No study is perfect. As an example there will be protocol violations in every study ranging from minor (patient missed a study visit window) to major (study drug mixed up between 2 patients). The regulatory agencies take this information into account. As I've said, this is probably a bigger issue for the ADCS-ADL than the ADAS-Cog unless the assessor was aware of side effects. I would be surprised if the protocol allowed the cognitive assessor to have this type of information, however, I would not be shocked if a self promoting PI lets info leak to the assessor.
Regulatory agencies do look at blinding and their guidance discusses these issues. They may never be aware of MacFarlane's blind-breaking statements. Lemtrada, an MS medication, was delayed a year by the FDA partially due to unavoidable blinding issues (bigger issue than Macfarlane's statements) while the EMA approved it. The data safety committees would certainly never release unblinded data during a study.
Market Data 
Markets 
AI
