Pediatr Blood Cancer 2005;44:29–32

Transient Leukemia in Newborns With Down Syndrome

Gita V. Massey, MD*

Down syndrome (DS) is the most common congenitalchromosomal disorder in humans, having an estimatedincidence of 1/600 to 1/1,000 live births. The numerousdevelopmental abnormalities and congenital defects ofDS are well recognized and include cardiac and gut mal-formations and various degrees of mental retardation.Children with DS also have a 10- to 20-fold increased riskof developing leukemia [1–3]. In the first 4 years of life themajority of cases of leukemia in these patients are acutenon-lymphocytic leukemia, particularly acute megakar-yocytic leukemia (AMKL). The incidence of AMKL is upto 500 times higher in DS than in normal children [4,5].After the age of 4 years, the ratio of acute non-lymphocyticleukemia to lymphocytic leukemia approaches that of thenormal population of children, however the incidenceof leukemia still remains about ten times higher in DSchildren than in normal children.

One form of leukemia occurs only in newborn infantswith DS (or trisomy 21 mosaics). Frequently referred to astransient myeloproliferative disorder (TMD) or transientleukemia (TL), the first case was reported by Schunk andLehman [6] in 1954. This disorder has historically beenassociated with a high incidence of spontaneous remis-sion. At presentation many infants may be clinically wellwith an incidental finding of circulating blasts in theperipheral blood. In some cases, however, the disease issevere and life-threatening, manifesting as hydrops fetalis,multiple effusions, and liver or multi-organ system failure[7–17].

Most of the information about the natural history andbiology of this TL has been gleaned from sporadic case

reports and/or retrospective reviews of medical and/orpathologic records [4,5,15,17]. Although the disease mayspontaneously disappear, a significant percentage ofpatients with TL go on to develop AMKL within the first4 years of life, which if left untreated rarely shows spon-taneous regression a second time [4,5,15,17–19]. ThePediatric Oncology Group Study 9481 was the first studydesigned to prospectively study the clinical presentationand natural history of DS children (or trisomy 21 mosaics)with TL, to determine the immunophenotypic, cytoge-netic, and clonal characteristics of the blast cells in thisdisorder, and to assess their prognostic significance forthe development of acute non-lymphocytic leukemia. Thedata from this study are currently undergoing analysis.Thus the remainder of this review will focus on infor-mation gathered from retrospective studies, reviews, andcase reports.

Of importance is that, to date, there have been noprospective population-based studies among patients withDS to determine the true incidence of TL. Although allinfants with DS are screened for congenital heart disease,no current recommendations exist for ‘‘screening’’ bloodcounts in these infants. It is, however estimated that the

Children with Down syndrome (DS) have a10- to 20-fold increased risk of developingleukemia, particularly acute megakaryocyticleukemia. Newborns with DS or trisomy 21mosaicism may exhibit a particularly uniqueform of leukemia that historically has beenassociated with a high rate of spontaneous re-mission. This transient leukemia (TL) has beenshown to be a clonal proliferation of blast cellsexhibiting megakaryocytic features. Its true inci-dence remains to be determined. At presenta-tion, many infants are clinically well with onlyan incidental finding of abnormal blood countsand circulating blasts in the peripheral blood.However, in approximately 20% of cases, the

disease is severe and life-threatening, manifest-ing as hydrops faetalis, multiple effusions, andliver or multi-organ system failure resulting indeath. Of those children who enter a sponta-neous remission, 13–33% have been found todevelop subsequent acute megakaryoblasticleukemia, usually within the first 3 years of life,which if left untreated is fatal. This unique TL ofthe DS newborn has been the subject of recentclinical cooperative group trials as well as manybiological and genetic research efforts. We sum-marize here the known clinical, biological, andcytogenetic features of TL associated with DS.Pediatr Blood Cancer 2005;44:29–32.� 2004 Wiley-Liss, Inc.

