We sought to determine the long-term outcome of renal allograft recipients that developed de novo DSA (dnDSA) in the setting of stable allograft function.
Methods: A retrospective single-center review of 150 consecutive renal allograft recipients transplanted between January 2008 and May 2009.  DnDSA was performed at 3, 6, 9, 12, and every 6 months thereafter. DnDSA were defined as HLA-A, B, Cw, DR, DQ, or DP antibodies directed against the donor, that were not present pre-transplant. All dnDSA had a median florescence index of greater than or equal to 2000 and were confirmed by repeated testing. Measurement of C1q binding was not available. Thirteen recipients were excluded because of pre-transplant DSA (n=11) or initial non-function (n=2).
Results: The study cohort included 137 recipients, with 58% male, mean age 49 ± 13 years, 86% primary transplants and 44% living donors. Median graft survival was 76 (range 6-110) months. There were 115 recipients with no DSA (Group 1), 17 recipients with dnDSA and stable graft function (Group 2), and 5 recipients with dnDSA associated with an acute rejection (Group 3). Median time to first dnDSA was 26 (range 6-96) months. Death censured graft survival was compared between groups 1 and 2, as shown below.

All 5 recipients of Group 3 lost their grafts at a median of 38 (range 37-49) months post-transplant. Mean serum creatinines of Groups 1 and 2 were 1.2±0.4 vs. 1.4±0.6 at 3 years, 1.2±0.4 vs. 1.3±0.5 at 5 years, and 1.3±0.8 vs.1.1±0.4 mg/dl at 7 years post-transplantation (p=ns).
Conclusion: In this cohort, development of dnDSA in a recipient with stable graft function had no detrimental impact on long-term graft survival, whereas dnDSA associated with an acute rejection universally resulted in graft loss.