Because the data show that they don't improve outcomes and do cause harm in some circumstances. My opinion (not based on data) is that the technology isn't sufficiently well developed to be useful.
Outcomes are improved when these tests are performed.
If they were scanned yearly and underwent a series of unnecessary medical interventions as a result, that wouldn't be considered to be "better".
Which is why I suggested that if you feel that it's helpful, then by all means. It my be. Outside of over-diagnosis, it is not harmful, and that's certainly something a prudent individual can mitigate. Unlike, for example certain kinds of CT scans which expose folks to levels of radiation that could cause direct harm.
I don't fall into the category of folks that lack the means to get one. As for insurance companies, they tend to get reamed for the cost of an MRI and so understandably would want to limit expenditures where not completely necessary. A full body MRI is generally a few thousand dollars. I got an MRI for a spinal issue a few years back for which the hospital billed my insurance roughly $10k. Friend of mine paid $200 out of pocket for a similar MRI in Thailand.
In support of your point, it is an interesting prevention tool, but so far, doesn't seem all that useful, at least to the extent that I would spend money on it. I'm hopeful that if its more broadly used, that it will eventually demonstrate an improvemenet in outcomes.