…as clarified the 80% for A is OS, which includes the AE related deaths.
That’s not how I interpret the problem statement in #msg-68027538; rather, I interpret the problem statement to mean that drug “A” cures disease (instantaneously) in 80% of the patients who take it, but 20% (not necessarily the other 20%) of patients who take drug “A” are quickly (but not instantaneously) killed by the side effects. The fact that the problem statement gives the mortality rate as 100% within one week is a trick insofar as the one-week time period is immaterial to the solution.
Unless my interpretation of the problem statement is wrong, I stand by the answer given in #msg-68053216 that drug B is the better choice.