Abstract
Management of the patient with transfusion-dependent anemia (TDA) is complex. Diagnosis is frequently difficult as numerous disorders may lead to TDA, including bone marrow failure syndromes, congenital dyserythropoietic anemias, or inherited hemolytic anemias. Assigning the diagnosis may be problematic as transfused blood or reticulocytosis confound diagnostic testing, or, mutant erythrocytes are so unstable, they are rapidly destroyed. Complications of chronic transfusion include iron overload, infection risk, alloimmunization, cost, and inconvenience. TDA is an excellent candidate for targeted next generation sequencing. There is significant genotypic variability and many of the associated genetic loci are very large, making traditional sequencing strategies cumbersome. We studied 21 patients with TDA using genome-wide targeted exon capture followed by high-throughput next generation DNA sequencing (whole-exome sequencing, WES) using a NimbleGen SeqCap EZ Exome v2.0 solution-based capture system followed by next-generation sequencing on a HiSeq 2000 with paired-end sequencing at 75bp read length. The male:female ratio was 13:8. Age at referral ranged from 2 months to 14 years. All patients were transfusion dependent by 6 months of age. Working diagnoses included possible marrow failure syndrome, congenital dyserythropoietic anemia, and possible enzyme or membrane defect. Variant analyses were performed using the GATK pipeline. Targeted filtering and annotation of protein changing variants in 154 erythrocyte disease genes were performed using the ANNOVAR algorithm. Variants were assessed by mutation prediction and conservation programs including PolyPhen2, Sift, LRT, and Mutation Taster. Variants were also assessed for occurrence and frequency Thousand Genomes, Exome Sequencing Project, dbSNP, on line and local mutation databases, and PubMed. Copy number variants were assessed by ExomeCount and visual inspection. Potential disease-associated variants were validated by Sanger sequencing of DNA from the proband and parents. Interpretation was made using historical, clinical, laboratory and genetic data. The most common diagnosis was hereditary spherocytosis due to alpha spectrin gene (SPTA1) mutations, found in 7 patients. Two patients had deleterious mutations in both SPTA1 alleles; one with nonsense mutations in trans died of liver failure associated with iron overload, the other with nonsense and splicing mutations in trans remains transfusion dependent. One patient homozygous for an SPTA1 missense mutation in a highly conserved, functionally important amino acid had a sibling homozygous for the same mutation die in the perinatal period due to complications of anemia. Finally, one patient with SPTA1 nonsense and missense mutations in trans became transfusion independent post splenectomy. Ten patients had defects in erythrocyte metabolism. Mutations in the pyruvate kinase gene PKLR were found in 6 patients; two of these patients had deletions in the PKLR gene locus suggested by WES and confirmed by Gap PCR and MLPA. Three patients had bi-allelic mutations in the glucose phosphate isomerase gene and one had bi-allelic mutations in the hexokinase gene. Homozygosity was found in 4 of 10 patients with metabolic gene mutations. A single patient had beta thalassemia major with homozygous beta-globin gene mutations. Confirmatory functional studies are underway in three patients. Two TD patients had bone marrow findings suggestive of hypoplastic anemia; one had a missense mutation in a highly conserved residue of RPS7, recently associated with Diamond Blackfan anemia; the other had a deleterious mutation in FANCI predicted to function as a dominant negative. Functional studies are underway in a third patient with likely deleterious, membrane-linked variants. Application of targeted WES to TDA allows precise diagnosis to guide appropriate therapy, e.g. splenectomy or transplant; it allows genetic counseling of associated family members, and permits diagnosis and expectant management of future pregnancies. Targeted WES is an excellent tool for application to monogenic hematologic diseases where genotypic variability, i.e. mutations in numerous genes, leads to the same clinical phenotype. Examples include bone marrow failure syndromes, hemolytic anemia, congenital neutropenia, and immunodeficiency syndromes.
Disclosures:
No relevant conflicts of interest to declare.
DeviCNV: detection and visualization of exon-level copy number variants in targeted next-generation sequencing data
