RE:RE:RE:RE:RE:RE:RE:RE:Rebranding the Sort1It's definitely trickier with small sample sizes. But I also figure they've all thought it out and are comfortable. I have seen many post-hoc analysis of P1 and P2 trials where they very clearly create 2, sometimes 3, cohorts based on various parameters to provide a clearer explanation.
Sometimes it's baseline parameters they segregate and sometimes its things like target expressors or high v. low expressors. So you can always get that clean read later. Unfortunately, it's a bit more meaningful if you have a larger group going through the trial than this one. I do recall, however, that type of expressor analysis from one of the other targeted drugs in trial - not sure which company though.
In the end, they just need to see much lower neuropathy with the first 6 and some signs of stability or better, then the next six need to see same low toxicity and more confirmed response along with SD.