Harnessing AI to Predict Rheumatoid Arthritis
Fan Zhang, PhD, an assistant professor at the University of Colorado Department of Medicine’s Division of Rheumatology, is using artificial intelligence (AI) to combat rheumatoid arthritis (RA). Zhang, also affiliated with the Department of Biomedical Informatics on the CU Anschutz Medical Campus, recently received a grant from the Arthritis Foundation to further her work in predicting the onset of RA. A new paper details the latest steps in her research.
Zhang’s research focuses on developing computational machine learning methods, using algorithms to learn from data to predict RA and other autoimmune diseases. She draws on large-scale clinical and preclinical single-cell datasets. According to Zhang, this work could drive targeted interventions and prevent the disease’s progression. “We focus on enhancing disease prediction, ultimately enabling early disease prevention.”
The Challenge of Early Prediction
While there’s been significant research into treating RA after diagnosis, there are fewer studies on developing preventive strategies and identifying who is at risk of developing RA in the coming years. RA is a chronic autoimmune disease affecting about 18 million people worldwide, 1.5 million in the United States. It causes the body’s immune system to attack healthy tissue, resulting in inflammation, pain, and often, joint stiffness. Treatments exist to reduce inflammation, but there are no effective preventative treatments or cures.
The cause of RA is uncertain, though linked to specific genes triggered by external factors. Research indicates that many people with preclinical abnormalities experience immunological abnormalities years before symptoms appear. However, the symptom-free period can vary significantly, and some never develop the disease.
Zhang explains that her work functions as a “bridge” between data science and translational medicine. “Our research is very interdisciplinary,” Zhang says. “We have large-scale data from patients with autoimmune disease, so that gives us the opportunity to apply our AI tools to various cohorts of patients.”
Pinpointing Key Immunological Changes
Zhang’s team analyzes data on genetics, genomics, epigenetics, protein, and other factors from individual cells over long periods, known as single-cell multi-modal sequencing.
The study, “Deep immunophenotyping reveals circulating activated lymphocytes in individuals at risk for rheumatoid arthritis,” published March 17 in The Journal of Clinical Investigation, lays the groundwork for her next phase of research, supported by a new $150,000 Arthritis Foundation grant. Zhang’s lab will apply its computational tools to complex datasets from a large preclinical trial called StopRA, strengthening her collaboration with CU rheumatologist Kevin Deane, MD, PhD. The goal is to pinpoint immune system changes associated with the progression from preclinical RA to symptomatic RA.
Zhang and her colleagues analyzed RNA and protein expression in cells, comparing people at risk of RA to those with symptoms and healthy individuals. They found significant differences in certain immune cells, particularly the expansion of specific T cell subtypes, in the at-risk group. Zhang says that those cells “could be a promising marker” for RA onset and could lead to better prevention strategies. She cautions markers are not yet reliable and will require larger, more geographically diverse datasets.
Zhang is the corresponding author of the publication. Jun Inamo, MD, PhD, her lab’s postdoctoral fellow, is the first author. Deane and V. Michael Holers, MD, a rheumatology colleague, are among the co-senior authors.
Zhang, who has been at CU Anschutz for just over three years, says the Aurora campus is ideally suited for this collaborative research, stating, “with all the expertise and resources surrounding you. This is one of the leading places for autoimmune disease research for translational impact.”
Journal Reference
Inamo, J., et al. (2025). Deep immunophenotyping reveals circulating activated lymphocytes in individuals at risk for rheumatoid arthritis. Journal of Clinical Investigation. doi.org/10.1172/jci185217.
