*Purpose: Acute rejection (AR) in kidney allograft recipients is a significant risk factor for reduced graft survival. Identifying individuals at risk for AR could facilitate more intense management and monitoring to reduce AR incidence. In a previous study, Steers and coworkers reported (NEJM 380:1918, 2019) that a kidney allograft recipient with the GG genotype (genotype frequency=0.15) of the SNP rs893403, who received an allograft with at least one copy of the A allele, were at significantly greater risk for AR (hazard ratio=1.63). The G allele, which is in linkage disequilibrium with a 1.5-kb deletion 3’ of the LIMS1 gene, results in a reduction of LIMS1 gene expression. For GG recipients, receiving a donor kidney with at least one copy of the LIMS1 expressing allele increases the risk of allosensitization against the LIMS1 protein (termed genomic collision).

*Methods: We attempted to validate this reported association of the rs893403 variant with first time biopsy proven acute T cell mediated rejection using data from two cohort studies, DeKAF and GEN03, of kidney allograft recipients (n=2,982) and donor recipient pairs (n=1,115). We tested both a recessive model (GG vs non-GG genotypes) in recipients and the genomic collision model in recipient-donor pairs, as done according to Steers et al. The cohorts were analyzed separately along with covariates including recipient age at transplantation, gender, the top principal component derived from GWAS data, donor status (living or deceased), PRA (positive or negative), prior transplant and the number of HLA mismatches. Our analyses was restricted to recipients with European ancestry.

*Results: The power analysis a hazard ratio of 1.6 for the DeKAF cohort was 0.93 for the recessive model and 0.88 for the collision model. In all cases the rs893403 SNP in LIMS1 was not statistically associated with increased risk of AR for either model (p>0.1).

*Conclusions: The HLA loci are not the only loci that can have mismatches in the amino acid sequence in proteins that can elicit an immune response. Two unrelated individuals can have thousands of coding variants creating mismatches between recipients and donors, resulting in an immune response of the recipient directed to donor tissue. We think that the use of a single variant, resulting in an amino acid difference for a single protein, is not enough to capture the potential total antigen mismatch between donor and recipient. A genome-wide analysis of protein mismatches, including HLA, creating a mismatch profile, will be necessary to create a continuum of risk for AR depending on the number of genetic mismatches between donor and recipient. We suggest an expanded analysis of this association prior to any implementation into clinical practice.

« Back to 2020 American Transplant Congress