NorthWest Biotherapeutics Inc (NWBO)

[radonculous]

First, the trials is an add-on to SOC. If it is "as effective" as SOC, then it is not effective at all. So could never be approved.

That's what I said

Second, the alpha allocation issue is fuzzy. Is it .02 for PFS with OS being only sequential at a subsequent .05 (which is how the protocol read to me). Have they changed it, or was this never the case? Nobody knows.

I have no idea what they have set alpha to in this study for PFS


And lastly, you also fall into the alpha trap. The P value does not state what you just asserted. It states that there would be a 97.5% chance that a placebo would not reproduce the same benefit (or larger) in the same trial. [Maybe your were just keeping it simple here. But this is one of my two pet peeves. Other being when people assert that a trial that failed to show a benefit showed no benefit]

What is an alpha trap? I don't think you know what a p-value is based off of your comment. I have no idea what you are talking about here with saying that a P-value states that there would be a "97.5% chance that a placebo would not reproduce the same benefit (or larger) in the same trial". What p-value? Where did you get that number? Are you referring to a specific trial or specific data? It's really not clear. I'm speaking about p-values and alpha (significance levels) in general.

Also, this sentence makes no sense: "Other being when people assert that a trial that failed to show a benefit showed no benefit".

I'm sorry if I hit on your pet peeves, but I think we are talking about completely difference things.

Let me explain what I was trying to say again:
When alpha is set to 0.05 (or 5%) this means there is a 5% probability that the null hypothesis is true (meaning that there is no association between two measured phenomena). When p is less than alpha, you can reject the null hypothesis and the alternate hypothesis is accepted. If alpha is switched to 0.1 this means there is a 10% probability that the null hypothesis is true.

Ronald Fisher initially set significance level at 0.05 and it stuck, but he never intended it to. There is a growing call for the end of statistical significance among statisticians. This was published in Nature a couple months ago: https://www.nature.com/articles/d41586-019-00857-9