If the p-value at n=170 isn't good enough to get an interim halt for efficacy, it very likely won't be at n=195 either. We don't want to change the total number of patients required (390), only the interim analysis point.

The difference is only striking above 30% placebo mortality, but the difference is not so big that it would yield a result above 0.0294 which is one boundary used for interim analyses to allow for a halt, if I see correctly.

That is if my calculations are correct; and of course depends on the test they use..