Sickle cell disease is the most common monogenic genetic disease in the United States. Mortality is higher in those with elevated pulmonary pressures by echocardiography or other markers of disease severity including frequent pain, iron overload and renal insufficiency. We have enrolled 795 subjects and 125 controls in a study of the prevalence and prognosis of subjects with sickle cell disease and suspected pulmonary hypertension. This represents approximately a 1% sample of patients with sickle cell disease living in the United States. All subjects are screened with transthoracic echocardiograms and the tricuspid regurgitant jet velocity (TRV) used to estimate the pulmonary artery systolic pressure. Additional phenotypic information related to the manifestations of sickle cell disease is also collected at evaluation. Suspected pulmonary hypertension was prospectively defined by a TRV >= 2.5 m/sec and severe pulmonary hypertension defined by a TRV >= 3.0 m/sec. Subjects have been followed for a mean of 6 years and censored at time of death or loss to follow-up. There were no new subjects enrolled at Howard University or the NIH during this year. Total enrolled at Howard University is estimated at 131 and the total enrolled at NIH is 789. Total enrollment for all sites is 920. We are conducting 2, 4, 6, 8, 10 and 12 year follow-up visits for comprehensive data collection. We also completed a comprehensive survival update on all subjects, the first since 2010. All subjects have corresponding biospecimens collected for genomic investigations. Our goal is to recruit at least 1000 subjects with sickle cell disease for initial exploratory genetic studies. Variants in the GCH1 gene, which have previously been identified as modulators of pain in other diseases, are associated with increased risk of severe vaso-occlusive pain crises in a case control study. We also replicated these findings in an independent cohort. GCH1 is the rate limiting step for the synthesis of BH4 (tetrahydrobiopterin) which is an essential co-factor for the synthesis of nitric oxide, catecholamines, serotonin, and phenylalanine metabolism. In vitro analysis of whole blood samples from subjects with sickle cell disease shows that this GCH1 haplotype is associated with higher BH4 levels. This result indicates that this is a unique African haplotype associated with increased BH4 and increased pain. This haplotype is also associated with altered vascular function during forearm blood flow studies in adults with sickle cell disease. Surprisingly, this haplotype is not associated with pulmonary hypertension in SCD despite the biological suggestion that the African haplotype might promote nitric oxide production. Variants in the MYH9 gene are associated with decreased glomerular filtration rate in SCD. Genetic markers in the region of MYH9 on chromosome 22 have been identified as major risk factors for focal segmental glomerular sclerosis (FSGS) in African American populations. This candidate gene association study extends this work to a disease specific patient population that is at high risk for renal insufficiency. At the same time, we are also using gene wide admixture markers to perform a genome wide association study for this phenotype by admixture mapping. These genome wide data will also allow for assessment of both local and global ancestry relative to the MYH9 locus. A genome wide association study several phenotypes in sickle cell anemia has been performed using ancestry informative markers that were identified from the HapMap project. We have designed a genome wide panel of >1500 ancestry informative markers and have completed genotyping in more than 1700 subjects with sickle cell disease and 2 HapMap study populations. Analysis of population structure using this panel demonstrates complex population sub-structure of SCD in the United States, where this admixed population is derived from 3 ancestral populations including European, African and Amerindian/Asian. Nearly 20% of SCD subjects in the eastern US have essentially no European or Amerindian/Asian ancestry, suggesting very little admixture. The western US SCD population has a lower proportion of overall African ancestry with more individuals with Amerindian/Asian ancestry. These results suggest that genetic studies of sickle cell disease may require additional statistical adjustments when combining cohorts. Importantly, the statistical power form genome wide studies is more limited than anticipated based upon these results, as non-admixed study subjects violate the assumption of admixture analysis. For genome wide association studies, we have performed a combined analysis with samples from the Bethesda Sickle Cell Cohort Study (Clinicaltrials.gov Identifier NCT00011648) and the WalkPhasst clinical trial screening population (Clinicaltrials.gov Identifier NCT00492531) using pain defined by hospitalizations and HbF as phenotypes. Over the past year, an admixture GWAS in the Bethesda Sickle Cell Cohort Study shows no loci associated with pain crisis events, while a scan using sleep disturbance defined by the Pittsburgh Sleep Quality Index is suggestive of linkage for markers on chromosome 13. Based upon the above results, several additional genome wide analyses have been performed using ventricular hypertrophy, a measure of nitric oxide bioavailability, renal function and red cell alloimmunization as categorical phenotypes. In particular, red cell transfusion is one of the standard therapies available for the treatment of sickle cell disease. Transfusion therapy is complicated by red cell antigen alloimmunization. Prior studies have suggested that patients who develop allo-antibodies may have a genetic predisposition, in addition to antigenic differences between red cell donor and SCD populations. Among more than 500 subjects, the prevalence of allo-immunization documented by a blood bank is 27.6 %. Using regression modeling, significant risk factors for antibodies are frequent transfusion and Amerindian/Asian ancestry, which suggests both environmental and genetic contributions to this SCD complication. Admixture analyses using several different risk models preliminarily indicate statistically significant genetic associations localizing to either chromosome 4 or 5. Overall, these results indicate that admixture mapping may be a cost effective means for complex trait mapping in admixed rare disease populations like sickle cell disease. Studies are also underway to also determine if gene expression profiling can also be used to identify sub-phenotypes of sickle cell disease. In a pilot study, we have isolated RNA from different peripheral blood preparations to determine which is the most informative for larger studies. For this pilot, subjects with SS sickle cell disease were compared to those with SC sickle cell disease because they each have distinct clinical manifestations. Preliminary analysis suggests that SS and SC have nearly identical gene expression profiles despite their clinical differences. In a larger follow-up study, we compared gene expression profiles for 2 sickle cell anemia sub-phenotypes defined by a plasma biomarker using exon arrays. We have identified candidate genes associated with this sub-phenotype, including several alternatively spliced transcripts. Finally, we have collected total blood RNA from a cohort of 30 sickle cell anemia subjects phenotyped for pain and prospectively followed for 12 months. MicroRNA profiles from 2 crisis events from the same subject are being compared to baseline profiles is a study designed to identify biomarkers for pain crisis events.
Showing the most recent 10 out of 25 publications
