Using Data to Advance Clinical Trials

From helping improve the patient experience to streamlining the process from drug development to approval, all stakeholders in the clinical trials process rely on the expertise of collaborative contributors. A connected technology ecosystem – spanning patients, research sites, and trial sponsors – and interoperability across systems and processes provides researchers with improved visibility to better manage trials in real time, eliminating the need for manual tasks and producing key insights faster. According to Richard Young, Vice President, Strategy, Veeva Vault CDMS at Veeva Europe, by embracing a digital approach, the industry is better positioned to deliver life-changing medicines more quickly to the patients who need them. Here, he explains some of the key breakthroughs in clinical trial technology over the last decade. 

How much data is generated in clinical trials now compared with 5–10 years ago?
 

Clinical trials vary greatly by nature. In the past, they were often assessed based on the quantity of data points collected, with a mere 20 per page. In recent years, it was more common to collect a million data points in a trial. Today, a device like a smartwatch can collect a million data points per patient per day. Some studies now even collect billions of data points per trial. 

There is now also a wide range of data sources available. In the past, a study with three to four data sources was deemed complex, whereas current studies easily incorporate 20 to 30 sources. 

What have been the major technological milestones?
 

The unprecedented amount and variety of data collected in trials today is a challenge for data managers. Over the past decade, there’s been a big shift from manual, paper-based data management to digital solutions, such as electronic data capture and electronic patient-reported outcomes. This has enabled far more effective collection and management of large datasets. Data managers are now using data lakes or warehouses to capture the full picture of clinical trial data. This has prompted a greater adoption of clinical databases, which aggregate and harmonize sources to provide clean, well-organized data for rapid analysis.

How have these data contributed to improved efficiency? 
 

With the vast volume of data produced, data managers can see a far more complete picture of a patient’s status in trials. With all that data now consolidated, it’s easier for data managers to process and query data to get to critical insights faster. 

Also, with access to the digital data generated from daily activities over longer periods, such as sleeping patterns and electrocardiogram (ECG) monitoring, we can now get more consistent insight into the patient’s health with the relevant context.

How is technology changing how patients and participants interact with trials?
 

Meeting patients’ needs and ensuring their safety in a real-world setting is important. True patient and site centricity is something the industry needs to embrace. For example, rather than needing to visit a site for regular testing, a patient can be equipped with a wearable device. This does not interfere with the patient’s everyday life and provides insights in real time.

Trials leveraging new technologies can often burden patients and sites with troubleshooting technical issues. Trust between the caregiver and patient is crucial, and technological advances should not be rushed or prioritized over comfort. Every patient will have different preferences regarding technology. Therefore, technology must be carefully considered before it becomes a part of the trial experience.

How do you foresee the technology trends developing?
 

Real-time data will remain essential and rely on successful technology deployment and integration within clinical trial design. Optimizing technology deployments in clinical trials is an ongoing and complex process. Effective data management is crucial for efficacy, compliance, and patient benefits, as well as how we improve, understand, and interpret the data.

Regarding the emergence of new technologies, AI holds immense potential for the sector. Though the opportunities are exciting, the increasing volume and variety of data in today’s trials are creating massive amounts of manual work. We need to first prioritize automation to support data aggregation and cleaning. Having this foundation of clean, quality data will be essential in feeding these AI models with usable data to ensure AI can meet its full potential.