School of Public Health

Eliza Blanchette

Eplet mismatches and de novo donor-specific antibody development in pediatric kidney transplant recipients

Background

The presence of HLA de novo donor-specific antibodies (dnDSA) is associated, probably causally, with failure of kidney transplants. In a pediatric kidney transplant population, we sought to determine the extent to which eplet mismatches (MM) are associated with the development of dnDSA.

Methods

We performed a retrospective cohort analysis of pediatric kidney transplant recipients from 2008 to 2014 who underwent surveillance dnDSA testing and who had at least 3 years of follow up from the time of transplant. We used the National Marrow Donor Program's Haplostats platform to impute high-resolution HLA alleles from recipient and donor low resolution HLA alleles selecting the most commonly matched high-resolution typing. We then generated eplet mismatches using HLA matchmaker version 2.0. Eplet mismatch exposure was modeled both on the continuous scale and with identified thresholds. We utilized Cox proportional hazards regression models to predict risk of dnDSA based on eplet MM, adjusting all models for age, donor type, and sex.

Results

A total of 133 subjects (median age 13 years, 64% male) met inclusion criteria. Of these, 44% developed persistent dnDSA (at least 2 positive tests) at a median of 19 months post-transplant. The presence of Class II eplet MM was associated with a 3% increase in risk for dnDSA per each additional eplet MM (HR 1.03, 95% CI 1.02-1.05); children with >5 MM had approximately a 3.5 fold increased risk for dnDSA development (HR=3.61, 95% CI = 1.70-7.67). The presence of Class I MM was not associated with risk of development of dnDSA to any appreciable degree. The presence of DR and DQ MM was associated with a 6-7% increased risk of dnDSA (DR: HR 2.02, 95% CI: 1.12-3.63; DQ: HR 1.07, 95% CI 1.03-1.10); those with DR >2 MM and DQ >3 MM were each associated with increased risk of dnDSA development (DR: HR 1.06, 95% CI: 1.03-1.10; DQ: HR 1.92, 95% CI: 1.08-3.41).

Conclusion

The presence of Class II, but not Class I eplet MM, is predictive of increased risk of dnDSA development in pediatric transplant recipients.