Well, for s/c trial it's just an interim analysis at n=51... not getting stat sig at that point/low sample size isn't bad, though it would be great if we could achieve it... for the trial to be stopped early at n=51 for overwhelming efficacy, we'd probably need a p-value below <0.01 (?) anyway, and for that we'd need a death rate in the placebo group that is higher than 1/3 and no death in the leronlimab arm. Most likely result is that the review board will recommend continuing the trial. Perhaps we can already achieve significance on certain secondary endpoints like the clinical improvement scale.

The m/m trial is the best shot at success; I am most interested in what the clinical improvement at Day 3, 7 (secondary endpoints) will look like, since at Day 14 a high recovery rate in the placebo arm might be a problem for achieving significance with n=75 too. For the primary endpoint (the 12 point scale), one could perhaps use the "mean function"of this tool (though it gives you the sample size required, not exactly what we need, but assuming a certain improvement you can at least see what trial size would be required and if it matches ours that would be a good sign): https://clincalc.com/stats/samplesize.aspx