I know the Russian space pencil story is a myth, but I think some of the logic applies here.

If you're sending a camera millions of miles away for a multi-year mission, or into orbit where repairs are infrequent, it's worth designing something custom, with appropriate shielding and redundancy. For example, from a description of the Curiosity rover's MAHLI camera:

The camera head electronics are laid out as a single rigid-flex printed circuit board (PCB) with three rigid sections. The sections are sandwiched between housings that provide mechanical support and radiation shielding; the interconnecting flexible cables are enclosed in metal covers.

I can't find offhand how much this camera alone cost, but I wouldn't be surprised if it was a decent fraction of the $700 million that went into the rover (out of $2.5 billion for the mission overall).

If, on the other hand, you're sending something along with humans that need to be regularly rotated out, that would be a waste of money. The effort and expense would be huge, and because projects like that take years, the results would probably be disappointing compared to current-generation consumer products. That MAHLI camera — the highest resolution camera on the Curiosity rover — produces 1600×1200 (e.g., 2 megapixel) images.

And that's just thinking about the image quality. There's a lot more than that which goes into making a camera — from design ergonomics to softare. Even at $20k per pound to get a new DSLR body to the space station, better to just keep sending up the (top end) mass-market cameras. (And hey, a new camera every 12-18 months sounds like the same schedule a lot of enthusiast photographers follow in any case!) It's like simply using a perfectly functional pencil rather than spending millions on a custom solution.