We have no interest in the question of whether the apollo astronauts who have already died will be alive when a human next walks on the moon.
The question is of the apollo astronauts who are still alive, what is the probability that they will still be alive when a human next walks on the moon.
The population under consideration is only astronauts who are alive now.
Survivorship bias would only be involved if you were considering some question that involved all apollo astronauts eg for example if we used the population who are alive now to make a prediction as of the completion of the programme.
This isn't survivorship bias, just imperfect information being sufficient. Will anyone be alive at X date can ignore the dead from the population as irrelevant.
The second question if they are an unusually healthy or sick group doesn't need to look at the dead either. Only ~20 percent of 35 year old men live to 88. Having at least 7 men out of 24 reaching that age is already an long lived group, though not necessarily statistically significant.
Looking at the full numbers gives a more precise number, but 12 of 24 vs 7(+) making it to 88 doesn't change the result. Further, even if it was exactly 7 of 24 again the answer doesn't change.