Key words: leukemia; Down syndrome; megakaryoblastic leukemia

——————Department of Pediatrics, VCU Health System, Medical College of

Virginia, Richmond, Virginia

*Correspondence to: Gita V. Massey, Department of Pediatrics, VCU

Health System, Medical College of Virginia, Richmond, Virginia, PO

Box 980121, Richmond, VA. E-mail: gvmassey@hsc.vcu.edu

Received 10 March 2004; Accepted 2 June 2004

� 2004 Wiley-Liss, Inc.DOI 10.1002/pbc.20141

incidence may range from 10 to 25% of newborns with DS[20]. Affected neonates may appear entirely healthy ormay have fulminant hepatic and multi-organ systemfailure leading to death in utero or in infancy [1,15–17,21,22].

Two retrospective studies exist which characterize theclinical findings associated with DS and TL. The first is areview of 95 cases by Homans et al. [15] collected from aquestionnaire survey of pediatric oncologists and from theliterature. The second, is a recent review of 69 neonateswith DS and perinatal leukemia from the literature byIsaacs [17]. The clinical and laboratory data from thesetwo studies are summarized in the Table I.

Although historically, TL has been shown to sponta-neously resolve in most cases, both Homans and Isaacsdata showed a significant mortality. In Homans review,11% of the patients died during the initial event withcauses of death including sepsis, hemorrhage, dissemi-nated intravascular coagulation, hyperviscosity, and heartfailure [15]. Isaacs review found an even greater mortalityof 55% of the patients dying during the newborn period[17]. This included five patients who were stillborn.Several case reports attest to the association of hydropsfetalis and neonatal leukemia in DS [1,23,24].

In the POG 9481 prospective study, 8 of 47 patients(17%) experienced early death at a mean of 90 days [16].All of the children who experienced early death ultimatelyhad liver failure and disseminated intravascular coagula-tion as terminal events. Ascitic, pleural, and pericardial

effusions were frequent complicating features. Liver biop-sies were done in two of these children and pathologi-cally demonstrated hepatic fibrosis with extramedullaryhematopoisis and leukemic infiltrate with megakaryocyticdifferentiation. This unusual diffuse liver fibrosis accom-panying TL in DS infants has also been previouslyreported [11–13,25]. It is speculated that the infiltratingmegakaryoblasts through the release of cytokines, such asplatelet derived growth factor and transforming growthfactor beta are the cause of the fibrosis [26,27]. Anothersuggested mechanism is that the blasts in TL originatefrom fetal liver, recapitulating fetal hematopoiesis, andcausing liver fibrosis [13]. Early therapy with low dosecytosine arabinoside may be effective in preventing thislethal complication [16].

Of those children who enter a spontaneous remission13–33% have been found to develop a subsequenthematologic disorder, most frequently acute megakaryo-blastic leukemia [15,17]. The subsequent leukemia wasusually diagnosed at less than 3 years of age with a meanage of 16 months [15]. Homans et al. concluded from theirstudy that there were no initial clinical or hematologicfeatures that predicted the development of subsequentleukemia [15]. However infants who subsequently de-veloped acute megakaryoblastic leukemia had a goodprognosis, provided they were treated with appropriatechemotherapy regimens that included cytosine arabino-side [28–31]. Increased expression of the chromosome 21localized genes cystathione-B-synthetase and superoxide

TABLE I. Clinical Characteristics at Presentation of Patients With Transient Leukemia andDown Syndrome

Homans et al. [15] (n¼ 95) Isaacs [17] (n¼ 69)

Age at onset 2 days (median) 5/69 (7%) Dx prenatally

Sex

Male 38/69 (55%)

Female 31/69 (45%)

Hepatosplenomegaly 69% 85%

Leukocyte count (�109/L) 5.0–384.0 9.1–1300.0

Hemoglobin (g/dl) 4.0–23.2 8.8–23.3

Platelets (per ml) 5,000–1,800,000

Type of blast cells

Megakaryoblasts 76.8%

Myoloblasts (NOS) 11.6%

Myoloblastsa 7.3%

Lymphoblasts 1.4%

Erythroblasts 1.4%

Undifferentiated 1.4%

Cytogenetics

Trisomy 21 only 91% 90%

Mosaics 20% 17%

Outcome

Neonatal mortality 11% 52%b

Relapse rate 29% 13%

NOS, not otherwise specified.aIncludes myoloblasts with lymphoblasts, megakaryoblasts, and erythroblasts.bIncludes five stillborn patients.

