Hereditary Myelodysplastic Syndrome (MDS) / Acute Myeloid Leukemia (AML) Panel

Myelodysplastic syndromes (MDS) are clonal hematopoietic stem cell disorders characterized by ineffective hematopoiesis, dysplastic bone marrow, and peripheral blood cytopenias (for review see Bannon and DiNardo. 2016. PubMed ID: 27248996). Progression of MDS to acute myeloid leukemia (AML) occurs in up to 40% of patients (Heaney and Golde. 1999. PubMed ID: 10341278). MDS and AML are most often sporadic, late-onset malignancies, but recent data indicate that hereditary MDS/AML may have a higher incidence than was thought previously, and may have a younger age of onset than sporadic cases (Bannon and DiNardo. 2016. PubMed ID: 27248996). MDS/AML predisposition is associated with several primary bone marrow failure disorders including Diamond-Blackfan anemia, Fanconi anemia, severe congenital neutropenia, Shwachman-Diamond syndrome, and dyskeratosis congenita (Owen et al. 2008. PubMed ID: 18173751; Auerbach 2009. PubMed ID: 19622403). Several additional MDS/AML predisposition syndromes have also been recognized that may or may not be associated with additional clinical features such as pancytopenia, MonoMAC, and hearing loss. Consequently, the presence of 'accessory' clinical features may increase suspicion of MDS/AML predisposition, but the absence of accessory phenotypes does not rule out an increased risk for MDS/AML. Genetic testing for germline predisposition for MDS/AML is important for establishing a diagnosis and for identifying potential bone marrow donors and at risk family members.

This test involves sequencing of familial MDS/AML predisposition genes. Familial forms of MDS/AML involve inheritance of a single abnormal copy of one of the genes listed below that encode transcription factors or other proteins important for hematopoiesis, protein translocation, or tumor suppression. Penetrance and age at onset of disease vary among the MDS/AML predisposition genes. Pathogenic variants in the familial MDS/AML predisposition genes have also been found in sporadic cases of MDS/AML. For example, RUNX1 pathogenic variants have been reported in up to 32% of sporadic cases of MDS/AML (West et al. 2014. PubMed ID: 24467820). It is often difficult to determine whether the variants found in hematopoietic malignancies are inherited or acquired when a sample such as peripheral blood or bone marrow is used for genetic testing. When testing for inherited MDS/AML, skin fibroblasts are the recommended specimen type for germline analysis (Nickels et al. 2013. PubMed ID: 23926458). Variants associated with MDS/AML are heterogeneous and include missense, nonsense, splicing, and deletions and duplications of various sizes. Missense and nonsense variants are the predominant types of variants found in patients. To date, large deletions/duplications have not been reported in the ANKRD26, CEBPA, DDX41, ETV6, SAMD9, SAMD9L, or SRP72 genes in patients with Hereditary MDS/AML. Large deletions/duplications and complex rearrangements have been reported in patients with GATA2, RUNX1, TERC, TERT, and TP53 related syndromes and comprise ~ 9%, 29%, 5%, 5%, and 5%, respectively, of the unique variants reported for these genes (Human Gene Mutation Database). For ANKRD26, nearly all pathogenic variants are found in the 5’ UTR.

ANKRD26—Thrombocytopenia 2 (THC2) is characterized by mild to severe thrombocytopenia, normal platelet size, and mild bleeding tendency (Gandhi et al. 2003. PubMed ID: 12890928; Drachman et al. 2000. PubMed ID: 10891439; Punzo et al. 2010. PubMed ID: 20626622; Noris et al. 2011. PubMed ID: 21467542). Most pathogenic variants in the ANKRD26 gene cluster in the 5' UTR and are associated with a high incidence of acute leukemia; studies show a 30 fold increase in the frequency of MDS and AML in patients with pathogenic ANKRD26 variants (Noris et al. 2011. PubMed ID: 21467542; Noris et al. 2013. PubMed ID: 24030261; Marquez et al. 2014. PubMed ID: 24628296). The function of the ANKRD26 protein is unknown, but it is found in several tissue types, including brain, liver, skeletal, and hematopoietic cells and is associated with the cytosolic portion of cell membranes where it may interact with signaling proteins (Bluteau et al. 2014. PubMed ID: 24430186).

