Pediatrics International (2019) 61, 222–229 doi: 10.1111/ped.

13776

Review Article

Recent advances in the understanding of transient abnormal

myelopoiesis in Down syndrome

Kenichiro Watanabe
Department of Hematology and Oncology, Shizuoka Children’s Hospital, Aoi-ku, Shizuoka, Japan

Abstract Neonates with Down syndrome (DS) have a propensity to develop the unique myeloproliferative disorder, transient
abnormal myelopoiesis (TAM). TAM usually resolves spontaneously in ≤3 months, but approximately 10% of
patients with TAM die from hepatic or multi-organ failure. After remission, 20% of patients with TAM develop
acute myeloid leukemia associated with Down syndrome (ML-DS). Blasts in both TAM and ML-DS have trisomy
21 and GATA binding protein 1 (GATA1) mutations. Recent studies have shown that infants with DS and no clinical
signs of TAM or increases in peripheral blood blasts can have minor clones carrying GATA1 mutations, referred to
as silent TAM. Low-dose cytarabine can improve the outcomes of patients with TAM and high white blood cell
count. A number of studies using fetal liver cells, mouse models, or induced pluripotent stem cells have elucidated
the roles of trisomy 21 and GATA1 mutations in the development of TAM. Next-generation sequencing of TAM
and ML-DS patient samples identified additional mutations in genes involved in epigenetic regulation. Xenograft
models of TAM demonstrate the genetic heterogeneity of TAM blasts and mimic the process of clonal selection
and expansion of TAM clones that leads to ML-DS. DNA methylation analysis suggests that epigenetic dysregula-
tion may be involved in the progression from TAM to ML-DS. Unraveling the mechanisms underlying leukemogen-
esis and identification of factors that predict progression to leukemia could assist in development of strategies to
prevent progression to ML-DS. Investigation of TAM, a unique pre-leukemic condition, will continue to strongly
influence basic and clinical research into the development of hematological malignancies.

Key words Down syndrome, GATA1, leukemia, transient abnormal myelopoiesis, trisomy 21.

Neonates with Down syndrome (DS) have a marked propen-
sity to develop the unique myeloproliferative disorder referred
to as transient abnormal myelopoiesis (TAM), transient myelo-
proliferative disorder, or transient leukemia.1 In the majority
of cases, TAM resolves spontaneously in ≤3 months, but
approximately 10% of patients with TAM die from hepatic or
multi-organ failure. Recent studies have identified risk factors
for infant death due to TAM and indicate the efficacy of low-
dose cytarabine for treatment of severe manifestations of
TAM.2–5
After remission, approximately 20% of patients with TAM
develop acute myeloid leukemia (AML) associated with Down
syndrome (ML-DS).2,3,5,6 Given that blasts in both TAM and
ML-DS have common genetic abnormalities, including tri-
somy 21 and GATA binding protein 1 (GATA1) mutation,7–13
TAM is considered a pre-leukemic disorder, and understanding
how TAM progresses to ML-DS may provide information cru-
cial to clarifying the mechanism involved in multi-step leuke-
mogenesis.14–17 Recent studies using a xenograft model and
Correspondence: Kenichiro Watanabe, MD PhD, Department of
Hematology and Oncology, Shizuoka Children’s Hospital, 860
Urushiyama, Shizuoka 420-8660, Japan. Email: wataken@kuhp.
kyoto-u.ac.jp
Received 10 May 2018; revised 8 December 2018; accepted 28
December 2018.
next-generation sequencing (NGS) have detected subclones
that emerge during the TAM phase and appear to expand to
develop into overt leukemia through clonal selection and addi-
tional genetic alterations.18,19
This review outlines the clinical features of TAM and
focuses on the latest findings that may provide clues to the
understanding of this unique disease. I also consider future
research directions in this field.
Definition and diagnosis of TAM
Transient abnormal myelopoiesis is a unique myeloprolifera-
tive disorder in newborns with DS who present with circulat-
ing blasts that are morphologically and phenotypically similar
to AML. The blasts have morphologic and immunologic fea-
tures of the megakaryocytic lineage.2,20–22 Clinical diagnosis
of TAM is based on the presence of characteristic blasts in
peripheral blood smears. Given that TAM blasts carry muta-
tions of GATA1 in addition to trisomy 21,8–13 positivity for
GATA1 mutation confirms the diagnosis of TAM, but the per-
centage of TAM cells varies enormously, both temporally and
between patients, and there are no clear diagnostic criteria for
TAM.23
Diagnosis of TAM is important, particularly because
infants with TAM have a higher likelihood of developing

© 2018 Japan Pediatric Society


ML-DS; children with no clear history of TAM, however, also
develop ML-DS. These patients may not have overt clinical
symptoms associated with TAM, or may present with normal
blood cell count and low blast count in the neonatal period.
Thus, current clinical diagnosis, based on the presence of
blasts in peripheral blood smears, is unsatisfactory for identifi-
cation of the group at high risk of ML-DS development.
The investigators of the Oxford–Imperial Down Syndrome
Cohort Study Group (OIDSCS) analyzed clinical findings,
blood counts and smears, and GATA1 mutation status in 200
DS neonates.23 They reported that all DS neonates had multi-
ple blood count and morphological abnormalities and that
97.5% had circulating blasts, indicating that all DS neonates
have at least some hematological abnormalities, regardless of
GATA1 mutation status. On conventional Sanger sequencing
and denaturing high-performance liquid chromatography,
GATA1 mutations were detected in 17 of 200 (8.5%) enrolled
DS neonates.23 Furthermore, infrequent GATA1 mutant clones
were detected using the more sensitive method of NGS in 18
of 88 (20.4%) DS neonates in whom GATA1 mutations were
not detected on conventional sequencing methods.23 The clini-
cal and hematologic features of these neonates were indistin-
guishable from those of DS neonates without GATA1
mutations detectable on NGS. The OIDSCS proposed the term
“silent TAM” for DS neonates with GATA1 mutation detect-
able only on NGS. Of note, one of 18 individuals with silent
TAM developed ML-DS, while none of the 70 neonates with-
out NGS-detected GATA1 mutations were documented to
develop the condition.23 That indicates that peripheral blood
smear and GATA1 mutation analysis should be performed for
all DS neonates, and that NGS-based methods can identify
neonates with minor GATA1 mutant clones.
