Industry has been asking a lot of questions lately about Elsa (see a great example from Martin Lush). Why is the FDA rushing in? I think that they are trying to leverage technology to build a scientific process for provoking the kind of questions that they want to ask.
If so, we can expect a deep sea fishing expedition of the safety data, powered by AI. They’re likely to ask questions about every considerable imbalance disfavoring study drug or any credible high-risk subgroup they can evoke from the data. All the more reason why you need to know your safety story – rock solid – when preparing your safety-related filing documents. And to have all of the necessary infrastructure and processes in place to be able to quickly and authoritatively respond to any questions that they may have that you haven’t quite covered. What’s more, FDA reviewers have the same pride in finding an interesting safety signal that we have in finding a novel treatment. If you’re completely caught by surprise, you’re going to have a hard time convincing that FDA reviewer that you can quickly understand and characterize that medical concept.
That’s why we need a framework for safety evaluation – aggregate safety monitoring and scientific evaluation of integrated safety data throughout the product development lifecycle – that is different from the framework for efficacy analyses:
- Engages critical team members: The team can best identify potential safety signals, evaluate them for causal association with study drug and develop risk management strategies, when it is comprised of safety clinicians and data scientists.
- Supports the iterative nature of the process: As this process is repeated over the development period, the safety profile of the product emerges and is captured in living documents, like a Safety Storyboard.
- Recognizes different data sources used across the life-cycle of the product: Throughout development and post-marketing, sources of data will evolve, requiring development of various strategies and tools for their evaluation.
The safety profile of the product emerges, is documented in the Development Core Safety Information (DCSI) and Investigator’s Brochure (IB), and is communicated out through the Development Safety Update Report (DSUR) and the Integrated Summary of Safety (ISS). The iterative process of learning and decision-making continues, is documented in the Company Core Safety Information (CCSI) and product label, and is communicated out through the Risk Management Plan (RMP) and Periodic Benefit Risk Evaluation Report (PBRER).
To be successful, cross-disciplinary scientific engagement will be needed to integrate medical judgment and quantitative expertise in the monitoring and management of the product safety profile. Success will require dynamic, interactive, cross-disciplinary collaboration between clinical and statistical scientists.
A ship in harbor is safe, but that is not what ships are built for.
– John A Shedd (1928)
Let’s prepare for rough waters. In my next series of blogs, I will discuss how this framework for interdisciplinary safety evaluation can be supported by the Aggregate Safety Assessment Planning (ASAP) process.