Post-unblinding changes to a statistical analysis plan doom even highly promising therapeutic signals.
Even when peer-reviewed publications validate clinical efficacy, regulatory agencies will reject a drug candidate if the sponsor altered the statistical analysis plan after unblinding.
The same conclusion keeps arriving from across the workspace's research — 1 topics independently instantiate this theme. Filter the evidence by where it came from:
Capricor's regulatory campaign faltered as the advisory committee rejected efficacy data tied to post-hoc, post-unblinding statistical modifications.
Changing the statistical analysis parameters post-unblinding to salvage an efficacy signal led the FDA to reject the trial as failed under its original plan.
Finding a statistical model error and having to revert to the pre-specified SAP weakened Capricor's cardiac efficacy signal, forcing them to pivot to an alternative skeletal muscle indication.
Capricor's BLA faces rejection because the developer modified the trial's statistical analysis plan prior to unblinding, creating a major dispute with the FDA over the data's validity.
An advisory panel rejected the candidate after the FDA flagged suspicious, late-stage statistical plan changes right before unblinding.
The advisory panel voted down the candidate because of disputes regarding post-unblinding statistical modifications.
The advisory committee voted against the drug's approval because of structural disputes and post-hoc changes to its pre-specified statistical analysis plan.
The FDA rejected Capricor's efficacy claims because the sponsor modified its statistical plan after the trial was already unblinded.