Correspondence 1192 7 Pyatt DW, Stillman WS, Yang Y, Gross S, Zheng JH, Irons RD. An apoptosis by inhibition of NF-kappaB. Nat Med 1999; 5: essential role for NF-kappaB in human CD34(+) bone marrow cell 412–417. survival. Blood 1999; 93: 3302–3308. 11 Lamberti A, Romano MF, Agosti V, Garbi C, Sun S-C, Turco MC et al. 8 Romano MF, Lamberti A, Bisogni R, Garbi C, Pagnano AM, Auletta P Regulation of cell survival in CD95-induced T cell apoptosis: role of et al. Amifostine inhibits hematopoietic progenitor cell apoptosis by NF-kB/Rel factors. Apoptosis 1999; 4: 179–185. activating NF-kB/Rel transcription factors. Blood 1999; 94: 4060–4066. 12 Nicoletti I, Migliorati G, Pagliacci MC, Grignani F, Riccardi C. A rapid 9 Adams j. Proteasome inhibition: a novel approach to cancer therapy. and simple method for measuring thymocyte apoptosis by propidium Trends Mol Med. 2002; 8: S49–S54. iodide staining and flow cytometry. J Immunol Methods 1991;139: 10 Wang CY, Cusack Jr JC, Liu R, Baldwin Jr AS. Control of inducible 271–279. chemoresistance: enhanced anti-tumor therapy through increased

Frequencies of the major subgroups of precursor B-cell acute lymphoblastic leukemia in Indian children differ from the West

Leukemia (2003) 17, 1192–1193. doi:10.1038/sj.leu.2402931 patients carrying BCR-ABL or E2A-PBX1 are older than patients carrying TEL-AML1 and the latter subgroup was associated with the TO THE EDITOR lowest WBC. However, other differences were evident. The four The presence of the major chromosomal translocations, that is, translocation subgroups together represented 19% of all Indian t(12;21), t(1;19), t(9;22) and t(4;11) defines clinicopathological precursor B-cell ALLs whereas in the West these ALLs, represent as subgroups of childhood precursor B-cell ALL (ALL), which represent much as 35%. The frequency of the TEL-AML1 subgroup in India about 30% of all cases in USA/Europe.1 Furthermore, these was significantly lower than in USA or Europe (Po0.005). In translocations have been used in risk stratification for treatment contrast, the other two translocations associated with an inter- purposes.2 However, their relative distribution and prognostic value mediate [t(1;19)] or poor prognosis [t(9;22)] are more commonly in other geographic regions are sparse. seen in India than in the West (Po0.005). It is still possible that rare Stratification of pediatric precursor B-cell ALL in India using standard clinical criteria suggested that an unusually large fraction of these patients fall into a high-risk disease category.3 This may Table 1 Relative proportions and clinical characteristics of the either be a result of late diagnosis or the low-risk subgroups are molecular subgroups of Indian childhood ALL under-represented in these children or a combination of both. It is possible that the distribution of molecular subtypes is not uniform in TEL- mBCR- E2A- MLL- Other/ different World regions. It is also possible that gene–environment AML1 ABL PBX1 AF4 none interactions, which are critical in leukemogenesis, may differently Number (%) 18 (7%) 14 (5%) 18 (7%) 0a 209 (81%) contribute in defining the relative proportions of molecular M:F ratio 17:1 4.7:1 4.5:1 F 2.7:1 subgroups in different geographic regions. Age We have therefore studied 259 newly diagnosed children with Range (years) 2–10 3–18 3–18 F o1–21 precursor B-cell ALL from India. In all, 201 consecutive samples Median (years) 6 9.5 8 7 9 F were obtained from Tata Memorial Hospital, Mumbai and 58 ALL WBC ( Â 10 /l) Range 1.5–82.6 3.5–400 3.3–254 0.5–820 samples were available at All India Institute for Medical Sciences, Median 10.7 30.05 20.3 11.05 New Delhi. Patients were o1–21 years old (63% were in the 1–9 range) with a median age of 7 years. Male-to-female ratio was 2.9:1. aOther MLL rearrangements were detected in two of 50 samples analyzed by Only 23% of the patients presented with WBC X50 Â 109/l Southern blot. (Table 1). cDNAs were prepared from frozen samples; their quality and integrity were tested by RQ-RT-PCR for GAPDH using the LightCycler instrument (Roche, Mannheim, Germany). The presence Table 2 Distribution of chromosomal translocations in childhood of the four translocations was assayed by real-time multiplex RT- precursor B-cell ALL in different World regions PCR as described.4 Leukemia-specific chimeric transcripts were detected in 50 of the 259 samples (19%). In all, 18 (7%) carried a TEL-AML1 mBCR-ABL E2A-PBX1 MLL-AF4 t(12;21); only one of them corresponded to the variant isoform that juxtaposes exon 5 of TEL to exon 3 of AML1. A total of 18 samples India (7%) demonstrated a t(1;19), while 14 (5%) showed a t(9;22) (ALL- n 18/259 14/259 18/259 0 type, p190). The absence of t(4;11) was remarkable since the %757 USA(a) primers used would amplify 85% of all known MLL-AF4 variants. Of n 170/773 168/7556 237/6301 67/5590 the 209 samples without any of the four translocations analyzed, we % 22 2.2 3.8 1.2 had enough DNA from 50 samples to analyze other 11q23 Range (%) 19–32 1.7–3.4 3.4–3.9 0.7–2.1 a abnormalities. Southern blot analyses identified two patients Europe( ) carrying a rearrangement in the MLL gene (4%). n 422/1838 160/8945 74/4516 63/3926 % 23 1.8 1.6 1.6 Table 1 summarizes molecular and clinical data from Indian Range (%) 19–25 1–3 0.5–2.5 1.1–3.3 patients. We compared these results with data compiled from P-value publications from USA/Europe (Table 2). Similar to the West, India–USA o0.005 o0.005 o0.01 NA P-value India–Europe o0.005 o0.005 o0.005 NA Correspondence: Dr K Bhatia, PO Box 3354, MBC #98-16, Riyadh 11211, Saudi Arabia; Fax: +966-1-246-5422 NA: not applicable. Received 4 December 2002; accepted 5 February 2003 aData compiled from Leukemia 2000; 14: 2196–2320.

