3 (or more) treatments
Would a third treatment add significant testing burden (and cost) to the trial? I assume it would be done only for those that went from CR to NR, but it could be more broad. It seems logical that this could improve outcomes, but safety profile and protocols would need to be updated/written.