Background: Immune checkpoint inhibitors (ICIs) targeting PD-1/PD-L1 increasingly cause rheumatic immune-related adverse events (irAEs) — inflammatory arthritis, myositis, vasculitis — in 5–10% of oncology patients. Current prediction relies on post-symptom clinical recognition, delaying rheumatology referral by weeks.
Hypothesis: Serial measurement of the serum soluble PD-1 (sPD-1) to soluble PD-L1 (sPD-L1) ratio, combined with CXCL9 and IL-2 receptor alpha (sIL-2Rα) trajectory slopes, will identify patients who develop rheumatic irAEs 4–10 weeks before musculoskeletal symptom onset with >80% sensitivity and >75% specificity.
Mechanistic Rationale: ICI-induced rheumatic irAEs arise from breakdown of peripheral tolerance in predisposed joints and muscles. sPD-1 accumulates as activated T cells shed their checkpoint receptors, while sPD-L1 reflects tumor and tissue counter-regulatory responses. A rising sPD-1/sPD-L1 ratio signals net checkpoint escape. CXCL9 marks interferon-γ-driven tissue inflammation (particularly synovial and muscle), while sIL-2Rα reflects T-cell activation amplitude. The combined kinetics should capture the immunological inflection point preceding clinical rheumatic irAE manifestation.
Testable Predictions:
- Patients developing rheumatic irAEs will show sPD-1/sPD-L1 ratio increase >2-fold from baseline within 4–10 weeks before symptom onset, versus stable ratios in irAE-free patients
- CXCL9 trajectory slope >15 pg/mL/week will distinguish musculoskeletal-targeted irAEs from dermatologic or hepatic irAEs
- A composite score (sPD-





