Leukemia Correspondence 1193 translocations reported in the West, for example involving 11q23, References may be more frequent in India. However, the absence of MLL-AF4 together with the preliminary 4% MLL rearrangements suggests that 1 Biondi A, Masera G. Molecular pathogenesis of childhood acute their frequency is not high. lymphoblastic leukemia. Haematologica 1998; 83: 651–659. Surprisingly, a notable low proportion of female patients with 2 Rubnitz JE, Pui CH. Molecular diagnostics in the treatment of leukemia. TEL-AML1 was observed (17M:1F) compared to 2.7–4.7M:1F in Curr Opin Hematol 1999; 6: 229–235. other ALLs. Whether this results from a referral bias or is indeed a 3 Advani S, Pai S, Venzon D, Adde M, Kurkure PK, Nair CN et al. Acute feature of ALL in India remains unknown. However, it is unlikely lymphoblastic leukemia in India: an analysis of prognostic factors using a single treatment regimen. Ann Oncol 1999; 10: 167–176. that such a referral bias will be observed only for the TEL-AML1 4 Siraj AK, Ozbek U, Sazawal S, Sirma S, Timson G, Al-Nasser A et al. Pre- subgroup. Data from USA/Europe does not support a gender clinical validation of a monochrome real-time multiplex assay for dependency.5 TEL-AML1 is associated with young age and it may translocations in childhood acute lymphoblastic leukemia. Clin Cancer constitute a significant proportion of the preschool peak, the Res 2002; 8: 3832–3840. magnitude of which correlates with socioeconomic status. The 5 Borkhardt A, Cazzaniega G, Viehmann S, Valsecchi MG, Ludwig WD, Burci L et al. Incidence and clinical relevance of TEL/AML1 fusion genes increase in the incidence of ALL during socioeconomic transition 6 in children with acute lymphoblastic leukemia enrolled in the German has been associated more prominently with females, and thus the and Italian multicenter therapy trials. Blood 1997; 90: 571–577. selective reduction of females in this TEL-AML1 cohort (one of 18) is 6 Hrusak O, Trka J, Zuna J, Polouckova A, Kalina T, Stary J. Acute of epidemiological interest. lymphoblastic leukemia incidence during socioeconomic transition: The data provide support for geographic differences in subgroups selective increase in children from 1 to 4 years. Leukemia 2002; 16: of otherwise similar malignancies. Since the etiology of ALL is 720–725. almost certainly dependent upon gene–environment interactions, it remains possible that differences in both the population and the environment dictate the relative distribution of these subtypes.

