What is being proposed is a small but important tweak to what currently occurs. Presently, these things are typically discussed at an end of phase II meeting. Plus there may be special protocol assessments. Any agreements regarding the trials must be reduced to writing, included in the approval package released by the FDA and may only be changed with regards to the size or design of the trial but not other aspects by the division director and only if justified. One thing that is understood and is included in the law on accelerated approval is that the any differences must be “clinically meaningful”. In other words a statistically significant improvement is not good enough for you can always get a statistically significant improvement if you enroll enough people but that doesn’t mean it means anything. Examples of this are with depression where in the past changes in the HAM-D not only had to be statistically different from placebo but also had to reach a value of 12 and be a 50% decrease from baseline. For mania this would be a difference of 4 on the YMRS, a difference of 2 simply isn’t good enough to make a practical difference. So even though these were kind of understood in the past and were done much of the time, I see no problem in making it a requirement. To avoid the times where there’s a problem, e.g. Aduhelm and Exondys 51.