Emerging Approaches in Data-Driven Drug Discovery for Rare Diseases.

Abstract

Rare diseases present unique challenges in drug discovery and development, primarily due to small patient populations, limited clinical data, and significant variability in disease mechanisms. The primary objective of this review is to examine the integration of pharmacokinetics (PK) and drug metabolism data into data-driven drug discovery approaches, particularly in the context of rare diseases. By incorporating advanced computational techniques such as Machine Learning (ML) and Artificial Intelligence (AI), researchers can better predict PK parameters, optimize drug candidates, and identify personalized therapeutic strategies. AI integration with genomic and proteomic data reveals previously unidentifiable pathways, fostering collaboration among researchers, clinicians, and pharmaceutical companies. This interdisciplinary approach reduces development timelines and costs while enhancing the precision and effectiveness of therapies for patients with rare diseases. This review highlights the critical role of absorption, distribution, metabolism, and excretion (ADME) in understanding drug behavior in genetically diverse populations, thereby enabling the development of tailored treatments for patients with rare diseases. Additionally, it evaluates the opportunities and limitations of integrating PK/PD (pharmacodynamics) models with multi-omics data to improve drug discovery efficiency. Key examples of enzyme-drug interactions, metabolic pathway analysis, and AIbased PK simulations are discussed to illustrate advancements in predictive accuracy and drug safety. This review concludes by emphasizing the transformative potential of integrating PK and metabolism studies into the broader framework of data-driven drug discovery, ultimately accelerating therapeutic innovation and addressing unmet medical needs in rare diseases.