F.O.R.W.A.R.D: A Data-Driven Framework for Network-Based Target Prioritization in Drug Discovery

Area of Science:

• Biomedical Informatics
• Computational Biology
• Drug Discovery

Background:

• Target-based drug development is expensive and imprecise, with AI offering potential improvements.
• Inflammatory Bowel Diseases (IBD) present a complex therapeutic challenge due to multifactorial origins.

Purpose of the Study:

• To introduce and validate F.O.R.W.A.R.D. (Framework for Outcome-based Research and Drug Development), a network-based target prioritization method.
• To assess F.O.R.W.A.R.D.'s utility in predicting drug efficacy for Inflammatory Bowel Diseases.

Main Methods:

• F.O.R.W.A.R.D. utilizes real-world outcomes and a machine-learning classifier trained on transcriptomic data from clinical trials.
• It defines a molecular signature of remission and integrates network connectivity to predict drug-target-gene interactions.
• The approach was benchmarked against 210 clinical trials involving 52 targets.

Main Results:

• F.O.R.W.A.R.D. demonstrated a perfect predictive accuracy of 100% across diverse targets, mechanisms, and trial designs.
• In-silico phase '0' trials indicated potential for informing trial design and re-evaluating failed drug candidates.

Conclusions:

• F.O.R.W.A.R.D. offers a powerful, data-driven approach to enhance drug discovery and development, improving precision and reducing costs.
• The framework's adaptability to other therapeutic areas and its potential to guide clinical decision-making promise to transform R&D.