I thought the trial was powered to show a survival difference of six months, assuming mOS of 20 months in the survival arm. If the control arm does better than anticipated, would the sample size still be sufficient to show a statistically significant difference of 6 months?
Market Data 
Markets 
AI