CEBPA—Pathogenic variants in CEBPA are associated with near complete penetrance for AML development with a variable latency period and no preceding hematologic or other clinical phenotypes (Smith et al. 2004. PubMed ID: 15575056; Owen et al. 2008. PubMed ID: 18173751). The CEBPA protein (CCAAT enhancer binding protein alpha) helps regulate normal granulocyte development. Biallelic CEBPA variants comprising an inherited 5' frameshift and a somatic 3' variant in the second allele have also been reported in several families, and in general, patients with CEBPA variants may have a more favorable prognosis than patients with variants in other MDS/AML predisposition genes (Smith et al. 2004. PubMed ID: 15575056; Owen et al. 2008. PubMed ID: 18173751; Pabst and Mueller 2009. PubMed ID: 19706798).

DDX41—DDX41 encodes an RNA helicase. Pathogenic variants in DDX41 have been identified in inherited and acquired forms of MDS/AML and are not associated with other preceding clinical symptoms (Polprasert et al. 2015. PubMed ID: 25920683; Lewinsohn et al. 2016. PubMed ID: 26712909). A long latency period with an average age of disease onset of 61 years is associated with DDX41 variants (Lewinsohn et al. 2016. PubMed ID: 26712909). As is the case with other familial MDS/AML predisposition genes, germline variants in DDX41 may predispose to secondary somatic variants in DDX41 and other cancer genes.

ETV6—Pathogenic variants in ETV6 have been identified in several families with autosomal dominant thrombocytopenia (Thrombocytopenia 5) and who have mild-to-moderate bleeding tendencies and an increased risk of developing hematologic malignancies including MDS, AML and acute lymphoblastic leukemia (ALL) (Zhang et al. 2015. PubMed ID: 25239263; Topka et al. 2015. PubMed ID: 26102509; Noetzli et al. 2015. PubMed ID: 25807284). ETV6 encodes a transcription factor important for hematopoietic stem cell survival (Zhang et al. 2015. PubMed ID: 25239263). Somatic rearrangements and fusions involving ETV6 along with point pathogenic variants are frequently found in hematopoietic malignancies (Wang et al. 2014. PubMed ID: 24997145), but only recently have germline missense variants been found to segregate with hematopoietic malignancies (Zhang et al. 2015. PubMed ID: 25239263). ETV6 pathogenic variants segregated completely with thrombocytopenia, but penetrance for malignancies is incomplete (Zhang et al. 2015. PubMed ID: 25239263; Topka et al. 2015. PubMed ID: 26102509; Noetzli et al. 2015. PubMed ID: 25807284).

GATA2—Pathogenic variants in the GATA2 gene are associated with a number of clinical phenotypes including primary lymphedema and Emberger syndrome (Ostergaard et al. 2011. PubMed ID: 21892158), combined immunodeficiencies termed, dendritic cell, monocyte, B and NK lymphoid deficiency (aka DCML) (Dickinson et al. 2011. PubMed ID: 21765025), or monocytopenia and mycobacterial infection syndrome (aka MonoMAC) (Hsu et al. 2011. PubMed ID: 21670465). GATA2 pathogenic variants have also been reported in patients with variable hematologic phenotypes including neutropenia, bone marrow failure, and familial MDS/AML with no 'accessory' phenotype (Pasquet et al. 2013. PubMed ID: 23223431; Hahn et al. 2011. PubMed ID: 21892162). GATA2 pathogenic variants are associated with early onset MDS/AML, but penetrance is incomplete (Hahn et al. 2011. PubMed ID: 21892162). A variety of chromosomal rearrangements, in particular monosomy 7, have also been found in patients with GATA2 - associated MDS/AML (Hahn et al. 2011. PubMed ID: 21892162).

RUNX1—Familial Platelet Disorder with predisposition to myeloid malignancy (FPD/AML) is characterized by mild to moderate bleeding tendencies and impaired platelet aggregation (Owen et al. 2008. PubMed ID: 18173751). The incidence of MDS/AML in patients with RUNX1 pathogenic variants is over 40% with wide ranging age of onset from childhood to adults in their 70s (Churpek et al. 2012. PubMed ID: 23258901). The RUNX1 gene encodes a transcription factor critical for normal hematopoiesis. Causative variants in RUNX1 are most often missense, nonsense, or frameshift variants resulting in premature protein truncation with possible dominant negative effects.

