Correlation of Longitudinal Gene-Expression Profiling (GEP) Score to Cytomegalovirus (CMV) Infection: Results from the Outcomes Allomap® Registry

Abstract

Cytomegalovirus (CMV) infection has been implicated in the pathogenesis of allograft rejection, which is reflected by high gene-expression profiling (GEP) scores. The effect of CMV infection on serial GEP scores in the absence of acute cellular rejection (ACR) is unknown. Data from 14,985 samples collected from 2288 heart transplant (HT) recipients enrolled in Outcomes AlloMap® Registry (OAR) were analyzed. The empirical distribution of GEP scores at the visit closest to CMV infection were compared to GEP scores obtained at other visits via Kolmogorov-Smirnov tests. AlloSure measurements from those who underwent concurrent dd-cfDNA testing were also reviewed. 311 samples were collected from 213 patients with CMV infection at the time of GEP testing. The mean time to CMV infection from transplant was 346 days. Median GEP score at the time of diagnosis of CMV infection was significantly higher when compared to GEP score obtained at other visits (34 v 30, p<; 0.0001) (Figure 1). For patients with samples assessed via AlloMap and AlloSure, there was no significant difference in the distribution of AlloSure measurements at the time of CMV infection (Figure 2). CMV infections in HT recipients are associated with an increase in GEP score independent of associated ACR. The likely mechanism is immune activation/modulation of one or more of the 11 genes in the GEP signature. Hence, CMV status should be considered when clinically interpreting high GEP scores. AlloSure testing in conjunction with AlloMap demonstrates AlloSure is invariant to CMV infection.