Now what? The future of clinical trials after the pandemic

Sheila Rocchio highlights four key areas life sciences must improve on as an industry to drive real change in how clinical trials are conducted.

This is definitely a time for the life sciences industry to be proud of our accomplishments. The way in which the industry came together to support the development of the COVID-19 vaccines has shown that it is equally important to determine how to apply the lessons of the last year to the future.

We need to be prepared to do it again. As an industry, we must take the principles that led to the successful development of vaccines in record time and apply them to other serious medical conditions that may lack the urgency of a pandemic but still affect millions of lives every year.

This means establishing a sustainable process that supports scientific discovery through a data-first framework that leverages technology and emphasises the development of new skills within our teams.

Increase in data sources and providers

Data is more diverse than ever before, with the volume, velocity and variety of data being collected and analysed reaching new highs. Some of this data is gained through continuous sensor devices, like those used to monitor glucose levels over an extended period of time.

Not only does this type of data help treat diabetes, but continuous sensor data also represents important markers of wellness that can have great value to improving human health overall. With the availability of all this rich data from specialty devices and labs, researchers can gain a variety of insights.

Unfortunately, this also creates layers of data complexity, leaving drug developers responsible for making sense of these disparate sources and insights. This challenge has led to significant delays in clinical studies, with a 40% increase in data cleaning cycle times over the last two years. Having data in one place is the first step in unlocking all of these possibilities that digital therapies can bring to treating people.

Implementing AI/machine learning models

AI and machine learning will be one of the key enablers to tap into the potential of this proliferation of data. As the electronic sources of data collection continue to grow, centralised monitoring, risk-based monitoring, utilising statistical insights and machine learning will become an important part of every trial to focus data review efforts. AI models can be used to monitor trial processes and automate tasks associated with data collection, review and analysis.

This emphasis on developing AI capabilities is notable, as highlighted in a recent report by Gartner, where "46% of life science CIOs state that BI/data analytics capabilities will receive the largest amount of new or additional investments". As life sciences companies invest in building the necessary skills and resources across the board, having a data sciences infrastructure to mine and learn from data, glean insights from this data with machine learning and automate more of the clinical development processes with learning models will become a critical core competency.

Data accessibility is now a strategic imperative, and the manual methods that are still heavily relied upon are no longer effective in this new environment. The organisations that will be competitive in the drug development industry will be those that enable AI capabilities. 

Building the right skill sets to implement new technologies

The promise of machine learning and AI can be fulfilled by combining them with data knowledge and data scientists who can apply their expertise in a repeatable fashion. This means that in order to advance our trial processes and increase the speed of drug discovery, the right skills within our research teams must be supported and developed. These skills include an understanding of how best to implement new technologies so that they can help optimise trials and make processes more efficient. Expanding on data science skills drives the benefit of these technologies even further, turning them into powerful tools that can be wielded intelligently to reduce trial inefficiencies. When clinical operations teams and data managers can supplement their efforts with technology and perform their tasks at scale, that’s when clinical research can significantly move forward at an accelerated pace.

Next generation of decentralised trials

Another important movement to come out of the pandemic has been the increased adoption of a decentralised trial approach. While this model isn’t new to the industry, the way in which it has been widely implemented has shown how critical this approach was to enabling patients’ continuation in trials during the pandemic. Access to finding the right patients to participate in trials is often cited as the number one obstacle for completing a clinical trial on time and within budget. Finding sites that want to enroll patients is also traditionally a challenge. Decentralised trials will help increase access to the possible pool of patients and allow a more diverse population to participate in research.

The incredible enthusiasm across the industry to build on the gains during this time have also resulted in the establishment of patient-focused groups, like The Decentralised Trials & Research Alliance (DTRA). By using a combination of technology and processes, including remote visits and diagnostics, decentralised approaches can lessen the hurdle of finding more representative patients and make the clinical trial product more accessible, easier to use and adaptable. It will mean, however, that infrastructure to support centralised, standardised data will become even more critical.

The momentum to shape the future

We have been talking about digital transformation and improving trial processes for years. That change has finally occurred with the pandemic. What have we learned? There was no single entity or initiative that led to the creation of the COVID-19 vaccines, but rather a collective effort. There was an unprecedented collaboration, fast adoption of new technologies and a single-minded focus unlike anything we have ever seen in the life sciences industry. If we want to carry this momentum into the successful development of therapies that treat other conditions, we must rethink our clinical development processes and commit ourselves to moving forward — not back — to a new and improved way of conducting clinical studies.

About the author

Sheila Rocchio is the chief marketing officer of eClinical Solutions and oversees the company’s marketing function.