In the era of high-throughput genome sequencing, the defi-
nition of TAM needs to be reconsidered and revised to reflect
the clinical behavior and the latest molecular insight into the
disease. It may be appropriate to define TAM according to the
presence of GATA1 mutations detected using a more sensitive
method such as NGS as well as conventional Sanger sequenc-
ing. The incidence and clinical significance of silent TAM
should be re-evaluated for separate cohorts, given that they
may vary depending on the sensitivity of the methods used to
detect GATA1 mutations.
Incidence of TAM
Transient abnormal myelopoiesis occurs in approximately 10%
of DS newborns. In Japan, during the 18 year period between
1980 and 1997, the overall prevalence of DS was 5.82 per
10 000 live births (8.3–9.7 per 10 000 after correction accord-
ing to the estimated ascertainment ratio, 60–70%).24 Accord-
ing to these data, the incidence of TAM in Japan is estimated
as approximately 0.58 per 10 000 live births, but, as afore-
mentioned, the diagnostic criteria for TAM differ between pre-
vious studies, and not all DS neonates may have been
carefully examined for TAM on thorough morphological
assessment of peripheral blood smears. One population-based
Review of TAM 223
study screened for GATA1 mutations using Guthrie cards from
590 DS infants and detected them in 3.8% of samples.25 In
contrast, the application of targeted NGS by OIDSCS found
GATA1 mutations in 20% of DS neonates who were GATA1
mutation negative on Sanger sequencing.23 Hence, the inci-
dence of TAM can vary depending on the clinical definition
used and the sensitivity of the methods for detection of
GATA1 mutations.
Clinical and hematological features of TAM
The clinical and hematological features of TAM are listed in
Table 1. Clinical presentation of TAM may vary from asymp-
tomatic to very severe disease. Previous studies report a med-
ian age at diagnosis of 3–7 days, with almost all cases
diagnosed in the 2 months after birth. Asymptomatic babies
with DS may be diagnosed with TAM on incidental detection
of blasts in peripheral blood smears with or without leukocyto-
sis. In contrast, neonates with TAM may present with severe
symptoms, such as marked hepatosplenomegaly, serous effu-
sions, and multi-organ failure, due to blast cell infiltration.
Patients with very severe disease are at high risk of early
death.2–5 TAM can present as hydrops fetalis and hep-
atosplenomegaly in fetuses with DS.26–29 The most severe
form of TAM is diagnosed on post-mortem in stillborn DS
fetuses.30–32 Infants with TAM may also have delayed-onset
hyperbilirubinemia, which is a sign of progressive liver fibro-
sis that can result in fatal liver failure.33–35 Pathogenesis of
liver fibrosis in TAM appears to be similar to that of marrow
fibrosis that is associated with acute megakaryoblastic leuke-
mia (AMKL). TAM blasts from patients with liver fibrosis
can exhibit high platelet-derived growth factor and transform-
ing growth factor-b1 expression,36,37 leading to the hypothesis
that TAM blasts may arise from the fetal hepatic hematopoi-
etic environment, rather than from the bone marrow.
Patients with TAM present with various hematological
abnormalities.2–5 Leukocytosis and thrombocytopenia are com-
mon, and white blood cell (WBC) count may exceed 100 9
109/L. TAM is characterized by a varying degree of increased
blast cells in the peripheral blood. The typical morphologic
features of blasts are similar to those of AMKL,2 but the
Table 1 Clinical and hematological features of TAM
Hepatosplenomegaly
Skin rash
Pericardial/pleural effusion, ascites
Hydrops fetalis
Jaundice
Coagulopathy
Liver failure (may occur in late neonatal period)
Thrombocytopenia
Leukocytosis
Anemia
Early death (approx. 10%)
Progression to ML-DS (approx. 20%)
MD-DS, myeloid leukemia associated with Down syndrome;
TAM, transient abnormal myelopoiesis.
© 2018 Japan Pediatric Society
224 K Watanabe
morphology of TAM blasts is variable in practice. Increased
neutrophils, myelocytes, monocytes, and basophils are fre-
quently observed in peripheral blood smears, and, interestingly,
there are cases of TAM involving prominent eosinophilia.38,39
Relatively small percentages of blasts are found in the bone
marrow of patients with TAM.2,3,5 Thus, bone marrow aspira-
tion or biopsy is not usually recommended for diagnosis or
evaluation of TAM.
The immunophenotype of TAM blasts is the same as that
of AMKL, with blast cells positive for stem cell markers
(CD34, CD117), myeloid markers (CD13, CD33), platelet gly-
coproteins (CD36, CD42, CD61), CD56, and CD7, with varia-
tion among cases.20–22
Coagulopathy is a potential complication of TAM,3,5 and
considered a risk factor for early death. Disseminated intravas-
cular coagulation (DIC) is usually observed in the majority of
severe cases of TAM with fulminant liver dysfunction.40
Management of TAM
Most patients with TAM have spontaneous regression, usually
within a few months, and do not need treatment, but some
individuals develop progressive, life-threatening disease, and
up to 20% of patients with TAM eventually die in early
infancy. Risk factors for early death in patients with TAM
include hyperleukocytosis, ascites, and prematurity, as well as
cholestasis, liver fibrosis, renal dysfunction, and DIC.2–5
The DS myeloblasts are highly sensitive to cytarabine,41
and low-dose cytarabine is effective in decreasing blast cells
in patients with TAM. Thus, low-dose cytarabine was used in
previous large prospective studies to treat the severe form of
TAM. In a Berlin-Frankfurt-M€ unster (BFM) group study,
low-dose cytarabine (0.5–1.5 mg/kg for 3–12 days) was rec-
ommended for TAM patients with thrombocytopenia, liver
dysfunction, or high white cell count (>50 9 109/L); 5 year
overall survival (OS) in the treated group was similar to that
in the untreated group (78%  8% vs 85%  3%; P = 0.44).3
In a Children’s Oncology Group (COG) study, cytarabine
(3.33 mg/kg/day continuous infusion for 7 days) was given to
patients with life-threatening symptoms. The patient survival
rate was 51%.14 Recently the Japan Children’s Cancer Group
(JCCG) TAM committee, conducted a large multi-institutional
prospective study of patients with TAM in Japan.42 In that
study high WBC count and anasarca were confirmed as risk
factors for early death in patients with DS diagnosed with
TAM. Although low-dose cytarabine could significantly
improve the survival rate in patients with a high WBC count,
it failed to change the prognosis of patients with anasarca.42
Overall, low-dose cytarabine is effective in reducing blasts,
thereby improving the outcome for a specific subset of
patients with TAM. Nevertheless, patients with very severe
disease, such as serous effusion, liver failure, and/or multi-
organ dysfunction, often fail to respond to low-dose cytarabine
and require other supportive treatment, including steroids,
exchange transfusion, and/or intensive care; the development
of novel treatment modalities for this group is imperative.