30 Massey

dismutase, the latter being associated with increasedapoptosis, are implicated in the increased sensitivity ofthese cells to cytosine arabinoside [32,33].

TL has also been shown to occur in phenotypicallynormal neonates with trisomy 21 mosaicism [15,17,19].These patients also usually undergo spontaneous resolu-tion of their blast cells. It was thought that these infantsmay be at lower risk for developing subsequent leukemia.None of the 19 trisomy 21 mosaic patients in Homans et al.series developed subsequent hematologic abnormalities[15]. However, in Isaacs study, two of ten patients withtrisomy 21 leukemic cell mosaicism developed AMKL 1and 2 years, respectively after neonatal TL [17]. Similarly,additional case reports of trisomy 21 mosaicism andrecurrent leukemia exist [19].

Many studies have attempted to characterize the bio-logic features of the blasts of TL. Case studies from thepast have suggested a mixed phenotype or even an un-differentiated leukemia [34,35]. Immunohistochemicalmarkers point toward megakaryocytic lineage as thesecells exhibit factor VIII, glycoprotein IIb/IIIa and plateletperoxidase by electron microscopy [36]. However, thepresence of dyserythropoiesis and basophilia has also beenreported in the blood smears and bone marrow aspirates ofthese patients, thus attesting to the blast cells’ capability ofdifferentiating along other cell lineages [37–39].

In cases where cytogenetics were obtained at diagnosis,the majority of patients exhibited trisomy 21 as the onlycytogenetic abnormality. However, the clonal nature ofthis disorder has been established by analysis of methyla-tion patterns of hypoxanthine phosphoribosyltransferase(HPRT) genes located on the X chromosome [40,41].Since TL appears to be restricted to patients with DS ortrisomy 21 mosaicism, the origin of TL must in some waybe related to a cytogenetic abnormality in chromosome 21[42]. Whether this is simply a result of increased genedosage, or a result of disomic homozygosity remains to bedetermined [43–45]. Candidate genes for leukemogenicpotential that are expressed on chromosome 21 include theFPDMM gene (associated with the autosomal dominantfamilial platelet disorder) [46], AML1 gene (associatedwith FAB M2 AML) [47], as well as IFN-a/B receptor(IFNAR), CFR2-4 (cytokine family 2–3), and phospho-ribosylglycinamide formyltransferase [42].

More recently several reports have implicated themutagenesis of hematopoietic transcription factor geneGATA 1 as an intitiating event in DS leukemogenesis. TheGATA 1 gene, located on the X chromosome encodesa zinc-finger transcription factor that is essential fornormal erythroid and megakaryocytic differentiation[48,49]. Somatic mutations in exon 2 of GATA 1 havebeen detected exclusively in trisomy 21 associated TL aswell as AMKL, but not in non-DS related AMKL [50–52].Its role in erythroid as well as megakaryocytic differen-tiation may help explain the multi-lineage features of the

TL blast cells and may also contribute to the increasedchemotherapy sensitivity of these blast cells [53].

There remain many unanswered questions about theunique TL of DS infants. From a clinical perspective, wehave yet to determine the true incidence of this disorder, aswell as why some infants are affected with a lethal form ofhepatic fibrosis shortly after birth or prenatally, whereasothers are virtually asymptomatic. How and who should betreated and why leukemia recurs in up to 30% of patientsare additional clinical challenges and questions for furtherclinical trials. The fact that the majority of cases resolvespontaneously is in itself intriguing. Further elucidation ofthe unique biologic features of TL and trisomy 21 (itsevolution and spontaneous regression) are thereforelikely to contribute to our understanding of the complexmechanisms of leukemogenesis as well as normalhematopoiesis.

