We know the Medicenna trial control arm will consist of 50 patients randomized within the trial plus 100 patients from Medidata's database of 9M patients. 100 would be enough for NWBO's ECA. Not clear how many matches would be available, but certainly it is not impossible to find 100.
No, I would not expect NWBO to go back and change anything. They spent a couple years deciding what comparator they could use to give them the best shot on goal.
And no, that the SAP was accepted does not in any way means the FDA considers it worthy of getting -:L approved. The vast majority of SAPs are not for approval, yet the FDA accepts them.