Correspondence 442 a patient treated with etoposide for Langerhans cell . 11 Lakhani S, Davidson RN, Hiwaizi F, Marsden RA. Razoxane and Br J Haematol 1994; 86: 887–889. leukaemia. Lancet 1984; II: 288–289. 8 Wiernik PH, Muse IM. Acute promyelocytic after treat- 12 Bhavnani M, Al Azzawi S, Liu Yin JA, Lucas GS. Therapy-related ment of malignant glioma in a patient with Von Recklinghausen’s acute promyelocytic leukaemia. Br J Haematol 1994; 86: 231– disease: case report and review of the literature. Leukemia 1996; 232. 10: 178–191. 13 Anticancer Drugs. Antitumor antibiotics. In: De Vita VT Jr, Hell- 9 Snowden JA, Laidlaw ST, Champion AE, Reilly JT. Acute promyel- man S, Rosenberg SA (eds). . Principles and Practice of ocytic leukaemia after treatment for seminoma with carboplatin. , 4th edn. JB Lippincott: Philadelphia: 1993; pp. 374– Lancet 1994; 344: 1361. 384. 10 Xue Y, Lu D, Guo Y, Lin B. Specific chromosomal translocations 14 Buckley JD, Gilchrist GS, Ruccione K, Sather HN, Woods WG, and therapy-related leukemia induced by bimolane therapy for Hammon GD. Multiple sclerosis in mothers of children with acute psoriasis. Leukemia Res 1992; 16: 1113–1123. lymphoblastic leukemia. Leukemia 1989; 3: 736–739.

The first report of a Philadelphia chromosome and BCR/ABL rearrangement positive myeloproliferative disorder in a child with thrombocythemia

TO THE EDITOR adrenalin, arachidonic acid and adenosine diphosphate, and was normal with ristocetin. Chronic myelogenous leukemia (CML) is a chronic myeloprol- Three months later, the patient still had 3000 × 109/l plate- iferative disorder (CMD) characterized by the presence of the lets, WBC 12.2 × 109/l (63% neutrophils, 25% lymphocytes, Philadelphia chromosome (Ph1) in more than 95% of patients. 4% , 5% basophils, 3% eosinophils) and hemoglo- The Ph1 chromosome is caused by reciprocal translocation bin 13.4 g/dl with no treatment, and began treatment with t(9;22)(q34;q11) which results in rearrangement of the BCR hydroxyurea. and ABL genes. In children under 5 years of age, Ph1 positive Cytogenetic analysis showed a t(9;22). The RT-PCR analy- CML is rare and accounts for 3% of childhood and sis, performed using published methods,6 revealed the pres- 10% of all CML. ence of P210 (b2a2 junction) (Figure 1a) and P190 (Figure 1b) Essential thrombocythemia (ET) is a chronic myeloprolifer- type of BCR/ABL mRNAs in peripheral blood and bone mar- ative disorder, characterized by the proliferation of a pluripot- row cells. The vast majority of CML cases exhibit a break of ent cell, leading to an increase in the number of platelets. the BCR gene in a 5.8 kb region named M-BCR which, Since ET does not have a distinct cytogenetic or molecular depending on the BCR gene breakpoint, produces two types marker, it has been defined by means of negative criteria, one of mRNA by the junction of exon 3 of the BCR gene to exon 1 of which is the lack of the Ph chromosome and/or BCR/ABL 2 of the ABL gene (b3a2 junction) or of BCR exon 2 to exon rearrangement. However, adult patients have been described 2 of ABL (b2a2 junction) and both chimeric mRNAs encode 1 with a disease resembling ET, but showing Ph chromosome a 210 kDa (P210). In our case Southern blot analysis 1–5 and/or BCR/ABL rearrangement. confirmed the BCR/ABL rearrangement with a breakpoint in 1 We report a girl with a Ph -positive and BCR/ABL-positive the M-BCR region (data not shown). . CMD presenting as ET. A 5 -year-old female, without previous Recently, Cervantes et al1 have suggested an association relevant history, came to our hospital because of fever up to ° between b3a2 junction and CML with thrombocythemia, in 39.5 C, abdominal pain, vomiting, and diarrhea. She was contrast to Kwong7 who has suggested that this association diagnosed as having intestinal infection and these symptoms may be fortuitous. Our case showed the b2a2 junction con- subsided in 48 h. Physical examination showed splenomegaly firmed by Southern blot analysis and could support the last 3 cm below the left costal margin and hepatomegaly 1 cm hypothesis. In addition to P210 mRNA, there was a P190 below the costal margin, with no lymphadenopathies or other BCR/ABL hybrid mRNA. The presence of low levels of mRNA abnormal signs. Peripheral blood counts showed: hemoglobin of the P190 protein in most cases of P210 BCR/ABL positive 13.6 g/dl, WBC 8.4 × 109/l (57% neutrophils, 38% lympho- patients has been recently described.6 cytes, 5% basophils) and 2200 × 109/l platelets. Neutrophil Sporadic cases have been reported of adult patients with alkaline phosphatase score was 40/400, ferritin 12 ng/ml, ␮ thrombocythemia higher than 1000 × 109/l and moderate leu- transferrin 325 g/dl, transferrin saturation 18%, and all other 1 biochemical parameters were normal. The bone marrow kocytosis, who presented Ph chromosome and/or BCR/ABL rearrangement.1–5 To our knowledge, the case described here aspirate showed a marked megakaryocytic hyperplasia (many 1 megakaryocytes, small in size, with hypolobulated nuclei), is the first pediatric patient with a Ph -positive CMD aggregates of platelet debris and 3–5% blast; the myeloid and presenting as ET. erythroid series were conserved. The study of platelet aggre- These cases might be considered as at the initial stage of gation showed a decrease of the aggregation with collagen, CML, although from the clinical point of view would fit with real ETs with a Philadelphia chromosome. However, the Poly- cythemia Vera Study Group establishes the presence of t(9;22) of the BCR/ABL hybrid gene as an exclusion criteria for the Correspondence: A Villegas, Department of Haematology, Hospital 1 Clinico San Carlos, Calle Pofesor Martin Lagos S/N, 28040, Madrid, ET diagnosis. It has been suggested that Ph positive throm- Spain bocythemias have clinicopathological characteristics different Received 12 June 1997; accepted 18 November 1997 from those of CML, mainly higher frequency in females, Correspondence 443