REFERENCES

1. Zipursky A, Rose T, Skidmore M, et al. Hydrops fetalis and neo-

natal leukemia in Down syndrome. Pediatr Hematol Oncol 1996;

13:81–87.

2. Broadbent VA. 1992. Malignant disease in the neonate. In:

Roberton NRC, editor. Textbook of neonatology. 2nd edn.

Edinburgh: Churchill Livingstone; 879 p.

3. Robison LL, Nesbit ME, Sather HN, et al. Down syndrome and

acute leukemia in children: A 10-year retrospective survey from

Children’s Cancer Study Group. J Pediatr 1984;105:235–242.

4. Zipursky A, Poon A, Doyle J. Leukemia in Down syndrome:

A review. Pediatr Hematol Oncol 1992;9:139–149.

5. Zipursky A, Brown E, Christensen H, et al. Leukemia and/or

myeloproliferative syndrome in neonates with Down syndrome.

Sem Perinat 1997;21:97–101.

6. Schunk GJ, Lehman WL. Mongolism and congenital leukemia.

JAMA 1954;155(3):250–251.

7. Weinstein HJ. Congenital leukemia and the neonatal myeloproli-

ferative disorders associated with Down syndrome: A report. Clin

Haematol 1978;7:147–154.

8. Wong KY, Jones MM, Svrivastava AK, et al. Transient myelopro-

liferative disorder and acute nonlymphoblastic leukemia in Down

syndrome. J Pediatr 1988;112:18–22.

9. Paolucci G, Rosito P. Neonatal myeloproliferative disorders in

Down syndrome and congenital leukemias. Haematologica 1987;

72:121–125.

10. Nakagawa T, Nishida H, Arai T, et al. Hyperviscosity syndrome

with transient abnormal myelopoiesis in Down syndrome. J Pediatr

1988;112:58–61.

11. Becroft DMO, Zwi LJ. Perinatal visceral fibrosis accompanying the

megakaryoblastic leukemoid reaction to Down syndrome. Pediatr

Pathol 1990;10:397–406.

12. Ruchelli ED, Uri A, Dimmick JE, et al. Severe perinatal liver

disease and Down syndrome: An apparent relationship. Hum

Pathol 1991;22:1274–1280.

13. Miyauchi J, Ito Y, Kawano T, et al. Unusual diffuse liver fibrosis

accompanying transient myeloproliferative disorder in Down

syndrome: A report of four autopsy cases and proposal of a

hypothesis. Blood 1992;80:1521–1527.

14. Foucar K, Friedman K, Llewellyn A, et al. Prenatal diagnosis of

transient myeloproliferative disorder via percutaneous umbilical

blood sampling. Report of two cases in fetuses affected by Down

syndrome. Am J Clin Pathol 1992;97:584–590.

Transient Leukemia in Newborns With DS 31

15. Homans AC, Verissimo AM, Vlacha V. Transient abnormal myelo-

poiesis of infancy associated with trisomy 21. Am J Pediatr

Hematol Oncol 1993;15:392–399.

16. Al-Kasim F, Doyle JJ, Massey GV, et al. Incidence and treatment of

potentially lethal diseases in transient leukemia of Down syn-

drome: Pediatric Oncology Group Study. J Pediatr Hematol Oncol

2002;24:9–13.

17. Issacs H. Fetal and neonatal leukemia. J Pediatr Hematol Oncol

2003;25:348–361.

18. Barnett PL, Clark AC, Garson OM. Acute nonlymphocytic after

transient myeloproliferative disorder in a patient with Down

syndrome. Med Pediatr Oncol 1990;18:347–353.

19. Doyle JJ, Thorner P, Poon A, et al. Transient leukemia followed

by megakaryoblastic leukemia in a child with mosaic Down

syndrome. Leuk Lymphoma 1995;17:345–350.

20. Zipursky A, Brown EJ, Christensen H, et al. Transient myelopro-

liferative disorder (transient leukemia) and hematologic manifes-

tations. Clin Lab Med 1999;19:157–167.

21. Baschat AA, Wagner T, Malisius R, et al. Prenatal diagnosis of a

transient myeloproliferative disorder in trisomy 21. Prenat Diagn

1998;18:731–736.

22. de Tar MW, Dittman W, Gilbert J. Transient myeloproliferative

disease of the newborn: Case report with placental, cytogenetic,

and flow cytometric findings. Hum Pathol 2000;31:396–398.

23. Donnenfeld AE, Scott SC, Henselder-Kimmel M, et al. Prenatally

diagnosed non-immune hydrops caused by congenital transient

leukemia. Prenat Diagn 1994;14:721–724.

24. Hendricks S, Sorensen TT, Baker E. Trisomy 21, fetal hydrops, and

anemia: Prenatal diagnosis of transient myeloproliferative dis-

order. Obstet Gynecol 1993;82:703–705.

25. Yagihashi N, Watanabe K, Yagihashi S. Transient abnormal myelo-

poiesis accompanied by hepatic fibrosis in two infants with Down

syndrome. J Clin Pathol 1995;48:973–975.

26. Breton-Gorius J, Bizet M, Reyes F, et al. Myelofibrosis and acute

megakaryoblastic leukemia in a child: Topographic relationship

between fibroblasts and megakaryocytes with alpha granule defect.

Leuk Res 1982;6:97–110.

27. Terui T, Niitsu Y, Mahara K, et al. The production of transforming

growth factor beta in acute megakaryoblastic leukemia and its

possible implication in myelofibrosis. Blood 1990;75:1540–1548.

28. Ravindranath Y, Abella E, Krischer J, et al. Acute myeloid

leukemia (AML) in Down syndrome is highly responsive to

chemotherapy: Experience on Pediatric Oncology AML study

8498. Blood 1992;80:2210–2214.

29. Cruetzig U, Ritter J, Vormoor J, et al. Myelodysplasia and acute

myelogenous leukemia in Down syndrome. A report of 40 children

of the AML-BFM study group. Leukemia 1996;1677–1686.

30. Lange B, Kobrinsky N, Barnard D, et al. Distinctive demography,

biology, and outcome of acute myeloid leukemia and myelodys-

plastic syndrome in children with Down syndrome: Children’s

Cancer Group studies 2861 and 2891. Blood 1998;91:608–615.

31. Kojima S, Sako M, Kato K, et al. An effective chemotherapeutic

regimen for acute myeloid leukemia and myelodysplastic syndrome

in children with Down syndrome. Leukemia 2000;14:786–791.

32. Taub J, Matheraly L, Stout M, et al. Enhanced metabolism of 1-B-

D-arabinofuranosylcytosine in Down syndrome cells: A contribut-

ing factor to the superior event free survival of Down syndrome

children with acute myeloid leukemia. Blood 1996;87:3395–3403.

33. Taub J, Huang X, Ge Y, et al. Cystathionine-beta-synthase cDNA

transfection alters the sensitivity and metabolism of 1-beta-D-

arabinofuranosylcytosine in CCRF-CEM leukemia cells in vitro

and in vivo: A model of leukemia in Down syndrome. Cancer Res

2000;60:6421–6426.

34. Yumura-Yagi K, Hara J, Kurahashi H, et al. Mixed phenotype of

blasts in acute megakaryocytic leukemia and transient abnormal

myelopoiesis in Down syndrome. Br J Haematol 1992;81:520–

525.

35. Hayashi Y, Eguchi M, Sugita K, et al. Cytogenetic findings and

clinical features in acute leukemia and transient myeloproliferative

disorder in Down syndrome. Blood 1988;72:15–223.

