We sat down with Chris Boone, PhD, vice president, global medical epidemiology and big data analysis, Pfizer Inc., and HIMSS20 Pharma Forum speaker, to discover the latest pharmaceutical innovation trends. He shares his insights on the strength and limitations of real-world evidence, how analytics will transform research and development across the industry and the power of predictive modeling.
From a real-world evidence and big data analytics perspective, there are some innovative trends occurring across the entire pharmaceutical value chain:
I’m hoping to see more movement toward a much more robust citizen science ecosystem and the use of patient-generated data to power precision medicine efforts.
One of the biggest misconceptions about real-world evidence is that it can address every question. Real-world evidence is not a one-size-fits-all approach to evidence generation, especially as it pertains to regulatory submissions. It has its strengths and limitations, and a degree of pragmatism is necessary to uncover hidden patterns in the data and address questions in our clinical discovery, development and commercialization efforts.
The strength of real-world evidence lies within its ability to demonstrate effectiveness in a variety of clinical practice settings and capture data from a large and homogenous patient population. However, there are technical limitations—such as data quality issues, bias control and the inability to test causality—resulting from residual confounding and reverse causation.
Another misconception is that pharma is the purveyor of all things real-world data. The reality is the pharmaceutical industry does not generate data at all, relying on data partnerships with providers, payers, tech and other third-party intermediaries to collect, aggregate and/or analyze real-world data to produce real-world evidence. As such, continuous education about the pertinent use cases for real-world evidence is necessary to keep practitioners and the general public informed.
We also need to remain diligent about the appropriate use cases for real-world evidence to support decision-making, continue efforts to establish data standards and improve system interoperability, and determine how best to engage pharma in the health data supply chain to ensure access to this relevant data.
There is reason to be excited about many things, but if I had to choose: transforming pharmaceutical research and development with real-world evidence and advanced analytics. One of the more exciting areas is the use of predictive modeling for clinical discovery.
In early clinical development stages, the chance of a drug reaching a patient is 10%, and it can take approximately 10–15 years for a drug to be developed and approved by the Food and Drug Administration. To that end, you’re seeing industry-wide initiatives to bolster innovation and efficiency in our research and development efforts.
The use of big data technologies—such as real-world evidence, AI and predictive modeling—create potential to improve clinical discovery and development processes that have relied heavily on prior experiences and assumptions. With the effective curation of real-world data, predictive modeling enables a deeper understanding of disease progression, identification of potential candidate molecules with a high probability of being successful, and even the ability to target specific patient populations based on genetic information, personality traits and disease status. It can be a gamechanger for clinical discovery, and I’m extremely excited about the possibilities.
The future rests on more open innovation across the industry. The effective and sustainable use of real-world evidence will require the establishment of clinical research networks—which involves forming cross-sector partnerships with providers, payers, patients and biopharma to pool resources and expertise in conducting clinical research. We should no longer continue to operate in our industry silos. The time is now to do what is best for patients.