1. Distinct Non-Human Leukocyte Antigen Antibody Signatures Correlate with Endothelial Crossmatch Status in Lung and Renal Transplant Recipients.
- Author
-
Alhamdan, Fahd, Coppolino, Antonio, Sheikh, Adil, Miele, Anna, Lee, Stefi, Gasiewski, Allison, Brescia, Peter, Wood, Isabelle, Venkat, Arvin, Thaniyavarn, Tany, Jacob, Selvin, Keshk, Mohamed, Meadowcroft, Stacia, Banday, Mudassir M., Khan, Mohd Moin, Hayes Jr., Don, Chandrekar, Anil, Goldberg, Hilary, Guleria, Indira, and Sharma, Nirmal S.
- Subjects
- *
CYTOSKELETAL proteins , *LUNG transplantation , *ANGIOTENSIN II , *KIDNEY transplantation , *FEATURE selection - Abstract
Non-HLA antibodies against heterogeneous targets on endothelial cells have been associated with allograft injuries. The endothelial cell crossmatch (ECXM) is used in the detection of non-HLA antibodies but remains non-discriminatory for specific antibody identification. The primary objective of this study was to delineate the specific non-HLA antibody signatures associated with ECXM positivity and to determine the correlation of ECXM status and non-HLA antibody signatures on allograft health. Serum specimens from 25 lung transplant recipients (LTRs) and 13 renal transplant recipients (RTRs) were collected as part of clinical evaluation, and testing for angiotensin II receptor type 1 (AT1R) and donor-specific MHC class I chain-related gene A (MICA) antibodies and ECXM was performed. Remnant sera were tested for non-HLA antibodies using the LABScreen™ Autoantibody (LSAUT) Group 1, 2, and 3 kits (One Lambda, Inc., Los Angeles, CA, USA). In both cohorts, the concordance of AT1R and MICA together or individually with ECXM+ status was poor (<0.7), suggesting the presence of other unaccounted antibodies. Autoantibody profiling revealed three distinct clusters targeting fibrotic products, cytoskeletal proteins, and cell signaling molecules. A comparative analysis of ECXM+ and ECXM− specimens identified nine and five differentially expressed antibodies in the LTR and RTR cohorts, respectively. Employing machine learning techniques (variable importance, feature selection, ROC-AUC), we derived a five-antibody panel (TNFα, collagen V, CXCL11, GDNF, GAPDH) and a two-antibody panel (TNFα, CXCL9) that effectively discriminated between ECXM+ and ECXM− status in the LTR and RTR cohorts, respectively. Distinct antibody signatures were identified in LTR and RTR cohorts that correlated with ECXM+ status and were associated with allograft dysfunction. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF