Comment by pintxo | Hacker News Reader

Comment by pintxo | Hacker News Reader

One big problem today is, that professionals have voted to reduce screening because of the side-effects of treatment. Especially on those, non-threatening, low risk cancers, which would (likely) never have posted a risk.

If removing the cancer label on those, and thus remove most of the useless treatment, this would be a net positive for all men. As we could start to role out society-wide screening. And really catch the problematic cases. Like men in their 40s or 50s getting prostate cancer.

Today we have voted (in most western countries) to not screen at all because the cost (in human suffering, not money, although that’s surely also a problem) of over-treatment is higher than the cost for those who die early because their cancer wasn’t found early enough. Which basically means we let people die, because we are not able to manage low risk cancer cases correctly.

Source: I’ve once been the tech part of a prostate cancer patient network startup for seven years. Turns out it’s hard to make it cash positive.

Can confirm.

Due to late/lazy screening, my cancer was caught late and now I'm fighting an uphill battle.

It doesn't seem like rocket science to screen early but also not over-treat.

I hope you win your battle.

Hey friend, I'm so sorry you're going through that and I wish you the best of luck in your journey.

I would argue though that it's not rocket science...currently, it's a lot harder with many more unknowns. Some of the issues we face are:

- Imperfect screening tests: the (very real) risk of both false positives and false negatives

- Lack of great prognostic models: if we think a certain cancer has a 25% chance of metastasizing within 5 years and I have 4 patients in front of me, how do I know which is the one?

- How do we identify if we're actually changing the trajectory of cancer with certain treatments? If a treatment fails, is it because we didn't choose the right treatment or is it reflective of underlying aggressive disease biology?

- Mass amounts of data from new technologies: this is obviously great and has allowed us to become granular with cancer in a way we haven't been able to before. It's important to recognize that widespread next generation sequencing is something we've had for less than a decade, and we have identified a lot of risk factors with this BUT it's also raised many questions (e.g. what mutations are actually significant?).

- Being human: every individual has their own values. Some patients will hear that it's a 1% chance this will become cancer and be okay with monitoring it. Some patients hear 1% and think it's too much of a risk and want intervention.

I think, with time and more data, we will become a lot better at this. But it's really not easy.

Can a doctor reliably tell when a prostate cancer is likely to become dangerous?

There are good indicators yes.

E.g. PSA (a blood value, easy to measure), if increasing very slowly over a long timeframe, is a good indicator that you may have cancer but that it is growing really slow, aka unlikely to be aggressive and to be an immediate problem. While a quick increase tells you the opposite.

Then there is DNA sequencing of cancer samples from a Biopsy. Which is evolving rapidly and should be soon cheap enough to be available at scale.

AFAIK there is currently know way to reliably tell from DNA sequencing data whether a prostate cancer will become dangerous. That may change, but I suspect there may be some inherent stochasticity in the transition of a cancer from benign to dangerous that will make a perfect classification possible. I would guess some form of monitoring may be required for all the currently non-dangerous cancers.

Yes, I had a glimpse at current research, it’s being worked on, but it’s not yet there. Also sequencing is still too expensive for more general application.

Genotyping assays for single nucleotide polymorphisms (SNPs) are relatively cheaper (I may not be up to date). In the space of 'polygenic risk score' of these genomice variants it is still a bit of debate how useful it is for screening.

The article addresses this. There's some risk of misclassification, but the answer is the same either way: you monitor the condition carefully to see how it behaves.

The problem then is, if you rename the "grade 1" condition "prostate dysplasia", some fraction of patients don't do the follow-ups, and then those misclassifications are a big deal. But if you don't rename it, some fraction of patients get unnecessary treatment that lowers their quality of life.

So it's an interesting dilemma.

3Blue1Brown explains it well. https://www.youtube.com/watch?v=lG4VkPoG3ko

If the population who do _not_ actually have serious cancer is orders of magnitudes larger than those who do then there will be far more people in the false positive group than the true positive group and thus the vast majority of operations will be unnecessary.

Good one.

The key here is how we manage the positive ones. And this is where the current system fails. We have yet to find a good way to manage them without doing unnecessary treatments causing more harm than good.

But instead of fixing this problem, we opted to stop screening. Which, for me, sounds stupid. But it sure is easier to do.

With all due respect to 3Blue1Brown, there are limits to wjat you can explain with statistics. Medicine applied on the individual pevel is one of those cases, because its purpose is to optimize the outcome for the individual.

A pandemic is different, there the goal is to optimize the outcome for the population.

Applying the latter to the former is at the very least cynical. The opposite is what got us anti-mask and anti-vaxxers.

Also, especially with cancer, people tend to be so affraid of the diagnosis, they look for excuses to not be examined. As if cancer onpy pops up when it is found or something. Which is stupid, if ypu have cancer you want to know as early as possible.

> Applying the latter to the former is at the very least cynical. The opposite is what got us anti-mask and anti-vaxxers.

I have no idea where you're going here. Wearing masks and getting vaccinated benefit both the individual and the population. (That actually contributed to it being politicized -- your choices affect my chances of getting a nasty disease).

At the individual level, you wear a mask when appropriate and get vaccinated, then you're less likely to get the disease and it's likely to be less severe if you do. Likewise, the population benefits from less disease spread, and less strain on the healthcare system as many of these cases become outpatient "take some Paxlovid" type deals.

Where is there any sort of basis for anti-mask and anti-vax theology here?

Exactly, case management is the key problem here. Re-naming is just a band-aid.

The testing itself is an issue as well. Most people think that you put a blood sample in a machine and it spits out "yes you have cancer" or "no you don't", but in reality in the earliest stages it's more akin to reading tea leaves. The chances of false positives are high, and can be equally or more harmful to the patient than false negatives.

This is also why doctors are increasingly recommending that young, healthy people not get routine annual checkups and blood work done unless they have any specific problems.

Docs don’t treat Gleason 3+3 so no risk there.

There's no hard facts separating a 3+3 from a 3+4, just one (or two if you're lucky) overworked pathologists and their subjective opinion on what they see on their slides.

Why isn’t it screened like breast cancer?

For the very same reason, breast cancer screening is under discussion.

We treat false-positives and low-risk patients that actually don't need treatment. As any treatment, at scale, invariably results in (averse) side-effects, that's a problem.

And the easiest way to get rid of false-positives is to not screen in the first place.

For breast cancer the statistics are the same as for prostate cancer, there is no society net-benefit to screening, as false-positive side effects are equal or worse to the benefits of the screening.

But, women are way better in handling (their) health than men, the public outcry over removing breast cancer screening would be load.

In other words, men are not load enough when it comes to (their) health. Which is no surprise I guess.

'loud'?

Thx, English is not my first language.

Also I feel like it's actually getting worse, sometimes I am wondering if this is a covid side-effect thing.

Not ony do treatments have risks, but so do tests! It doesn't take all that lopsided a ratio where if the odds of cancer are low, you have greater lifetime cancer risk from the additional xrays.

For Prostate Cancer screening, no x-rays are involved.

Reply on news.ycombinator.com