SAMD9—SAMD9, and SAMD9L are parologous genes oriented head to tail on chromsome 7q21.2 (Li et al. 2007. PubMed ID: 17407603). Both SAMD9 and SAMD9L are tumor supressors found in many different tissue types. Pathogenic variants in SAMD9 are associated with autosomal dominant MIRAGE syndrome and predisposition to myelodysplastic syndrome (MDS). In patients with germline gain-of-function missense SAMD9 or SAMD9L variants, loss of the variant allele via somatic aberrations, such as partial or complete chromosome 7 loss, can occur that gives rise to clonal expansion and progression towards MDS (Davidsson et al. 2018. PubMed ID: 29535429; Tesi et al. 2017. PubMed ID: 28202457; Schwartz et al. 2017. PubMed ID: 28487541). Consequently, patients with monosomy 7 and MDS are good candidates for SAMD9 / SAMD9L germline sequencing.

SAMD9L—SAMD9L encodes a tumor suppressor; heterozygous gain of function variants in SAMD9L have been identified in families with ataxia-pancytopenia, immunodeficiency, neurological symptoms and predisposition to MDS (Chen et al. 2016. PubMed ID: 27259050; Tesi et al. 2017. PubMed ID: 28202457). Germline pathogenic variants in SAMD9L are associated with somatic monosomy 7 (See SAMD9 above for more information).

SRP72—Pathogeic variants in the SRP72 gene are associated with aplastic anemia (AA), pancytopenia, and congenital nerve deafness (Kirwan et al. 2012. PubMed ID: 22541560). SRP72 encodes a component of the signal recognition particle (SRP) responsible for transport of membrane bound and excreted proteins to the endoplasmic reticulum. To date, few AA/MDS families have been identified with pathogenic SRP72 gene variants. Reported pathogenic variants include one small deletion and one missense variant (Kirwan et al. 2012. PubMed ID: 22541560).

TERC and TERT—Dyskeratosis congenital (DC) is a bone marrow failure syndrome caused by deficiency in proteins responsible for maintaining telomere length. TERT encodes a reverse transcriptase and TERC the RNA component of the telomerase enzyme complex. Pathogenic variants in DC-associated genes cause telomere shortening; pathogenic variants in TERC show anticipation characterized by shorter telomeres through generations (Vulliamy et al. 2006. PubMed ID: 16332973). The triad of clinical presentation in DC patients is abnormal skin pigmentation, nail dystrophy and mucosal leucoplakia with over 80% of DC patients developing progressive bone marrow failure by age 30 (Dokal 2011. PubMed ID: 22160078). Age of onset and severity of disease vary, even among family members, making diagnosis difficult.

TP53—Li-Fraumeni Syndrome is a hereditary cancer syndrome that predisposes individuals to multiple neoplasms at an early age including leukemias (Owen et al. 2008. PubMed ID: 18173751). The average age of malignancy for individuals with LFS is typically between 20 and 45, which is at least 2-3 decades sooner than reported for the general population (Nichols et al. 2001. PubMed ID: 11219776). P53 encodes the well described cellular tumor p53 antigen (Soussi 2010. PubMed ID: 20930848). p53 is a ubiquitously expressed DNA-binding protein that plays a major role in the regulation of cell division, DNA repair, programmed cell death, and metabolism.

See individual gene test descriptions for additional information on clinical features, gene function, and spectra of pathogenic variants.

In a recent study, pathogenic variants were identified in 29% of families with predisposition to myelodysplastic syndromes/acute myeloid leukemia (MDS/AML) (Churpek et al. 2015. PubMed ID: 26492932). The gene panel used in this study, however, did not include ANKRD26, SAMD9L, SRP72, DDX41, or ETV6 suggesting that 29% is a minimum value.

This test is performed using Next-Gen sequencing with additional Sanger sequencing as necessary.

This panel provides 100% coverage of all coding exons of the genes plus 10 bases of flanking noncoding DNA in all available transcripts along with other non-coding regions in which pathogenic variants have been identified at PreventionGenetics or reported elsewhere. We define coverage as ≥20X NGS reads or Sanger sequencing. PGnome panels typically provide slightly increased coverage over the PGxome equivalent. PGnome sequencing panels have the added benefit of additional analysis and reporting of deep intronic regions (where applicable).

Dependent on the sequencing backbone selected for this testing, discounted reflex testing to any other similar backbone-based test is available (i.e., PGxome panel to whole PGxome; PGnome panel to whole PGnome).

This test is designed for detecting germline variants and is not designed specifically for detecting low levels of acquired variants in somatic MDS /AML. Skin fibroblasts are the recommended specimen type for germline testing for MDS / AML (Nickels et al. 2013. PubMed ID: 23926458).