Convergence of AI and Digital Measures for Seamless Preclinical to Clinical Translation

Szczepan Baran
Chief Scientific Officer

A recent workshop on the "Convergence of AI and Digital Measures for Seamless Preclinical to Clinical Translation" brought together, for the first time, experts from clinical and preclinical into one room to explore the transformative potential of combinatorial power of digital measures and AI technologies in drug development. Through focused sessions and case studies, we explored the current implementation and validation processes for AI/ML-derived digital measures and addressed the challenges and obstacles to their adoption. Our goal was to identify strategies for integrating digital measures and AI to enhance the translation from preclinical to clinical stages, improve data precision, and create a synergistic effect for a comprehensive understanding of these technologies. 

This symposium had two primary goals:

  1. Formulate a strategic approach for incorporating digital measures and AI/ML into the entire drug development process, aiming to create a paradigm shift through innovative synergies. 
  2. Build a collaborative community committed to achieving the vision outlined during the symposium, utilizing collective expertise to drive the field forward. 

In the first session, we explored the fundamental integration of AI/ML in the preclinical and clinical phases of drug development. This session demystified digital measures and AI/ML, highlighting their pivotal role in advancing pharmaceutical innovation. Participants gained valuable insights into the harmonious synergy between AI/ML technologies and digital measures, fostering a comprehensive understanding of their joint impact. This foundational knowledge set the stage for deeper discussions on technical challenges and innovative solutions in the subsequent sessions.

The second session delved into the intricate technicalities of employing AI/ML in identifying and validating novel preclinical digital measures. The comparative analysis of AI/ML applications revealed the potential of these technologies to provide high-resolution, continuous preclinical data that enhances animal welfare and clinical translatability. Furthermore, the discussion extended to multiple AI/ML methodologies, addressing critical issues such as data heterogeneity and algorithm robustness. This session provided a comprehensive overview of the current challenges and innovative strides in this rapidly evolving field.

The breakout sessions were instrumental in identifying strategies for integrating digital measures and AI to enhance preclinical to clinical translation. Participants addressed the challenges arising from the disconnect between preclinical and clinical assessments, emphasizing the need for input from clinical colleagues to enhance preclinical study designs. Additionally, discussions focused on the biological and clinical relevance of digital measures and AI/ML predictions, proposing strategies to enhance the correlation and translation between preclinical and clinical data. These collaborative sessions fostered a deeper understanding of the integration opportunities and laid the groundwork for a comprehensive framework to adopt novel preclinical digital measures.

Bridging the gap between preclinical and clinical digital measures is crucial for achieving successful drug development outcomes. Here are couple of examples how integration of these digital measures can positively impact drug development:

  • Enhanced Early Detection of Efficacy and Safety Signals - In preclinical studies, digital measures can reveal early signs of toxicity or therapeutic efficacy. By seamlessly integrating these preclinical digital measures with clinical data, we can better predict how a drug will perform in humans. This integration helps in identifying potential safety concerns or efficacy signals earlier in the development process, allowing for timely adjustments to study designs or dosing regimens, thereby reducing the risk of late-stage clinical trial failures.
  • Improved Patient Stratification and Personalized Medicine - Digital measures can be used to identify specific biological signatures associated with disease progression or treatment response. In preclinical models, AI algorithms can analyze these biomarkers to identify patterns that predict which subsets of patients are most likely to benefit from a new therapy. When these digital biomarkers are validated and integrated into clinical trials, they can be used to stratify patients more effectively, ensuring that the right patients receive the right treatments. This approach not only increases the chances of clinical trial success by targeting the appropriate patient populations but also paves the way for personalized medicine, where treatments are tailored to individual patients based on their unique digital biomarker profiles.

This workshop highlighted the transformative potential of integrating preclinical and clinical digital measures with AI technologies, inspiring innovative ideas and laying the groundwork for their practical application. The key recommendations and actions from the workshop are poised to advance the field, with follow-up initiatives, including a manuscript in a high-impact journal and a workshop with the FDA at White Oaks Campus, further solidifying the workshop's outcomes. The convergence of AI and digital measures holds the promise of enhancing translational predictability, improving data precision, and ultimately accelerating the drug development process.

Check out our events page to see other upcoming presentations.