ArchivesofDisease inChildhood 1992; 67: 962-966

CURRENT TOPICS

Myelodysplastic syndromes

Ian M Hann

Myelodysplasia is one of the Cinderella subjectsof paediatric haematology and oncology.Although cases are rare and probably do notaccount for more than 3% of haematologicalmalignancies, patients presenting with thesedisorders pose a very serious challenge tosuccessful diagnosis and management withwhich we are only just coming to grips. In fact,such is the previous state of confusion that wehave no idea of the prevalence of the componentdisorders and there is little doubt that underdiagnosis was itself prevalent. The first tentativesteps have now been taken out of the darknessand probable cures are being achieved inpreviously almost uniformly fatal conditions.The first moves in the right direction wereachieved by abandoning the paediatriciansinherent love of syndromes and concentrating,with our adult haematology colleagues, ongetting some sense and science into classificationof the disorders.

tric literature, juvenile chronic myeloidleukaemia (JCML) and infantile monosomy 7syndrome. Both of these disorders will pre-sumably take their place under more appropriateheadings when more systematic studies arecompleted, for example, JCML shows somefeatures akin to a well described adult counter-part, chronic myelomonocytic leukaemia(CMML). Table 2 shows a recommended list ofinvestigations required for the diagnosis andclassification of myelodysplasia.

I would like to expend a little space on thespecific features of the disorders listed in table 1.However, there is insufficient space to dealadequately with chronic myeloid leukaemia,which has been extensively reviewed else-where.5-10 Essential thrombocythaemia andpolycythaemia are excessively rare in childhoodand the few authenticated cases have also beenreviewed elsewhere. 113 Malignant myelo-fibrosis as seen in adults'4 1l is not seen inchildren and modern techniques have shown

ClassificationThe basic concept of myelodysplasia is a verysimple one: the bone marrow is active butineffective. Thus, there are functional mor-phological abnormalities and/or pancytopenia.The syndromes thus include the disorderspreviously called refractory anaemia, smoulder-ing leukaemia, and preleukaemia all of whichhad a strong predilection to progress to acutemyeloid leukaemia at varying pace. One seriesof childhood acute myeloid leukaemia has sug-gested that 17% of cases had a preleukaemicphase.'The French-American-British cooperative

morphology group has led the way in providingthe systematic groundwork for a classificationsystem.2 Subsequently the Morphologic,Immunologic and Cytogenetic Cooperative StudyGroup has provided additional cytogeneticinformation which has allowed further refine-ment.3 The conventional approach has been todistinguish myeloproliferative disorders (suchas chronic granulocytic leukaemia, essentialthrombocythaemia, myelofibrosis, and poly-cythaemia) from myelodysplasia. This is asomewhat artificial distinction, bearing in mindthat all are clonal proliferations and all mayevolve to a frank acute leukaemia. However, itforms the basis for the best attempts at classi-fication currently available for childhoodmyelodysplasia4 (see table 1) and eventually abetter classification will hopefully come out inthe biological wash. This includes the twocommonest syndromes described in the paedia-

Table I Classification of paediatric myelodysplasia

* Primary myelodysplasia:Juvenile chronic myeloid leukaemiaInfantile monosomy 7 syndromeRefractory anaemiaRefractory anaemia with excess of blastsRefractory anaemia with excess of blasts in transformation

* Proliferative myelodysplasia:Down's syndrome leukaemoid reactionsMyelodysplasia with eosinophiliaPhiladelphia positive chronic myeloid leukaemiaAtypical chronic myeloid leukaemiaFamilial myelodysplasiaEssential thrombocythaemiaPolycythaemia rubra vera

* Secondary myelodysplasia:FamilialTherapy induced

Adapted from Chessells.4

Table 2 Investigation of myelodysplasia

* Blood.Haemoglobin+red cell indicesWhite cell count+differentialBlood filmFetal haemoglobin concentrationImmunoglobulins

* Bone marrow:Aspirate+trephine biopsyCytochemistry and iron stainCytogenetics

* Additional tests:Neutrophil functionPlatelet functionLymphocyte subsetsTissue culture marrow colony assaysAntinuclear antibodiesFetal globin glycine:alanine ratio Juvenile chronicI:i red cell antigen ratio myeloid leukaemiaCarbonic anhydrase

Adapated from Chessells.4

Haematology andOncology Department,Hospitals forSick Children,Great Ormond Street,London WC1N 3JHCorrespondence to:Dr Hann.

962

on July 1, 2021 by guest. Protected by copyright.

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.67.7.962 on 1 July 1992. Dow

nloaded from

http://adc.bmj.com/

Myelodysplastic smdromes

that primary myelofibrosis is in fact acutemegakaryoblastic leukaemia wherein theproliferating megakaryocytes induce extensivemarrow fibrosis.'6

Juvenile chronic myeloid leukaemiaJCML does appear to be a distinct diseaseentity, the pathognomonic feature of which is areversion to a truly fetal haemopoiesis withraised concentrations of fetal haemoglobinabove 10%, 7 a fetal pattern of red cell en-zymes,'8 and in one reported case the persistenceof epsilon globin chains that normally disappearby 3 months of age.'9 The disorder appears tobe more common in boys, and children withneurofibromatosis show a higher number ofcases than expected of JCML and chronicmyeloid leukaemia.20 21 The usual clinicalpresentation is with pallor, splenomegaly, andbleeding due to thrombocytopenia. The latter(along with the higher fetal haemoglobin con-centration) distinguishes this disorder frommonosomy 7 myelodysplasia (see below). 17Cutaneous xanthomata occasionally occur20 andled to the outdated description of 'xantho-leukaemia'. There is often a facial rash in abutterfly distribution that may also involve thetrunk, predate the other findings, and lead to amisdiagnosis of lupus erythematosus, especiallyas antinuclear antibodies may be present.22

Since the original reports of this disorder23it has become standard practice to compareand contrast this disorder with Philadelphiachromosome positive chronic myeloid leu-kaemia. This is a largely pointless exercise andin fact the morphological and some otherfeatures, for example the immune disturbance,22are more akin toCMML as previously describedin adults. Further evidence is provided bythe presence of monocyte specific antigenson the surface of mononuclear cells23 andoutgrowth of macrophage colonies in tissueculture.24 However, the bone marrow may showvarious cytogenetic abnormalities or be normaland thus there are no diagnostic cytogeneticfindings.25 This is in contrast to monosomy 7myelodysplasia which can otherwise be in-distinguishable,26 although JCML is morefrequently associated with a lower platelet countas well as the raised fetal haemoglobin.27 Aswith many of the myelodysplastic disorders inchildhood, more can be gleaned from anexamination of the blood film than the bonemarrow aspirate. The white cell count isusually only moderately raised and rarely above5Ox 109/1, anaemia is common, the plateletcount reduced, and there is usually a prominentmonocytosis often associated with dysplasticgranulocytes and some blast cells. The bonemarrow findings are similar but usually muchless noticeable. The prognosis is very poor andover the last 20 years we have seen 14 childrenat Great Ormond Street, 11 boys and three girlswith a median age of 3 years and mediansurvival of only six months. These patients didnot develop a terminal leukaemic 'blast crisis' asis sometimes described, but developed progres-sive weight loss, nodal enlargement, and someincrease in blasts and normoblasts.

