Novel Loci Associated with Acute Chest Syndrome in Sickle Cell Disease Patients

Sickle cell disease (SCD) results from a genetic mutation in the beta hemoglobin gene. Despite the same underlying causal mutation, patients exhibit highly heterogeneous clinical outcomes. Acute chest syndrome (ACS) is the leading cause of mortality among individuals with SCD, accounting for 25% of all deaths, and is the second most common reason for hospitalizations after vaso-occlusive pain. ACS is associated with extended hospital stays, higher risk of respiratory failure, and the potential for developing chronic lung disease after recurrent episodes. Many previous genome-wide association studies (GWAS) have failed to identify risk loci associated with ACS due to its multifactorial etiology and the absence of a well-defined, consensus phenotype.

To understand the underlying disease pathophysiology of ACS, we first performed a GWAS with 196 adult SCD patients from the Outcome Modifying Genes in SCD (OMG-SCD) cohort. Using a standardized definition of ACS, history of ACS was collected in 535 patients at five comprehensive sickle cell centers. We then selected the extreme phenotype, whereby patients who had never had any episode of ACS were categorized as controls (n=114), while patients who had had more than 5 episodes of ACS were categorized as cases (n=82). All patients included in the study had either HbSS or HbSβ⁰ genotype. Genotype data were generated by whole genome sequencing by Trans-Omics for Precision Medicine – TOPMed. We performed the GWAS using the R package SAIGE, incorporating treatment (hydroxyurea) status and 2 genomic principal components as covariates. Linkage-disequilibrium clumping was performed using PLINK 1.9 to identify independent single nucleotide polymorphisms (SNPs). We then performed functional analyses to understand the biological implications of the top SNPs. Using quantitative predictions of transcription factor binding changes due to sequence variants (QBiC), we identified whether associated SNPs resulted in alterations to transcription factor binding sites. We also performed SNP-association tests (using snpStats in R) to determine other clinical outcomes that may be associated with the ACS-associated SNPs.

175 SNPs with False Discovery Rate (FDR) adjusted q-values of less than 0.05 were identified. Some associated SNPs are in genes involved in cadherin expression & function as well as in growth factor regulation pathways. Our top ACS-associated SNP was rs7319269 on chromosome 13, reaching the genome-wide threshold of 5x10^-8 (q = 0.002). This intergenic SNP lies between SNORA107 and LINC00375, both of which give rise to noncoding RNAs, which as a class have a broad role in protein expression and cellular homeostasis. Interestingly, another of our top SNPs (rs1356744 on chr 2, q = 0.008) has been previously associated with severe acute respiratory syndrome in COVID-19, suggesting a common biological pathway in ACS and COVID-19. This SNP is also predicted to induce regulatory changes (unbound to bound state) in the NKX2 transcription factor family. Functionally, NKX2-1 may play a role in lung development and surfactant homeostasis, while NKX2-8 is overexpressed in some lung cancers. ACS-associated SNPs were also associated with other clinical outcomes experienced by SCD patients. For example, SNP rs16931960 on chr 12 (q = 0.018) is an intronic variant in WNK1, a gene that is also associated with the occurrence of gall stones (p = 0.00005). WNK1 is thought to be a key regulator of blood pressure and plays a role in endothelial permeability and barrier function, a process important to pneumonia pathophysiology. It has also been shown to selectively localize in polarized epithelia in gallbladder. In our dataset, occurrence of ACS is also independently associated with occurrence of gallstones (p = 0.01). This association may be due to the role of free Hb in both ACS and formation of bilirubin gallstones in SCD.

Our study highlights the importance of a well-defined phenotype with distinct cases and controls for successful identification of risk loci through GWAS. It also emphasizes the need to understand clinical outcomes in relation to each other in order to gain a holistic view of disease pathophysiology. Our goal is to decipher the intricate and interactive biological mechanisms and pathways involved in severe SCD, so that we can provide valuable insights into the molecular profiles specific to severe SCD and potentially improve targeted treatment approaches.