Very few published is damn right. I think megadungeons are not very well suited to publication.
Well, just what are you going to publish?
The whole thing is always in flux, maps and keys, so all you can ever get is a snapshot.
Notes to oneself are a long way from making great reading for anyone else (even if they are not simply cryptic).
Oh, but that's just the dungeons used in actual campaigns, eh? Why not pay someone to design one specifically as a text for publication?
At most, that would be just the
start of the work of making a good dungeon. The
mise en scene before play is in a sort of sterile vacuum, as opposed to the fertile soil that produces organic growth once players start to interact with it. Personally, I think that even half a dozen levels may be too much preliminary construction unless they are going into the context of a campaign already well underway (so that none will be long awaiting action).
It would be sort of like offering a still photograph
as a movie. It might be nice, but it's not the same thing. An old-style dungeon is a
process, and -- unlike the dynamics of a movie -- a point of it is spontaneity.
Another point is
mystery. D&D very basically started as a game of limited information, and the dungeon game is that in spades.
What becomes of that when DMs are running the same danged dungeon? Not a one-shot "module", but a full campaign dungeon?
The ratio of work to value is pretty low here, I think. How many people are really going to buy Standard Issue Dungeon, and how much will they pay, versus how much you had to invest to produce it?
Given the basic selection pressure of the game itself in appealing to creative types, probably
not enough.
From a business standpoint, "enough" basically would be at least as profitable as making the same investment in small scenarios or in rules supplements. Has that situation ever emerged?