How AI is Minimizing Clinical Trial Protocol Amendments

In this interview, we sit down with Catherine (Cat) Hall, Head of GXP Quality at Egnyte, to discuss how AI is transforming clinical research and development approaches. Cat shares her insights on integrating AI with regulatory and ethical standards. This discussion explores the balance between automation and human oversight, interoperability challenges, and AI’s role in minimizing protocol amendments.

Moe: With AI’s crucial role in development, how do Egnyte and Espero ensure AI insights align with regulatory and ethical standards in clinical trials?

Cat: I believe in the “human in the loop” approach. In our regulated industry, human oversight is essential to ensure AI tools are used responsibly. I treat AI like an employee, assessing its background, training, and experience to ensure it fits our needs. At Egnyte and Espero, we are focused on understanding the use case and improving models so that human evaluators can confidently proceed with AI-generated insights. This approach ensures AI meets regulatory standards and aligns with ethical considerations, providing a robust framework for clinical trials. By maintaining this balance, we can harness AI’s potential while safeguarding the integrity of our research processes.

Moe: Automation minimizes errors, but protocol design needs unique perspectives. How do Egnyte and Espero balance automation with human oversight for efficiency and rigor?

Cat: Securing data as intellectual assets is crucial. I’ve spent two decades in the industry and I’ve seen the importance of preventing competitive leaks. For instance, when designing protocols, ensuring data isn’t emailed around is vital, which could lead to security breaches. We maintain security and efficiency without compromising scientific rigor by integrating tools that simplify protocols into new regulatory standards. This approach protects intellectual property and streamlines the process, allowing experts to focus on the nuances of protocol design. Doing so ensures that automation complements human expertise, enhancing both efficiency and scientific rigor in our trials.

Moe: Many stakeholders use disparate systems. How does your partnership address interoperability challenges for seamless data exchanges across CROs, sponsors, and agencies?

Cat: Collaboration is at the heart of our industry. Egnyte’s platform ensures everyone has access to the correct version of data, fostering confidence in shared information and facilitates the transition from traditional document storage to a more digital, modern approach, ensuring all parties are on the same page. By providing a unified platform, we address interoperability challenges, allowing seamless data exchanges and enhancing stakeholder collaboration. This improves efficiency and ensures that all parties are working with the most accurate and up-to-date information. By fostering this level of cooperation, we can drive innovation and efficiency in clinical trials.

Moe: Protocol amendments are a pain point. How do AI-powered solutions identify potential inconsistencies to minimize costly amendments?

Cat: AI and machine learning can analyze past data to identify potential inconsistencies. For example, by examining previous studies, we can learn from their design and avoid unnecessary complexities. At a recent conference, a speaker highlighted the high cost of amendments, which can ripple through the clinical supply chain. By designing adaptive trials with AI insights, we can reduce the need for amendments, ultimately saving time and resources. This proactive approach minimizes disruptions and enhances the overall efficiency of clinical trials. By leveraging AI, we can anticipate challenges and design more robust protocols.

Moe: What KPIs will Egnyte and Espero use to measure the impact of this integration on trial efficiency, and how do you plan to drive adoption among hesitant sponsors?

Cat: Success will be measured by faster trial starts, fewer amendments, and better protocol understanding by investigators. The industry is increasingly embracing AI for its potential to improve fundamental trial elements. For instance, AI can enhance enrollment processes and streamline protocol design, leading to more efficient trials. We aim to encourage sponsors to adopt AI-driven workflows by demonstrating these benefits. The focus is on tangible improvements, such as reduced protocol deviations and enhanced investigator understanding, which can significantly impact trial outcomes. By showcasing these successes, we hope to build confidence and drive broader adoption of AI in clinical trials.

Moe: AI-driven content and management insights are prevalent. What validations have you implemented to ensure high-quality results?

Cat: Validation involves ensuring explainability. We need to be able to trace AI-generated answers back to specific data sets. This transparency allows humans to evaluate the context and accuracy of AI insights, ensuring they meet industry standards. As a member of the International Society of Pharmaceutical Engineers, I’ve seen the importance of a risk-based approach to validation. By ensuring that AI models can explain their outputs, we provide the necessary assurance to sponsors and regulatory bodies, fostering trust in AI-driven solutions. This approach ensures compliance and builds confidence in the reliability of AI insights.

Moe: Is there anything else you’d like to add, Cat?

Cat: While there’s excitement about AI’s potential, having clean, structured data is crucial. Without it, AI models can’t be trusted. Understanding data origins and maintaining version control is essential to leveraging AI effectively in clinical trials. A lot of pre-work is involved, and skipping these steps can lead to unreliable outcomes. It’s important to remember that AI is only as good as the data it processes, and ensuring data quality is fundamental to achieving meaningful insights. We can fully realize AI’s potential in transforming clinical trials by prioritizing data integrity.