> Thus any approvals would require having a particular condition to get a prescription (or an unethical doctor), and this aren't immediately useful.
That's not how any of this works (US specific to follow). FDA approval to market is a constraint on how the vendor markets a drug or device, not how a physician uses it. This is known as indications for use.
So if I sell a drug approved for blood pressure, I can never say it may improve longevity. If there are papers out there claiming good effects on longevity I as a vendor have to be very careful to never make a public claim about that.
However, if you as a physician have read the papers you can prescribe the drug for other uses based on your personal expertise. This is known as off-label (i.e. for a use not covered by the indications for use), and is very common. There are limits of course, but at the end of the day the limits are based on professional conduct/ malpractice - not the FDA. The FDA explicitly does not regulate the practice of medicine.
Note that this doesn't just apply to somewhat speculative uses. For example a significant amount of pediatric care is done off-label, because setting up the clinical trials for pediatric indications is a pain. This may only be a dose change, for example, but is very common.