Signs of cancer can appear long before diagnosis, study shows

For example, a lot of breast cancer screening has been highly controversial, because of either A) false positive diagnosis B) slowly growing cancers that are too aggressively treated. To properly assess the utility of a cancer screening technique, you need to look at specific long term metrics, such as mortality reduction, that can only be quantified using randomized controlled trials. On the other hand, there's many pitfalls for over treatment. Patients may die during surgery. Biopsy isn't a fun procedure to go through on the receiving end. There's a lot of emotional stress involved. Adverse reactions from chemotherapy/biological agents used to treat the disease. And all this is in the face of the fact that you may die from other causes such as a heart attack.

Certainly I share the optimism with many that we can develop breakthrough cancer screening techniques, but it's irresponsible to let it interfere with rational decision making based on proper studies.

I've also heard some doctors suggest that fasting during cancer treatment improves the effectiveness of chemotherapy.

> “What’s extraordinary is how some of the genetic changes appear to have occurred many years before diagnosis, long before any other signs that a cancer may develop, and perhaps even in apparently normal tissue,” said Clemency Jolly, a co-author of the research based at the Francis Crick Institute in London.

One issue with this might be selection bias towards cancer. What is the baseline rate of finding these type of mutations in people that do not develop cancer? There are cells going bad all the time in the human body, but because of factors such as the immune system, they actually don't cause any problems.

One example of this kind of thing is prostate cancer. Prostate cancer used to be aggressively treated often with a lot of side effects such as not being able to control urine and sexual dysfunction. However, further research showed that a lot of these cancers either don't continue to progress or progress so slowly that the patient will die with the cancer rather than from the cancer, and so not doing anything is the best course of action.

Overall this is an exciting development, but it will take a lot more research to know what should be done with that information, and to avoid over-treating which has its own downsides.

Medicine today seems a lot like running a web app and never knowing if anything is broken unless users complain. It sort of works, but with good monitoring and alerting set up you can catch issues earlier or prevent them altogether. And the stakes with your body are much higher than just a web app going down, the difference between catching something like cancer a few months earlier is literally life or death.

From what I've read in threads like these, almost everyone in the medicine field is very opposed to moving in this direction, even seemingly in principle, and I really can't understand why. I get that doing certain tests can be invasive on its own, so the cost/benefit has to be considered on a case by case basis. This argument also as an analogy to operations - you get false positives on your webservers too, and sometimes people get woken up in the middle of the night for no reason. But we work to fix noisy alerts one by one, and things generally improve over time. Why is this not possible in medicine?

Findings based on tumors sound like 20/20 hindsight!

If your crystal ball tells you where in a body tumor will develop five years from now, then if you look at the cells there, you might see some subtle signs that you know are not false positives.

Somehow, I don't think I'm holding my breath for new screening tests.