"Receiver-operator characteristic analyses revealed that the probability of an improver at any level coming from the PBT2 250 mg group was significantly greater, compared to placebo, for Composite z-scores (Area Under the Curve [AUC] =0.76, p=0.0007), Executive Factor z-scores (AUC =0.93, p=1.3 x 10(-9)), and near-significant for the ADAS-cog (AUC =0.72, p=0.056)."
The previous analysis method was not appropriate because the study was small, no possibilities to use power calculations and leastsquare means differentials are too crude a yardstick to measure small differences in short interval trials. This is why the reanalysis used the ROC statistical analysis. Now in the Image study it is possible to do the power calculations and the treatment time is longer but I think that ROC analysis could even in this new study to be better (?).
In the first Ph2 study nobody knew that each of the patient had AD because they had no PET scan done, even AD was very likely. Now in the Image study we know even the speed of accumulating amyloid plaques and everybody has positive PET scan. That is why I would expect the difference already at 3 months to be bigger in the Image study than in the first ph2 study. But as you can see the p values are on very strong level already in the previous study after 3 months on PBT2. Now when the follow-up is 1y, the difference will (IMO) be much bigger.