haematologica online 2004

Leukaemia or Leukaemoid, Down Syndrome or not?

Case Report Baby WM was the first live born child of the family. Her 18-year-old mother had swollen ankles, lower abdominal pain and reduced fetal movements at 31st weeks of gestation. Ultrasound scan showed fetal cardiomegaly and early hydropic changes including hepatomegaly and swollen placenta. No maternal-fetal haemorrhage could be demonstrated by Kleihauer test. A baby girl was delivered one week later by Caesarian section. Her birth weight was 1830 grams. Apgar scores were nine at first minute and ten at fifth minute. She was phenotypically normal. She had moderate pallor but no lymphadenopathy. There were no blue-moffin rashes. A grade 2/6 ejection systolic murmur was heard over the left sternal border and of characters in keeping with flow murmur. Her abdomen was distended and the liver edge was felt at 6 cm below the costal margin. The spleen was palpable at 4 cm below the costal margin She had respiratory distress that necessitated nasal CPAP support. CXR showed cardiomegaly with streaky lung fields. There was no pleural effusion. Echocardiogram showed a structurally normal heart except a small patent ductus arteriosus. There was no pericardial effusion. Blood count showed high white cell count at 73109/L, with blast cell predominance (47109/L). The haemoglobin was 9.7 g/dL and platelet count was 135109/L. Smear found the blast cells were morphologically undifferentiated. Circulating nucleated red cells showed features of dyserythropoiesis. The blast cells were negative for myeloperoxidase and Sudan black B on cytochemical staining. Immunophenotyping of the blast cells showed expression of megakaryocytic markers CD42b and CD61, as well as CD7. A panel of myeloid, B-cell and other T-cell lineage markers were negative. Cytogenetic study of her peripheral blood detected 47,XX,+21[16] on overnight culture and a mosaic pattern of 47,XX,+21[4]/46,XX[14] on PHA-stimulated culture. She had severe unconjugated hyperbilirubinaemia and successfully controlled by a double volume exchange transfusion performed at 32 hours of life. The clinical course was further complicated by ileal perforation with pneumoperitonium on day 3 of life. Ileostomy was performed. Histology showed focal inflammation and haemorrhage at perforation site with suspicious leukaemic infiltration. Liver biopsy performed during the operation showed mild to moderate polymorph infiltration at portal tracts and evidence of extramedullary haemopoesis, with focal collections of blast cells in sinusoidal areas. She was thought to be either a mosaic Down syndrome with transient abnormal myelopoiesis (TAM) or a normal baby with acquired +21 due to a leukaemic process. She was managed conservatively without any form of chemotherapy. She became pancytopenic on day 14. The platelet count numbered 9109/L, while absolute neutrophil count was 0.5109/L on day 22. Bone marrow aspiration and trephine biopsy were performed. The peripheral blood showed 15% blasts while marrow blood showed 39% blasts with similar morphology. The marrow blast cells on immunophenotyping showed expression of CD45, CD61, CD7 and CD33. A repeat cytogenetics study on bone marrow cells again showed mosaicism of 47,XX,+21[9]/46,XX[3]. While FISH with

Figure 1. A) Circulating blast cells and nucleated red cell at presentation Wright x 1000. B & C) Interphase FISH with 21q probe, showing cells with 3 signals i.e. +21 (arrows). D) Collection of blasts and a megakaryocyte (arrow) in liver biopsy. E) Interphase FISH on buccal epithelial cells showing disomy 21. F) Bone marrow aspirate showing return of normal haemopoiesis and resolution of TAM. Wright Giemsa x 1000.

21q probe (Vysis, Downers Grove, IL) showed 50.4% of interphase cells in peripheral blood harboured +21, PHAstimulated culture showed +21 in only one cell out of 100 metaphases examined, which might represent the occasional residual blast cell in the latter culture system. The number of circulatory blast cells slowly declined and was not detected in peripheral blood after day 40 of life, concurrent with normalization of blood count. Blood marrow aspiration repeated on day 84 showed normal haemopoietic activity. Blood count was completely normal and she was discharged on day 95. Buccal epithelial cells and cultured skin fibroblasts (each 500 cells analysed) showed absence of +21 as detected by FISH. At 8 months old, we repeated the interphase FISH study on peripheral blood and confirmed persistent absence of +21. When she was seen at 18 months old, she had completely normal haematological parameters. Discussion The diagnosis is a phenotypically normal but premature baby girl with TAM. The +21 abnormality was restricted to the haemopoietic cells. Cytogenetics and interphase FISH study showed presence of normal and +21 cells in peripheral blood and marrow, the latter clone disappearing with disease resolution. The failure to detect a single cell with +21 among 500 cells metaphases and interphases from stimulated lymphocytes, bone marrow and skin fibroblasts excludes 1% or more mosaicism. In this case, some early hydropic changes could be detected. In Fact, Baschat et al. reported that prenatal diagnosis of TAM is possible as early as 26 weeks of gestation.1 Fetal hydrops and and hepatosplenomegaly may indicate an underlying haematopoietic disorder. TAM is a condition clinically resembling congenital acute myelogenous leukaemia. As in acute leukemias the blast cell population in this disorder may have either normal or abnormal chromosomal complement. Transient myeloproliferative disorder is well recognized in neonates with a complete or mosaic Down syndrome.2-4 Phenotypically and cytogenetically normal infant with this syndrome in whom only blast cells showed trisomy 21 had been previously reported.5-7 Trisomy was apparently restricted to the leukemic clone and could be detected in neither phytohemagglutinin-stimulated peripheral blood cells or bone marrow in myeloid progenitors cells after resolution of the TAM. These infants just like the mosaic Down and constitutional Down could have durable spontaneous remissions.
haematologica/the hematology journal | 2004; 89(online) | 111 |

Marco H. K. Ho et al.

