Rare diseases are defined as conditions that affect fewer than 200,000 people in the US, or fewer than one in 2,000 people in the EU. There are over 7,000 known rare diseases, affecting more than 300 million people worldwide. However, only about 5% of rare diseases have approved treatments, and many of them are incurable and life-threatening. Therefore, finding new and effective therapies for rare diseases is a major unmet medical need and a global challenge.
Drug repurposing, also known as drug repositioning or drug reprofiling, is a strategy that aims to identify new indications for existing drugs, which have already been tested for safety and efficacy in humans. Drug repurposing can offer several advantages over traditional drug discovery, such as lower cost, shorter time, higher success rate, and broader applicability. Drug repurposing is especially suitable for rare diseases, as it can overcome some of the barriers and limitations that hinder the development of novel drugs, such as lack of funding, small patient populations, and complex pathophysiology.
Artificial intelligence (AI) is a branch of computer science that deals with creating machines or systems that can perform tasks that normally require human intelligence, such as reasoning, learning, and problem-solving. AI can play a crucial role in drug repurposing for rare diseases, as it can help to accelerate and improve the discovery of new therapies, by leveraging the vast and diverse data sources and methods available in the biomedical domain. AI can also help to address some of the challenges and gaps that exist in the current drug repurposing process, such as data quality, data integration, data analysis, and data interpretation.
In this blog post, we will explore some of the ways that AI can help in drug repurposing for rare diseases, and some of the examples and applications of AI in this field.
How AI can help in drug repurposing for rare diseases?
AI can help in drug repurposing for rare diseases in various ways, such as:
- Data mining and knowledge extraction: AI can help to mine and extract relevant and useful information and knowledge from various data sources, such as scientific literature, clinical trials, electronic health records, genomic data, proteomic data, metabolomic data, and phenotypic data. AI can use various techniques, such as natural language processing (NLP), machine learning (ML), and deep learning (DL), to analyze and interpret the data, and to generate new hypotheses and insights about the disease mechanisms, drug targets, drug effects, and drug-disease associations.
- Data integration and network analysis: AI can help to integrate and analyze the heterogeneous and complex data from different sources and levels, and to construct and explore the biological networks and pathways that underlie the rare diseases and the drug actions. AI can use various techniques, such as graph theory, network science, and systems biology, to model and simulate the interactions and dynamics of the biological entities and processes, and to identify the key nodes and edges that represent the potential drug targets and repurposing candidates.
- Data prediction and validation: AI can help to predict and validate the efficacy and safety of the repurposed drugs for rare diseases, and to optimize and personalize the drug dosages and regimens. AI can use various techniques, such as supervised learning, unsupervised learning, and reinforcement learning, to train and test the predictive models and algorithms, and to evaluate and compare the performance and outcomes of the repurposed drugs. AI can also use various techniques, such as computer-aided drug design, molecular docking, and pharmacokinetic/pharmacodynamic modeling, to design and optimize the drug structures and properties, and to estimate the drug absorption, distribution, metabolism, excretion, and toxicity.
Examples and applications of AI in drug repurposing for rare diseases
AI has been applied and demonstrated in several drug repurposing projects and initiatives for rare diseases, such as:
- Healx: Healx is a UK-based company that uses AI to discover and develop new treatments for rare diseases. Healx uses its AI platform, Healnet, which integrates and analyzes various data sources, such as scientific literature, clinical trials, patient registries, and genomic data, and uses ML and NLP techniques to generate and rank drug repurposing hypotheses. Healx also collaborates with patient groups, researchers, and clinicians to validate and advance the repurposed drugs to clinical trials.
- Findacure: Findacure is a UK-based charity that aims to empower and support the rare disease community. Findacure uses AI to identify and prioritize drug repurposing opportunities for rare diseases, by using NLP techniques to analyze the scientific literature and clinical trial data, and by using phenome-wide association studies (PheWAS) to identify the drug-disease associations. Findacure also works with academic and industry partners to test and translate the repurposed drugs to the clinic.
- Vanderbilt University Medical Center: Vanderbilt University Medical Center (VUMC) is a US-based academic medical center that conducts research and innovation in various biomedical fields. VUMC uses AI to discover and develop new therapies for rare diseases, by using ML and DL techniques to analyze various data sources, such as electronic health records, genomic data, and phenotypic data, and by using PheWAS and network analysis techniques to generate and validate drug repurposing hypotheses. VUMC also collaborates with patient groups, regulators, and pharmaceutical companies to advance the repurposed drugs to clinical trials.
Conclusion
Drug repurposing is a promising and attractive strategy for finding new and effective therapies for rare diseases, which are often neglected and underserved by the traditional drug discovery process. AI is a powerful and versatile technology that can enhance and facilitate the drug repurposing process, by leveraging the vast and diverse data sources and methods available in the biomedical domain, and by addressing some of the challenges and gaps that exist in the current drug repurposing process. AI can help in drug repurposing for rare diseases in various ways, such as data mining and knowledge extraction, data integration and network analysis, and data prediction and validation. AI has been applied and demonstrated in several drug repurposing projects and initiatives for rare diseases, such as Healx, Findacure, and VUMC.
We hope you enjoyed reading this blog post, and learned something new and interesting about the role of AI in drug repurposing for rare diseases.