© 2018 Japan Pediatric Society
It should be noted that low-dose cytarabine may cause sig-
nificant myelosuppression. In the COG study, grade 3/4
myelosuppression was detected in 96% of treated patients,
likely due to continuous infusion of relatively high-dose
cytarabine.5 The most common side-effect of cytarabine in a
recent study was neutropenia (grade 3–4, 59%).42 Although
low-dose cytarabine is more frequently applied to TAM
patients with potential risk factors, there has been no study
that defines clear criteria for its application and treatment
schedule for TAM. A prospective intervention study is
planned, to establish a standard for treatment with cytarabine
for patients with TAM.
Low-dose cytarabine therapy has been hypothesized to pre-
vent progression to ML-DS by eliminating pre-leukemic
clones, but no studies have shown that cytarabine reduces the
risk of leukemia development.3,5 Most recently, a multicenter,
non-randomized, historically controlled study, TMD 2007,
indicated that low-dose cytarabine was insufficient to prevent
progression to ML-DS.43
It is recommended to follow up patients with TAM every
1–3 months to screen for the development of ML-DS. Throm-
bocytopenia often proceeds to manifestation of overt acute
leukemia. At such an early phase, it may be difficult to detect
a significant percentage of blasts on smear slides obtained by
bone marrow aspiration, while the identification of clusters of
leukemic cells in bone marrow biopsy specimens may be more
feasible.
Pathogenesis of TAM
A schematic diagram of the pathogenesis of TAM and pro-
gression to ML-DS is shown in Fig. 1. TAM is a unique con-
dition that occurs only in blood cells with trisomy 21. Blast
cells of both TAM and subsequent ML-DS have acquired
GATA1 mutations, in addition to trisomy 21.8,10,12 In the
majority of cases, TAM resolves spontaneously after birth,2–5
seemingly along with the shift in the site of hematopoiesis
from fetal liver44 to bone marrow. TAM can cause massive
infiltration of the liver and hydrops fetalis in utero.26–29 Based
on these observations, it has been considered that acquisition
of an N-terminal GATA1 mutation in hematopoietic cells car-
rying trisomy 21 gives rise to TAM clones in the fetal liver.
Abnormal hematopoiesis associated with trisomy 21
Several studies of primary fetal liver cells demonstrate abnor-
mal hematopoiesis in fetuses with trisomy 21 by the second
trimester. Hematopoietic stem cells (HSC) and megakary-
ocyte–erythroid progenitors are markedly increased,45 whereas
granulocyte–macrophage progenitors are reduced in primary
fetal liver cells with trisomy 21 compared with primary dis-
omic controls.46,47 In clonogenic assays, HSC and megakary-
ocyte–erythroid progenitors with trisomy 21 have higher
proliferation rates, with increased megakaryocyte, megakary-
ocyte–erythroid, and replatable blast colonies. Furthermore,
megakaryocyte lineage differentiation is hampered in fetal

Fetal liver
Clonal selection/expansion
TAM Additional genetic events:
Trisomy 21→
Perturbed gene mutations affecting
cohesin/CTCF, EZH2, etc.
hematopoiesis
Large clones Epigenetic dysregulation
Acquired Regression
GATA1mutations
GATA1mutant
clones Clonal extinction Remission
Silent
TAM
Small clones
Birth 6 months 1 year
Fig. 1 Schematic diagram of the pathogenesis of transient
abnormal myelopoiesis (TAM) and progression to myeloid leuke-
mia associated with Down syndrome (ML-DS). CTCF, CCC
TC-binding factor; EZH2, enhancer of zeste homolog 2; GATA1,
GATA binding protein 1.
hematopoietic cells with trisomy 21. Notably, these character-
istics of perturbed hematopoiesis are observed in the absence
of GATA1 mutation.46,47 This suggests that trisomy 21 itself is
an initial event of TAM, prior to acquisition of GATA1 muta-
tions.48 Recent studies using embryonic stem and induced
pluripotent stem cells (iPSC) with trisomy 21 also demon-
strated impaired erythro-megakaryopoiesis, consistent with the
observations in fetal liver cells.49–51
The notion that trisomy 21 alone causes abnormal hemato-
poiesis prompted us to consider that the gene dosage effect of
trisomy 21 may be responsible for the TAM phenomenon. A
number of studies have attempted to identify genes or regions
of chromosome 21 that could be responsible for hematological
phenotypes in DS patients. Previous genotype–phenotype stud-
ies identified subregions in chromosome 21 necessary and suf-
ficient to produce the clinical phenotype of DS, referred to as
the “Down syndrome critical region” (DSCR).52,53 The DSCR
contains several genes involved in hematopoiesis, including
BACH1, RUNX1, ETS2, ERG, and DYRK1A,11,54–57 but the
role of increased doses of those genes in the development of
TAM and ML-DS has not been fully determined. DS mouse
models, engineered to have extra chromosomal segments con-
taining almost all of murine chromosome 16, which shows
synteny with human chromosome 21, fail to develop condi-
tions similar to either TAM or ML-DS.58–60 A recent study
establishing systematic disease models using human iPSC and
genome/chromosome-editing technologies suggests that
RUNX1/ETS2/ERG, in association with GATA1 mutation, are
critical for the hematopoietic abnormalities observed in DS.51
Using genome editing, the investigators generated a panel of
human iPSC with different combinations of complements of
chromosome 21 (disomy, trisomy, or partial trisomy) and
GATA1 status (wild-type, short, knockout). Comparative analy-
sis of these engineered iPSC demonstrated that trisomy 21
affected hematopoietic development through the enhanced pro-
duction of early hematopoietic progenitors and upregulation of
Review of TAM 225
mutated GATA1, resulting in the accelerated production of
ML-DS aberrantly differentiated cells. These effects were mediated by
dosage alterations of RUNX1, ETS2, and ERG, which are
located in a critical 4 Mb region of chromosome 21.51
GATA1 mutations in TAM
GATA1 is a key hematopoietic transcription factor with an
essential role in megakaryocyte–erythroid differentiation.61,62
ML-DS Both TAM and ML-DS cells harbor acquired N-terminal trun-
cating GATA1 mutations.7–13 GATA1 mutations are primarily
detected in exon 2, and result in premature stop codons, lead-
5 years
ing to the production of a short form of the GATA1 protein,
referred to as GATA1s.7,10 Given that GATA1 is located on
the X chromosome, cells with GATA1 mutations exclusively
express GATA1s and no full-length GATA1 protein.7 Individ-
uals without trisomy 21 who have constitutional GATA1 muta-
tions that produce only GATA1s develop thrombocytopenia,
but not TAM or ML-DS.63,64 Thus, GATA1 mutation alone is
not sufficient to transform cells into pre-leukemic or leukemic
clones.
The roles of GATA1s in the development of TAM and
ML-DS have not been fully elucidated. Forced expression of
GATA1s in fetal hematopoietic progenitors from mice with
wild-type GATA1 causes marked expansion of megakary-
oblastic progenitors.65 Toki et al. also found that GATA1s has
a more profound effect in promotion of megakaryocyte prolif-
eration than that induced by GATA1 deficiency,66 supporting
a gain-of-function mechanism for GATA1s. GATA1s is
thought to perturb GATA1-mediated regulation of other tran-
scription factors, including GATA2, MYB, MYC, and
IKAROS family zinc finger 1 (IKZF1), in fetal megakary-
ocytes.67 Transduction of GATA1s into GATA1 knockdown
mice results in abnormal accumulation of CD41+ megakary-
ocytic progenitors.68 Loss of expression of full-length GATA1
may also influence the development of TAM by impairing
the growth and differentiation of murine fetal/embryonic
hematopoietic progenitors.69 Overall, the expression of
GATA1s, along with GATA1 dysfunction, appears to be
essential for the development of TAM. A recent study using
genetically engineered human iPSC showed that GATA1s is
upregulated by trisomy 21, leading to impaired megakary-
opoiesis, which indicates that a synergistic interaction between
trisomy 21 and GATA1 mutation is important for development
of TAM.51
Progression from TAM to ML-DS
Factors predictive of ML-DS
It is a fascinating idea that the development of overt leukemia
could be predicted in the pre-leukemic phase. If risk factors
for ML-DS were identified in infants with DS, it could
become possible to design a more adequate follow-up sched-
ule and devise effective methods to prevent progression to
ML-DS. Previous large clinical trials by Pediatric Oncology
© 2018 Japan Pediatric Society
226 K Watanabe
Group (POG), BFM, and COG reported candidate risk factors
for ML-DS.2,3,5 The POG trial reported that additional chro-
mosome abnormalities at the TAM phase were risk factors.2
The BFM group demonstrated that pleural effusion at diagno-
sis of TAM was predictive,3 while the COG trial found that
longer time to regression of TAM was prognostic.5 In these
studies, the use of cytarabine to treat TAM was not identified
as a significant factor predicting development of ML-DS.
Kanezaki et al. showed that the expression level of
GATA1s varied according to the type of GATA1 mutation.70
Moreover, patients with GATA1 mutations leading to low
GATA1s expression had significantly higher risk of progres-
sion to ML-DS.70 This suggests that the type of GATA1 muta-
tion may be a predictive factor for ML-DS, and this potential
association is currently under investigation in a prospective
study conducted by the JCCG.
Technologies such as quantitative polymerase chain reac-
tion (PCR) and NGS enable detection of minimal populations
of clones with GATA1 mutations.18,19,23 Tracing GATA1
mutant clones by such sensitive methods in infants with DS
would help to clarify the behavior of pre-leukemic clones and
assess clonal evolution to ML-DS, which would allow identifi-
cation of patients at high risk of progression to ML-DS.
Additional genetic events involved in progression from
TAM to ML-DS
The mechanism of progression from TAM to ML-DS is an
important research topic, because this unique condition is con-
sidered a useful model for investigation of how full-blown leu-
kemia evolves from the pre-leukemic phase. Investigators
have hypothesized that genetic events additional to trisomy 21
and GATA1 are responsible for the transformation of TAM
cells to ML-DS blasts.
The ML-DS blast cells often carry additional chromosomal
abnormalities, such as dup(1q), del(6q), del(7p), dup(7q), +8,
+11, and del(16q).2,40,71 Moreover, several mutations in cancer-
related genes, including JAK1/2/3, TP53, FLT3, and MPL, have
been identified in cells from patients with ML-DS. Recently,
Yoshida et al. performed comprehensive genetic analysis on
samples from patients with TAM and ML-DS using massively
parallel sequencing.19 The mean number of non-silent mutations
per TAM sample was extremely small compared with that in
other human cancers. Other than common GATA1 mutations,
no other recurrent mutations were identified in TAM samples,
suggesting that development of TAM does not require genetic
changes additional to those of GATA1 mutation and trisomy 21.
In the majority of cases, ML-DS was accompanied by newly
acquired mutations not present in the original TAM samples.
