Chart, page-3854

It did and a larger trial is needed. In saying that though, p-value calculations account for sample-size (with good reason).

i.e. It's harder to achieve a nice low p-values (as we did) on small samples.

So, in spite of the small sample, I'm personally left confident by the PII data - it's highly unlikely that we observed this data by chance.

That said, I'm self-aware enough to know that I'm the dunce (computer scientist) from a family of statisticians; there's no substitute for actual expertise. I'm left far more confident by Prof Voit's proclamations than my own understanding here

Good luck!