Aiding Drug Repurposing Investigations with AI

Developing a drug is expensive and time-consuming. Drug repurposing offers a cost-effective alternative which can streamline parts of the development process, with the potential of expediting research timelines by 6-7 years.¹ Recent estimates indicate about one-third of approvals are for drug reuse, contributing to around 25% annual revenue for the pharmaceutical industry.² Moreover, the ability to leverage safety profiles and side effects of existing drugs from all biomedical literature is a competitive advantage in the race-to-market.

Data-driven strategies play an important role in drug repurposing, harnessing the power of big data to unveil new therapeutic applications for existing drugs.³ Notably, technological advancements in AI for the large-scale analysis of biomedical data are helping researchers identify novel drug-disease connections.⁴ Such insights can enhance the success of repurposing efforts by opening up new avenues for treatments in unexplored indications. Here, we show how Causaly’s AI can aid investigations into drug repurposing.

We investigated the potential of Exenatide, a glucagon-like peptide-1 (GLP-1) receptor agonist used to treat type II diabetes,⁵ for drug repurposing applications. Causaly identified  almost 300 diseases treated by Exenatide, the majority of which were nervous system disorders (94 disorders in total), as illustrated in Figure 1. Alzheimer’s and Parkinson’s Diseases were among the most studied diseases.

To identify new areas for indication expansion, we assessed diseases treated with Exenatide based on their novelty, aiming to understand the current focus of research relating to this drug. For example, a 2024 study explored the promising role of exendin-4 as a therapeutic agent for the treatment and management of Alzheimer’s and Parkison’s Diseases.⁶ Additional evidence for the treatment of pseudotumor cerebri⁷ and liver cancer,⁸  was also extracted from 2024 research.

To investigate further, we used Causaly to identify potential mediators of the drug-disease effect. Around 750 genes and proteins which may mediate the effect of Exenatide on Alzheimer’s Disease were unveiled; almost 580 were identified for Parkinson’s Disease. This can be visualized as a network view of results, as shown in Figure 2, which can be used to generate hypotheses.  For example, it can be hypothesized that microtubule-associated protein tau (MAPT) could be one of the proteins that mediate the drug effect on both neurological conditions.

Drug repurposing offers a cost-effective and efficient pathway to discovery new therapeutic uses for existing treatments. AI can advance this process by rapidly analyzing large-scale biomedical data and scientific texts to identify key drug-disease relationships and unveil applications in novel indications. Causaly’s human-centric AI removes the signal from the noise, extracting relevant insights and uncovering hidden connections with 100% source transparency, taking scientific exploration and hypothesis generation to a new level.

Want to find out more?