Deal With This As You Would Any Player Overstep
As it happens, I have had players try to insert real-world but inappropriate cultural influences into a setting of mine. In my case it was Japanese influences into a very European-themed fantasy world.
But this was not so different from the time someone tried to bring Japanese influences into an actual European-historical setting, long before Europeans had made independent, direct contact with Japan. (Same player.)
And honestly, it was not so different from the time someone tried to bring a Dune Mentat into my utterly, insipidly generic medievaloid fantasy ElfDwarfHuman setting. (Same. Player! I have other examples with different players, but they veer farther and farther from your basic problem.)
These examples are superficially different from each other, even though they get farther away from your exact premise-- from wrong culture in a culture-inspired game, to wrong culture in a historical game, to science fiction characters into fantasy games. They are all cases of the player trying to override the GM's setting and genre judgment.
What I Did Then:
(These were all long ago, but formative experiences that I remember fairly well and don't much want to repeat.)
In the first two cases, I just said no. Especially, "No, there are no secret ninja clans wandering around 10th century Europe. There just aren't. No, you did not inherit any of the exotic Japanese weapons you want. If you want to run a game on these themes, cool, I'll play in it. But this ain't that game."
Twenty minutes of pouting is a good description of the response. Then he got over it and played the game. But as you can see, there can be a serious persistence to this mentality that carries on from game to game. I had to be really clear and not give an inch.
In the third case (the Mentat) since the setting was wide open, I just made him stick to the actual rules of an established class, and let him design his little sect of Fantasy Mentats, with some oversight. It wasn't a complete failure, but it was a mistake because there simply was no opportunity for him to do all the cool stuff he wanted to do; the game world did not, and really could not, engage his character. As time went by he reverted more and more to a generic wizard of that setting.
What I Do Now:
What I do now is much more successful, and seems to head these issues off at the pass.
I'll get about halfway through my conceptualization process and run something similar to but not quite, a Session Zero. At this point, I have my general premise and concept, I have a very rough draft of how I'd expect a campaign to go, I have some definite ideas about what the world contains and how it works, but nothing set in stone.
Then I canvass the players and see if they're interested in that, and what types of players they might want to play. In a generic fantasy game, this is when I can more easily move things around to accommodate player background preferences. And in general, if I can, I usually do because I want the players to be playing characters they enjoy.
But sometimes I just can't (i.e., some alien influence from some completely mismatched background or genre) and in those cases, I can at least give the player a lot more time to get over it.
This has worked very well for me. There will always be problem children who require kid gloves or refusals, but getting that done early makes it much easier.
To Summarize Your Options:
- Just Say No: You're allowed to do this. Yes, you run the risk of the player pouting, or quitting, or continuing to try to subvert the game. But you can do it, and sometimes it is the right choice.
- Acquiesce: This is always, inherently, a judgment call on your part. Sometimes it works, but I historically don't have much luck with it because I run games with high levels of GM design and specificity. But if someone wants to bring a Slavic warrior into your Germanic themed campaign, that's maybe not a huge stretch.
- Early Feedback Cycles: Get feedback from your players early, before slight to moderate changes feel like backbreaking effort that scraps half your work. It may be too late for this game for you, but it is the best tool in my toolbox for this very common problem.