While anything is possible, I think it is highly unlikely that Cubase will end up on Linux because it is difficult to make a good business case for it.
If Steinberg were considering porting to Linux there are 2 key questions that need to be answered.
1. How much would adding a market segment for Linux increase development and support costs.
2. How much additional revenue would having Cubase on Linux generate.
To be viable the answer to 2 has to be greater than the answer to 1 plus a reasonable profit margin.
Looking at costs, the actual code development costs likely wouldn't increase too much as that's probably done on an OS independent platform. However the cost to validate the code likely goes up much more because instead of having a list of things that need to be checked/verified that has to be gone through twice it would need to be gone through three times. Also as we know from experience some problems only occur on a Mac or PC but not the other. You'd have to expect that Linux only problems would crop up which would increase support costs somewhat. Additionally there is a hidden cost related to development & validation. These would increase the development workload which would lengthen the the product development cycle. Right now Steinberg gets $150 or so about every 2 years from many (most?) of us for upgrades. If, for example, the development cycle increases to 2.5 years that results in a 25% reduction in their revenue stream.
Even without access to Steinberg's actual numbers, it makes sense that their costs would increase by a non-trivial amount. But of course if they make a ton more money then those costs would be a wise investment.
Looking at increased revenue, it is dependent on how much the customer base expands due to offering Cubase on Linux. Now a bunch of folks on this thread have very enthusiastically stated how much they'd love to run Cubase on Linux. Guess what, you probably don't count on the revenue side of the equation. Since you are on the forum, presumably you already are using Cubase. If you switch from using Cubase on a PC to using it on a Linux box you aren't expanding the customer base. Rather you are shifting from one market segment to another (which is revenue neutral assuming the produce is priced the same on all platforms) . The only customers who'd really increase revenue are those folks who A) would buy Cubase on Linux but not buy it on a PC or Mac or B) buy a license to run on Linux in addition to having a license on a PC or Mac. That's the entire pool of potential extra revenue and I'd bet it is pretty small. Note that this is different than what percentage of the installed OS market is on Linux.
Finally that potential pool of new customers is further reduced because all DAWs need a technical ecosystem in order to thrive. Audio interfaces, controller interfaces, third-party audio plug-ins and VSTi's, dongles/security etc. are necessary tools that we all require. So now you've gotta find that guy who will only use Cubase on Linux and doesn't care that Waves, RME, NI, UAD and all the rest are not available. Sure you can hope to convince Waves et. al. to support Linux but that's unlikely. First all those companies would need to confront the increased-costs vs. additional revenue issues themselves. Second there is a chicken-and-the-egg problem - who is going to develop Linux plugs if there isn't a platform to run them on; and who is going to develop that platform if there aren't any good plugs for use on that platform. Heck, Steinberg can't even convince developers to support note-expression which is much less of an "ask."
While Cubase on Linux might be technically and aesthetically attractive, it's hard to see how it is economically viable.