Hacker News new | past | comments | ask | show | jobs | submit login

> You see so many stories about new cancer treatments or prevention strategies all the time, and it’s so easy to discount it as just more spam, but we’re getting extremely good at making people better, even in comparison to as recently as a decade ago.

IF we catch the cancer early, we are much better at treatments.

What we are getting good at is treatment with early detection because we can just cut it out with surgery. What we are only marginally better at is treatment with late detection.

For cancers that we don't have early detection, the success rates are far, far lower and the treatments are far more limited.

For example, a significant fraction of women get diagnose with lobular breast cancer--it doesn't form lumps. So by the time you detect it, you are almost always already Stage 3 or worse.

Radiation, in particular, is much better thanks to computing. The only advance in surgery I have seen since the 1970s is that they use a radioctive marker to trace where the lymph nodes are--it helps with lympedema but not the cancer. Chemotherapy is only marginally better than 20 years ago--they're a little better at managing side effects but chemo is still horrible.

Medicine does well on things that can be fixed by physical means--cutting things out or sewing things together. Medicine still doesn't do all that well from a chemical standpoint.




This really hits the nail on the head, so many people miss this.

Didn't realize myself until I was diagnosed with a sarcoma (1% of adult cancer diagnoses), MPNST (a single digit percentage of sarcomas).

Doctors scrambled to figure out what to do - first we should just cut it out, then, well maybe we should do chemo + radiation. A day before getting a port put in receive a phone call that "oops, chemo doesn't work, just cut it out and radiation afterwards."

Luckily I was in a good position to get multiple opinions and read the data myself.

Almost 2 years cancer free now and am so grateful, but looking at the data, it's all seems like a toss up unless you can just cut it out with clean margins.

I really can't help but think modern medicine is really missing the mark with the reactionary approach to disease in general, but of course I have no clue what to do about it.


> I really can't help but think modern medicine is really missing the mark with the reactionary approach to disease in general, but of course I have no clue what to do about it.

There are two hard problems here:

1) Medicine doesn't always have anything good to actually do

As you point out, there was nothing other than cut it out and do radiation for your case. Those are very physical processes. Chemical processes that either do something directly to problematic cells or cause the body to do something to problematic cells are very lacking. We don't have very many knobs that do something to diseased cells that don't also do horrible things to the patient.

2) The source of being "reactionary" is that statistically rare diseases show up as false positives way more often than real positives without some Bayesian prior.

Medical testing is a balance between causing too much harm in testing and not enough detection.

Lobular breast cancer only shows up on MRI. So, women should get an MRI with their mammogram, right? Well, except that the MRI requires a gadolinium contrast which bioaccumulates. And you needed an IV for the contrast which can get infected. And, you will find a lump of something with a modern MRI. Now what? You really should leave it alone and watch it, but some women will be unable to deal with the anxiety so you will need to biopsy it. Biopsies have their own problems. etc.

Testing for infrequent events is hard.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: