Novel machine learning identifies distinct antibody signatures in kidney transplant rejection, suggesting new risk stratification tools.
Key Points
Antibody-mediated rejection is a leading cause of kidney transplant failure, affecting over 50% of patients with donor-specific alloantibodies.
High-resolution multiplexed technology provides the most comprehensive profile of alloantibodies to date, identifying crucial signatures for rejection.
A novel machine learning algorithm uncovers significant roles for IgM and glycosylation patterns in early and late transplant rejection.
Developed risk score based on antibody signatures predicts late rejection with high sensitivity and specificity, enhancing diagnosis for clinicians.