Mutations affecting cohesin/CCCTC-binding factor (65%),
other epigenetic regulators, including enhancer of zeste homo-
log 2 (45%), and RAS/signal transduction molecules were fre-
quently detected as additional mutations in ML-DS.19 Given
that these genes have important roles in human leukemogenesis
or tumorigenesis, such mutations may be essential genetic
events for transformation of TAM blasts.
© 2018 Japan Pediatric Society
It was recently reported that dysregulation of DNA methyla-
tion may be involved in progression from TAM to ML-DS.72
Given that ML-DS is frequently associated with additional
mutations in epigenetic regulators, it is reasonable to speculate
that epigenetic dysregulation could have a critical role in the
transformation of TAM to ML-DS.
Clonal evolution from TAM to ML-DS
It has long been hypothesized that a linear sequential acquisi-
tion of genetic alterations induces disease progression in
TAM/ML-DS.73 Recent studies using high-throughput geno-
mic technology indicate that evolutionary trajectories are more
complex and branching in other cancers and leukemias. In this
schema, genomic instability in founder cells gives rise to
heterogeneous mutant subclones, and, under selective pressure,
some subclones expand, resulting in disease progression, while
others become extinct or remain dormant. Thus, leukemic
clones may evolve and emerge through the complex interac-
tion of selectively advantageous “driver” mutations, additional
advantageous “cooperating” mutations, neutral “passenger”
mutations, and deleterious mutations.74,75 Recent studies indi-
cate that this schema may also be applicable to the progres-
sion from TAM to ML-DS.
Multiple distinct GATA1 mutant clones can occur in a sin-
gle patient with TAM. For example, Xu et al. reported a case
of ML-DS derived from a minor clone with a GATA1 muta-
tion that differed from the mutation in the major clone present
at the TAM phase.76 Moreover, a recent study using whole
exome/genome sequencing of paired TAM and ML/DS sam-
ples showed that ML-DS can arise from minor, as well as
major, TAM clones.19 These observations suggest that hetero-
geneous TAM clones harboring distinct GATA1 mutations are
present during the TAM phase and that subsequent ML-DS
originates from one of multiple TAM subclones through selec-
tion.
A robust xenograft model of TAM was recently estab-
lished using highly immunodeficient NOG mice. This model
enabled the successful observation of clonal selection and
expansion of minor mutant TAM clones.18 In that study, pri-
mary TAM cell samples from three of 11 patients were
engrafted into NOG mice. Interestingly, serial engraftment to
subsequently transplanted recipients was successful using
cells obtained from one patient, indicating that some TAM
clones have long-term self-renewal capacity. Of note, that
patient later developed ML-DS at the age of 1 year. ML-DS
cells often have additional chromosome abnormalities or
copy number alterations (CNA), while TAM cells are rarely
associated with such genetic changes. Extensive serial trans-
plantation was performed using TAM cells from that patient
and CNA analyzed in the propagating cells. The primary
samples had no CNA, other than trisomy 21. Serial trans-
plantation indicated the emergence of subclones with various
additional CNA and diverse repopulating capacities as xeno-
grafts. The genomic structure of a del(16q22) rearrangement
found in one of the engrafted cell samples was determined,

and breakpoint-specific PCR was performed to detect the
presence of cells carrying the del(16q22) change. Surpris-
ingly, on breakpoint-specific PCR a minor clone with del
(16q22) was already present in the primary samples. Exten-
sive transplantation also indicated a distinct GATA1 mutant
that was not a major clone in the primary sample. Yet, geno-
mic analysis showed that the subclone with the distinct
GATA1 mutation was present as a minor clone among the
primary cells, consistent with clinical observations of the
development of ML-DS from a minor TAM clone.18 This
xenograft model is likely to mimic the early phase of the
leukemic evolutionary process, because the TAM blasts have
striking genetic heterogeneity, and subclones with leukemia-
initiating potential may already reside among them during
the pre-leukemic phase. These findings support the hypothe-
sis that ML-DS develops from a pool of heterogeneous
minor clones through clonal selection, illustrating the early
evolutionary process of leukemia.77
As mentioned, it was recently shown that dysregulation of
DNA methylation may be involved in progression from TAM
to ML-DS.72 Analysis using primary blood samples indicated
distinct DNA methylation profiles in TAM and ML-DS
cells.72 Consistent with this finding, the global DNA methyla-
tion profiles of engrafted cells from first-generation mice were
similar to those of TAM patients, whereas those from the
fourth, fifth, and sixth generations approached the profiles of
ML-DS patients.72 Further investigation is ongoing to clarify
the precise mechanisms involved in the development of TAM
and ML-DS.
Future directions
Recent developments in genomic technology have facilitated
the revelation that DS infants with neither apparent clinical
signs of TAM nor increased blasts in peripheral blood may
carry minor GATA1 mutant clones, a condition referred to as
silent TAM. The incidence of silent TAM may vary according
to the sensitivity of the method used to detect GATA1 mutant
clones. It follows that the clinical significance, that is, the
potential risk of developing ML-DS from silent TAM, may
also depend on the methodology used for GATA1 mutation
analysis. Thus, the incidence and significance of silent TAM
requires investigation in different cohorts.
Low-dose cytarabine is effective in improving the outcome
for TAM patients with high WBC count, but a standard dose
and treatment schedule need to be established to ensure safe
and effective use of the drug for neonates with DS. Moreover,
development of novel therapies is required for successful treat-
ment of the most severe form of TAM, complicated with ser-
ous effusion, multi-organ dysfunction, and liver fibrosis,
which results in fulminant hepatic failure.
A number of studies using fetal liver cells, mouse models,
or iPSC have elucidated the roles of trisomy 21 and GATA1
mutations in the development of TAM. Nevertheless, the pre-
cise mechanisms underlying the expansion and regression of
TAM cells remain to be clarified.
Review of TAM 227
Next-generation sequencing of samples from TAM and
ML-DS patients identified additional mutations in genes
involved in epigenetic regulation that potentially transform
pre-leukemia to full-blown leukemia. The xenograft model
demonstrates the genetic heterogeneity of TAM blasts and
mimics how TAM clones expand or become extinct through
clonal selection, leading to progression to ML-DS.18 DNA
methylation analysis, using this model and clinical samples,
suggests that epigenetic dysregulation may be involved in pro-
gression from TAM to ML-DS.
Unraveling the mechanisms involved in leukemogenesis
and identification of factors predictive for the development of
ML-DS may lead to strategies to prevent progression to ML-
DS. Investigation of TAM, a unique pre-leukemic condition,
will continue to result in deep impacts on basic research and
the development of new methods for treatment of hematologi-
cal malignancy.
Acknowledgments
This article is based on research performed in collaboration with
Professor Etsuro Ito and Dr Akira Watanabe, who are supported
by AMED under Grant Number JP17 cm0106407h0002. TAM-
10 was conducted by JCCG as a multi-institutional prospective
study.
Author contribution
K.W. conceived of this review and wrote the manuscript.
Disclosure
The author declares no conflict of interest.
References
1 Zipursky A, Poon A, Doyle J. Leukemia in Down syndrome:
A review. Pediatr. Hematol. Oncol. 1992; 9: 139–49.
2 Massey GV, Zipursky A, Chang MN et al. A prospective
study of the natural history of transient leukemia (TL) in
neonates with Down syndrome (DS): Children’s Oncology
Group (COG) study POG-9481. Blood 2006; 107: 4606–13.
3 Klusmann JH, Creutzig U, Zimmermann M et al. Treatment
and prognostic impact of transient leukemia in neonates with
Down syndrome. Blood 2008; 111: 2991–8.
4 Muramatsu H, Kato K, Watanabe N et al. Risk factors for
early death in neonates with Down syndrome and transient
leukaemia. Br. J. Haematol. 2008; 142: 610–5.
5 Gamis AS, Alonzo TA, Gerbing RB et al. Natural history of
transient myeloproliferative disorder clinically diagnosed in Down
syndrome neonates: A report from the Children’s Oncology
Group Study A2971. Blood 2011; 118: 6752–9; quiz 996.
6 Zipursky A, Brown E, Christensen H, Sutherland R, Doyle J.
Leukemia and/or myeloproliferative syndrome in neonates
with Down syndrome. Semin. Perinatol. 1997; 21: 97–101.
7 Wechsler J, Greene M, McDevitt MA et al. Acquired
mutations in GATA1 in the megakaryoblastic leukemia of
Down syndrome. Nat. Genet. 2002; 32: 148–52.
8 Greene ME, Mundschau G, Wechsler J et al. Mutations in
GATA1 in both transient myeloproliferative disorder and
© 2018 Japan Pediatric Society
228 K Watanabe
acute megakaryoblastic leukemia of Down syndrome. Blood
Cells Mol. Dis. 2003; 31: 351–6.
9 Groet J, McElwaine S, Spinelli M et al. Acquired mutations
in GATA1 in neonates with Down’s syndrome with transient
myeloid disorder. Lancet 2003; 361: 1617–20.
10 Hitzler JK, Cheung J, Li Y, Scherer SW, Zipursky A. GATA1
mutations in transient leukemia and acute megakaryoblastic
leukemia of Down syndrome. Blood 2003; 101: 4301–4.
11 Rainis L, Bercovich D, Strehl S et al. Mutations in exon 2 of
GATA1 are early events in megakaryocytic malignancies
associated with trisomy 21. Blood 2003; 102: 981–6.
12 Xu G, Nagano M, Kanezaki R et al. Frequent mutations in the
GATA-1 gene in the transient myeloproliferative disorder of
Down syndrome. Blood 2003; 102: 2960–8.
13 Ahmed M, Sternberg A, Hall G et al. Natural history of
GATA1 mutations in Down syndrome. Blood 2004; 103:
2480–9.
14 Gamis AS, Smith FO. Transient myeloproliferative disorder in
children with Down syndrome: Clarity to this enigmatic
disorder. Br. J. Haematol. 2012; 159: 277–87.
15 Roy A, Roberts I, Vyas P. Biology and management of
transient abnormal myelopoiesis (TAM) in children with
Down syndrome. Semin. Fetal Neonatal. Med. 2012; 17: 196–
201.
16 Bhatnagar N, Nizery L, Tunstall O, Vyas P, Roberts I.
Transient abnormal myelopoiesis and AML in Down
syndrome: An update. Curr. Hematol. Malig. Rep. 2016; 11:
333–41.
17 Saida S. Evolution of myeloid leukemia in children with
Down syndrome. Int. J. Hematol. 2016; 103: 365–72.
18 Saida S, Watanabe K, Sato-Otsubo A et al. Clonal selection in
xenografted TAM recapitulates the evolutionary process of
myeloid leukemia in Down syndrome. Blood 2013; 121:
4377–87.
19 Yoshida K, Toki T, Okuno Y et al. The landscape of somatic
mutations in Down syndrome-related myeloid disorders. Nat.
Genet. 2013; 45: 1293–9.
20 Girodon F, Favre B, Couillaud G, Carli PM, Parmeland C,
Maynadie M. Immunophenotype of a transient
myeloproliferative disorder in a newborn with trisomy 21.
Cytometry 2000; 42: 118–22.
21 Karandikar NJ, Aquino DB, McKenna RW, Kroft SH.
Transient myeloproliferative disorder and acute myeloid
leukemia in Down syndrome. An immunophenotypic analysis.
Am. J. Clin. Pathol.. 2001; 116: 204–10.
22 Langebrake C, Creutzig U, Reinhardt D. Immunophenotype of
Down syndrome acute myeloid leukemia and transient
myeloproliferative disease differs significantly from other
diseases with morphologically identical or similar blasts. Klin.
Padiatr. 2005; 217: 126–34.
23 Roberts I, Alford K, Hall G et al. GATA1-mutant clones are
frequent and often unsuspected in babies with Down
syndrome: Identification of a population at risk of leukemia.
Blood 2013; 122: 3908–17.
