As COVID-19 vaccines and other new treatments have come to market at record speed, pharmaceutical companies have faced a deluge of adverse event reports. According to Ernst & Young,¹ large pharma companies currently have to deal with an average of 700,000 adverse event cases a year and IDC estimates² that caseloads are increasing by 30-50% annually, due in large part to the arrival of COVID-19 vaccines.

Companies are constantly challenged to keep costs down to optimize their investment, focus on patient safety, and the benefit/risk analysis of products. The hard truth is that pharmacovigilance is an additional cost that doesn’t drive revenue, but it’s an essential requirement to doing business. The view of pharmacovigilance as just a cost of doing business is being replaced with a new perspective of how PV data can be leveraged as an asset to inform discovery, protocol design, competitive differentiation, and more.

With so many new ways to report adverse events companies today have access to more data on drug safety than ever before. As adverse events rise, it’s critical to have a system in place that can provide fast, high-quality insights at scale to drive the company—and industry—forward.

To handle the spike in safety reports and improve general efficiencies with adverse event reporting, the industry is turning to newer technologies, specifically machine learning and natural language processing (NLP).

A well-designed, automated system using the right technology can eliminate tedious and repetitive tasks, reduce data entry errors, and even enable “predictive” signal detection. In fact, according to Ernst and Young,³ companies could reduce their pharmacovigilance costs by nearly 45% by automating manual steps.

One such company is CSL Behring which manufactures plasma-derived and recombinant therapeutic products. The company processes more than 50,000 adverse events per year using the Oracle Argus cloud platform. Now, it is developing robotic process automation processes around non-value-added activities to allow doctors to focus on benefit/risk management. They are also using NLP around Medical Dictionary for Regulatory Activities (MedDRA) coding to limit work done on routine tasks.

“COVID has taught us a lot about adverse events and how they are handled—it’s brought pharmacovigilance into the mainstream,” said Richard Wolf, head of pharmacovigilance operations at CSL Behring. “Now, companies around the world are embracing and making reporting a high priority. And, just as importantly, technology has made the reporting much easier and more efficient.”

As companies like CSL Behring look towards the future, they are exploring how to best integrate AI and machine learning technologies with their workforce. Key to this process is understanding how tools can be more efficient and effective, both in crunching data and enabling staff to focus on different areas of their jobs that allow them to be more creative and innovative. By applying machine learning and data science approaches, companies can automate mundane, repetitive tasks and quickly gain new information and insights about patients and therapies from the abundance of data.

Focusing on adverse event case intake, AI can be applied to a wide range of data types such as forms with a defined structure to images. It is also possible to extract and analyze data from unstructured sources like journal articles or emails. Once the documents have been automatically structured and processed, they can be separated into “routine” cases that can be handled entirely by software, and “high-priority” cases that require a closer review by safety specialists.

The insights provided by AI also enable safety evaluators to make more informed observations, for example, with new techniques such as neural signal detection, multimodal signal detection, and predictive signal detection.
Such insights help specialists make more informed benefit-risk evaluations, cross-checking patterns of data to better understand the safety profiles of their products, and extend the boundaries of scientific research.

There is no doubt, change is difficult, iterative, and mass adoption won’t happen overnight. One way that new techniques are being shared and tested is through consortiums. These organizations, like the Oracle Health Sciences Safety Consortium provide arenas for testing new ideas that can help the industry. Consortiums such as this provide an even playing field where members feel confident that they are moving together—and in the right direction.

Throughout the past 15 years a major priority has been on optimizing the entirety of the safety case management process from end to end. But it’s been a challenge to get there. The complexity of applying medical judgement to individual cases continues and is essential for patient safety. The prospects and potential in applying robotics and AI to routine aspects of the pharmacovigilance process is very exciting and with forward leaning organizations such as CSL Bering making inroads, we will see the rest of the industry follow. 

References

1. How robotics is reshaping the biopharma value chain, Ernst & Young, 2018.
2. IDC MarketScape: Worldwide Life Science Drug Safety Services 2019–2020 Vendor Assessment — Building for Innovation, IDC, 2019.
3. How robotics is reshaping the biopharma value chain, Ernst & Young, 2018.