Advera Health has long embraced the fact that safety signals can arise from a wide variety of data sources. The authors of GVP IX agree. The module specifies that common datasets include spontaneous reporting (MAH’s own database, national databases such as FDA Adverse Event Reporting System (FAERS), EudraVigilance, and VigiBase), active surveillance, studies, and scientific literature. All of these data sources should be considered depending on characteristics of the product and based on clinical judgement.
Signal detection should include manual review of Individual Case Safety Reports (ICSRs), statistical analyses, or a combination of both. Statistical methodologies are discussed in GVP IX Addendum I - Methodological aspects of signal detection from spontaneous reports of suspected adverse reactions. In short, disproportionality analysis in combination with additional data summaries based on both statistical and clinical considerations should be used.
It is this type of analysis that the Evidex platform provides for our clients. With analytics based on Advera Health’s optimized FAERS dataset, and expanded to VigiBase and clients’ own ICSR databases, Evidex allows for easy-to-use ICSR viewing, case series generation, and custom data mining. Clinical considerations such as labeling status, seriousness (both designated medical events (DME) as well as IME Seriousness mappings), and disease relatedness are provided out-of-the-box. Internal documents such as a Core Common Data Sheet (CCDS) and risk monitoring event lists can be imported and displayed.
Evidex Signal Management further automates signal detection as it relates to GVP IX. After working with the Advera Health team to customize desired prioritization thresholds based on any criteria and data that a user has access to, new potential safety signals are automatically pushed into a dashboard for review. After an initial assessment, a safety reviewer can then choose to add the signal into a formal signal validation and confirmation process or dismiss the observation. This combination of automation and human review provides a ‘check’ in order to ensure an observation does in fact warrant a validation review.
Evidex Signal Management users can also add a signal into the evaluation process manually from any drug-adverse event analysis report, or by entering a new signal from scratch. The latter is helpful for situations where the signal isn’t generated by a dataset currently connected to Evidex, or for regulatory authority inquiries that need to be tracked in the same fashion as a signal.
After entering the evaluation process, the next step will be signal validation to determine if prioritization and further assessment is warranted.