Associations Between Genome-Wide Native American Ancestry, Known Risk Alleles and B-Cell ALL Risk in Hispanic Children

Letters to the Editor 2416 5 Division of Hematology and Oncology, Shizuoka 2 Tsukimoto I, Tawa A, Horibe K, Tabuchi K, Kigasawa H, Tsuchida M et al. Risk- Children’s Hospital, Shizuoka, Japan; stratified therapy and the intensive use of cytarabine improves the outcome in 6 Department of Pediatrics, Shiga University of childhood acute myeloid leukemia: the AML99 trial from the Japanese Childhood Medical Science, Otsu, Japan; AML Cooperative Study Group. J Clin Oncol 2009; 27: 4007–4013. 7Department of Pediatrics, Mie University School of 3 Creutzig U, Zimmermann M, Bourquin JP, Dworzak MN, von Neuhoff C, Sander A Medicine, Tsu, Japan; et al. Second induction with high-dose cytarabine and mitoxantrone: different 8Department of Pediatrics, Okayama University, Okayama, Japan; impact on pediatric AML patients with t(8;21) and with inv(16). Blood 2011; 118: 9Department of Pediatrics, Hirosaki University 5409–5415. 4 Mayer RJ, Davis RB, Schiffer CA, Berg DT, Powell BL, Schulman P et al. Intensive Graduate School of Medicine, Hirosaki, Japan; 10 postremission chemotherapy in adults with acute myeloid leukemia. Cancer and Division of Pediatrics, Department of Reproductive and Leukemia Group B. N Engl J Med 1994; 331: 896–903. Developmental Medicine, Faculty of Medicine, 5 Bloomfield CD, Lawrence D, Byrd JC, Carroll A, Pettenati MJ, Tantravahi R et al. University of Miyazaki, Miyazaki, Japan; Frequency of prolonged remission duration after high-dose cytarabine intensifi- 11Department of Pediatrics, St Marianna cation in acute myeloid leukemia varies by cytogenetic subtype. Cancer research University School of Medicine, Kawasaki, Japan; 1998; 58: 4173–4179. 12Department of Pediatrics, Saiseikai Yokohama City Southern 6 Lowenberg B, Ossenkoppele GJ, van Putten W, Schouten HC, Graux C, Ferrant A et al. Hospital, Yokohama, Japan; High-dose daunorubicin in older patients with acute myeloid leukemia. NEnglJMed 13Department of Pediatrics, Fukuoka-Higashi Medical Center, 2009; 361: 1235–1248. 7 Fernandez HF, Sun Z, Yao X, Litzow MR, Luger SM, Paietta EM et al. Anthracycline National Hospital Organization, Koga, Japan; 14 dose intensification in acute myeloid leukemia. NEnglJMed2009; 361: 1249–1259. Department of Hematology/Oncology, Saitama 8 Kremer LC, van der Pal HJ, Offringa M, van Dalen EC, Voute PA. Frequency and risk Children’s Medical Center, Saitama, Japan; factors of subclinical cardiotoxicity after anthracycline therapy in children: a 15 Department of Hematology/Oncology, Kanagawa systematic review. Ann Oncol 2002; 13: 819–829. Children’s Medical Center, Yokohama, Japan; 9 Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) 16Department of Hematology and Oncology, classification of the myeloid neoplasms. Blood 2002; 100: 2292–2302. Hyogo Children’s Hospital, Kobe, Japan; 10 Gibson BE, Wheatley K, Hann IM, Stevens RF, Webb D, Hills RK et al. Treatment 17Department of Laboratory Medicine, Tokai strategy and long-term results in paediatric patients treated in consecutive UK University School of Medicine, Isehara, Japan; AML trials. Leukemia 2005; 19: 2130–2138. 18 11 Shimada A, Taki T, Tabuchi K, Tawa A, Horibe K, Tsuchida M et al. KIT Center for iPS Cell Research and Application, mutations, and not FLT3 internal tandem duplication, are strongly associated with Kyoto University, Kyoto, Japan and a poor prognosis in pediatric acute myeloid leukemia with t(8;21): a study of the 19 Human Health Sciences, Kyoto University, Kyoto, Japan Japanese Childhood AML Cooperative Study Group. Blood 2006; 107: 1806–1809. E-mail: [email protected] 12 Burnett AK, Hills RK, Milligan D, Kjeldsen L, Kell J, Russell NH et al. Identification of patients with acute myeloblastic leukemia who benefit from the addition of gemtuzumab ozogamicin: results of the MRC AML15 Trial. J Clin Oncol 2011; 29: 369–377. REFERENCES 13 Castaigne S, Pautas C, Terre C, Raffoux E, Bordessoule D, Bastie JN et al. Effect of 1 Rubnitz JE, Inaba H, Dahl G, Ribeiro RC, Bowman WP, Taub J et al. Minimal residual gemtuzumab ozogamicin on survival of adult patients with de-novo acute disease-directed therapy for childhood acute myeloid leukaemia: results of the myeloid leukaemia (ALFA-0701): a randomised, open-label, phase 3 study. Lancet AML02 multicentre trial. Lancet Oncol 2010; 11: 543–552. 2012; 379: 1508–1516. Associations between genome-wide Native American ancestry, known risk alleles and B-cell ALL risk in Hispanic children