Congenital leukemia should be differentiated from transient leukemoid reaction or TAM, which is commonly noted in Down syndrome. Bresters et al reviewed Dutch patients and those described in the literature with congenital leukaemia in the past 25 years.8 This shed lights on how to differentiate leukaemia from leukemoid. Obviously, lymphoblastic blasts pointed towards leukaemia but its incidence was less common than AML (21% ALL vs 64% AML). Myeoloid blasts other than M7 type pointed towards leukaemia. For congenital leukaemia patients, most had a high leukaemic cell load with hepatosplenomegaly, leukaemia cutis and hyperleucocytosis; whereas leukemoid cases were much more stable and had fewer signs.8,9 Cytogenetic abnormalities were more commonly found in the majority of the leukaemia patients tested (72%); 11q23 abnormalities were found in less than half of them (42%). To the contrary, TAM or leukemoid patients had few cytogenetic abnormalities apart from +21. Outcome in congenital leukemia is poor, but spontaneous remissions have been described. To date, 18 cases of congenital leukemia showing spontaneous remission have been described in the literature, almost exclusively myeloid leukemia (FAB M4 and M5).8 Because of that, some experts suggested cytostatic treatment should start only if the malignancy interferes with vital parameters. In case of relapse or progression, initial postponement of chemotherapy in these frail neonates will result in less toxicity and probably a better survival. This is even more true for patient who has + 21, be it constitutional, mosaism or from the neoplastic clone. +21 cytogenetics together with M7 type of immunophenotyping, give much higher chance of spontaneous resolution. The association of Down syndrome and leukaemia has been documented for over 50 years. Incidence of leukaemia in Down patients are 10-20 fold higher than that in the general population. All cytogenetic types of Down syndrome apparently predispose to leukaemia. A significant proportion of Down patients with TAM after initial resolution may subsequently develop AML. Normal child with TAM also developed subsequent AML in isolated case report but the incidence is largely unknown.12 TAM with +21, most probably a leukaemia-specific abnormality in this disorder, exists in three groups of neonates: Downs syndrome, constitutional mosaic Downs syndrome in which non-disjunction occurs early in embryogenesis with +21 appearing in different cell lines, and +21 restricted to blood (neoplastic) cells. The +21 abnormality may be considered as acquired in the last category, akin to other types of leukaemia. It will be most interesting to see whether the risk of progression to acute megakaryoblastic leukaemia is different among these categories.11-13

Marco H. K. Ho, M.B.B.S., M.R.C.P., Shau-Yin Ha, M.B.B.S., F.R.C.P. Godfrey, C.F.Chan, M.D., F.R.C.P Department of Paediatrics and Adolescent Medicine, The University of Hong Kong Edmond S. K. Ma, M.D., F.R.C.Path Division of Haematology, Department of Pathology, The University of Hong Kong

References
1. Baschat AA, Wagner T, Malisius R, Gembruch U. Prenatal diagnosis of a transient myeloproliferative disorder in trisomy 21. Prenat Diagn 1998;18:731-6 2. Zipursky A, Brown E, Christensen H, Doyle J. Transient myeloporliferative disorder (transient leukemia) and hematologic manifestations of Down syndrome. Clin Lab Med 1999; 19: 157 67. 3. Brodeur GM, Dahl GV, Williams DL, Tipton RE, Kalwinsky DK. Transient leukemoid reaction and trisomy 21 mosaicism in a phenotypically normal newborn. Blood 1980; 55: 691 3. 4. Slayton WB, Spangrude GJ, Chen Z, Greene WF, Virshup D. Lineage-specific trisomy 21 in a neonate with resolving transient myeloproliferative syndrome. J Pediatr Hematol Oncol 2002; 24: 224 6. 5. Seibel NL, Sommer A, Miser J.Transient neonatal leukemoid reactions in mosaic trisomy 21. J Pediatr. 1984; 104:251-4. 6. Kalousek DK, Chan KW. Transient myeloproliferative disorder in chromosomally normal newborn infant. Med. Pedi. Onco.1987; 15: 38-41 7. Ridgway D, Benda GI, Magenis E, Allen L, Segal GM, Braziel RM, Neerhout RC.Transient myeloproliferative disorder of the Down type in the normal newborn. Am J Dis Child. 1990;144:1117-9. 8. Bresters D, Reus AC, Veerman AJ, van Wering ER, van der Does-van den Berg A, Kaspers GJ. Congenital leukaemia: the Dutch experience and review of the literature Br J Haematol. 2002;117:513-24. 9. Lampkin BC.The newborn infant with leukemia. J Pediatr. 1997;131:176-7 10. Stark B, Jeison M, Preudhomme C, Fenaux P, Ash S, Korek Y, Stein J, Zaizov R, Yaniv I. Acquired trisomy 21 and distinct clonal evolution in acute megakaryoblastic leukaemia in young monozygotic twins. Br J Haematol 2002; 118: 1082 6. 11. van den Berg H, Hopman AH, Kraakman KC, de Jong D.Spontaneous remission in congenital leukemia is not related to (mosaic) trisomy 21: case presentation and literature review. Pediatr Hematol Oncol. 2004; 21:135-44. 12. Brissette M.D., Duval Arnould B.J., Gordon B.G, Cotelingam J.D.. Acute megakaryoblastic leukemia following transient myeloproliferative disorder in a patient without Down syndrome.Am J Hematol. 1994 ;47 : 316-9 13. Ma SK, Wan TSK, Chan GCF, Ha SY, Fung LF, Chan LC. Relationship between transient abnormal myelopoiesis and acute megakaryoblastic leukaemia in Downs syndrome. Br J Haematol 2001; 112: 824 5.

| 112 | haematologica/the hematology journal | 2004; 89(online)