Genetic variants are different from drugs. For one things, there are billions of genetic variants.
Currently the FDA is trying to apply drug validation studies to genetic variants. That's like requiring cars to have proper horse shoes.
So, umm, this is not the case. Do you have any idea how the non-ag side of the FDA actually operates?
When you go for approval you define the criteria. "All" you have to show is that you will show statically valid evidence that your claims are plausible. Sometimes the Agency wants some specific proof (e.g. they may ask that you use a specific animal to verify a specific procedure or organ consequence) or may want an additional test because they know some classes of can compounds have characteristic issues. But in general they don't dictate anything. Oh yeah, for a treatment there are outcome criteria, but you are mostly the origin of those as well.
Diagnostics are simpler to approve than devices, which tend to be simpler to approve than drugs, which to some degree are simpler than biologics. Each division has its own specialists. And the so-called "billions" of genetic cases are a red herring -- just consider the complex consequences of introducing organic compounds into the body.
Disclaimer: I am not a geneticist, and have not gone for any diagnostic approval, though I am an advisor to a couple of companies that have. But apart from my own experience in the small molecule domain, I do know most of 23&me's regulatory team socially, and we do talk about work. So I know generally what issues they face.