This would require a test that is (a) cheap (so that it can be applied to basically everyone) and (b) has extremely low false positive ratio (because a particular kind of cancer is a rare thing). Repeat for every different kind of cancer.
For these reasons I'm not sure that it'll end up being a lower hanging fruit than vastly better treatment of late detected cancer for the kinds that are usually detected very late now.
It's not that they are not trying. There are numerous studies looking for bio-markers to detect the presence of cancer, but from what I remember, it does not yield good results.
For these reasons I'm not sure that it'll end up being a lower hanging fruit than vastly better treatment of late detected cancer for the kinds that are usually detected very late now.