It depends on a lot of things. First is how easy it is to evolve a way around a particular defense. For example, it will be easier to achieve an vaccine escape

It depends on a lot of things. First is how easy it is to evolve a way around a particular defense. For example, it will be easier to achieve an vaccine escape when only one gene is required to mutate compared to ten thousand. If you have a vaccine that disrupts something fundamental about a pathogen's lifecycle, it's better than one that targets something superficial. Another question is what fraction of the population is exposed to the selective pressure, and to what extent the selective pressure confers a benefit or cost to all other members of the population. If responding to the pressure makes the population more fit in other hosts, then we should expect it to become widespread eventually. On the other hand, if it causes the disease to become less efficient in other hosts, you would not expect the mutation to become widespread. For example, imagine if there were a monster that lived in a small part of Alaska that tended to eat humans with small fingers, we would probably just not go to that part of Alaska. On the other hand, if such monsters were present throughout the world, you'd probably see humans with longer fingers over the generations. Does that make any sense? It's a strained metaphor but I hope it gets the point across. Another thing that can happen is speciation, but we'll leave that for another day.