Development and Validation of a Multiclass Model Defining Molecular Archetypes of Kidney Transplant Rejection: A Large Cohort Study of the Banff Human Organ Transplant Gene Expression Panel

Gene expression profiling from formalin-fixed paraffin-embedded (FFPE) renal allograft biopsies is a promising approach for feasibly providing a molecular diagnosis of rejection. However, large-scale studies evaluating the performance of models using NanoString platform data to define molecular archetypes of rejection are lacking. We tested a diverse retrospective cohort of over 1400 FFPE biopsy specimens, rescored according to Banff 2019 criteria and representing 10 of 11 United Network of Organ Sharing regions, using the Banff Human Organ Transplant panel from NanoString and developed a multiclass model from the gene expression data to assign relative probabilities of 4 molecular archetypes: No Rejection, Antibody-Mediated Rejection, T Cell–Mediated Rejection, and Mixed Rejection. Using Least Absolute Shrinkage and Selection Operator regularized regression with 10-fold cross-validation fitted to 1050 biopsies in the discovery cohort and technically validated on an additional 345 biopsies, our model achieved overall accuracy of 85% in the discovery cohort and 80% in the validation cohort, with ≥75% positive predictive value for each class, except for the Mixed Rejection class in the validation cohort (positive predictive value, 53%). This study represents the technical validation of the first model built from a large and diverse sample of diagnostic FFPE biopsy specimens to define and classify molecular archetypes of histologically defined diagnoses as derived from Banff Human Organ Transplant panel gene expression profiling data.