>i believe the point of [gofish’s] question was the onerous size of a study required to show (statistically) a 5% difference in efficacy (compared with a much smaller sample size needed to reach statistical significance if the relative svr rates were larger)<
If the mathematical model NVS and IDIX are using says that NM283+SoC will boost the SVR by only 5 percentage points vs Soc alone, they should probably drop the program and not even bother with phase-3.
Market Data 
Markets 
AI