36. Zipursky A, Christensen H, De Harven E. Ultrastructural studies of

megakaryoblastic leukemias of Down syndrome. Leuk Lymphoma

1995;18:341–347.

37. Worth L, Zipursky A, Christensen H, et al. Transient leukemia with

extreme basophilia in a phenotypically normal infant with blast

cells containing a pseudodiploid clone, 46,XY I(21(q10)). J Pediatr

Hematol Oncol 1999;21:63–66.

38. Suda T, Suda J, Miura Y, et al. Clonal analysis of basophil

differentiation in bone marrow cultures from a Down syndrome

patient with megakaryoblastic leukemia. Blood 1985;66:1278–

1283.

39. Ito E, Kasai M, Hayashi Y, et al. Expression of erythroid specific

genes in acute megakaryoblastic leukemia and transient myelo-

proliferative disorder in Down syndrome. Br J Haematol 1995;

90:607–614.

40. Kurahashi H, Hara J, Yumura-Yagi K, et al. Monoclonal nature

of transient abnormal myelopoiesis in Down syndrome. Blood

1991;77:1161–1163.

41. Miyashita T, Asada M, Fujimoto J, et al. Clonal analysis of transient

myeloproliferative disorder in Down syndrome. Leukemia 1991;5:

56–59.

42. Cavani S, Perfumo C, Argusti A, et al. Cytogenetic and molecular

study of 32 Down syndrome families: Potential leukaemia pre-

disposing role of the most proximal segment of chromosome 21q.

Br J Haematol 1998;103:213–216.

43. Abe K, Kajii T, Niikawa N. Disomic homozygosity in 21-trisomic

cells: A mechanism responsible for transient myeloproliferative

disorder. Hum Genet 1989;82:213–216.

44. Lorber BJ, Freeman SB, Hassold T, et al. Characterization and

molecular analysis of nondisjunction in 18 cases of trisomy 21 and

leukemia. Genes Chromosomes Cancer 1992;4:222–227.

45. Shea J, Williams B, Zipursky A, et al. Cytogenetic and molecular

studies of Down syndrome individuals with leukemia. Am J Hum

Genet 1995;56:915–925.

46. Ho C, Otterud B, Legare R, et al. Linkage of a familial platelet

disorder with a propensity to develop myeloid malignancies to

human chromosome 21q22.1-22.2. Blood 1996;87:5218–5224.

47. Okuda T, van Deursen J, Hiebert S, et al. AML 1, the target of

multiple chromosomal translocations in human leukemia, is es-

sential for normal fetal liver hematopoiesis. Cell 1996;84:321–

330.

48. Martin D, Zon L, Mutter G, et al. Expression of an erythroid

transcription factor in megakaryocytic and mast cell lineages.

Nature 1990;344:444–447.

49. Romeo P, Prandini M, Joulin V, et al. Megakaryocytic and erythro-

cytic lineages share specific transcription factors. Nature 1990;344:

447–449.

50. Wechsler J, Greene M, McDevitt MA, et al. Acquired mutations in

GATA1 in the megakaryoblastic leukemia of Down syndrome. Nat

Genet 2002;32:148–152.

51. Mundschau G, Gurbuxani S, Gamis AS, et al. Mutagenesis of

GATA1 is an intitating event in Down syndrome leukemogenesis.

Blood 2003;101:4298–4300.

52. Hitzler JK, Cheung J, Li Y, et al. GATA1 mutations in transient

leukemia and acute megakaryoblastic leukemia of Down syn-

drome. Blood 2003;101:4301–4340.

53. Ge Y, Tatman-Lily D, Stout ML, et al. Mutations of the GATA1

transcription factor alter the cytosine Arabinoside (ara-C) sensi-

tivity of Down syndrome (DS) megakaryoblasts. Proc AACR

2004.

32 Massey