24 Hoshi N, Hattori R, Hanatani K et al. Recent trends in the
prevalence of Down syndrome in Japan, 1980-1997. Am. J.
Med. Genet. 1999; 84: 340–5.
25 Pine SR, Guo Q, Yin C, Jayabose S, Druschel CM, Sandoval
C. Incidence and clinical implications of GATA1 mutations in
newborns with Down syndrome. Blood 2007; 110: 2128–31.
26 Strobelt N, Ghidini A, Locatelli A, Vergani P, Mariani S,
Biondi A. Intrauterine diagnosis and management of transient
myeloproliferative disorder. Am. J. Perinatol. 1995; 12: 132–4.
27 Zipursky A, Rose T, Skidmore M, Thorner P, Doyle J.
Hydrops fetalis and neonatal leukemia in Down syndrome.
Pediatr. Hematol. Oncol. 1996; 13: 81–7.
© 2018 Japan Pediatric Society
28 Hojo S, Tsukimori K, Kitade S et al. Prenatal sonographic
findings and hematological abnormalities in fetuses with
transient abnormal myelopoiesis with Down syndrome.
Prenat. Diagn. 2007; 27: 507–11.
29 Tamblyn JA, Norton A, Spurgeon L et al. Prenatal therapy in
transient abnormal myelopoiesis: A systematic review. Arch.
Dis. Child. Fetal Neonatal Ed. 2016; 101: F67–71.
30 Heald B, Hilden JM, Zbuk K et al. Severe TMD/AMKL with
GATA1 mutation in a stillborn fetus with Down syndrome.
Nat. Clin. Pract. Oncol. 2007; 4: 433–8.
31 Ishigaki H, Miyauchi J, Yokoe A et al. Expression of
megakaryocytic and myeloid markers in blasts of transient
abnormal myelopoiesis in a stillbirth with Down syndrome:
Report of histopathological findings of an autopsy case. Hum.
Pathol. 2011; 42: 141–5.
32 Federmann B, Fasan A, Kagan KO, Haen S, Fend F. Transient
abnormal myelopoiesis/acute megakaryoblastic leukemia
diagnosed in the placenta of a stillborn Down syndrome fetus
with targeted next-generation sequencing. Leukemia 2015; 29:
232–3.
33 Yagihashi N, Watanabe K, Yagihashi S. Transient abnormal
myelopoiesis accompanied by hepatic fibrosis in two infants
with Down syndrome. J. Clin. Pathol. 1995; 48: 973–5.
34 Schwab M, Niemeyer C, Schwarzer U. Down syndrome,
transient myeloproliferative disorder, and infantile liver
fibrosis. Med. Pediatr. Oncol. 1998; 31: 159–65.
35 Park MJ, Sotomatsu M, Ohki K et al. Liver disease is
frequently observed in Down syndrome patients with transient
abnormal myelopoiesis. Int. J. Hematol. 2014; 99: 154–61.
36 Ogawa J, Kanegane H, Tsuneyama K et al. Platelet-derived
growth factor may be associated with fibrosis in a Down
syndrome patient with transient myeloproliferative disorder.
Eur. J. Haematol. 2008; 81: 58–64.
37 Hattori H, Matsuzaki A, Suminoe A, Ihara K, Nakayama H,
Hara T. High expression of platelet-derived growth factor and
transforming growth factor-beta 1 in blast cells from patients
with Down Syndrome suffering from transient
myeloproliferative disorder and organ fibrosis. Br. J.
Haematol. 2001; 115: 472–5.
38 Maroz A, Stachorski L, Emmrich S et al. GATA1s induces
hyperproliferation of eosinophil precursors in Down syndrome
transient leukemia. Leukemia 2014; 28: 1259–70.
39 Minakata S, Sakata N, Wada N et al. Liver fibrosis with
hypereosinophilia causing transient abnormal myelopoiesis.
Pediatr. Int. 2016; 58: 1222–5.
40 Hayashi Y, Eguchi M, Sugita K et al. Cytogenetic findings
and clinical features in acute leukemia and transient
myeloproliferative disorder in Down’s syndrome. Blood 1988;
72: 15–23.
41 Ge Y, Jensen TL, Stout M et al. The role of cytidine
deaminase and GATA1 mutations in the increased cytosine
arabinoside sensitivity of Down syndrome myeloblasts and
leukemia cell lines. Cancer Res. 2004; 64: 728–35.
42 Muramatsu H, Watanabe T, Hasegawa D et al. Prospective
study of 168 infants with transient abnormal myelopoiesis
with Down syndrome: Japan Pediatric Leukemia/Lymphoma
Study Group, TAM-10 Study. Abstract 1311; 57th Annual
Meeting of the American Society of Hematology, 2015. Blood
2015; 126: 1311.
43 Flasinski M, Scheibke K, Zimmermann M et al. Low-dose
cytarabine to prevent myeloid leukemia in children with Down
syndrome: TMD Prevention 2007 study. Blood Adv. 2018; 2:
1532–40.
44 Chou S, Lodish HF. Fetal liver hepatic progenitors are
supportive stromal cells for hematopoietic stem cells. Proc.
Natl Acad. Sci. USA 2010; 107: 7799–804.

45 Chou ST, Opalinska JB, Yao Y et al. Trisomy 21 enhances
human fetal erythro-megakaryocytic development. Blood
2008; 112: 4503–6.
46 Tunstall-Pedoe O, Roy A, Karadimitris A et al. Abnormalities
in the myeloid progenitor compartment in Down syndrome
fetal liver precede acquisition of GATA1 mutations. Blood
2008; 112: 4507–11.
47 Roy A, Cowan G, Mead AJ et al. Perturbation of fetal liver
hematopoietic stem and progenitor cell development by
trisomy 21. Proc. Natl Acad. Sci. USA 2012; 109: 17579–84.
48 Liu B, Filippi S, Roy A, Roberts I. Stem and progenitor cell
dysfunction in human trisomies. EMBO Rep. 2015; 16: 44–62.
49 Chou ST, Byrska-Bishop M, Tober JM et al. Trisomy 21-
associated defects in human primitive hematopoiesis revealed
through induced pluripotent stem cells. Proc. Natl Acad. Sci.
