That was my thought: Legally they would have to disclose, material data can't be 'insider info'.
But--let's say the DSMB met and told them, the data is blinded, but we see a pattern of change in six cases, nothing in the other four. We just don't know who is who. That's a hint of a possible signal, but since it's still blinded, they can't be sure. That seems like it would not have to be disclosed, but it could be informally, part of a partnering discussion. Or if the DSMB says--we see just two cases with some change, even though it's blinded, you can be pretty sure not a lot is happening.
In either case, if the prospective partner says: OK, we'd like to see the unblinded data, and we'll agree to XX terms if it is consistent with what this looks like--but we do not want to get into a bidding war with someone who steps in at the last minute based on a pres release.
If I were Cortex, and I was OK with the XX terms, I think I'd find a way to not disclose the SA results in an entirely timely fashion.
But this is total speculation--in all respects.
NeuroInvestment