monitor in my room that can detect airborne viral particles which might tell me I have a cold, flu or COVID before I even know about it. Then I know to treat and isolate before spreading it to others.

If we had cheap, instant, accurate tests for infectious diseases then I’d imagine we could begin eradicating some of them.

Breakthrough medicinal research is great, but we should always remember that even most promising results - see the COVID vaccines/treatments we've been hearing about over the past few months - end up not viable at the late stages.

So, cautious optimism, not great hope and sensationalism, is the right attitude to take here.

  “We are still a ways away from having an accurate blood-based ‘pan-cancer’ screening test.
  But it is not impossible to achieve,” Pritchard says.

I think the article itself got you covered there.

Reply on news.ycombinator.com