Abstract Rationale Critically ill pediatric patients exhibit variability in their inflammatory response to red blood cell (RBC) transfusion. Understanding host genetic contributors to this heterogeneity may inform more personalized transfusion practices and interventions. Methods We conducted a genome-wide association study (GWAS) in a multicenter cohort of critically ill children (n = 151) who received RBC transfusions. A panel of 249 inflammatory biomarkers was measured pre and within 24 hours post-transfusion. Biomarker changes were calculated as the log2 fold change between pre- and post-transfusion values. A composite inflammatory response phenotype based on pre and post transfusion change was derived via principal component analysis (PCA) to reduce dimensionality and produce phenotype principal components (PCs) that summarize major trends in the biomarker change data. Whole genome sequencing was performed using Illumina NovaSeq, and post-processing was conducted with Illumina’s BaseSpace DRAGEN pipeline. Filtering of biallelic single nucleotide polymorphisms (SNPs) was done with Hardy-Weinberg equilibrium threshold (1x10^-6), minor allele frequency (MAF = 0. 05), and linkage disequilibrium (LD = 0. 2), leaving 3. 1 million SNPS. SNPs were annotated to their nearest genes using the hg38 reference gene database. Association testing was conducted using linear regression, adjusting for ancestry and clinical covariates with the top phenotype PC (PC1) which explain 20. 9% of total variance and captures the overall magnitude and direction of post-transfusion inflammatory responses. Similar analyses were repeated with the top 4 phenotype PCs 1-4, which explains 47% of total variance, as the outcome. Results Multiple loci demonstrated a clear genotype-phenotype relationship, with increasing alternate allele dosage associated with greater deviation in inflammatory biomarker scores. Twenty-nine SNPs exceeded the genome-wide Bonferroni significance threshold (p 5 × 10−8). Among these, six loci with minor homozygous count ≥5 across genotype groups are depicted below See Figure. Notably, GRM3 (PC1 p = 2. 8e-10, PCs 1-4 p = 1. 5e-10) and SRGAP2C (PC1 p = 1. 7e-9, PCs 1-4 p = 3. 4e-9) were significant in both GWAS analyses with PC1 and combined PCs 1-4. Conclusions This study identifies host genomic variants associated with the early inflammatory response to RBC transfusion in critically ill children. The implicated genes suggest mechanistic links to pathways regulating cellular signaling, protein trafficking, metabolic stress response, chromatin organization, and cytokine-mediated repair. Specifically, GRM3 is known to play a role in reducing the expression of pro-inflammatory markers IL6 and TNFα. These results highlight the potential of integrating genomic profiling into transfusion decision-making, offering a step toward precision medicine in pediatric critical care. This abstract is funded by: RO1HD092472
Abiram et al. (Fri,) studied this question.