An Elevated Tricuspid Regurgitant Jet Velocity in Sickle Cell Disease Is Associated with Polymorphisms in Genes Impacting Innate Immunity

Abstract 514

An elevated tricuspid regurgitant jet velocity (TRV) by echocardiography, associated with increased mortality risk, may be reflective of pulmonary and systemic vasculopathy in sickle cell disease (SCD). The epidemiologic associations between an elevated TRV and other sub-phenotypes of SCD, such as leg ulcers, priapism, proteinuria and increased rate of hemolysis, suggest a common pathogenic link across vascular beds. We were interested in identifying genetic modifiers of SCD and hypothesized that an elevated TRV is, in part, genetically determined. Previous candidate gene studies identified single nucleotide polymorphisms (SNPs) in activin A receptor, type II-like 1 (ACVRL1), bone morphogenetic receptor 2 (BMPR2) and bone morphogenetic protein 6 (BMP6) associated with an elevated TRV suggesting a role for the TGF-b pathway in pathogenesis. We performed a genome-wide association study (GWAS) to identify novel genes and pathways that might be associated with an elevated TRV. We first used Illumina 610K arrays in a GWAS on 340 sickle cell disease patients recruited as part of the observational arm of the Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy (Walk-PHaSST) clinical trial (discovery set) and then validated these results using a replication set consisting of 56 patients enrolled in the studies of TRV and SCD at Duke and Boston Universities. All subjects were 17 years or older and had the HbS only phenotype. We analyzed the data using TRV as a continuous variable and we adjusted for potential confounding effects of age and gender. In addition, as there were 9 different clinical centers in the discovery set, we used center-adjusted TRV values for the analysis. We identified 3 SNPs in CUB and sushi multiple domains 1 (CSMD1) (rs12674750, p=9.5 × 10⁻⁶, rs7008391, p=1.8×10⁻⁵, rs4433172, p=4.1×10⁻⁵) and 1 in sorting nexin 31 (SNX31) (rs1609, p=8.2×10⁻⁵), also on chromosome 8q, which were also identified in our replication set (p=0.04 for CSMD1 SNPs and 0.01 for SNX31). CSMD1 is a very large gene of more than 2 Mb and the 610 array includes 1710 SNPs in this gene. However, the probability that we observe 3 SNPs replicated in this gene in a sample size of 56 by chance is only 0.0005. Additionally, 2 SNPs in the mediator of innate immunity, lymphocyte antigen 86 (LY86) (rs3827783, p=3.89×10⁻⁶, rs9502483, p=1.02×10⁻⁵), were identified in our discovery set. We evaluated the top 5 SNPs originally identified in the candidate gene studies and found that 4 were not present on the 610K arrays. The remaining one had a p-value of 0.17 but is in linkage dysequilibrium with another SNP in BMP6 (rs7450930), a rare variant, with a p-value of 2.5 × 10⁻⁵. CSMD1 inhibits the classical pathway of complement activation and complement-mediated hemolysis in rats. This is of interest in SCD because of recent data linking cell-free heme and immune modulation. These findings suggest that sickle cell disease patients with an elevated TRV have altered innate immunity and that sickle vasculopathy may be associated with dysregulation of inflammation.