Figure 1 (a) cDNA amplification by RT-PCR for study of P210 mRNA. Lane 1: molecular weight marker (phiX174 DNA digested with HaeIII); lane 2: patient with P210 negative acute lymphoblastic leukaemia; lane 3: study in bone marrow of the patient described, showing P210 (b2a2 junction) positivity; lane 4: study in peripheral blood of the patient described, showing P210 (b2a2 junction) positivity; lane 5: positive control (CML positive for the b3a2 junction of the P210 type of BCR/ABL); lane 6: negative control; lane 7, 8 and 9: amplification of non-rearranged ABL genes from the same cDNAs as lanes 2, 3 and 4 which were used as internal controls. (b) cDNA amplification by RT-PCR for study of P190 type of BCR/ABL mRNA. Lane 1: molecular weight marker (phiX174 DNA digested with HaeIII); lane 2: patient with P190 positive acute lymphoblastic leukaemia (ALL), used as positive control; lane 3 and 4: P190 negative patients with ALL; lane 5: study in bone marrow of the patient described, showing P190 positivity; lane 6: negative control; lanes 7, 8 and 9: amplification of non-rearranged ABL genes from the same cDNAs as lanes 3, 4 and 5 which were used as internal controls.

absence of splenomegaly and variable clinical evolution.2,4,5 E Anguita11Department of Haematology, and It will be necessary to describe and follow-up more such cases F Valverde22Paediatric Department, Hospital ´ to see whether they have clinico-biological features in com- FA Gonzalez1 Universitario San Carlos, mon and behave differently to typical CML. C Gil2 and Department of Human Genetics, ´ M Mateo1 Hospital Ramony Cajal, MT Ferro3 Madrid, Spain Acknowledgements A Villegas1

This work has been supported in part by the grant FIJC-96/ESP ´ from the Jose Carreras International Leukemia Foundation, Spain. Correspondence 444 References BCR rearrangement in apparent essential thrombocythaemia. Br J Haematol 1993; 85: 625–626. 1 Cervantes F, Colomer D, Vives-Corrons JL, Rozman C, Montserrat 5 Stoll DB, Peterson P, Exten R, Laszlo J, Pisciotta AV, Ellis JT, White E. Chronic with thrombocythemic onset: a CML P, Vaidya K, Bozdech M, Murphy S. Clinical presentation and subtype with distinct hematological and molecular features? Leu- natural history of patients with essential thrombocythemia and the kemia 1996; 10: 1241–1243. Philadelphia chromosome. Am J Hematol 1988; 27: 77–83. 2 Kwong YL, Chiu EKW, Liang RHS, Chan V, Chan TK. Essential 6 Saglio G, Pane F, Gottardi E, Frigeri F, Buonaiuto M, Guerrasio thrombocythemia with BCR/ABL rearrangement. Cancer Genet A, De Micheli D, Parziale A, Fornaci MN, Martinelli G, Salvatore Cytogenet 1996; 89: 74–76. F. Consistent amounts of acute leukemia-associated P190BCR/ABL 3 Michiels JJ, Prins ME, Hagermeijer A, Brederoo P, van der Meulen transcripts are expressed by chronic myelogenous leukemia J, van Vliet HHD, Abels J. Philadelphia chromosome-positive patients at diagnosis. Blood 1996; 87: 1075–1080. thrombocythemia and megakaryoblast leukemia. Am J Clin Pathol 7 Kwong YL. Association between b3a2 BCR/ABL fusion and 1987; 88: 645–652. chronic myeloid leukemia with thrombocythemic onset: fortuitous 4 Richards EM, Bloxham DM, Nacheva E, Marcus RE, Green AR. or real? Leukemia 1997; 11: 617–618.

COMMENT ON PUBLISHED PAPER

Discordant erythropoiesis in CML

TO THE EDITOR plicities should be observed with different recombinant cyto- kines but otherwise identical culture conditions.4 We enjoyed reading the recent leading article by Clarkson et Finally, Clarkson et al1 have highlighted the response of al1 concerning ‘New understanding of the pathogenesis of normal and CML progenitor cells to c-kit ligand (KL). In both CML: a prototype of early neoplasia’ and broadly agree with cases, addition of KL to Epo increased the size of the colonies their ‘Discordant Maturation’ hypothesis. However, we were formed by BFU-E. Similarly, we have found that KL plus Epo, surprised by their view that our published data on BFU-E in compared with Epo alone, increases the cellularity of normal CML2 were not in line with their thinking. They have found BFU-E (7.5-fold; P = 0.0001)4 although subcolony multiplicity that the majority of erythrocyte progenitors in CML are more was not increased. In CML, in addition to increased subcolony mature and less capable of extensive proliferation than com- cellularity, there was a moderate increase in subcolony multi- parable normal erythrocyte progenitors. We reported that plicity (unpublished data). BFU-E in CML formed bursts consisting of more subcolonies than normal BFU-E.2 Assuming, as we do, that commitment SB Marley Department of Haematology to subcolony formation during the growth of an erythroid burst JL Lewis Imperial College School of Medicine marks a terminal differentiation event, these two statements JM Goldman Hammersmith Hospital, London, UK imply exactly the same relationship between early and late MY Gordon erythroid progenitors. That is, in order for a given cell popu- lation to outnumber its progenitors disproportionately, there must be an increase in its productive capacity either by References increasing its probability of self-renewal or by increasing the number of cell doublings separating the two compartments. 1 Clarkson BD, Strife A, Wisniewski D, Lambek C, Carpino N. New We would like to take this opportunity to comment further understanding of the pathogenesis of CML: a prototype of early neoplasia. Leukemia 1997; 11: 1404–1428. on the interpretation of our methodology, which relies on the 2 Marley SB, Lewis JL, Goldman JM, Gordon MY. Abnormal kinetics number of subcolonies formed per erythroid burst, ranging of colony formation by erythroid burst-forming units (BFU-E) in from one to the maximum observed. According to Clarkson chronic myeloid leukaemia. Br J Haematol 1996; 93: 878–883. et al,1 this phenomenon results from the breaking up of the 3 Marley SB, Amos TAS, Gordon MY. Kinetics of colony formation colonies into subunits at later culture times. Against this by BFU-E grown under different culture conditions in vitro. Br J interpretation, we have made sequential video recordings Haematol 1996; 92: 559–561. 4 Lewis JL, Marley SB, Blackett NM, Szydlo R, Goldman JM, Gordon which show quite clearly that the cells giving rise to the sub- MY. Interleukin-3 (IL-3) but not stem cell factor (SCF) increases 3 colonies separate prior to the onset of subcolony formation. self-renewal by human erythroid burst-forming units (BFU-E) in Another criticism that has been levelled at our methodology vitro. Cytokine (in press). is that subcolony formation is simply a function of the density of the supporting methylcellulose. However, if this were cor- rect, it is difficult to understand why different subcolony multi-

Correspondence: SB Marley; Fax: 44 181 740 9679 Received 10 September 1997; accepted 25 November 1997