One of the most critical functions of the US Food and Drug Administration (FDA) involves the evaluation of safety data in its review of marketing applications for drugs and biologics. However, the processes for reporting safety data during clinical trials have not been optimized for premarket safety analysis in the current era of targeted therapies and biologics. Therefore, as published in Drug Safety, the Office of New Drugs (OND) re-examined its earlier classifications of adverse events (AEs) and developed 104 new custom groupings of AE terms, known as OND Custom Medical Queries (OCMQs), to streamline and standardize clinically meaningful groupings of AEs (eg, safety signals) for use by OND staff in premarket drug safety evaluation.
More than 80 FDA clinical reviewers from the Center for Drug Evaluation and Research and the Center for Biologics Evaluation and Research were included in the development of the OCMQs. OCMQ groupings are categorized as either Narrow or Broad. OCMQ Narrow groupings aim to ensure a high level of confidence in the association between the AE and the OCMQ Narrow grouping, while Broad category terms are designed to be more inclusive, providing greater sensitivity at the cost of specificity, and aim to enhance detection sensitivity of the OCMQ Narrow grouping. The Broad category terms may serve as the basis for requesting additional information or pursuing additional analyses but are not anticipated to lead directly to regulatory action.
OCMQs attempt to capture all instances of a medical concept, regardless of the etiology or potential to be drug related, which the authors propose may be appropriate for multiple reasons. First, etiology specified by an AE may represent the mechanism by which a medication causes an adverse reaction. Second, study participants experiencing a particular AE may be more susceptible to a medication that causes the adverse reaction. Third, clinical trial investigators may misattribute the cause of an AE. Consideration of all related AEs regardless of assessed causality is important to support a comprehensive evaluation of data for safety signals. Algorithmic OCMQs were also discussed in the publication. Algorithmic OCMQs use a set of rules to leverage additional information, such as combinations of AEs, laboratory data, concomitant medications, medical history, or timing information, to provide a more comprehensive view of the AEs occurring during a trial.
High level
Investigators should be aware that the AE reporting practices for their trials will likely change in the near future and that there might be some retroactive tasks required to recategorize AEs in ongoing trials. Proactive planning for supportive staff assistance might be prudent. The OCMQs were validated across a total of 459 unique clinical trials, demonstrating that they are broadly successful at detecting safety signals in clinical trial data for drugs and biologics. Additional validation work may include investigation of the use of OCMQs by review teams during the conduct of clinical reviews. Be aware that there will be a change in how AEs are presented in regulatory literature, and this will likely be different from how doctors are accustomed to seeing them (eg, with Medical Dictionary for Regulatory Activities preferred terms). For manufacturers, it is important to note that the regulatory standard for including adverse reactions (ARs) in drug labels remains the same. If the results of an OCMQ analysis conclude that there is an AR, medical and regulatory judgment should be used to determine whether the AR table should include individual AEs or simply list the OCMQ name in the label. If the OCMQ name is included in the table and individual AEs qualify as ARs, the individual AEs will be listed in the table under the OCMQ or included as a footnote.
Ground level
The FDA anticipates that further improvement and implementation of the OCMQ strategy described in this report will improve the OND’s ability to detect safety signals and distinguish and quantify ARs in clinical trial data. When incorporated into drug labels, this may help to improve identification of ARs and support informed clinical decision-making as new treatment options are approved. Be aware that regulatory documentation of AEs might look different from the way in which this information was presented previously.
