Search

BACKGROUND Recent data have shown an increased risk for rejection, fibrosis progression, and death in liver transplantation (LT) recipients with preformed or de novo HLA donor-specific alloantibodies (DSA). However, the role of non-HLA autoantibodies and the interaction between HLA DSA and non-HLA autoantibodies remains uncharacterized. METHODS We analyzed 1269 primary LT recipients from 1 of 2000 to 4 of 2009 with known HLA DSA status for angiotensin II type-1 receptor and endothelin-1 type A receptor autoantibodies pre-LT, and year 1 post-LT. RESULTS Preformed non-HLA autoantibodies alone did not impact outcomes. In multivariable modeling, the combination of preformed non-HLA autoantibodies and HLA-DSA were associated with an increased risk for death (hazard ratio [HR], 1.66; P = 0.02) especially if the HLA DSA was of the IgG3 subclass (HR, 2.28; P = 0.01). A single de novo non-HLA autoantibody was associated with an increased risk for T cell-mediated rejection or antibody-mediated rejection (68% vs 41%, P = 0.01) and fibrosis progression (HR, 1.84; P = 0.02). Biopsies with de novo non-HLA autoantibodies revealed a new sinusoidal C4d staining pattern when compared with HLA DSA (71% vs 3%; P < 0.001). Liver sinusoidal endothelial cell activation and stellate cell activation was increased in patients with non-HLA autoantibodies in the location of C4d positivity. CONCLUSIONS A non-HLA autoantibody combined with a preformed HLA DSA is associated with an increased mortality risk. Isolated de novo anti-angiotensin II type-1 receptor and anti-endothelin-1 type A receptor autoantibodies are associated with an increased risk of rejection and fibrosis progression. The novel location of C4d staining in proximity to liver sinusoidal endothelial cell capillarization and stellate cell activation demonstrates allograft injury in proximity to non-HLA autoantibody binding.

A proposed chronic antibody-mediated rejection (AMR) score has recently predicted 50%10-year death-censored allograft loss in patients with donor-specific alloantibodies (DSA) mean florescence intensity (MFI) greater than 10 000 and requires confirmation in patients with lower MFI (1000-10 000).All patients who underwent liver transplantation from January 2000 to April 2009, had DSA (MFI ≥1000) in serum 10 to 14 months postliver transplantation, and had a protocolized liver biopsy were evaluated (n = 230). The previously proposed chronic AMR (cAMR) score was used to risk-stratify putative chronic AMR in DSA+ patients with MFI from 1000 to 10 000.The MFI distribution of DSA+ recipients were as follows: 66% had MFI 1000 to 4999, 14% had MFI 5000 to 10 000, and 20% had MFI greater than 10 000. The cAMR score distribution on 1-year protocol liver biopsy found that 41% had a score less than 13; 27% a score of 13 to 27.5, and 32% a score greater than 27.5. MFI correlated with 1-year cAMR category (13, 46% vs 21% and27.5, 29% vs 42% when MFI was 1000-10 000 vs MFI10 000; P = 0.047). In patients with a cAMR score less than 13, 10-year death-censored allograft survival was 96% to 100% regardless of MFI (P = NS). The risk of allograft loss increased in patients with a cAMR score greater than 13 (P = 0.004) in DSA+ patients with MFI 1000 to 10 000. DSA MFI greater than 10 000 versus MFI 1000 to 10 000 at 1 year was also more likely to persist at 5 years (95% vs 68%; P0.0001).Validation of the previously proposed cAMR score in a separate cohort predicts death-censored long-term allograft failure in DSA+ patients regardless of MFI, and higher MFI at 1 year predicts DSA persistence at 5 years.