Mutations and Polymorphisms Influencing Hemolysis in Hemoglobin Genes and Risk of Pulmonary Hypertension in Sickle Cell Disease: Effect of Hemoglobin SC.

Abstract

Pulmonary hypertension (PH) is an increasingly recognized phenotypic manifestation of sickle cell disease (SCD) and a leading cause of death among adults with SCD. Chronic intravascular hemolysis is a central pathologic event with release of hemoglobin into plasma and leads to nitric oxide consumption. A resulting NO deficiency state is hypothesized to produce vasoconstriction within the pulmonary vasculature, ultimately becoming manifest as symptomatic PH. We have hypothesized that genetic factors may influence the risk for developing PH in SCD. As an initial step in identifying PH susceptibility loci, the allelic spectrum of mutations and polymorphisms in the hemoglobin genes have been examined with respect to PH in adults with SCD. Echocardiography was used to phenotype 261 unrelated subjects for PH, where cases were prospectively defined by a triscuspid regurgitant jet velocity > 2.5 m/s. The entire HBB gene was sequenced, and genotypes for SNPs flanking the β and α-globin loci were determined. HBB sequence analysis identified 66 (25.3%) compound heterozygotes for 1 of 13 different mutations (16.1% Hb SC or 9.2% β+ thalassemia, β0 thalassemia, or another hemoglobinopathy). A significant difference in genotype distributions between PH cases and cohort controls (3 x 2 table, P=0.026) suggested that individual mutations within the locus may be associated with PH. When analyzed individually, the prevalence of HbSC between cases (7.4%) and controls (21.1%) suggested that the HbC allele is associated with protection from PH (Odds Ratio=0.35, 95% CI 0.15–0.78, P=0.009). HbSS is only a mild risk factor (OR=2.01, 95% CI 1.07–3.78, P=0.026). In contrast, there was no association between 5′ or 3′ β globin locus haplotypes defined by combinations of 16 common SNPs and PH in patients with either SCD or HbSS. When the α-globin locus was examined using 3 haplotype tagged SNPs flanking the HBA2 and HBA1 genes, no PH associations were observed. An analysis of α-thalassemia mutations has not yet been completed, thus no definitive conclusions can be made for the influence of the α-globin locus on PH. Finally, the HbSC association was explored further by comparing laboratory markers of hemolysis and survival data within the HbSC group. Between SC PH cases and controls, anemia was the only significant characteristic of PH (P=0.0027) without other labs suggestive of excessive hemolysis. In contrast, other clinical findings observed in association with SCD PH, including hepatic congestion, iron overload and renal insufficiency, were present in HbSC PH group. While an overall association with protection from PH was observed for HbSC, a preliminary 40 month survival analysis suggests that PH in SC patients remains a significant marker associated with a risk for death (Hazard Ratio=7.89, P=0.007) which is comparable to overall SCD survival with PH. We conclude that the HbSC genotype is associated with protection from the development of PH in SCD, possibly due a smaller relative contribution of intravascular hemolysis. However, this diagnosis remains a strong predictor for death in SCD independent of HBB genotype. Further study of large populations of HbSC patients are warranted for defining optimal PH therapies and may be useful for elucidating the genetic basis for the non-hemolytic mechanisms contributing to secondary PH in SCD.