Leukemia (2013) 27, 2416–2419; doi:10.1038/leu.2013.130 in genome-wide association studies of European-ancestry populations (IKZF1, CDKN2A, PIP4K2A, ARID5B, CEBPE) were more common in individuals with greater levels of Native American Hispanic children have a 10–30% greater incidence rate of ALL ancestry. Finally, we quantified the contribution of these validated than non-Hispanic whites, and nearly double the rate observed in risk loci to the increased ALL incidence observed in Hispanics African-Americans.1 Ethnic differences in ALL incidence may be relative to populations of European or African ancestry. explained by population-level differences in the frequency of Study participants were Hispanic children from the CCLS, whose genetic risk factors, including those first discovered in genome- recruitment and enrollment procedures have been described in wide association studies of European-ancestry populations.2–5 As detail previously (Supplementary Table S1).7,8 Cytogenetic Hispanics are an admixed population with European, African and characteristics of included cases are shown in Supplementary Native American ancestry, differences in ALL incidence observed Table S1. DNA was isolated from dried bloodspots collected at in Hispanics may be attributable to genetic risk factors associated birth and archived by the California Department of Public Health. with Native American ancestry. Samples were genotyped using the Illumina OmniExpress plat- Increased Native American ancestry has been linked to form, assaying 730 525 single-nucleotide polymorphism (SNPs). increased risk of relapse among Hispanic children with ALL,6 but Samples with genotyping call rateso98%, with discordant sex no study has yet investigated the contribution of genome-wide information (reported versus genotyped sex), or showing evidence Native American ancestry to ALL incidence. Using genome-wide of cryptic relatedness were excluded from analyses. To exclude SNP data from 298 Hispanic children with B cell ALL and 456 poorly genotyped SNPs, SNPs with genotyping call rates o98% or matched controls from the California Childhood Leukemia Hardy–Weinberg Equilibrium P-value o1 Â 10 À 5 in controls were Study (CCLS), we investigated whether genome-wide Native- removed from analyses. American ancestry was associated with increased risk of B-cell ALL. A linkage-reduced set of 63 303 autosomal SNPs, evenly Additionally, we assessed whether the risk alleles at loci identified distributed across the genome, was extracted from the

Accepted article preview online 25 April 2013; advance online publication, 17 May 2013

