Innovation + Partnerschaften

AI in pharma: Technology-driven disease prediction to advance patient care

By Mike Devoy


Mike DevoyMike Devoy
Member of the Board Management of Bayer AG
President Pharmaceuticals

We are in the early stages of what some are calling the fourth industrial revolution. This era will be defined by new technologies that come from the fusion of digital, physical and biological worlds. These are exciting times and as technology and access to data sources continue to increase, there will be incredible opportunities to better serve patients and society in how we discover, develop and deliver healthcare solutions.

In this context, we are seeing an explosive growth of data in the healthcare space. It is estimated that 750 thousand terabytes of healthcare data are created every day.*

To learn from and to take full and productive advantage of this data, we will need to use and apply artificial intelligence (AI) and machine learning to this Real World Data (RWD). Such data relating to patient health status and/or delivery of healthcare is routinely collected. It is available from a variety of sources, such as electronic medical records, health insurance claims, genomic data and data from health apps, wearables and other biometric devices.

From this data, we will increasingly be able to generate meaningful Real World Evidence (RWE). The U.S. Food and Drug Administration (FDA) refers to RWE as “the clinical evidence about the usage and potential benefits or risks of a medical product derived from the analysis of RWD”.**

There is a lot of talk and sometimes a degree of hype about what is happening and what is possible in this area. However, beneath the hype there are already practical examples of how these technologies can and are delivering value as well as the potential for much more in the future. We see practical engagement from regulatory agencies such as FDA and EMA who are working, communicating and issuing guidance about the potential role and benefits of RWE in the drug development process and for post-approval activities.

Applying new technologies in clinical trials

At Bayer, we are learning and experimenting with these techniques across a range of areas to better inform ourselves about what is possible. As part of this process, we are developing knowledge, partnerships and better outcomes for patients.

In the area of clinical trials, RWD sources provide a platform to potentially run clinical trials in the real-world setting, as well as to better inform the study design and feasibility. We are supporting a real-world randomized registry study in Scandinavia to allow us to answer a clinical question regarding NOAC patients in an efficient way. We are also working with partners to explore the use of technology to support decentralized (virtual) clinical trials enabled by telemedicine, home delivery of clinical supplies, e-consent, the use of wearables, and live 24/7 data monitoring. This will allow patients to participate in trials without disruption to their daily life and should allow for clinical trials that are faster, less costly and more patient centric.

We are also using the combination of data and AI-driven solutions to better characterize and stratify disease and the appropriate patient populations. We are in active collaborations in the Cardiovascular and Oncology therapeutic areas with external partners (e.g., the BROAD institute) to achieve this. Today, no company can act on its own and success requires partnership and collaboration in order to access data and expertise to achieve the best results.

While new technology may be exciting, the focus of healthcare must be on the patient. Care will become increasingly personalized as the particular, often unique issues of individuals are recognized. I believe that if we apply the new technologies as we are able to, then this will make it easier to put patients at the heart of all decision making about their health - a combination of technology and humanity.