AI-based Drug Repurposing Overview

Interest in drug repurposing has been increasing since the COVID-19 outbreak. Drug discovery is a time-consuming process that requires several stages, including target identification, lead identification, clinical studies, and approval. The process of bringing a drug to market can take 17 years, can cost $2 billion, and can fail at any stage in the clinical study. Drug repurposing, or drug repositioning, identifies novel therapeutic uses for already-approved drugs. This approach shortens the approval time, lowers the failure rate, and uses approved drug safety data and pharmacological profiles, thereby lowering development time and cost.

Drug repurposing has emerged as an appealing and successful strategy for finding novel therapeutic applications for already-approved medications. The shorter timeframe makes the approach attractive for pharmaceutical industries and patients. Almost 30% of repurposed medications eventually reach patients, which is a significant advance over the 10% success rate of conventional processes.

Drug repurposing has the following 4 applications: indication expansion, identification of new uses of drugs in different therapeutic areas, repurposing of failed or discontinued drugs, and combination therapies.

Technologies such as ML, DL, natural language processing (NLP), predictive AI, predictive modeling, and generative AI are revolutionizing drug repurposing by analyzing a large amount of data from different sources, such as scientific literature, claim data, electronic health records (EHRs), and bioinformatics data. These technologies can identify drug–protein interactions at the molecular level by analyzing millions of data points and identifying drugs companies’ use for different disease indications. They analyze EHRs and claim data to provide information on the drugs people use off-label. AI, therefore, helps establish connections between drug targets, disease mechanisms, and novel diseases that companies can target with established drugs. Clinical trials have already confirmed their safety, thus shortening the time these drugs take to reach patients. The process will be especially beneficial for orphan diseases with few or no other treatment options.

AI will speed up the drug repurposing process and uncover the additional potential therapeutic uses of existing drugs. While the process presents challenges, such as a lack of available data for older drugs and the necessity to conduct more studies to apply repurposed drugs for new disease indications, AI could significantly speed up the drug repurposing process, providing patients with novel therapeutic options.

Research Scope

• Drug repurposing offers numerous speed and cost benefits over traditional drug discovery.
• This report covers the different technologies, such as ML and deep learning (DL), that companies use as the architectural basis for AI-based drug repurposing.
• The report covers the applications of AI-based drug repurposing across different diseases, including rare and metabolic diseases.
• The report illustrates the landscape of the industry’s most common AI technologies.
• Biopharma and AI companies operating in this space focus on rare disease development.
• The report covers the industry’s main participants operating in the space.

AI-based Drug Repurposing Segmentation

Drug-centric Approach:

• Small Molecules (All main participants covered in this report are doing small molecules)

Disease-centric Approach

• Oncology
• Neurodegenerative diseases
• Infectious diseases
• Rare diseases
• Autoimmune diseases
• Metabolic diseases

AI-based Drug Repurposing Growth Drivers

1. Need for efficiency
2. Increased AI adoption
3. Need to address emerging threats
4. Accelerated pipeline for rare diseases

AI-based Drug Repurposing Growth Restraints

1. Limited data
2. Interpretability of AI models
3. Regulatory issues
4. High infrastructure costs