Taking Stock of Change in Clinical Trial Data and Operations

Throughout the COVID-19 pandemic, people have continued to make predictions, whether about “the great return” to on-site work or live conferences, or the move to decentralized clinical trials. So far, most of them have been inaccurate.

Instead of making binary choices, we have become comfortable with more complex and diverse models, and even with some uncertainty. Off-site and on-site work both flourish; online meetings coexist with traditional events; and more clinical trials include both remote and on-site elements.

It is interesting to look back on DIA 2022, a key industry event which took place on site for the first time after three years of major change. It showed how pharma is adapting and how far it has come in reimagining the future of clinical trials.

A more skeptical, balanced approach to technology

With so many discussions focused on patient-centric and decentralized clinical trials (DCTs), we must remember that these models are still at a very early stage of development. Although we now have hybrid digital trials, true patient centricity – which would allow patients to take part in clinical trials flexibly and based on their changing preferences – will take time to achieve.

However, there are encouraging signs that business leaders are looking deeper into what is missing from the data foundations for this work. After experiencing the limitations of new technology designed for DCTs, more have a healthy skepticism about what technology alone can achieve. Many have also learned that simply implementing new applications won’t make the patient’s journey any easier if it burdens research sites. At this point, many sites have become de facto “help desks” for patient technology, and are operating at a fraction of the capacity they supported before the pandemic began.

External data and wearables

Companies are evaluating and experimenting with new approaches for integrating data. For many, clinical data strategy has become the focal point, with a move toward aggregating, cleaning, and storing not only EDC data, but the large volumes of multisource clinical data now used for most trials. They are also working to find better ways to collect and aggregate eCOA/ePRO data with traditional data sources. Both efforts will be essential to establishing the kind of data foundation required for patient-centered trials.

Those of you who attended the more recent Society of Clinical Data Management conference heard how these ideas are driving the transformation of clinical data management into clinical data science. More on that in a future blog.

New data sources such as wearable sensors, which enable the use of digital trial endpoints, are also becoming increasingly important in cardiovascular, oncology, and rare diseases trials. At DIA, Alison Ritchie, global head of sensor solutions at Parexel, shared results of a 2020 survey of 146 industry professionals. At the time, 67% were already using wearables, mainly for heart rate, blood pressure, and temperature, and 25% were planning to introduce them in 2022.

Fatigue with DCT messages; challenges with patient diversity

Attendees openly expressed fatigue with the continued emphasis of DCTs in corporate marketing messages, and instead showed more interest in hearing about practical implementations with concrete examples. They discussed long-standing concerns, such as the need to improve patient diversity in clinical trials, but also shared challenges that have intensified since the pandemic, such as the impact of the “great resignation.” Companies are finding it more difficult to recruit and retain the best people for clinical work, and some are working on smarter approaches to onboarding and training, such as solutions that involve augmented reality.

This suggests an urgent need to avoid burdening specialists with recordkeeping and IT chores by eliminating disconnected data and paper-based processes. Drug safety, for example, highlighted at DIA and closely connected to clinical operations and outcomes, has long been managed using paper and disconnected legacy IT systems. As AbbVie and Sanofi executives showed, sponsors and CROs are moving safety operations away from manual and paper-based processes to centralized digital data, and toward automation. Clinical operations and data may be farther along this path, but continued progress will be vital.

Need for study velocity, not just speedy data collection

The explosion in clinical and third-party data still challenges the industry, which has gained speed but not velocity in addressing it, according to Saran Kukreja, senior director of business engagement and technology at Vertex Pharma, who participated in a panel discussion on digital trials at DIA. At this point, he and other panelists noted, clinical data is coming from different directions and different data sets are not talking to each other.

Adaptive trials, which allow protocols to be updated as new information is learned, were the focus of another panel discussion. They represent a more scientific way to run today’s complex trials, but taking this approach poses challenges for many companies. They can either choose to optimize the science behind the trial or the trial’s execution with patients, because every trial change requires unbolting each data source, amending data, then rebolting it.

There is a need for a central control location that accepts data, and a way to have disparate trial data come together seamlessly. We can expect to see more solutions addressing these needs in the future. As DIA reminded us, adaptive approaches are already being used, and can lead to better study results and shorter patient recruitment times — 81% likelihood of [post-trial] drug launch, compared with 68% for non-adaptive trials, and three vs. seven months for patient recruitment, as Mark Kaufman, director of RTSM and customer adoption at Medidata noted.

Experts shared best practices for making adaptive trials more successful, including starting projects off with the end in mind and planning adaptations based on what is most likely to happen. Simulation can be useful in these efforts, noted Song Wang, senior statistical science director at Thermo Fisher Scientific’s PPD Division, who also presented case studies showing how to take a ‘go/no-go’ approach to deciding whether to proceed with adaptive trials.

The need for clean, accessible, and connected data

In our presentations and conversations at DIA 2002, the Veeva team emphasized the need for data connectivity and digital trials that connect the key clinical trial stakeholders: patients, sites, and sponsors/CROs. Veeva has recently introduced two new products to help improve results on both fronts. One is Veeva Clinical Database (CDB), which allows patient data from Vault EDC and third-party clinical data sources to be aggregated, harmonized, and cleaned in a single place. This will take the focus off EDC and redefine clinical data management as the entire process of collecting and cleaning data, and ensuring that it is always ready for analysis. Another is Veeva ePRO, which research sites can use with a bring your own device (BYOD) model for patients. This new approach will give sites more control and enable them to connect more easily with sponsors.

For more insights from industry leaders on ways to advance operations to become more patient- and site-centric, read ‘Mastering Patient Centricity’s Moving Parts: Innovators Share Real-World Examples’.