So if we need a 10% survival improvement and the response rate is 50%, then of those who respond, they need a 20% improvement. Is that correct?
How does the math change btwn considering the test group as one group with 10% improvement vs 2 equal groups with 0% and 20% improvements? (Magnitude and direction of change)
I assume if modeling survival for the two groups would be y=(.5)(a)ln(x)+b +(.5)(c)ln(x)+d, where a and b are the same as the SOC group.
Actually it was a 100% response rate, the other half had no growth which assumes to provide better surgery results. They can't see and note a reaction if the tumor doesn't shrink, but the effect of zero growth produces support for surgery working better, therefore also adding to the survival period and bettering the SOC