Today’s trials are more complex than ever before – studies are targeting more specific patient populations and collecting a higher volume of data from more sources, including lab values, genomic markers, sensor data, images and patient-reported outcomes. In this environment, it’s no wonder that the majority of studies fail to meet enrollment timelines. Overcoming these challenges requires arming innovators with the necessary tools and insights. At Medidata, we’ve been doing this for more than two decades, becoming the first to have supported more than 30,000 clinical trials representing more than 9 million study volunteers on our platform. With that experience comes a lot of lessons. Here are some of the most important:
We can use technology to go faster than we thought
As data managers are responsible for the collection of more and different kinds of information, they need tools to integrate it, identify anomalies and quality issues, and create clean data sets to support regulatory submissions. Properly managed data can unlock new insights, help design safer, more successful trials and bring new therapeutics to market faster.
Speed was the imperative when the world was gripped by the race to develop a COVID-19 vaccine. That now serves as a model of leveraging real-time data collection and analysis. It played a pivotal role in identifying potential risks and making timely adjustments, resulting in the rapid delivery of safe and effective vaccines used by billions of people.
By using technology and analytics, we were able to start studies faster, accelerate the integration and reconciliation of data from diverse sources, identify anomalies sooner and automate locking of patient data. What we learned can and is being applied to a broader range of trials today.
Getting a 360 degree view of the patient delivers new insights
More and different types of data collected before, during and after clinical trials means we can gain a broader, holistic view of the patient. With this data, we are rapidly moving to a future where trial designs can be optimized and outcomes can be simulated before a patient is ever enrolled.
Even today, we are using AI to design better studies, predict outcomes and make trials safer for patients. For instance, we’ve used AI models analyzing historical clinical trial data to predict biomarker differences associated with severe cytokine release syndrome (CRS) in patients undergoing CAR-T therapy. (CAR-T therapy involves engineering a patient’s own immune cells to treat their cancer and CRS can be a life-threatening complication.) By leveraging AI insights, safer trials can be designed for patients receiving these innovative therapies, improving patient safety and trial success.
In another example of personalized medicine, Medidata collaborated with the University of Pennsylvania, Castleman Disease Collaborative Network, and Every Cure to identify drugs that can be repurposed to treat idiopathic multicentric Castleman disease (iMCD), a rare and life-threatening condition. AI algorithms analyzed proteomic data, leading to the discovery of adalimumab as a potential treatment for iMCD. This previously unknown application demonstrates how AI can help uncover new uses for existing medications.
We can make trials better for patients by putting them first
In addition to the design of studies, we can minimize the operational burden of trials on clinical sites and patients. Our models show that when the trial burden on patients is reduced – for example, by minimizing visits or painful procedures – enrollment increases and patient retention improves.
When data from numerous sources are combined, including clinical, genomic and laboratory data, clinicians can identify patients best suited for a clinical trial. This helps to enroll patients faster but also recruit populations who are most likely to respond positively to treatment, resulting in better outcomes.
And by harnessing the available data from previous trials and patients, important insights into patient populations, treatment effects, and likely outcomes can be derived. Synthetic Control Arms® (SCAs) can act as a ‘virtual twin’ by using data from past participants as the control arm in a clinical trial. This removes the need to enroll all or part of that population. SCAs are becoming more accepted by regulators in certain disease areas where it is hard to recruit into a control arm due to the rare nature of the disease, or where it may be unethical. The more that we can help patients get the experimental treatment that they're really hoping for, the better – and SCAs will increasingly play a vital role in facilitating this.
Conclusion
The importance of high fidelity data sets and new technologies for the benefit of patients, clinical trial participants, study sponsors and contract research organizations cannot be overstated. These are already transforming the life science industry and delivering enormous public health benefits. We still have lots of work to do, but experience tells us that we can deliver a future of accelerated research, personalized medicine and improved healthcare for all.
