This is precisely the problem with presenting generic DCs, or scaling DCs. Rolling with a 65% chance of success (or any other number) isn't a meaningful challenge; whatever prompted the PC to make the decision to accept the risk provided by the roll is the challenge. Going from "35 is the hard DC at the PC's level" to looking at the table and spotting that DC 35 locks are usually made of mithral to putting putting a mithral lock in the dungeon is not actually showing progress, it's just shuffling adjectives around so you can maintain the same gameplay loop of rolling to open the locked door.
A high level challenge must take in to account things like "the PCs cannot be stopped by locked doors" and "the athletic one can spiderman climb on ceilings" and so forth. There's a general understanding about PC capability as it relates to spells, we understand PCs can deploy flight and teleportation and scrying and all that, but we tend to try and make the skill game exactly the same; instead PC capability actually needs to expand, thus that they have different tools to resolve problems. It's insufficient to simple change the adjectives describing the challenge, it's underlying structure needs to change.
This is, of course, a lot of hard work. I've lately come to believe that the actual high level differentiator is about the PC's agency. The move to higher level has them moving more and more from a reactive to proactive position. At lower levels you enter the dungeon and deal with what it throws at you, and then slowly you get the ability to ignore more and more problems outright or mitigate them via preparation/bringing the right tool. Eventually you can't be forced onto a reactive footing at all, and will start setting the terms on which you'll engage with your foes and problems. I think that shift in adventure/campaign structure generally needs more attention. We've talked about domain management, which is one manifestation and part of it, but that's not the only way it can go, or even the most common one in modern D&D; more likely you need to shift from presenting problems the players need to react to, to letting them decide what they want to do. At the most basic level, the PCs will stop encountering meaningfully threats in their immediate areas and will actively have to go to places that are dangerous to be challenged, which means giving them new motivators to be proactive.