Study purpose: Besides HLA antibody characterization and workup of biopsies (Bx) molecular assessment of gene expression may considerably improve ABMR diagnosis. The objective of the present study, which was performed within an RCT (BORTEJECT Trial, NCT01873157), was to determine late ABMR in a prevalent longterm kidney Tx population, laying focus on the association between serological, morphological and molecular data. Methods: We screened 741 adult kidney allograft recipients for the presence of DSA, identifying 15% DSA+ stable patients. A cross-sectional approach of performing protocol Bx (n=86) after identification of DSA was carried out. Microarrays were performed in 83 Bx and a previously published validated score for the molecular diagnosis of ABMR generated by the Edmonton reference Bx set (N=1212) was applied. Results: Of 83 Bx from DSA+ patients in which microarrays were performed, Banff'13 ABMR was diagnosed in 54%. Using a ≥0.2 cutoff for the molecular ABMR score, an AUC of 0.92 for the diagnosis of Banff'13 ABMR was computed. In a set of variables that might predict the ABMR score or a combination of molecular and histologic ABMR, IgG MFI revealed the highest importance in a multivariable random forest model. IgG MFI showed a crossvalidated peak accuracy of 0.71 (MFI 1,679) to predict a molecular ABMR score ≥0.2 and 0.70 (MFI 2,395) to combined positive molecular ABMR+histology. Conclusion: The ABMR score is tightly associated with ABMR histology in late DSA+ Bx. Searching for variables able to predict a positive molecular ABMR score, DSA MFI showed a moderate prediction. However, in a set of variables including proteinuria and kidney function, IgG MFI remained most important to diagnose molecular ABMR.

CITATION INFORMATION: Eskandary F, Reeve J, Yoo D, Regele H, Hidalgo L, Böhmig G, Halloran P. The Prediction of Molecular ABMR by Anti-HLA Donor-Specific Antibody Properties. Am J Transplant. 2016;16 (suppl 3).

« Back to 2016 American Transplant Congress