Can I just clarify whether you consider an RCT to be a minimum...

Can I just clarify whether you consider an RCT to be a minimum requirement for the FDA to approve Rem-L, or whether you think the FDA may be able to come to a compromise with MSB on a trial design that is not an RCT at the Type A meeting?

Given that MSB intends to treat the worst Adult sufferers with a mortality rate of 80% within 100 days, what would you consider to be the appropriate confounders or factors affecting prognosis that MSB will have to control for to justify an RCT? Also, is there any evidence for these factors? I know several people have mentioned confounders, but no one has actually presented any evidence of confounders in the worst-to-treat population.

For a trial to be blinded it requires a placebo and to limit selection bias it needs to be randomised. Block randomization helps to allocate patients based on matching criteria to limit bias whereas stratification divides on the basis of subgroups. So if there are no confounders known to have a large effect on prognosis, then wouldn't you only need to ensure that selection bias is controlled for? So therefore the trial won't need to be double-blinded and placebo-controlled, but in fact it just needs to be randomised with Rem-L compared to Rux as the appropriate control group.