Pharmacogenomics of intravenous immunoglobulin response in Kawasaki disease

Abstract

Introduction: Kawasaki disease (KD) is a diffuse vasculitis in children. Response to high dose intravenous gamma globulin (IVIG), the primary treatment, varies according to genetic background. We sought to identify genetic loci, which associate with treatment response using whole genome sequencing (WGS).

Method: We performed WGS in 472 KD patients with 305 IVIG responders and 167 non-responders defined by AHA clinical criteria. We conducted logistic regression models to test additive genetic effect in the entire cohort and in four subgroups defined by ancestry information markers (Whites, African Americans, Asians, and Hispanics). We performed functional mapping and annotation using FUMA to examine genetic variants that are potentially involved IVIG non-response. Further, we conducted SNP-set [Sequence] Kernel Association Test (SKAT) for all rare and common variants.

Results: Of the 43,288,336 SNPs (23,660,970 in intergenic regions, 16,764,594 in introns and 556,814 in the exons) identified, the top ten hits associated with IVIG non-response were in FANK1, MAP2K3:KCNJ12, CA10, FRG1DP, CWH43 regions. When analyzed separately in ancestry-based racial subgroups, SNPs in several novel genes were associated. A total of 23 possible causal genes were pinpointed by positional and chromatin mapping. SKAT analysis demonstrated association in the entire MANIA2, EDN1, SFMBT2, and PPP2R5E genes and segments of CSMD2, LINC01317, HIVEPI, HSP90AB1, and TTLL11 genes.

Conclusions: This WGS study identified multiple predominantly novel understudied genes associated with IVIG response. These data can serve to inform regarding pathogenesis of KD, as well as lay ground work for developing treatment response predictors.

Keywords: IVIG refractory; Kawasaki disease; Pharmacaeconomics; ancestry; whole genome sequencing.

Figure 1

Manhattan plots of Whole Genome Sequencing (WGS) association analysis for IVIG refractory. Plots for Combined and Ancestry-specific Whites, Asians and Hispanics separately are shown.

Figure 2

Genotype distribution and OR (95% CI) of the allelic additive models for minor alleles for the top hits in the chromosomal regions for IVIG refractory (A) Protective Alleles (B) Risk Allele.

Figure 3

Functionally significant genes: (A) 3D chromatin interaction (Hi-C) and eQTL analysis for genomic risk locus in chromosome 17. The most outer layer displays the rsID of the top SNPs in each risk locus and other SNPs with P < 0.05 from Manhattan plot in the genomic risk loci. SNPs in genomic risk loci are color-coded as a function of their maximum r² to the one of the independent significant SNPs in the locus, as follows: red (r² > 0.8), orange (r² > 0.6), green (r² > 0.4) and blue (r² > 0.2). SNPs that are not in LD with any of the independent significant SNPs (with r² ≤ 0.2) are grey. The second layer is the chromosome ring and the Genomic risk loci are highlighted in blue. Only mapped genes by either chromatin interaction and/or eQTLs (conditional on user-defined parameters) are displayed. If the gene is mapped only by chromatin interactions or only by eQTLs, it is colored orange or green, respectively. When both map the gene, it is colored red. The third layer shows the genomic coordinate that aligns with the position of the SNPS and the genes (B) Tissue-specific expression of all 49 genes. Colors reflect average expression (log2 transformed) from highest (red) to lowest/absent (blue).