Lorentz Transformations, I believe, still maintain a certain sense of realism, and I don't believe the "truth of the matter" of a question like "is this star still alive?" in relativity relies on the information that star is dead reaching the people who you're asking this question relative to.
So let me just lay out what I'm saying more explicitly: let's say some civilization is 80million years in the future and they know, quite clearly - because they have observational data - that the star died. They know when it died. The know it died in the year 2000, but they also know that because it's 50m lightyears away, people in 2024 wouldn't have known that it died 24 years ago relative to them. So you can frame the question like this: is it reasonable for these people, 80m years in the future, to say "In the year 2024, that star was objecctively dead, but the people on Earth didn't know it?"
I think the answer is, yes. If you calculate the truth-of-the-universe as it was in 2024, knowing what they know in 80m years in the future, and calculate it from the reference frame of a person on earth in 2024 - doig whatever lorenz transform is necessary - that truth-of-the-universe will include that star being dead in that reference frame, but the information of its death not having reached earth yet.
I know naive realism can be a bit boring for philosophers, but I think it really is as simple as that. Even in relativity.