Racial Disparities in Gene Expression Profiling but Not Donor-Derived Cell-Free DNA after Heart Transplant

Purpose Recipient race remains a powerful risk factor for mortality after heart transplantation (HT), with African Americans (AA) having worse survival than Caucasians. The aim of this analysis was to evaluate racial differences in gene expression profiling (GEP) and donor derived cell free DNA (dd-cfDNA) after HT using the Surveillance HeartCare Outcomes Registry (SHORE). Methods SHORE is an ongoing registry with >1200 patients enrolled at 60 HT centers in the United States. The registry includes HT recipients aged ≥15 years and ≥55 days post-transplant who had routine GEP and dd-cfDNA testing for rejection surveillance. Since patient sampling is from multiple centers, a linear mixed effects model, an extension of the linear model, was used to accommodate both fixed and random effects in the data. Results Among 956 recipients included in the analysis, 198 (20.7%) were AA, median age 57.0 years [0, 74.0] and 73.9% male. Overall, median GEP values demonstrated a borderline difference (p=0.07, Kolmogorov Smirnov test) for AA versus non-AA patients (Figure). This is evident by the cumulative distribution function, where a slightly higher rise in slope is observed for the non-AA cohort. Using linear mixed effects models, there was a statistically significant difference in GEP score when grouped by months post-transplant, demonstrating an early difference between AA and non-AA patients between months 2-5, but not subsequently. For dd-cfDNA analysis, median values were similar between AA (0.12%; Range 0.1-3.9%) and non-AA (0.12%; 0.1-3.5%) patients within the first year post-transplant (p=0.87, 2-sided K-S test) (Figure). Conclusion AA race is associated with higher GEP scores early post-transplant. This may be related to differential expression of the MARCH8 and FLT3 genes following HT, as previously described. Further analysis of specific GEP clusters and covariates across racial groups may shed light on these observed differences. In contrast, dd-cfDNA values do not appear to be influenced by race.