Monosomy 7 myelodysplasiaMonosomy and partial deletion of chromo-some 7 is found in primary myelodysplasia asdescribed here, but also in de novo acutemyeloid leukaemia, secondary myelodysplasia,and secondary acute myeloid leukaemia. In eachof these disorders there is a defect in neutrophilchemotaxis and killing leading to a particularsusceptibility to bacterial infection.28 Thepresence of the deletion in biphenotypicleukaemias adds weight to the proposition thatmonosomy 7 is associated with transformationof a common lymphoid-myeloid progenitorcell.29 The variability of clinical features inassociation with abnormalities of chromosome 7are well illustrated by the experience at thishospital over the last 20 years. During thisperiod we have seen four cases of de novo acutemyeloid leukaemia, one of secondary acutemyeloid leukaemia, one of myelofibrosis withmyelodysplasia, one of refractory anaemia withexcess of blasts with transformation (RAEB-t),and 13 children with the myeloproliferativesyndrome. It is important to note that there arefamilies with an apparent predilection to thedevelopment of acute myeloid leukaemia ormyelodysplasia and this may be associated withchromosome 7 abnormalities.30 The 13 childrenwe have looked after with monosomy 7 myelo-dysplasia presented at a median age of 10months but the oldest was aged 8 years and hadan indolent process of recurrent infections,hypogammaglobulinaemia, and monocytosisover a six year period. Three of the 13 developedacute myeloid leukaemia and one developedmassive splenomegaly and marrow fibrosis. Thetime to the development of acute myeloidleukaemia varied from three months to twoyears. A third of the cases survived more thantwo years and thus this disorder is associatedwith longer survival than JCML with a pre-dilection to the development of acute myeloidleukaemia.

Other myelodysplasias: primaryThe current lack of knowledge does not allow acalculation of the prevalence of these disorders.Several series have highlighted the paucity ofchildren who present with the types of myelo-dysplasia associated with a better prognosis,that is refractory anaemia (RA) and refractoryanaemia with ringed sideroblasts (RARS).3' 32We have seen only two cases of RARS, both ofwhich followed a chronic course. Thus, the vastmajority of children with myelodysplasia unfor-tunately fall into the bad risk categories ofRAEB with or without transformation towardsacute myeloid leukaemia. All series consistentlyshow a less than 20% chance of survival to fiveyears and there are very few long term survivors.However, there are encouraging reports demon-strating a survival of up to 50% after bonemarrow transplantation for those who have ahistocompatible sibling.33 Attempts have beenmade to identify features associated withprognosis in adults. Features predictive of longsurvival include pancytopenia, more than5% blasts in the blood, abnormal localisation ofimmature precursors on a trephine biopsy

963

on July 1, 2021 by guest. Protected by copyright.

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.67.7.962 on 1 July 1992. Dow

nloaded from

http://adc.bmj.com/

Hann

specimen, and a karyotype showing complexchromosomal abnormalities.34

It has been known for many years thatchildren with Down's syndrome are at risk of aneonatal myeloproliferative syndrome resembl-ing acute leukaemia, which usually undergoesspontaneous remission. Some of these childrenlater develop acute leukaemia after an intervalranging from six months to three years.35 Inaddition, children with Down's syndrome andacute myeloid leukaemia can have a prodromalmyelodysplastic phase lasting several monthswith occasional blasts seen on the blood film.There is an obvious need to follow up a series ofbabies with Down's syndrome in order todetermine the true incidence of this transientabnormal myelopoiesis and the risk of develop-ment of acute leukaemia. Clinically the disorderresembles congenital leukaemia with hepato-splenomegaly but without the skin infiltrationseen in the former.36 There is a blood leuko-cytosis with many blast cells, which haverecently been shown to be megakaryoblasts.37Although these children may develop hyper-viscosity problems due to the high white bloodcell count, they do not develop severe bonemarrow failure although they may need somered cell and platelet transfusions. The manage-ment is conservative with supportive care onlyand the most important thing is to follow up thepatients carefully in case they develop acutemyeloid leukaemia.A number of children have presented with

hepatosplenomegaly, dramatic eosinophilia,myelodysplasia, and cytogenetic abnormalityinvolving the area of chromosome 5q31 thatcontains the gene for interleukin-5 (IL-5'eosinopoietin').38 We have a patient in whomthere is a prolonged history ofa severe infiltrativeskin disorder and dramatic eosinophilia.There is a 'ragbag' of other very rare myelo-

proliferative disorders in childhood. Severalhave been shown to involve translocations ofchromosome llpl5.39 Rarely, cases otherwiseidentical to chronic myeloid leukaemia butwithout the Philadelphia chromosome and withno bcr-abl gene rearrangement34 are seen inchildhood. A morphological review of childhoodcases is required and it is possible that they willbe classified with CMML. The prognosis of thisdisorder is said to be poor.

Other myelodysplasias: secondaryChildren with congenital bone marrow dis-orders such as Fanconi's anaemia and Shwach-man's syndrome can present with or developmyelodysplasia as well as acute myeloid leu-kaemia.4U 41 Unfortunately we are also seeing anincreased incidence of acute myeloid leukaemiaand myelodysplasia in patients treated formalignant disease, most commonly Hodgkin'sdisease but also non-Hodgkin's lymphoma andother solid tumours. The alkylating agents aremost clearly implicated in this susceptibility.42The risk of secondary leukaemia and myelo-dysplasia after treatment for acute lymphoblasticleukaemia is not clear but so far there have onlybeen a few cases reported in the UK.

Cytogenetic abnormalities are found in

almost all children with secondary myelo-dysplasia/acute myeloid leukaemia and arefrequently multiple, the majority involvingchromosomes 5 and 7 with 3 and 17 being lessfrequently involved. The chromosome 5abnormalities always involve 5q23-32, a regioncontaining the genes for several growth factorsand their receptors.43

Clearly the most important point to be madehere is that the planning of treatment formalignant disease should take into account theneed to try to avoid the use of alkylating agentswhere alternative treatments are possible andespecially to try to avoid their concomitant usewith radiotherapy. This has proved possiblewith regimens for low grade non-Hodgkin'slymphoma. The published results of treatmentfor secondary myelodysplasia make depressingreading. Patients with chromosome 5 abnor-malities did particularly badly in the past.Remission duration, if achieved at all, tends tobe extremely short in this whole group ofchildren before the era of bone marrow trans-plantation.42

Biology of myelodysplasiaVarious modern techniques have demonstratedthat the myelodysplasias are clonal disordersarising in a multipotent or pluripotent stemcell,' and the blood cells produced oftensurvive for a short time and have abnormalfunction. Disease progression is a function ofthe expansion of the abnormal clone withprogressively abnormal blood count and/ordevelopment of acute myeloid leukaemia. Thisprocess is frequently associated with immuno-logical abnormalities including raised immuno-globulins, reduced numbers of natural killercells, and an abnormal helper:suppressor T cellratio.45 A variety of abnormalities of bonemarrow growth in colony cultures has beendescribed. In children one of the commonestphenomena is spontaneous growth of colonyforming unit-granulocyte/macrophage andother patients show reduced growth of multi-potent progenitor cells.26A large number of attempts have been made

to identify factors associated with disease pro-gression towards acute myeloid leukaemia. Onepromising line of inquiry relates to the ras groupof genes which are involved in the regulation ofcell proliferation. Preliminary studies have infact shown that these genes are activated inpatients with myelodysplasia and possibly moreso in those who develop acute myeloid leu-kaemia.' The other main line of scientificinvestigation involves the possible role ofputative repressor genes in the regions ofchromosomes 5 and 7 frequently involved inmyelodysplasia. The hypothesis is that loss ofthe normal allele could result in the unmaskingof a recessive mutant allele on the remainingchromosome, as occurs in retinoblastoma. Therelevant chromosomal regions certainly containthe genes for a tantalising array of growthfactors and other important compounds-forexample, 5q23-3 1 contains granulocyte/macro-phage-colony stimulating factor (GM-CSF),interleukin 3, macrophage-colony stimulating

964

on July 1, 2021 by guest. Protected by copyright.

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.67.7.962 on 1 July 1992. Dow

nloaded from

http://adc.bmj.com/

Myelodysplastic sydromes

factor (M-CSF), and c-fms (the M-CSF receptor).Similarly 7q is the site of the erythropoietingene, the met proto-oncogene, and the multidrugresistance gene (MDR-1).f6

Treatment optionsHaving completed the investigations detailed intable 2, one is then faced with several dilemmas.Most children with myeloproliferative/myelo-dysplasia have an extremely poor prognosisalthough occasionally there is a long prodromalphase during which the patient can be very well;Philadelphia chromosome positive chronicmyeloid leukaemia is the best example of thisphenomenon. It is in fact quite easy sometimesto be lulled into a false sense of security when infact a knowledge of the extremely poor prog-nosis should stimulate a search for possiblecurative options. The management of chronicmyeloid leukaemia has been very well reviewedelsewhere"'0 and basically comes down tolooking for one type of bone marrow trans-plantation or another. Looking after thesechildren is a great challenge for paediatrichaematologists and I will briefly review thelimited therapeutic armamentarium.

Various chemotherapeutic regimens havebeen tried in myelodysplasia with palliative andcurative intent. Other than using intensivechemotherapy for patients who effectively haveacute myeloid leukaemia these have beensingularly unsuccessful in achieving the lattergoal.47 However, we have now given six childrenwith infantile monosomy 7 myelodysplasiacombination chemotherapy and two remain inremission four and five years from chemo-therapy.48 The lesson to be learnt here, applic-able in other myelodysplasias, is that earlytreatment can be successful but waiting untilthe development of acute myeloid leukaemiaappears to reduce the likelihood of a goodoutcome. A reasonable palliative effect can beachieved with combinations such as cytarabineand mercaptopurine49 with some reduction ofbulk disease and requirements for bloodproduct support. These patients usually requireintensive supportive care with antimicrobialagents, blood, and platelets especially if the planis to attempt a curative bone marrow trans-plantation.There is a current vogue for cytokine treat-

ment and many consequent publications.50 Themost commonly used at present are GM-CSFand granulocyte-colony stimulating factor. Tocut a long story short, it is usually possible -toimprove neutropenia but not thrombocytopenia.The worry of course is that the neutrophildysfunction will not be improved and theunderlying disease will be 'cranked up'.Randomised trials are in progress so we shouldeventually know if these expensive compoundswill help. It is also interesting to note thatpatients with JCML have abnormal cytokineproduction and regulation, which opens up thepossibility of treatment with inhibitors or regu-lators of cytokines when these become available.

Bone marrow transplantationAs previously stated, myelodysplasia with a

good prognosis (basically refractory anaemiaand RARS) is a rarity and the main considerationis, having improved the child's general health asfar as possible, whether a donor for bonemarrow transplantation is available and if notwhether a matched unrelated donor can befound on the various donor registries. Thedecision to go ahead with bone marrow trans-plantation using a matched unrelated donor canbe difficult especially in a currently well childbecause of the relatively poor results and highmortality at present. However, techniques areimproving and for almost all patients it will betheir only chance of cure.There are few published reports of the

outcome of bone marrow transplantation formyelodysplasia and even fewer related tochildren. Matched allogeneic transplants havebeen associated with survival beyond threeyears of approximately 45%. The evidence islimited but bone marrow transplantation isundoubtedly the most effective treatment formyelodysplasia in adults and children. What isnot known yet is whether one can get away witha chemotherapy 'conditioning' regimen ratherthan the conventional cyclophosphamide plustotal body irradiation approach which is associ-ated with major endocrine, growth, and otherlate effects. Preliminary studies in childhoodchronic myeloid leukaemia suggest that thechemotherapy only approach may be suitableespecially for very young patients.5' Successesin bone marrow transplantation for JCML havebeen reported from Seattle with three of sixchildren receiving a graft from a compatiblesibling and three of eight using a donor with a1-3 antigen mismatch alive and in remissionafter the use of cyclophosphamide and totalbody irradiation.52 In this situation one casereport suggests that a chemotherapy only regi-men may not eradicate the malignant clone.53We have also performed a successful bonemarrow transplantation in a child with mono-somy 7 myelodysplasia using a histocompatiblesiblingand a chemotherapy/total body irradiationregimen.The problem is that only one in three

children will have a histocompatible siblingdonor and now great efforts are being made toimprove the chances of finding a matchedunrelated donor rapidly and carrying out asuccessful bone marrow transplantation. Resultsin children with chronic myeloid leukaemiashow an acturial event free survival of 30% at 30months,54 suggesting that this approach shouldbe considered for all patients who do not have asibling donor.

ConclusionI hope that I have made it clear that themyelodysplastic syndromes are worthy of moreattention both scientifically and clinically. Webadly need more studies designed to determinethe real prognosis and incidence of these dis-orders and to test the efficacy of moderntreatments such as bone marrow transplantationand cytokine therapy. At a more fundamentallevel scientific investigations should lead to abetter basic understanding of leukaemogenesis.

965

on July 1, 2021 by guest. Protected by copyright.

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.67.7.962 on 1 July 1992. Dow

nloaded from

http://adc.bmj.com/

966 Hann

At long last we are beginning to understandmore and to cure some of the patients but forthe foreseeable future any child presenting withmyelodysplasia will continue to tax our under-standing and management skills to the limit.

Many thanks are due to Professor Judith Chessells, withoutwhose help and guidance this article could not have beenproduced.

1 Blank J, Lange B. Preleukemia in children. J Pediatr1982;98:565-8.

2 Bennett JM, Catovsky D, Daniel MT, et al. Proposals for theclassification of the myelodysplastic syndromes. Br JHaematol 1982;51:189-99.

3 Third MIC Co-operative Study Group. Recommendations fora morphologic, immunologic and cytogenetic (MIC)working classification of the primary and therapy-relatedmyelodysplastic disorders. Cancer Genet and Cytogenet1988;32:1-10.

4 Chessells JM. Myelodysplasia. In: Hann IM, Gibson BES,eds. Clinical haematolog paediatric haematology. Vol 4,No 2: 459-482. London: Bailliere-Tindali, 1991.

5 Goldman JM, ed. Chronic myeloid leukaemia. London:Bailliere-Tindall, 1987.

6 Klingebiel T, Creutzig U, Dopfer R, et al. Bone marrowtransplantation in comparison with conventional therapy inchildren with adult type chronic myelogenous leukemia.Bone Marrow Transplant 1990;5:317-20.

7 Offit K, Burns JP, Cunningham I, et al. Cytogenic analysis ofchimerism and leukemia relapse in chronic myelogenousleukemia patients after T cell-depleted bone marrowtransplantation. Blood 1990;75:1346-55.

8 Hirsch-Ginsberg C, LeMaistre AC, Kantarjian H, et al. RASmutations are rare events in Philadelphia chromosome-negative/bcr gene rearrangement-negative chronic myelo-genous leukemia, but are prevalent in chronic myelomono-cytic leukemia. Blood 1990;76:1214-9.

9 Tien HF, Wang CH, Chen YC, et al. Chromosome and bcrrearrangement in chronic myelogenous leukaemia and theircorrelation with clinical states and prognosis of the disease.Br J' Haematol 1990;75:469-75.

10 Nelkin BD, Przepiorka D, Burke PJ, et al. Abnormalmethylation of the calcitonin gene masks progression ofchronic myeloid leukaemia. Blood 1991;77:2431-4.

11 Fernandez-Robles E, Vermylen C, Martiat P, et al. Familialessential thrombocythemia. Pediatr Hematol Oncol 1990;7:373-6.

12 Chintagumpala MM, Steuber CP, Mahoney DH, et al.Essential thrombocythemia in a child: management withanagrelide. Am J Pediatr Hematol Oncol 1991;13:52-6.

13 Schwartz CL, Cohen HJ. Preleukaemic syndromes and othersyndromes predisposing to leukaemia. Pediatr Clin NorthAm 1988;35:853-71.

14 Donti E, Tabilio A, Bocchini F, et al. Partial trisomy lq inidiopathic myelofibrosis. Leuk Res 1990;14:1035-40.

15 Lambertenghi-Deliliers G, Orazi A, Luksch R, et al. Myelo-dysplastic syndrome with increased marrow fibrosis: adistinct clinico-pathological entity. BrJ Haematol 1991;78:161-6.

16 Bennett JM, Catovsky D, Daniel M-T, et al. Criteria for thediagnosis of acute megakaryocytic leukaemia (M7). AnnIntern Med 1985;103:460-2.

17 Weatherall DJ, Edwards JA, Donohoe WTA. Haemoglobinand red cell enzyme changes in juvenile chronic myeloidleukaemia., BMJ 1968;i:679-81.

18 Travis SF. Fetal erythropoiesis in juvenile chronic myelocyticleukaemia. Blood 1983;62:602-5.

19 Weinberg RS, Leibowitz D, Weinblatt ME, et al. Juvenilechronic myelogenous leukaemia: the only example of trulyfetal (not fetal-like) erythropoiesis. BrJ Haematol 1990;76:307-10.

20 Merys JA, Beerhout RC, Bagby GC, Koler RD. Juvenilechronic granulocytic leukaemia. AmJI Dis Child 1980;134:654-8.

21 Bader JL, Miller RW. Neurofibromatosis and childhoodleukaemia. J Pediatr 1978;92:925-9.

22 Cannat A, Seligmann M. Immunological abnormalities injuvenile myelomonocytic leukaemia. BMJ 1973;i:171-4.

23 Hardisty RM, Speed DE, Till M. Granulocytic leukaemia inchildhood. Br J Haematol 1964;76:1-6.

24 Gualtieri RJ, Emanuel PD, Zuckerman KS. Granulocyte-macrophage colony stimulating factor is an endogenousregulator of cell proliferation in JCML. Blood 1989;74:2360-7.

25 Kaneko Y, Maseki N, Sakurai M, et al. Chromosome patternin juvenile chronic myelogenous leukaemia, myelodysplastic

syndrome and acute leukaemia associated with neuro-fibromatosis. Leukemia 1989;3:36-41.

26 Brandwein JM, Horsman DE, Eaves AC, et al. Childhoodmyelodysplasia: suggested classification as myelodysplasticsyndromes based on laboratory and clinical findings. Am JPediatr Hematol Oncol 1990;12:63-70.

27 Sieff CA, Chessells JM, Harvey BAM, et al. Monosomy 7 inchildhood: a myeloproliferative disorder. Br J Haematol1981;49:235-49.

28 Ruutu P, Ruutu T, Vuopi I, et al. Defective chemotaxis inmonosomy 7. Nature 1977;265:146-7.

29 Chan LC, Sheer D, Drysdale HC, et al. Monosomy 7 andmultipotential stem cell transformation. Br J Haematol1985;61:531-9.

30 Carroll WL, Morgan R, Glader BE. Childhood bone marrowmonosomy 7 syndrome: a familial disorder? J Pediatr1985;107:578-80.

31 Wegelius R. Preleukaemic states in children. ScandinavianJournal of Haematology 1986;36:133-9.

32 Creutzig U, Cantu-Rainoldi A, Ritter J, et al. Myelodysplasticsyndromes in childhood. Report of 21 patients from Italyand West Germany. Am J Pediatr Hematol Oncol 1990;9:324-30.

33 Guinan EC, Tantravali R, Weinstein HJ. Bone marrowtransplantation for children with myelodysplastic syn-dromes. Blood 1989;73:619-22.

34 Mufti GJ, Galton DAG. Myelodysplastic syndromes: naturalhistory and features of prognostic importance. Clinics inHaematology 1986;15:953-71.

35 Bain B. Down's syndrome-transient abnormal myelopoiesisand acute leukaemia. Leukemia and Lymphoma 1991;3:309-17.

36 Barnett PLJ, Clark ACL, Garson OM. Acute non-lymphocyticleukemia after transient myeloproliferative disorder in apatient with Down's syndrome. Med Pediatr Oncol 1990;18:347-53.

37 Eguchi M, Sakakibara H, Suda J, et al. Ultrastructural andultracytochemical differences between TAM and mega-karyoblastic leukaemia in Down's syndrome. BrJ Haematol1989;73:315-22.

38 Darbyshire PJ, Shortland DB, Swansbury JG, et al. Amyeloproliferative disorder in two infants associated witheosinophilia and chromosome t(1;5) translocation. Br J7Haematol 1987;6:483-6.

39 Inaba T, Hayashi Y, Hanada R, et al. Childhood myelo-dysplastic syndromes with llpl5 translocation. CancerGenet Cytogenet 1988;34:41-6.

40 Auerbach AD, Weiner MA, Warburton D. Acute myeloidleukaemia as the first haematologic manifestation ofFanconi anaemia. Am J Hematol 1982;12:289.

41 Woods WG, Roloff JS, Lukens JN, Krivit W. The occurrenceof leukaemia in patients with Shwachman syndrome.J Pediatr 1981;99:425-8.

42 Levine EG, Bloomfield CD. Secondary myelodysplasticsyndromes and leukaemias. Clinics in Haematology 1986;15:1037-80.

43 Le Beau MM, Albain KS, Larson RA, et al. Clinical andcytogenetic correlations in 63 patients with therapy-relatedMDS and AML. J Clin Oncol 1986;4:325-45.

44 Tefferi A, Thibodeau SN, Solberg LA. Clonal studies in themyelodysplastic syndromes using X-linked restriction frag-ment length polymorphisms. Blood 1990;75:1770-3.

45 List AF, Ganewal HS, Sandberg AA. The myelodysplasticsyndromes: biology and implications for management.J Clin Oncol 1990;8:1424-41.

46 Jacobs A. Oncogenes in the myelodysplastic syndrome. BloodRev 1989;3:105-9.

47 De Witte T, Muus P, De Pauw B, Hannen C. Intensiveantileukaemic treatment of patients younger than 65 yearswith MDS and secondary AML. Cancer 1990;66:831-7.

48 Evans JPM, Czepulkowski B, Gibbons B, et al. Childhoodmonosomy 7 revisited. BrJ Haematol 1988;69:41-5.

49 Lilleyman JS, Harrison JF, Black JA. Treatment of juvenilechronic myeloid leukaemia with sequential subcutaneouscytarabine and oral mercaptopurine. Blood 1977;49:559-62.

50 Cheson BD. The myelodysplastic syndromes: currentapproaches to therapy. Ann Intemn Med 1990;112:932-41.

51 Tutschka PJ, Copelan EA, Klein JP. Bone marrow trans-plantation for leukaemia following a new busulfan andcyclophosphamide regimen. Blood 1987;70:1382-8.

52 Sanders JE, Buckner CD, Thomas ED, et al. Allogeneicmarrow transplantation for children with juvenile chronicmyeloid leukaemia. Blood 1988;71:1144-6.

53 Urban C, Schwinger W, Slavc I, et al. Busulfan/cyclophos-phamide plus BMT is not sufficient to eradicate themalignant clone in juvenile chronic myeloid leukaemia.Bone Marrow Transplant 1990;5:353-6.

54 McGlave PB, Beatty P, Ash R, Hows JM. Therapy forchronic myelogenous leukaemia with unrelated donor bonemarrow transplantation: results in 102 cases. Blood 1990;75:1728-32.

on July 1, 2021 by guest. Protected by copyright.

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.67.7.962 on 1 July 1992. Dow

nloaded from

http://adc.bmj.com/