Phenotype and long-term outcome of the patients with histology of ABMR but without detectable donor-specific HLA antibodies

A. Senev, V. van Sandt, L. Daniëls, E. Lerut, D. Kuypers, M. Coemans, M.P. Emonds, M. Naesens

Thursday 15 march 2018

12:00 - 12:10h at Van Rijck/Ruys Zaal

Categories: Clinical, Session (parallel)

Parallel session: Parallel session 5: Clinical

Over the last two decades significant progress has been made in the diagnosis of antibody mediated rejection (ABMR), resulting in continuous adjustments of the Banff criteria. However, the Banff classification from 2015 defines ABMR as well as a separate category “suspicious for ABMR”, with an incomplete phenotype.So far, a direct comparison of how these two groups differ in terms of graft injury and clinical outcome is lacking.

The aim of this study was to investigate the histological lesions in allograft biopsies in relation to allograft survival in patients who met the first two Banff histopathology criteria for ABMR (antibody-negative ABMR), in patients with definite ABMR (with circulating DSA), and in patients without ABMR. This study included 947 single kidney transplant recipients transplanted in one centre between 2004 and 2014. All 4390 post-transplant renal biopsies were rescored according to the current 2015 Banff criteria. The HLA antibodies in the recipient sera were determined for donor specificity (DSA) at Single Antigen Bead resolution using Luminex technology.

A total of 210 patients (22.2%) met the first 2 histological criteria for ABMR during the follow-up period. Of those, 80 patients had DSA (DSA_posABMR group), while in 130 patients DSA were not detected (DSA_negABMR). The DSA_posABMR group had a higher frequency of females and of repeat transplants (p=0.03 and p<0.0001, respectively). Patients in the DSA_negABMR group were more likely to have received an older kidney (mean: 51.1±14 vs. 45.8±14.5 years; p=0.02). The DSA_negABMR group had a significantly better graft outcome than the ABMR group in 10-year death-censored graft survival analysis (73.6% vs. 55.7%; p=0.003 by log-rank analysis). Compared to controls (ABMR negative group), the Cox model, adjusted for donor age, showed almost a fivefold increased risk for transplant failure in the DSA_posABMR group (HR=4.7; p<0.0001), but not for the DSA_negABMR group (HR=1.7, p=0.086). The only difference found in the histology profile between the two groups was that the C4d positivity was more common in DSA_posABMR group (mean Banff grade 1.44±1.4 vs. 0.83±1.2; p=0.001).

According to our results, patients with histological lesions of ABMR but without detectable DSA represent a different phenotype with superior graft survival compared to those having DSA and fully developed ABMR by the Banff criteria.Further work is necessary to elucidate the pathophysiology of histology of ABMR in absence of DSA.