Presentient has received comments from FDA providing clarity on how sponsors can use BRAKES

Presentient has received comments from the US Food and Drug Administration (FDA) in regard to BRAKES, the first real-time interim analysis solution. These comments from the Center for Drug Evaluation and Research (CDER) at the FDA were in response to Presentient’s recent submission of a Letter of Intent to the FDA ISTAND (Innovative Science and Technology Approaches for New Drugs) program which included evidence detailing BRAKES’ context of use, architecture and performance characterisation.

The FDA comments clarify how the FDA anticipate BRAKES being deployed; sponsors seeking to use BRAKES in the context of a specific clinical trial are to engage with the FDA via individual IND submissions. Presentient previously also had a positive meeting discussing the BRAKES platform with the European Medicines Agency’s (EMA) Innovation Task Force.

The FDA further noted that it may be possible for a Data Monitoring Committees to use BRAKES to start reviewing trial data if BRAKES indicates the trial has met its prespecified boundary probability and where BRAKES’ completion pack may not necessarily require an additional set of summary results prepared by an independent statistician. We welcome this clarification of how BRAKES could trigger a Data Monitoring Committee meeting in a streamlined way.

BRAKES’ deterministic & reproducible approach offers potential advantages over conventional methods across several key areas. Internal testing of BRAKES on an empirical dataset (of 10,000 simulated trials that emulate outcome distributions from ClinicalTrials.gov) show BRAKES had a false negative rate (Type II error) of 1.0% vs 4.8% for standard Conditional Power. This suggests that the 50+ successful trials that are being stopped erroneously each year by interim analyses using conditional power could be reduced five-fold by using BRAKES instead. On this same dataset, BRAKES detected 41% of futile trials within the first third of trial information (vs conditional power that was run at midpoint).

BRAKES detects the inflection point in unblinded trial data where the forecast of success becomes highly likely (1% false success detection error rate). BRAKES does this without spending alpha because BRAKES isn’t a stopping rule for efficacy nor should it be used to initiate early stopping for efficacy. Instead, BRAKES triggers key administrative steps of the next stage like manufacturing which could cut many, (potentially over 12) months of whitespace time between trials.

Presentient is committed to bringing realtime clinical trials into existence by modernising clinical trial decision infrastructure and eliminating operational bias.

If you would like more information, please contact us on via our webform or email us at info@presentient.com.