My previous blog summarized KOL views (in 2016) of the state of drug safety quantification in our industry. As part of the KOL interviews, there was an FDA focus group on the IND Safety Reporting final rule. The next few blogs will be on this focus group.
The FDA clearly states a preference for unblinded safety assessments of ongoing clinical trials. Review of blinded data can be effective and is not discouraged, however, causality determinations cannot be made on blinded data. In addition, the importance of protecting trial integrity is emphasized.
One approach mentioned: If there are serious anticipated events that could possibly result in reporting requirements and/or aggregate comparisons at a trial level indicate imbalance, then these need to be referred to a Safety Assessment Committee (SAC) for review [note, the 2021 draft guidance uses the less austere term of ‘entity’]. If upon review these are identified as safety signals, the review should be expanded to a program level. Once these issues seem clinically important and related they should be reported.
The FDA is not prescriptive about what an appropriate way to review would be or what would need to be reviewed in aggregate. However, there was emphasis on pre-specification and upfront planning to ensure a firewall, appropriate unblinding, and protection against false positives and negatives.
The 2015 draft guidance outlines principles about organizational structure, the SAC, and building in a Safety Surveillance Plan (SSP), which may help in the development of some processes. The importance of reviews being performed by an unblinded, firewalled, expert team who is external to study conduct is stressed. Note, this could be a CRO for smaller companies. The FDA is very concerned with the trial integrity issue. Looking at unblinded data has to be appropriately firewalled. The typical approach is to make aggregate comparisons. An alternative approach (the ‘trigger approach’) is to only perform an unblinded comparison of event rates across treatment groups if the overall rate for all treatment groups of a specific event is substantially higher than a predicted rate.
The FDA experts/representatives stressed that these new types of reviews and approaches may require organizational structure changes. SACs [or other assessment entities] are recommended to perform the reviews. Smaller organizations may be impacted more heavily here, relying on possible outsourcing of the work due to resource and organizational constraints. However, larger organizations are encouraged to develop internal SAC(s) that are firewalled from trial activities. Paramount in either approach is the preservation of trial integrity simultaneous with protecting patient safety.