NorthWest Biotherapeutics Inc (NWBO)

[exwannabe]

It's a subtle point but changing endpoints itself does not invalidate statistical conclusions

It depends on when. how and why the endpoints were changed. See this paper for a start.

The principle consideration when evaluating whether to modify an endpoint is whether the decision is independent of the data obtained from the trial to date.

NWBO new the IA happened and knew of the results. Many would think that would factor into the decision to not use it.

Box 1. Proposal for Handling Changes in Endpoints in Clinical Trials
Questions to Ask:
What is the source of the new information that triggers consideration of a change in endpoints? Mixed
Have interim data on the endpoint (or related data) been reviewed? Yes
Who is making the decision to change endpoints? Are trial sponsors involved, or is there an independent external advisory committee? Unknown, but no disclosure of an independent decision

Documenting the Endpoint Changes:
Update the protocol in a formal protocol amendment. Not done until 9/23/20
Update the clinical trial registry record.Not done until 5/15/22
Revise the statistical analysis plan. Not done until 9/23/20

Reporting the Trial Results:
Include a clear statement describing the changes in endpoints, and the information obtained after the start of the trial that led to these changes.Include a description of the reasons for these changes and the decision-making procedure.No disclosure that the IA factored into the change even though anybody can see it would. No disclosure of who decided what or when
Consider the potential biases that may have come about as a result of the change in endpoints.
Consider including a disclaimer that the results should be interpreted with caution and may need to be confirmed in future trials. No dsiclosure
Report the reasons for excluding endpoints from the analyses and whether this was independent of trial data Provided reasons, but did not include the IA.

As far as timeline on changes, the external information was known 7 years ago. They waited until they had better blended data and also had the full ECA data