Though as a corollary, such an event would be civilisation-ending in any regard.
It's also ... highly rare. Somewhere in the 1 in a million to 100 million year range, which is to say, not only longer than the planning horizon of most human institutions, but well outside the existence of the human species for the most part.
It's also well beyond the available supply of terrestrially-minable uranium,[1] and possibly of ocean-extracted uranium,[2] and thorium[3]. Breeders might be another option, though I've seen no substantive estimates of total fuel availability for that case either.[4]
But as an argument favouring nuclear over other energy options ... this is pretty silly, really.
About the only viable defence against such a risk would be, and I say this as someone who's markedly pessimistic on space colonisation, independent and self-sufficient habitations off-Earth: YACP.[5]
________________________________
Notes:
1. If we relied on naturally-occurring terrestrially-mined uranium for all present human energy consumption ... supplies would last fewer than two decades. This is seldom mentioned by nuclear advocates.
2. This offers a potentially much larger supply, as uranium is present as a solute in seawater, but viable industrial-scale extraction is unproven and would require filtering vast quantities of seawater.
3. Thorium is reasonably abundant, though I'm not quite sure how abundant. Thorium-based reactors are not much used, and the concept of molten-salt reactors (MSR) which gained some popularity in the past decade ... faces some very significant engineering hurdles. Managing high-temperature highly-corrosive radioactive salts is ... challenging.
4. Breeders produce plutonium. And now you have two, or more, problems.