AK Siraj1 1King Fahad National Centre for Children’s Cancer S Kamat2 and Research, Riyadh, Saudi Arabia; MI Gutie´rrez1 2Tata Memorial Hospital, Mumbai, India; S Banavali2 3All India Institute of Medical Sciences, New Delhi 1 India; G Timpson 4 S Sazawal3 International Network for Cancer Treatment and M Bhargava3 Research, Belgium, Brussels S Advani2 M Adde4 I Magrath4 K Bhatia1,4

A novel real-time RT-PCR assay for quantification of OTT-MAL fusion transcript reliable for diagnosis of t(1;22) and minimal residual disease (MRD) detection

Leukemia (2003) 17, 1193–1196. doi:10.1038/sj.leu.2402914 patients because of an associated myelofibrosis or the low number of blasts present in the samples. TO THE EDITOR Translocation t(1;22) has been shown to result in the in-frame fusion of the OTT(RBM15) on 1p13 and MAL(MKL1) on chromo- Acute megakaryoblastic leukemia (AMKL), M7 in the FAB nomen- 5–6 clature, is a heterogeneous subtype of acute myeloid leukemia some 22q13. Since the OTT-MAL fusion transcripts can be (AML) whose exact frequency ranges from 3 to 14% of AMLs, detected in the cells that underwent the translocation, we developed depending on the different institutions,1 and is higher in children a real-time quantitative RT-PCR (RQ-RT-PCR) in order to assess and than in adults. It is the most frequent type of AML occurring in young quantify the OTT-MAL fusion transcript associated with the t(1;22). patients with Down’s syndrome (DS),2 in which it is associated with Using this approach we analyzed BM samples taken at diagnosis acquired GATA1 mutations.3 AMKL may also occur as a de novo and during the treatment, of seven children with AMKL referred to leukemia in the absence of DS or as a secondary malignancy, our hospital since May 1996. The diagnosis was established on the following myelodysplasia or therapy for a prior malignancy.1 basis of the FAB criteria by studies of cell morphology and Accordingly, a large variety of chromosomal abnormalities can be cytochemistry, and was validated according to the immunological observed in AMKL. The most specific and recurring chromosomal criteria recommended by the European Group for the Immunolo- abnormality in pediatric AMKL is the t(1;22)(p13;q13) translocation, gical Classification of Acute Leukemias (EGIL). Five patients (median which is present in one-third of de novo AMKL, mainly in infants.4 age: 38 months) presented with a ‘de novo’ AMKL and two patients The t(1;22) translocation is routinely detected by standard cytoge- were diagnosed with AMKL following another hematopoietic netic analysis but could remain undetected, in part because of the disorder: a recurrent and severe neutropenia in one case and an difficulty to obtain accurate bone marrow (BM) aspirates of these hypoplastic anemia in the other. BM cells were obtained at the time of the diagnosis and during cytological remission and cryopreserved before use for molecular analysis. Cytogenetic studies were carried out on BM cells at the diagnosis of each patient and the classical Correspondence: Dr P Ballerini, Service d’He´matologie Biologique, Hoˆpital A Trousseau, 26-28 Avenue du Dr A Netter, 75012 Paris, t(1;22)(p13;q13) was found in only one case. For the molecular France; Fax: +33 1 44 73 63 33; studies we performed total RNA extraction and cDNA preparations 7 E-mail address: [email protected] as previously reported. The primers and probe used for real- Received 20 September 2002; accepted 22 January 2003 time quantitative PCR amplification (RQ-PCR) of OTT-MAL fusion

Leukemia