Exploring the Disease Pathophysiology of Rheumatoid Arthritis

Understanding the pathophysiology of a disease is pivotal in comprehending its cause and progression and facilitating the identification of novel targets for therapeutic intervention. However, piecing together disease mechanisms and molecular interactions is challenging, particularly given the increasing number of articles being published each year.¹

Data-driven strategies are essential in navigating this complexity, facilitating a deeper understanding of disease pathophysiology, which can be leveraged to develop more effective treatments. In this setting, AI emerges as a powerful tool for uncovering critical disease insights from the wealth of biomedical data. Here, we show how Causaly’s AI can be used to explore the disease biology of rheumatoid arthritis (RA).

Rheumatoid Arthritis (RA) is a chronic autoimmune disease impacting 1.3 million in the U.S. and requiring long-term management.² While drugs (such as DMARDs and biologics) exist, many patients do no respond adequately.³ In addition, side effects associated with long-term use can be significant. Developing drugs that can address this unmet need is therefore critical. Despite advancements, the exact causes and mechanisms of RA are not fully understood.⁴ Moreover, further research into the disease pathogenesis of RA could open up new avenues for targeted drug development.

Using Causaly, almost 4,000 genes and proteins (supported by 23,000+ documents) have been associated with RA. To delve into those with genetic evidence, the results were prioritized by GWAS data, refining the search to around 530 genes and proteins, around 10% of which have been reported in 2023-2024. From the dendrogram view of results, protein-coding gene ERBB3 was among those with compelling evidence.

To explore the potential for ERBB3 further, we utilized Causaly’s hypothesis generation tool to uncover ~150 proteins which may mediate the effect of ERBB3 on RA. This data can be visualized as a network view of results, as illustrated in Figure 1. For example, ERBB3 is involved in the activation of the PI3K pathway⁵; PI3K has also been implicated as a target for RA.⁶ Such insights can provide the foundation for hypothesis generation in drug discovery.

Targeting drugs at the main pathophysiological process is the key to success in drug discovery.⁷ But, identifying relevant genes, proteins, and pathways from the overload of biomedical data proves challenging. In this context, AI emerges as a powerful tool for rapidly extracting meaningful insights to better understand disease pathophysiology.