I don’t understand what you mean by a 20-40% premium in this context—please clarify.
The point of this approach (getting approval first in an ultra orphan setting - then only later getting the more general approval) is that by first getting the ultra-orphan you later get a higher price than you otherwise would have if you had just gotten your initial approval in the broader population. My WAG was an extra 20-40%.
1) I don't know what supporting studies have been for this exact implementation of Anchoring. But Anchoring itself is ironclad - however counter-intuitive it might be that it should work.
2) Similar kinds 'irrational' pricing traps exist all over the place - so it is certainly plausible in that sense.