Children with Down syndrome (DS; trisomy 21) have an increased risk of acute myeloid leukemia (ML-DS) in the first 5 years of life. In most cases ML-DS is preceded by Transient Abnormal Myelopoiesis (TAM), a fetal/neonatal pre-leukemic disorder unique to DS which regresses after birth. Both TAM and ML-DS harbor acquired N-terminal mutations in the hematopoietic transcription factor gene GATA1. In a prospective study of 200 DS neonates, we recently showed that 29% had acquired GATA1 mutations including 17/200 (8.5%) with clinical or hematologic evidence of TAM; the remaining 20.5% were clinically and hematologically 'silent', with smaller mutant GATA1 clones and lower blast frequency compared to overt TAM. The reasons why some DS neonates develop overt TAM and the factors which determine mutant GATA1 clone size are unknown. To address this, we analysed data from neonates in the prospective Oxford-Imperial DS Cohort Study and investigated the impact of 30 clinical and hematologic factors on clone size using statistical and mathematical modelling. Mutant GATA1 clones were determined in 54 neonates by targeted next generation sequencing of GATA1 exon 2 (mutation detection limit 0.3%). Clone size was determined by analysing original unprocessed reads using less stringent filtering parameters and counting reads containing mutated v total sequence. Correlation analysis identified 4 hematologic variables correlated with mutant GATA1 clone size: circulating nucleated red cells (r=+0.5003; p=0.0001), platelets (r=+0.436; p=0.001), total leukocytes (r=+0.7094; p<0.001) and % blasts (r=+0.7292; p<0.001); clone size was higher in neonates with megakaryocyte fragments (p=0.0024) and platelets >150x109/L (p=0.019). Numbers of neutrophils, monocytes, basophils, eosinophils and lymphocytes did not correlate with GATA1 clone size. Clinical variables significantly correlated with clone size were hepatomegaly (p=0.0016), splenomegaly (p=0.0001) and rash (0.0174). The only pregnancy-related variables affecting mutant GATA1 clone size were intrauterine growth restriction and maternal diabetes (p=0.0156). Linear regression to determine the joint impact of all 30 variables on clone size (r2=0.88) followed by Lasso penalization identified the same 4 hematologic variables (nucleated red cells, platelets, total leukocytes and % blasts); Lasso penalized regression with these 4 variables gave a coefficient of determination of 0.63. Together these data suggest that chronic intrauterine hypoxia may affect expansion/differentiation of mutant GATA1 clones in DS. Consistent with this, nucleated red cells from 3 neonates with TAM all harbored GATA1 mutations identical to those in total circulating nucleated cells. Since neither perinatal infection nor gestational age at birth correlated with mutant GATA1 clone size, infection-related cytokines and the timing of acquisition of a mutant GATA1 clone during fetal development may not play a major role in determining clone size. Finally, a hierarchical model to investigate the impact of GATA1 mutation on hematopoietic stem and progenitor (HSPC) differentiation in DS neonates using a Bayesian approach also predicted increased erythroid cell output from GATA1 mutated HSPC v HSPC without a GATA1 mutation. In conclusion, in neonates with DS the size of the mutant GATA1 clone correlates with the presence of clinical signs of hepatomegaly, splenomegaly and skin rash; mutant GATA1 clone size correlates with the numbers of circulating nucleated red cells, platelets and blast cells suggesting that GATA1 mutant HSPC retain the ability to differentiate down the erythroid and megakaryocyte lineage; intrauterine hypoxia may be one of the factors driving expansion and/or maturation of the GATA1 mutant clone during fetal life in DS.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution