Everyone loves to talk about how quickly technology evolves, and in the biopharma industry, that speed is essential for keeping up with the equally rapid advancements in the world of science. In the past few years alone, advances in immunotherapies, biologics and more targeted diagnostic tools have given new hope and better lives to millions of patients around the world.
As our understanding of science evolves, it is important for technology to keep up as well. The rapid pace of change in clinical research demands more efficient technology and devices that can store, manage and interpret data at a faster pace with more reliable outcomes. At the same time, regulatory bodies must also adapt to these new technologies and innovations in order to leverage these new capabilities in a timely manner.
Integrated data streams
A key component of the science and technology evolution is our ability to create horizontal systems that capture and analyze data across multiple data sets to support clinical research and development. This has been one of the biggest technological challenges for our industry because it is built on a foundation of vertical systems in which each department and project team captures and stores their data in isolated silos, locked-off from the rest of the world. This disconnected approach limits our ability to gain efficiencies, automate solutions, or implement real innovation from a data analytics perspective, because the information is so difficult to access.
In response, many IT experts in the biopharma industry are building new tools that can cross these barriers and create end-to-end solutions to support scientific evolution and add efficiency to the research process. We are already seeing many of these tools coming to market, and being implemented in a number of exciting ways.
Some of the most innovative applications of technology are in the clinical research and treatment environment, in which sensor technology, wearables and even long lasting implantable devices are changing the way physicians gather information about their patients’ conditions. Consider the Continuous Glucose Monitor (CGM), which is a tiny sensor embedded in a patient’s abdomen that continuously tracks their blood glucose levels and transmits that data back to a receiver; or wireless heart monitors that can monitor a patient’s heart rate for three days without needing to be recharged.
These innovative tools are less invasive and enable physicians to collect a real-time flow of data — rather than the snapshot in time they collect at infrequent in-person visits, which enables them to make more accurate data-driven decisions about their care.
We are also seeing the application of technology through apps that help track patient health and even recruit for clinical trials. In a notable example, the MyHeart Counts app, which was developed using the Apple ResearchKit platform, reportedly recruited more than 40,000 participants in a matter of weeks for a cardiovascular trial run by Stanford Medicine. The app then used iPhone motion sensors to track participants' physical activity, which was fed back to the study. The ResearchKit platform has seen similar success with apps recruiting patients to studies for autism, epilepsy, skin cancer and others.
And on the software side, we are seeing new applications of machine learning and automation that allow researchers to scan multiple disparate data sources to identify patterns or highlight anomalies. These kinds of innovations can drive real and meaningful efficiency gains, while enabling us to take a more proactive approach to safety.
In some cases, the automated reviews outperform humans. Google's DeepMind project, for example, is using automated machine learning tools to analyze more than one million anonymous eye scans, creating algorithms that can detect early warning signs of eye disease that humans might miss. "If you have diabetes you’re 25 times more likely to go blind,” DeepMind co-founder Mustafa Suleyman told The Guardian. “If we can detect this, and get in there as early as possible, then 98% of the most severe visual loss might be prevented."
These tools are also being used to do more proactive risk-based monitoring by using predictive analytics to gain a holistic view of site data and determine whether an anomaly is a true risk or simply an outcome of other factors, such as exposure time, or enrolment numbers. All of these advances promise to cut the time and cost of trials while improving safety and ease of use for patients.
However, for all of these advances there are many obstacles. As we integrate more data into the clinical research process and harness new sources of data and cloud-based storage, we face the increasing need to manage risk and ensure the information we collect and use is secure and privacy is maintained. As the number of data sources used expands, so too do these risks and questions about our ability to maintain data security.
Despite the rapid pace of progress, regulators are more cautious in the face of change, requiring extensive proof that these technologies are trustworthy, safe, ethical and necessary to the research environment before approving their use. As an innovator, it can be frustrating to watch regulations lag behind progress, but it is a necessary balance, and eventually the rules will catch up.
There is also the challenge of ownership and cost. By their nature these tools strive for a more collaborative transparent research environment, which often drives the question of who pays for these advances? Who owns them? And who carries the risks? These are questions we need to address so that we can begin to build more collaborative partnerships to further advance these technologies.
No single stakeholder within the industry can do it alone. Even tech giants like Google need to partner with industry organizations to achieve their vision for a technology-enabled healthcare environment. If we as an industry can embrace the change that these tools bring and work together, we can demonstrate what is possible, and make those possibilities a reality that benefits everyone — from vendors and biopharma companies down to the patients we all serve.