Incidence and prediction of early antibody-mediated rejection due to non-human leukocyte antigen-antibodies

Background. Antibody-mediated rejection (AMR) is responsible for a large proportion of early allograft losses. While preformed donor-specific human leukocyte antigen (HLA)-antibodies (HLA-DSA) are accountable for the majority of these episodes, non-HLA-DSA are also involved. However, data on the incidence of early AMR due to non-HLA-DSA are currently lacking. Methods. This study evaluated (i) the incidence of early AMR due to non-HLA-DSA-defined by exclusion of circulating HLA-DSA detected by flow beads- and (ii) the association with donor-specific major histocompatibility complex class I chain-related gene (MICA) -antibodies (MICA-DSA) and angiotensin-receptor antibodies. A retrospective cohort (n = 279) risk stratified by complement-dependent cytotoxicity crossmatches (CDC-YM era) and a prospective cohort (n = 154) risk stratified by virtual crossmatching using flow beads (virtual-XM era) were investigated. Results. In the CDC-XM era 25/279 patients (9%) developed early AMR, but only 3/154 patients (2%) in the virtual-XM era (P = 0.004).The incidence of early AMR due to HLA-DSA was significantly higher in the CDC-XM era than in virtual-XM era (18/279 patients [6.5%] vs. 0/154 patients [0%]; P = 0.0005). However, the incidence of early AMR presumably due to non-HLA-DSA remained unchanged in these two cohorts (7/279 patients [2.5%] vs. 3/154 patients [2%]; P = 1.0) consistent with a persisting gap in the ability to identify preformed DSA. Overall, 10/433 patients (2.3%) experienced early AMR presumably due to non-HLA-DSA. None of these 10 patients had angiotensin-receptor antibodies, at most 3/10 patients had MICA-DSA, while the antibodies remained unexplained in 7/10 cases. Conclusion. Early AMR due to non-HLA-DSA is a rare event, which is still difficult to predict by currently available assays.