Acute chest syndrome (ACS), a leading cause of morbidity and mortality in sickle cell anemia (HbSS), is thought to be multi-factorial in etiology. Prior small studies have implicated genes related to endothelial dysfunction in this process. We performed a candidate gene analysis on 1442 subjects with HbSS, with or without coincident α-thalassemia, enrolled in the Cooperative Study of Sickle Cell Disease, to identify genes associated with an increased risk of ACS. The etiology and clinical course of ACS for patients younger than 4 years is thought to be different from that in older ages. To account for inherent differences in ACS between adults and young children we divided the population conservatively into Group 1: pediatric (aged < 5 years) and Group 2: older children and adults (aged > 5 years). All subjects underwent a complete history, including medical record review, and were followed for approximately 5 years. ACS was defined as a new infiltrate on chest x-ray or pleuritic chest pain with a positive radionuclide lung scan. Controls had no history of ACS. Genotyping of haplotype tagging single nucleotide polymorphisms (htSNPs) of candidate genes was carried out on the Sequenom mass spectrometry SNP genotyping system or the ABI SNPlex system. We tested 847 SNPs in 354 candidate genes, including inflammatory mediators, modulators of nitric oxide (NO) biology, vasoregulatory molecules, cellular adhesion molecules and genes involved in the TGF-β signaling pathway. Time to first ACS event was analyzed using Cox proportional-hazards regression. Analyses were adjusted for age at entry, gender and baseline leukocyte count. Additional adjustments for reticulocyte and platelet count were made in Group 2. Correction for multiple testing was performed using the false discovery method based on the Benjamini and Hochberg algorithm. There were 170 ACS cases and 884 controls in Group 1, and 388 cases and 819 controls in Group 2. Two SNPs were significantly associated with ACS in both groups. In both populations, the most significant SNP was rs284157 located in TGFBR3 (transforming growth factor, beta receptor III) (Group 1: p=5.53 × 10–7, Group 2: p=6.40 × 10–38). The second most significantly associated SNP in both groups (rs736839) is in an unknown gene but is in linkage disequilibrium with the gene SMAD7, also of the TGF-β pathway (Group 1: p= 0.000792, Group 2: p=3.8 × 10–6). In Group 1 there were 2 additional SNPs significantly associated with ACS, both in PIK3CG (rs1526083, located at an intron/exon boundary, p= 0.0004 and rs12536620, p= 0.00007). This gene is a member of the PI3/PI4-kinase family and is involved in cell-cell adhesion. Six additional SNPs were associated with ACS in Group 2: rs2068991 in SMAD1 (p= 3.18 × 10–8), two SNPs in KLOTHO (KL), which plays a role in NO production (rs2149860, p=6.24 × 10-5 and rs656525, p=0.001138) and one each in the genes NRCAM, SMAD3 and STARD13, the latter of which is very near KL. In conclusion, 10 SNPs in 8 genes were significantly associated with development of ACS. Several of these genes are involved in the TGF-β signaling pathway, which has been implicated in other complications of sickle cell anemia. Confirmation of these results in other populations and further studies of these genes will likely lead to a greater understanding of the mechanisms involved in the development of ACS.

Author notes

Disclosure: No relevant conflicts of interest to declare.