FDA, page-58

Arnies – a few thoughts about this.

I haven’t looked at NEU for a year or two … and so I may be out of date with things.

Yes the effect size is modest. But it is a little bit more complicated than that because of the design of the trial with two co-primaries. If you remember this meant that both primary outcomes had to be positive for the trial to be positive.

So this flows through to the powering. Where co-primaries (A and B) are not correlated (as is most likely the case here) power is calculated from power(A)* power (B). So for example if with 100 subjects power for A was 0.80 and power for B was 0.80 for the co-primaries power is .8*.8 = 0.64. In other words, there is a loss of power from the co-primary design; hence with this design the trial was always going to have a larger sample size.

The larger sample in turn can produce a statistically significant result on one of the primaries from a small effect size that on its own you might think has questionable clinical significance.

But the point is you can’t evaluate the strength of one endpoint ignoring that the other exists. And here the strength is not simply from having two co-primaries per se it arises from having two different reporters – the caregiver and the clinician. So it’s a tough bar … both have to get on the same page; on average – not for each individual child.

So in theory the “clinical significance” argument is taken care of in the design … provided that Acadia and the FDA were in agreement about this.

The potential pinch points I recall thinking were the blinding and to what extent the P3 should be seen as replicating the P2b results. This replication was potentially problematic because the RSBQ was “added” to the P2b trial – and not prospectively registered.

I should study it all a lot more again before saying too much. But a few old posters here will remember that I was not the most glass half full type of guy when it came to those p2a and p2b trial results. But with the P3 results ... I’m much more positive … the best I think I have seen from a ASX company in ages; but of-course this opinion doesn’t come with a money back guarantee.