In this issue of Blood, Wlodarski and colleagues demonstrate that as many as 72% of adolescents diagnosed with myelodysplastic syndrome (MDS) and monosomy 7 harbor germline mutations in GATA2.1 Although pediatric MDS is a very rare diagnosis, occurring in 0.8 to 4 cases per million,2 Wlodarski et al screened >600 cases of primary or secondary MDS in children and adolescents who were enrolled in the European Working Group on MDS consortium over a period of 15 years. The overall frequency of germline GATA2 mutations in children with primary MDS was 7%, and 15% in those presenting with advanced disease. Notably, mutations in GATA2 were absent in patients with therapy-related MDS or acquired aplastic anemia.
Germline mutations in GATA2 were previously described in several congenital disorders, including MonoMAC3 and Emberger syndromes.4 MonoMAC syndrome is characterized by monocytopenia, frequent opportunistic infections, and a predisposition to developing MDS and acute myeloid leukemia. In 2011, several groups used next-generation sequencing technologies to screen kindreds with MonoMAC syndrome, identifying heterozygous germline mutations in GATA2, predominantly affecting the zinc finger (ZF) regions of this essential hematopoietic and endothelial transcription factor.3,5,6 Subsequently, germline GATA2 mutations were found to cause certain forms of congenital neutropenia as well as Emberger syndrome, which is characterized by lymphedema and a predisposition to MDS/acute myeloid leukemia.4 Although the majority of mutations occur in the coding region of the ZF1 and ZF2 domains, Wlodarski et al also describe recurrent noncoding mutations in the +9.5kb regulatory intronic region of GATA2 upstream of exon 4.
GATA2 is an essential zinc finger transcription factor that is critical for early developmental hematopoiesis. Gata2−/− mice die at E10.5 because of lack of early hematopoiesis,7 and conditional knockout mice also die later in gestation of edema and hemorrhage because of failure of fetal liver hematopoiesis and defective lymphatic development.8 There is broad evidence that haploinsufficiency of GATA2 leads to a loss of hematopoietic stem cells, and in humans, preferentially leads to depletion of dendritic, monocyte, B, and natural killer cells.3,4,6 Rarely, gain-of-function GATA2 mutations have also been described as secondary events in certain cases of blast crisis chronic myelogenous leukemia.9
In screening this large cohort of children and adolescents with primary and secondary MDS, Wlodarski et al reveal a number of interesting findings that lead to further questions. First, the incidence of GATA2 mutations was significantly higher in those with advanced MDS (15%, 13/85) compared with lower-grade MDS (4%, 15/341) (P < .01). Strikingly, 70% of patients with GATA2 mutations had monosomy 7 vs only 11% without (P < .01). Second, patients with GATA2 mutations were also older at diagnosis compared with those without (median age, 12.3 vs 10.3 years) (P < .01). Together, these observations question whether additional acquired mutations contribute to the more advanced phenotypes. Indeed, as one example, ASXL1 lesions have been reported to occur recurrently in one-third of patients with germline GATA2 lesions.10 Finally, no mutations were recorded in the 82 patients with treatment-related MDS, implying that additional unidentified loci are responsible for conferring a risk of developing this generally fatal secondary late effect.
Although the initial kindreds with MonoMAC syndrome demonstrated an autosomal dominant pattern of inheritance, only 12 of the 53 GATA2mut patients with an available family history were found to have an affected family member in the current study. The lack of family history in the majority of patients highlights the importance of screening children and adolescents for GATA2 mutations with MDS. GATA2 mutational status did not affect overall survival, nor were the overall rates of bacterial, viral, fungal, or parasitic infections different among GATA2wt and GATA2mut patients undergoing hematopoietic stem cell transplant. However, identifying affected patients will be essential to predict those who would not benefit long term from supportive care or immunosuppressive therapy and thus be candidates for swift hematopoietic stem cell transplant. Screening affected family members would obviously also be critical for identifying appropriate allogeneic donors.
This important study also raises several important questions about the function of the identified variant alleles. Specifically, noncoding mutations in intron 4 were found in both patients with MDS as well as in nonaffected family members. This was not seen with coding mutations detected in either of the ZF regions. In addition, in distinction to MonoMAC patients who often present with monocytopenia,3 pediatric patients with GATA2mut MDS frequently had monocytosis, which may or may not be related to the high proportion of coexisting presentation of advanced MDS, monosomy 7, or cooperative genetic events.
Wlodarski and colleagues have made an important contribution to the field of pediatric MDS with these findings. Although primary MDS in children is still considered an idiopathic disorder, GATA2 should be recognized as a common predisposing factor for the development of primary advanced MDS in childhood and even more so in adolescence. Monitoring for cytogenetic evolution, secondary mutations and signs of advanced disease such as worsening cytopenias will also be important to identify patients in need of swift transplantation. This important paper serves as another reminder of the protean nature of GATA2-related disorders and will affect the clinical care of children and adolescents affected with primary MDS.
Conflict-of-interest disclosure: The authors declare no competing financial interests.