I think this will be a fascinating dilemma for the FDA, particularly if BMY didn't file a second amendment to the SPA to convert their alpha spend to any ipi arm from the ipi+gp100 arm. One wonders whether Fleming will be trotted out again by Pazdur to lecture on the evils of unplanned analyses or whether Pazdur will just let this one slide by.
There is no issue with respect to alpha spend per se because the revised SPA provided more than enough alpha for both OS comparisons. (The OS p-values were 0.0004 and 0.0026 for Ipi+gp100-vs-gp100 and Ipi-vs-gp100, respectively.)
There was no second SPA amendment to change the primary analysis—it remained the comparison of Ipi+gp100 vs gp100. Thus, the technical issue in the BLA is not alpha spend per se, but rather is whether the FDA will entertain a BLA for Ipi monotherapy based on what the revised SPA specified as a secondary analysis (Ipi monotherapy vs gp100).
“The efficient-market hypothesis may be the foremost piece of B.S. ever promulgated in any area of human knowledge!”