USA 2012; 109: 17573–8.
50 Maclean GA, Menne TF, Guo G et al. Altered hematopoiesis
in trisomy 21 as revealed through in vitro differentiation of
isogenic human pluripotent cells. Proc. Natl Acad. Sci. USA
2012; 109: 17567–72.
51 Banno K, Omori S, Hirata K et al. Systematic cellular disease
models reveal synergistic interaction of trisomy 21 and
GATA1 mutations in hematopoietic abnormalities. Cell Rep.
2016; 15: 1228–41.
52 Delabar JM, Theophile D, Rahmani Z et al. Molecular
mapping of twenty-four features of Down syndrome on
chromosome 21. Eur. J. Hum. Genet. 1993; 1: 114–24.
53 Ronan A, Fagan K, Christie L, Conroy J, Nowak NJ, Turner G.
Familial 4.3 Mb duplication of 21q22 sheds new light on the
Down syndrome critical region. J. Med. Genet. 2007; 44: 448–51.
54 Bourquin JP, Subramanian A, Langebrake C et al.
Identification of distinct molecular phenotypes in acute
megakaryoblastic leukemia by gene expression profiling. Proc.
Natl Acad. Sci. USA 2006; 103: 3339–44.
55 Stankiewicz MJ, Crispino JD. ETS2 and ERG promote
megakaryopoiesis and synergize with alterations in GATA-1
to immortalize hematopoietic progenitor cells. Blood 2009;
113: 3337–47.
56 Malinge S, Bliss-Moreau M, Kirsammer G et al. Increased
dosage of the chromosome 21 ortholog Dyrk1a promotes
megakaryoblastic leukemia in a murine model of Down
syndrome. J. Clin. Invest. 2012; 122: 948–62.
57 Ichikawa M, Yoshimi A, Nakagawa M, Nishimoto N, Watanabe-
Okochi N, Kurokawa M. A role for RUNX1 in hematopoiesis and
myeloid leukemia. Int. J. Hematol. 2013; 97: 726–34.
58 Kirsammer G, Jilani S, Liu H et al. Highly penetrant
myeloproliferative disease in the Ts65Dn mouse model of
Down syndrome. Blood 2008; 111: 767–75.
59 Carmichael CL, Majewski IJ, Alexander WS et al.
Hematopoietic defects in the Ts1Cje mouse model of Down
syndrome. Blood 2009; 113: 1929–37.
60 Alford KA, Slender A, Vanes L et al. Perturbed
hematopoiesis in the Tc1 mouse model of Down syndrome.
Blood 2010; 115: 2928–37.
Review of TAM 229
61 Weiss MJ, Orkin SH. Transcription factor GATA-1 permits
survival and maturation of erythroid precursors by preventing
apoptosis. Proc. Natl Acad. Sci. USA 1995; 92: 9623–7.
62 Ferreira R, Ohneda K, Yamamoto M, Philipsen S. GATA1
function, a paradigm for transcription factors in hematopoiesis.
Mol. Cell. Biol. 2005; 25: 1215–27.
63 Crispino JD, Horwitz MS. GATA factor mutations in
hematologic disease. Blood 2017; 129: 2103–10.
64 Freson K, Wijgaerts A, Van Geet C. GATA1 gene variants
associated with thrombocytopenia and anemia. Platelets 2017;
28: 731–4.
65 Salek-Ardakani S, Smooha G, de Boer J et al. ERG is a
megakaryocytic oncogene. Cancer Res. 2009; 69: 4665–73.
66 Toki T, Kanezaki R, Kobayashi E et al. Naturally occurring
oncogenic GATA1 mutants with internal deletions in transient
abnormal myelopoiesis in Down syndrome. Blood 2013; 121:
3181–4.
67 Li Z, Godinho FJ, Klusmann JH, Garriga-Canut M, Yu C,
Orkin SH. Developmental stage-selective effect of somatically
mutated leukemogenic transcription factor GATA1. Nat.
Genet. 2005; 37: 613–9.
68 Shimizu R, Kobayashi E, Engel JD, Yamamoto M. Induction
of hyperproliferative fetal megakaryopoiesis by an N-
terminally truncated GATA1 mutant. Genes Cells 2009; 14:
1119–31.
69 Stachura DL, Chou ST, Weiss MJ. Early block to
erythromegakaryocytic development conferred by loss of
transcription factor GATA-1. Blood 2006; 107: 87–97.
70 Kanezaki R, Toki T, Terui K et al. Down syndrome and
GATA1 mutations in transient abnormal myeloproliferative
disorder: Mutation classes correlate with progression to
myeloid leukemia. Blood 2010; 116: 4631–8.
71 Forestier E, Izraeli S, Beverloo B et al. Cytogenetic features
of acute lymphoblastic and myeloid leukemias in pediatric
patients with Down syndrome: An iBFM-SG study. Blood
2008; 111: 1575–83.
72 Saida S, Toki T, Nishinaka-Arai Y et al. Dysregulation of
DNA methylation involves in progression of myeloid
leukemia in Down syndrome. Abstract; 57th Annual Meeting
of the American Society of Hematology, 2015. Blood 2015;
126: 89.
73 Khan I, Malinge S, Crispino J. Myeloid leukemia in Down
syndrome. Crit. Rev. Oncog. 2011; 16: 25–36.
74 Nowell PC. The clonal evolution of tumor cell populations.
Science 1976; 194: 23–8.
75 Welch JS, Ley TJ, Link DC et al. The origin and evolution of
mutations in acute myeloid leukemia. Cell 2012; 150: 264–78.
76 Xu G, Kato K, Toki T, Takahashi Y, Terui K, Ito E.
Development of acute megakaryoblastic leukemia from a
minor clone in a Down syndrome patient with clinically overt
transient myeloproliferative disorder. J. Pediatr. Hematol.
Oncol. 2006; 28: 696–8.
77 Greaves M, Maley CC. Clonal evolution in cancer. Nature
2012; 481: 306–13.
© 2018 Japan Pediatric Society