The use of secondary endpoints with nominal p-values
-
If a product would like to include secondary endpoint data from one of their pivotal trials that is not included in the TMA, but the endpoint does meet the requirements for inclusion as per the PAAB guidance document on the use of secondary endpoints (i.e., it is a validated instrument/in another product's TMA/etc.), but it only has a nominal p-value associated with the results, could it be used in APS? If it were to be presented secondary to the primary endpoint, it was clearly disclosed that the p-value was nominal and in a claim-neutral manner?
-
If a product would like to include secondary endpoint data from one of their pivotal trials that is not included in the TMA, but the endpoint does meet the requirements for inclusion as per the PAAB guidance document on the use of secondary endpoints (i.e., it is a validated instrument/in another product's TMA/etc.), but it only has a nominal p-value associated with the results, could it be used in APS? If it were to be presented secondary to the primary endpoint, it was clearly disclosed that the p-value was nominal and in a claim-neutral manner?
Hey @laraholmes
The term ‘nominal’ is often used for a p-value that is not adjusted for multiplicity but there are other scenarios. We would need to assess the study design and nature of the statistical outcome. As a general rule of thumb, a claim neutral presentation, given statistical significance of the primary, and if meeting all other aspects of the Code, could be considered. Clear disclosure of the nature of the p-value would be required.
-
Hey @laraholmes
The term ‘nominal’ is often used for a p-value that is not adjusted for multiplicity but there are other scenarios. We would need to assess the study design and nature of the statistical outcome. As a general rule of thumb, a claim neutral presentation, given statistical significance of the primary, and if meeting all other aspects of the Code, could be considered. Clear disclosure of the nature of the p-value would be required.
@jennifer-carroll Amazing, thank you so much!
