Clinical Researcher—April 2024 (Volume 38, Issue 2)

PRESCRIPTIONS FOR BUSINESS

Malia Lewin, JD

When I was working in the technology space, it was understood that if a sponsor’s contract research organization (CRO) had a role in supporting the implementation of our product at sites, it meant the end of its usage and adoption in those settings. Why would that be? Well, CROs are responsible for protecting study timelines, budgets, and quality. This means that isolating and eliminating variables that could distract sites with perceived time-drainers such as technology training, requirements for additional passwords, risks to data quality, or other contributions to site burden is part of the job.

My tech friends often ask me how to effectively partner with CROs and my answer is “give me something to improve clinical trial outcomes while driving faster timelines with reduced costs.” CROs don’t hate all technology, just point solutions that contribute to user burden without significant benefit or integration into already complex workflows and data management challenges.

Legacy technologies are no longer fit-for-purpose, and we must follow the lead of other sectors like manufacturing and consumer packaged goods in adopting generative artificial intelligence (AI) and other technology and data advancements to support pain points around patient experience, data integrity, quality, and the supply chain. A recent MIT report of senior executives in the healthcare and life sciences sector showed that 38% of us consider our use of generative AI and other advancements to be very slow to moderate.{1} This is simply not good enough.

Things to Pack for the Journey

In the journey from drug discovery to approval, a significant amount of information is collected. The persistent problem is generating and actioning insights from these data in a manner that allows us to take quick, decisive action. About 61% of those surveyed by MIT are increasing investment in data and AI analytics up to 25% over the next year and 38% expect an even larger increase.{1} The large majority (about 72%) will leverage these investments to support streamlining workloads and accessing real-time information. Others see greater potential to push substantive scientific breakthroughs and data mining in areas of high growth, such as biomarker identification, genetic variant targeting, and personalized medicine applications.{1}

Another timeless challenge up for a good technology solution is the persistent issue of data integration across the diverse network of sites and systems on a given clinical trial. More than half (52%) of the industry’s respondents in the MIT survey said having a single system for structured and unstructured data used for AI is “very important” to achieving their organization’s technology goals, and yet one third of the MIT respondents say their organizations support 10 or more data and AI systems.{1}

AI and natural language processing will help bridge multiple datasets and allow for the democratization of insights across stakeholders. Clarity around the location of real-time information with a user interface allows all study stakeholders to query the data with minimal training and hassle. This can have a positive impact on patient outcomes and the time, quality, and cost of running clinical research.

Leveraging AI methods like machine learning and natural language processing can also help meet the complex challenge of reducing bias in clinical research and ensuring equitable accessibility for all communities of stakeholders across therapeutic indications. Hispanic/Latinx patients make up 18.5% of the population but only 1% of typical trial participants{2}; African American/Black patients make up 13.4% of the population but only 5% of typical trial participants.

Further, between 2011 and 2020, 60% of vaccine trials{3} did not include any patients over 65, even though 16% of the U.S. population is over 65.{4} To fill diversity gaps like these, companies like Johnson & Johnson are  leveraging AI to identify new sites for accessing underserved populations{5} and others are using AI to build screening, enrollment, and retention models to close diversity, equity, and inclusion (DEI) gaps and meet data traceability and transparency goals aligned with U.S. Food and Drug Administration guidance and regulations.{4}

Conclusion

Leveraging technology in clinical research to drive advanced analytics, integrate and query data across multiple data sources, and improve DEI in clinical research are unifying causes for CROs and the technology sector. If we put our heads together on effective and efficient solutions that leverage AI, as other sectors outside of healthcare and life sciences have done, we will drive a faster, cheaper, safer path from discovery to clinic and benefit millions of patients globally.

References

1. Bringing Breakthrough Data Intelligence to Industries. 2023. MIT Technology Review Insights.
2. Representation in Clinical Trials: A Review on Reaching Underrepresented Populations in Research. 2020. Clinical Researcher.
3. Assessment of the Inclusion of Racial/Ethnic Minority, Female, and Older Individuals in Vaccine Clinical Trials. 2021. Journal of the American Medical Association.
4. Leveraging Machine Learning and AI to Improve Diversity in Clinical Trials. 2023. IBM.
5. Big Pharma Wants to Use AI to Increase Diversity in Clinical Trials. 2023. Bloomberg Law.

Malia Lewin, JD, is Senior Vice President for Global Commercial Operations at Lexitas Pharma Services. In a life sciences career spanning more than 25 years, she has held executive leadership positions with Teckro, QbDVision, Veeva, Rivermark (an IQVIA company), Ascension Health, and the International Psoriasis Council. She has served in a variety of mission critical roles, from lobbyist and public policy strategist to global strategy head and commercial lead.