Abstract 3228

Introduction:

Stroke is one of the most catastrophic acute complications of sickle cell anemia (SCA), occurring in 11% of patients before 20 years of age. A further 20 to 30% of children with SCA will develop less clinically overt cerebrovascular disease events such as transient ischemic attacks (TIA) and silent infarcts. There is a definite need for biomarkers that could determine the cause of these irreversible cerebrovascular events and which might predict children at greatest risk. Previous studies of sibling pairs have shown that there is a genetic component to cerebrovascular disease development but few genetic modifiers have been validated as having a substantial effect on risk of stroke. The aim of this study was to perform an unbiased whole genome search for genetic modifiers of stroke risk in SCA.

Methods:

Pediatric patients with SCA and documented primary stroke (n=177) were compared to a pediatric control non-stroke group with SCA (n=335). All control patients were over 5 years old and without previous clinical stroke prior to beginning any clinical treatment. Genome wide association studies (GWAS) were performed using genotype data obtained from Affymetrix SNP6.0 arrays. A pooled DNA approach was used to perform whole exome sequencing (WES) by Illumina next generation sequencing of pooled control (n=104) and pooled stroke (n=120) groups.

Results:

From the Affymetrix SNP6.0 GWAS data, 139 single nucleotide polymorphisms (SNP) were identified with stroke association. From the WES, 294 non-synonymous mutations were found to be significantly associated with stroke. In combination, 11 mutations identified by WES were located within 250kb of a SNP identified by GWAS (Table 1). These 11 mutations represent key areas of the genome that are targets for further in depth study. To next validate the genetic variants identified by WES with association with risk of stroke, 21 candidate mutations were genotyped in an independent cohort of control (n=231) and stroke (n=57) patients with SCA. One mutation in GOLGB1 (Y1212C) was corroborated as having significant association with lower risk of stroke (p=0.02).

Conclusion:

This mutation in GOLGB1 is predicted to effect the golgi associated function of the encoded protein and future studies will focus on how this functional mutation may protect against development of cerebrovascular disease in the context of SCA.

Table 1.

Variants identified by whole exome sequencing that are located within 250kb of a SNP marker identified by GWAS.

Gene SymbolC/somePosition (bp)Amino Acid ChangeWES p-valueDistance to GWAS SNPGWAS p-value
PON1 chr7 94937446 Q192R 9.0×10−4 1kb 4.3×10−5 
ENPP1 chr6 132172368 K173Q 7.7×10−5 3kb 7.1×10−5 
CYP4F2 chr19 16008388 W12G 4.1×10−4 44kb 6.0×10−5 
INPPL1 chr11 71948536 A1083G 8.6×10−7 76kb 8.8×10−5 
GOLGB1chr3 121415720 T1212C 5.6×10−4 127kb 9.3×10−5 
RSAD1 chr17 48557326 A119T 6.5×10−5 129kb 3.3×10−5 
PKD1L3chr16 71983772 S1176R 8.7×10−7 193kb 6.1×10−5 
LOC100289165 chr8 124658210 R128C 7.3×10−4 194kb 7.6×10−5 
ARGFXchr3 121305086 T196I 1.4×10−4 238kb 9.3×10−5 
KRT5 chr12 52908917 S528G 2.1×10−4 243kb 1.9×10−5 
HSD3B1chr1 120057246 T367N 1.8×10−7 248kb 2.7×10−5 
Gene SymbolC/somePosition (bp)Amino Acid ChangeWES p-valueDistance to GWAS SNPGWAS p-value
PON1 chr7 94937446 Q192R 9.0×10−4 1kb 4.3×10−5 
ENPP1 chr6 132172368 K173Q 7.7×10−5 3kb 7.1×10−5 
CYP4F2 chr19 16008388 W12G 4.1×10−4 44kb 6.0×10−5 
INPPL1 chr11 71948536 A1083G 8.6×10−7 76kb 8.8×10−5 
GOLGB1chr3 121415720 T1212C 5.6×10−4 127kb 9.3×10−5 
RSAD1 chr17 48557326 A119T 6.5×10−5 129kb 3.3×10−5 
PKD1L3chr16 71983772 S1176R 8.7×10−7 193kb 6.1×10−5 
LOC100289165 chr8 124658210 R128C 7.3×10−4 194kb 7.6×10−5 
ARGFXchr3 121305086 T196I 1.4×10−4 238kb 9.3×10−5 
KRT5 chr12 52908917 S528G 2.1×10−4 243kb 1.9×10−5 
HSD3B1chr1 120057246 T367N 1.8×10−7 248kb 2.7×10−5 

For all variants with significant association with stroke, the chromosomal position of each variant identified by WES (n=300, p<0.001) was compared to the location of all SNP markers (n=139, p<0.0001). We identified 11 variants by WES where there was at least one SNP marker within 250kb. These variants all represent excellent regions of the genome for future study. The four variants highlighted with a asterisk (*) are variants predicted by PolyPhen2 or SIFT to be deleterious.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution