A SHARMA1,2, A TAVERNITI1, JR LEWIS1-4, N GRAF5, WH LIM3,4, A DURKAN1, SI ALEXANDER1, J CRAIG1,2, G WONG1,2.
1Centre for Kidney Research, Children’s Hospital at Westmead, Westmead, NSW; 2School of Public Health, Sydney Medical School, The University of Sydney, Sydney, NSW; 3University of Western Australia School of Medicine and Pharmacology, University of Western Australia, Perth, WA; 4Department of Renal Medicine, Sir Charles Gairdner Hospital, Perth, WA; 5Department of Histopathology, Children’s Hospital at Westmead, Westmead, NSW
Aim: To determine the association between eplet HLA-mismatches at transplantation, development of de novo DSA (dnDSA) and antibody mediated rejection (AMR) in paediatric kidney transplant recipients.
Background: Whilst associations between eplet mismatches and dnDSA production, as well as dnDSA production and poor graft outcomes have been demonstrated, a direct association between eplet mismatches and poor graft outcomes is yet to be investigated.
Methods: Adjusted logistic regression analyses were conducted to determine associations between class specific eplet HLA mismatches, dnDSA production and AMR.
Results: Fifty-seven (41 living, 16 deceased donors) paediatric first transplant recipients between 2005-2016 at The Children’s Hospital Westmead were included. The mean (±SD) age at transplantation was 9±5 years and mean follow up was 4±2 years. The mean Class I and class II eplet mismatches were 13±7 and 23±16. Nineteen patients developed dnDSA [Class I (n=3), Class II (n=7), and both (n=9)]. An increased risk of dnDSA production was observed with increasing Class I (per 10 increase in eplet mismatches, adjusted odds ratio (aOR) 4.1 95%CI 1.3-13.2, P=0.02), and Class II eplet mismatches (aOR 1.7 95%CI 1.0-2.8, P=0.04). Presence of Class I or II dnDSA were associated with increased risk of AMR (aOR 29.8 95%CI 3.5-255.0, P<0.01 and aOR 9.8 95%CI 1.5-63.8, P<0.01 respectively). There was no direct association between class I or II eplet mismatches and AMR (aOR 2.1 95%CI 0.6-7.3, P=0.23 and aOR 1.4 95%CI 0.7-2.5, P=0.31 respectively).
Conclusions: We did not observe any direct association between increasing number of eplet mismatches and AMR despite demonstrating associations between eplet mismatches and dnDSA and dnDSA and AMR. Further work should examine whether inclusion of eplet matching may better predict long-term graft outcomes.