I need to calculate the required sample size given the following information:
  • Power=80%
  • Coefficient of variation for Cmax and AUC = 40%
  • True ratio of the geometric means=0.95(1.05)
  • Probability that the 90% CI of the ratio falls into the range 0.80 to 1.25 is 80%

In addition, we actually want to test two different treatments against the original, so I guess I'll also need to account for multiplicity.

I am only accustomed to estimating sample sizes for bioequivalence studies that are based on cross-over studies, and am at a loss as which method I need to use here.

For extra help, it would be greatly appreciated if you could advise of the appropriate SAS.