Risk Management and Minimization in Early Clinical Development:
“Designing with the End in Mind” in a Setting with Little Data
Ideally, risk management begins early in clinical development, to ensure that the safety profile of a drug is explored at a point where it is easier to make changes to adapt to biological reality. Given the limited number of subjects studied, using all available data to identify potential signals is key. However, the tension between understanding a drug’s safety profile and providing enough evidence of efficacy to warrant further study can inhibit detailed analysis. In contrast to the post-marketing setting, in early trials detailed and accurate safety data can be obtained, but the small number of subjects limits statistical approaches. Furthermore, relevant data may reside in different functions, requiring effective cross-functional collaboration. Patient safety is directly impacted by resulting decisions.
The new FDA draft guidance, “Sponsor Responsibilities - Safety Reporting Requirements and Safety Assessment for IND and Bioavailability/Bioequivalence Studies Draft Guidance for Industry” released in June 2021 adds both complexities and solutions to this enduring dilemma. It suggests comparing the incidence of ‘anticipated events’ to that of an untreated population to trigger an Aggregate Analysis to determine the presence of a signal that would require expedited reporting. However, our industry survey showed none of the small or startup companies having begun planning for implementation. We present a use case that combines patient exposure and the incidence of anticipated events to set a trigger for Aggregate Analysis.
Results of the proposed approach showed that statistical modeling of exposure versus ‘anticipated event’ frequency provides the number of SAEs for a given medical condition that would indicate a >80% probability of increased risk, allowing for focused analysis of clinical data sets even with the limited data sets in early development. The resulting Aggregate Analysis combines medical and statistical approaches to focus on potential signals. Prospectively determining incidence rates for ‘anticipated events’ remains a significant challenge. Use of all available data for Aggregate Analysis is key to understanding the emerging safety profile.
By using the concept of ‘anticipated events’ with appropriate statistical modeling, as well as aggregate analysis; the challenges of early clinical development can be addressed to assure patient safety and effective drug development with smaller data sets. This approach focuses on early signal detection, employs appropriate statistical modeling within a traditional pharmacovigilance framework, and addresses the challenges put to industry by the June 2021 FDA draft guidance.

**Risk Management and Minimization in Early Clinical Development:**
**“Designing with the End in Mind” in a Setting with Little Data**
Ideally, risk management begins early in clinical development, to ensure that the safety profile of a drug is explored at a point where it is easier to make changes to adapt to biological reality. Given the limited number of subjects studied, using all available data to identify potential signals is key. However, the tension between understanding a drug’s safety profile and providing enough evidence of efficacy to warrant further study can inhibit detailed analysis. In contrast to the post-marketing setting, in early trials detailed and accurate safety data can be obtained, but the small number of subjects limits statistical approaches. Furthermore, relevant data may reside in different functions, requiring effective cross-functional collaboration. Patient safety is directly impacted by resulting decisions.
The new FDA draft guidance, “Sponsor Responsibilities - Safety Reporting Requirements and Safety Assessment for IND and Bioavailability/Bioequivalence Studies Draft Guidance for Industry” released in June 2021 adds both complexities and solutions to this enduring dilemma. It suggests comparing the incidence of ‘anticipated events’ to that of an untreated population to trigger an Aggregate Analysis to determine the presence of a signal that would require expedited reporting. However, our industry survey showed none of the small or startup companies having begun planning for implementation. We present a use case that combines patient exposure and the incidence of anticipated events to set a trigger for Aggregate Analysis.
Results of the proposed approach showed that statistical modeling of exposure versus ‘anticipated event’ frequency provides the number of SAEs for a given medical condition that would indicate a >80% probability of increased risk, allowing for focused analysis of clinical data sets even with the limited data sets in early development. The resulting Aggregate Analysis combines medical and statistical approaches to focus on potential signals. Prospectively determining incidence rates for ‘anticipated events’ remains a significant challenge. Use of all available data for Aggregate Analysis is key to understanding the emerging safety profile.
By using the concept of ‘anticipated events’ with appropriate statistical modeling, as well as aggregate analysis; the challenges of early clinical development can be addressed to assure patient safety and effective drug development with smaller data sets. This approach focuses on early signal detection, employs appropriate statistical modeling within a traditional pharmacovigilance framework, and addresses the challenges put to industry by the June 2021 FDA draft guidance.

Risk Management and Minimisation in Early Clinical Development: “Designing with the End in Mind” in a Setting with Little Data

DIA Europe 2022

A global, member-driven non-profit organization mobilizing life sciences and healthcare. DIA fosters innovation to improve health and well-being worldwide. Professionals from 80 countries continue to engage with DIA through our unparalleled membership network, educational offerings, and professional development opportunities.