Wall:
from iv:
>Now comes one of the controversial parts of the BLA package, namely the Cox regression analysis for 9902A, which lowered the p value from 0.33 to 0.023. We do know that the same five final prognostic factors were used in 9902A's Cox regression. If you're a committed DNDN long, then you have to believe Dr. Higano and DNDN mgmt when they say that the Cox regression exercise in 9902A was done both backwards (starting with these same exact five covariates) and forwards (starting with the 20 prognostic factors in the nomogram and eliminating the non-stat sig factors). If you're a committed DNDN long, then you have to trust Dr. Higano and DNDN mgmt when they say that the Cox regression p value in 9902A was stat sig regardless of which way the exercise was run. <
We know this is not true. The backwards and forwards / any which way reference by Gold was referring to the fact that if they took the 5 distilled covariates from 01 and applied them to the 01 dataset, then the p value was stat sig; or if they took the 5 distilled covariates from 01 and applied them to the combined 01+02A dataset, then it was stat sig again.
Although DNDN may have started with the 20 prognostic factors and used the data from 02A to distill them, they did not present that data to the public (that I'm aware of). They only presented this outcome for 01, not 02A.
fwiw...
AI
