Are we sure that early detection really improves survival times or does it just look like this in the statistics because we detect the cancers in an earlier stage?
It depends upon a fact about what we mean by survival rates.
In most cases survival is defined in terms of fraction of population who are alive after a set period of time from diagnosis.
Suppose Bob is going to die of Foozle cancer in 2025. If I diagnose Bob in 2019 he gets more than five years survival. If my diagnosis method only catches it in 2023 the survival rate looks much worse. But all that changed is when I told Bob he's going to die.
In contrast if I develop a treatment for Foozle cancer, maybe Bob doesn't die until 2030. He's survived more than five years after diagnosis again - but this time it's because I was able to actually help.
You're forgetting a much simpler scenario that could wildly throw off survival statistics: early detected cancer might have a chance of never actually killing the patient even without treatment. Therefore early detection increases survivability in the statistics because it's counting more "duds".
Well, my father died of pancreatic cancer three years ago. It was detected in the last stage due to very strange positioning. He only had a few months to live at that point.
I really wish we had better methods to detect that cancer. Maybe before it metastasised he could be saved?
Pancreatic cancer has very poor survival. 5-year survival for stage 0 to II (i.e. early detection) is 34%. Most pancreas tumors are KRAS mutants - undruggable and aggressive.
Or in an extreme scenario, have certain organs be removed that have the deadliest forms of cancer. ie. Pancreas and gallbladder. Humans can live without either.
>It seems likely that, in the next 50 years, instead of actually curing cancer, we will just get really really good at early detection and prevention.
You got it _very_ right.
Many years ago on youtube I heard a non-main stream cancer doc who said that, the battle against cancer has been stagnant in 40 years ( so 50 years sounds about right). Main stream medicine does lot of massaging of numbers/statistics to give the impression that we are making progress on cancer. To add to the illusion that there is progress, early detection means a person diagnosed with cancer is likely to survive well past the 5 years mark. The 5 years mark is one of the parameters that is used to define the cancer 'cure rate'.
Wake me up when toilets tells us anything we need to know about our health.
Depending on implementation it could even be scary. A companion app that gives you AmazonFresh ads: A little low on zinc today eh? Why not try some Pumpkin Seeds™
Is anyone working on something like a smart toilet? Sounds like something papa Bill Gates would bank roll
There is some trial work going on to get kits for you to use to test Faecal calprotectin at home and use the camera and an app in your phone for analysis. Obviously there is a hurdle of actually having to handle faeces and put a sample in a tube so it is not something people would love to do. But it could be something people with IBD could do in the future to catch flare ups of their disease early.
If this works, it is going to put a lot of GI doctors out of business.
The guidelines for screening after 50, plus the ACA's provision that makes colonoscopy "free" has led to a large increase in the number of screenings over the last 20 years [1], and I suspect that a lot of GI docs are making out like bandits.
I don't understand the comparison between the unpleasantness of a colonoscopy vs. collecting a stool sample. Anyone with a dog is collecting a stool sample every day. It is not prohibitively high on the unpleasantness scale.
Is the whole point of their work the incremental improvement of procuring blood vs. stool?
23 comments
[ 2.9 ms ] story [ 62.4 ms ] threadWhen early detection influences survival rates fairly consistently, it’s basically the same thing as curing. See also: vaccines.
...because that (somewhat overdone) narrative already presupposes higher survival rates for cancers that are detected early.
In most cases survival is defined in terms of fraction of population who are alive after a set period of time from diagnosis.
Suppose Bob is going to die of Foozle cancer in 2025. If I diagnose Bob in 2019 he gets more than five years survival. If my diagnosis method only catches it in 2023 the survival rate looks much worse. But all that changed is when I told Bob he's going to die.
In contrast if I develop a treatment for Foozle cancer, maybe Bob doesn't die until 2030. He's survived more than five years after diagnosis again - but this time it's because I was able to actually help.
I really wish we had better methods to detect that cancer. Maybe before it metastasised he could be saved?
I'm not suggesting it's a good option though.
You got it _very_ right.
Many years ago on youtube I heard a non-main stream cancer doc who said that, the battle against cancer has been stagnant in 40 years ( so 50 years sounds about right). Main stream medicine does lot of massaging of numbers/statistics to give the impression that we are making progress on cancer. To add to the illusion that there is progress, early detection means a person diagnosed with cancer is likely to survive well past the 5 years mark. The 5 years mark is one of the parameters that is used to define the cancer 'cure rate'.
Depending on implementation it could even be scary. A companion app that gives you AmazonFresh ads: A little low on zinc today eh? Why not try some Pumpkin Seeds™
Is anyone working on something like a smart toilet? Sounds like something papa Bill Gates would bank roll
https://en.wikipedia.org/wiki/Faecal_calprotectin
Having it actually built in to your toilet would of course be better.
I’ve seen their pitch for commercializing it and it seems legit (without having done a ton of due diligence).
The guidelines for screening after 50, plus the ACA's provision that makes colonoscopy "free" has led to a large increase in the number of screenings over the last 20 years [1], and I suspect that a lot of GI docs are making out like bandits.
[1]https://progressreport.cancer.gov/detection/colorectal_cance...
Yes, unless Kaiser Permenente is your insurer. Oh, the almighty dollar.
Is the whole point of their work the incremental improvement of procuring blood vs. stool?