Multiscale, mechanistic model of Rheumatoid Arthritis to enable decision making in late stage drug development

Area of Science:

• Immunology
• Computational Biology
• Pharmacometrics

Background:

• Rheumatoid Arthritis (RA) is a prevalent chronic autoimmune disease with limited treatment benefits for many patients.
• Novel therapies exist, but optimizing treatment response and trial design remains a challenge.
• Identifying patient subpopulations and enabling reverse translation from clinical trials are crucial for advancing RA therapy.

Purpose of the Study:

• To develop an open-source, mechanistic multi-scale model of Rheumatoid Arthritis (RA).
• To simulate patient responses to standard-of-care treatments and identify subpopulations.
• To enable community evaluation and adoption of mechanistic models in autoimmune disease drug development.

Main Methods:

• Constructed a multi-scale mechanistic model of RA incorporating immune cell and mediator interactions.
• Developed a Virtual Population (Vpop) simulating treatment-naive and Methotrexate (MTX)-inadequate responder subpopulations.
• Modeled clinical read-outs like American College of Rheumatology (ACR) score and Disease Activity Score (DAS28-CRP).

Main Results:

• The model accurately simulates responses to Methotrexate (MTX) and Adalimumab (ADA) in a virtual population.
• A specific subpopulation demonstrated appropriate response to Tocilizumab (TCZ) therapy.
• Validated the model by predicting TCZ response in Adalimumab non-responders.

Conclusions:

• The developed mechanistic model provides a valuable tool for understanding RA pathophysiology and treatment response.
• This open-source model facilitates improved clinical trial design and drug development for autoimmune diseases.
• Sharing the model's code and equations encourages wider adoption and evaluation within the scientific community.