Leukemia (2013) 2376 – 2424 & 2013 Macmillan Publishers Limited Letters to the Editor 2417 case-control data and the Human Genome Diversity Project group allele frequencies using previously described methods.11 (HGDP) data. The genetic structure of study subjects was Additional information on samples, genotyping and statistical evaluated using Structure v2.3.1 to estimate percent membership procedures is available in the Supplementary Methods. in three distinct founder populations: sub-Saharan African, A total of 297 cases and 454 controls passed all quality control European and Native American.9 Founder population allele filters. Four SNPs identified as ALL risk factors in previous genome- frequencies were defined using SNP data from 372 unrelated wide association studies were significantly associated with ALL risk HGDP individuals, including 111 Africans, 107 Native Americans in our Hispanic sample (Supplementary Table S2). The strongest and 154 Europeans.10 association was at rs7089424 in ARID5B (odds ratio (OR) ¼ 2.33, Logistic regression was used to determine if Native American 95% confidence interval (CI): 1.85-2.92, P ¼ 2.6 Â 10 À 14). As ancestry was associated with case-status, with adjustment for sex, previously reported,2–4 this effect was stronger in hyperdiploid age and risk SNPs (where indicated). Logistic regression was also cases (OR ¼ 2.91, 95% CI: 2.05–4.12, P ¼ 2.1 Â 10 À 10). SNP used to determine if these SNPs were associated with case-status, rs2239633 in CEBPE was also more strongly associated with after adjustment for sex and age. We report results for the five hyperdiploid B-cell ALL (OR ¼ 2.07, 95% CI: 1.44–2.98, SNPs (one in each risk locus) that achieved genome-wide P ¼ 8.9 Â 10 À 5) than with B-cell ALL not stratified by subtype significance in a previously published genome-wide association (OR ¼ 1.35, 95% CI: 1.09–1.68, P ¼ 6.6 Â 10 À 3). Although study and which were successfully genotyped on our Illumina rs7088318 in PIP4K2A was not statistically significantly associated platform. Although the array data provides genotypes for with B-cell ALL risk in our sample (OR ¼ 1.16, 95% CI: 0.92–1.49, additional SNPs in these regions, we believed it important to P ¼ 0.21), the association approached significance among analyse Native American ancestry in relation to risk loci first hyperdiploid cases (OR ¼ 1.37, 95% CI: 0.96–1.96, P ¼ 0.084). Risk identified in populations of European-ancestry. alleles at rs4132601 (IKZF1) and rs3731217 (CDKN2A) were also Correlations between Native American ancestry and number of strongly associated with B-cell ALL risk in our case-control sample risk alleles in IKZF1, CDKN2A, PIP4K2A, ARID5B and CEBPE were (OR ¼ 1.46, 95% CI: 1.16–1.83, P ¼ 1.3 Â 10 À 3 and OR ¼ 1.76, 95% assessed using Pearson’s correlation coefficient. The contribution CI: 1.17–2.65, P ¼ 4.6 Â 10 À 3, respectively). of known susceptibility loci to ethnic incidence rate ratios were Compared with controls, cases had higher levels of Native calculated according to varying genotypic relative risks and ethnic American ancestry and lower levels of European ancestry

Table 1. Correlation coefﬁcients for number of risk alleles at known ALL risk loci and percent membership in each of three ancestral populations among CCLS controls, cases and combined sample

rs4132601-G rs3731217-T rs7088318-A rs7089424-C rs2239633-G (IKZF1) (CDKN2A) (PIP4K2A) (ARID5B) (CEBPE)

rPr Pr P r P rP

Controls only % Native American Ancestry 0.0088 0.85 0.12 0.014 0.18 7.4 Â 10 À 5 0.13 0.0060 0.098 0.027 % European Ancestry À 0.0018 0.97 À 0.11 0.018 À 0.18 7.5 Â 10 À 5 À 0.11 0.015 À 0.097 0.039 % African Ancestry À 0.021 0.65 0.001 0.91 0.034 0.46 À 0.027 0.56 À 0.003 0.94

Cases only % Native American Ancestry À 0.12 0.040 0.070 0.23 0.16 0.0044 0.082 0.16 0.031 0.60 % European Ancestry 0.067 0.25 À 0.058 0.32 À 0.12 0.039 À 0.077 0.19 À 0.055 0.35 % African Ancestry 0.14 0.018 À 0.029 0.62 À 0.11 0.051 À 0.0083 0.89 0.072 0.22

Cases and controls % Native American Ancestry À 0.035 0.35 0.11 0.0036 0.18 2.1 Â 10 À 5 0.13 0.00060 0.081 0.027 % European Ancestry 0.016 0.65 À 0.10 0.0061 À 0.16 4.6 Â 10 À 5 À 0.12 0.0013 À 0.087 0.017 % African Ancestry 0.054 0.14 À 0.0037 0.92 À 0.025 0.50 À 0.080 0.83 0.033 0.37

Nominally signiﬁcant P-values (o0.05) appear in bold.

Table 2. SNP effect size, risk allele frequency and contribution to B-cell ALL ethnic incidence rate ratios (IRR) by established susceptibility loci

Risk allelea SNP effect size Risk allele frequencyc Hispanic-Caucasian Hispanic-African (95% CI)b IRR (95% CI) IRR (95% CI) Caucasian African Hispanic Native American

rs4132601-G (IKZF1) 1.46 (1.16–1.83) 0.301 0.212 0.160 0.224 0.904 (0.843–0.953) 0.953 (0.912–0.987) rs3731217-T (CDKN2A) 1.76 (1.17–2.65) 0.863 0.907 0.880 1.000 1.003 (0.981–1.026) 0.974 (0.952–1.005) rs7088318-A (PIP4K2A) 1.16 (0.92–1.49) 0.588 0.243 0.727 0.939 1.041 (0.993–1.091) 1.163 (0.976–1.362) rs7089424-C (ARID5B) 2.33 (1.85–2.92) 0.304 0.235 0.380 0.570 1.110 (1.005–1.212) 1.236 (1.149–1.343) rs2239633-G (CEBPE) 1.35 (1.09–1.68) 0.522 0.832 0.610 0.612 1.031 (1.002–1.067) 0.893 (0.845–0.953) Abbreviations: CI, confidence interval; IRR, incidence rate ratio; SNP, single-nucleotide polymorphism. aSNPs have previously been reported to increase risk for B-cell ALL in a published genome-wide association study. bOR are derived from the CCLS Hispanic case-control study, comparing 297 B-cell ALL cases to 454 controls, adjusted for age, sex and the first five principal components. cAllele frequencies for Caucasians, Africans and Native Americans are from Human Genome Diversity Panel data. Allele frequencies for Hispanics are from HapMap data.

& 2013 Macmillan Publishers Limited Leukemia (2013) 2376 – 2424 Letters to the Editor 2418 (Supplementary Table S1 and Supplementary Figure S1). After and African-American populations would be expected to have adjustment for age, sex and percent African ancestry, each 20% higher ALL incidence than European populations. As African- increase in Native American ancestry was associated with a 1.20- Americans have lower ALL incidence than Europeans, it appears fold increase in risk of B-cell ALL (OR ¼ 1.20, 95% CI: 1.00–1.45, the Native American component of Hispanic ancestry may be P ¼ 0.048) (Supplementary Table S2). The association between a risk factor, and not that the European component is a protective genome-wide Native American ancestry and ALL risk was factor. This is further corroborated by our observations that modestly attenuated when controlling for genotype at known risk alleles in CDKN2A, PIP4K2A, CEBPE and ARID5B were rs3731217 (CDKN2A), rs7088318 (PIP4K2A) and rs2239633 (CEBPE) all significantly associated with increased Native American (1, 2.5 and 4.2% decreases, respectively), and was further ancestry. attenuated when conditioned on genotype at rs7089424 (ARID5B, In conclusion, we demonstrate that increased genome-wide 6.6% decrease) (Supplementary Table S2). These SNPs, in Native American ancestry is associated with an increased particular rs7089424, may contribute to the observed association risk of B-cell ALL in Hispanic children, and trace this to the effects between Native American ancestry and ALL risk. of at least three genes. Additional questions remain as to whether Further support for this was shown when correlations were the known risk loci can account for all of the increased B-cell ALL calculated between Native American ancestry and number of risk risk observed in Hispanics, or if additional risk loci can be alleles at the five ALL risk SNPs. The number of risk alleles at four identified though further study of this high-risk population. of these SNPs was positively and significantly correlated with increased Native American ancestry (Table 1). The strongest of these associations were with ARID5B and PIP4K2A SNPs (r ¼ 0.13, CONFLICT OF INTEREST P ¼ 6.0 Â 10 À 4 and r ¼ 0.18, P ¼ 2.1 Â 10 À 5, respectively). The number of risk alleles at rs3731217 (CDKN2A) and rs2239633 The authors declare no conflict of interest. (CEBPE) was also positively correlated with increased Native American ancestry (r ¼ 0.11, P ¼ 3.6 Â 10 À 3 and r ¼ 0.081, P ¼ 0.027, respectively). These associations were consistent when ACKNOWLEDGEMENTS analyses were restricted to control subjects, indicating that these This work was supported by National Institutes of Health grants: R25CA112355 associations reflect population structure, independent of case- (KMW), R01CA155461 (JLW, XM), R01CA126831 (JKW) and R01ES009137 (APC, status (Table 1). LH, CM, GVD, MLL, KB, LFB, JLW, and PAB). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the We next assessed whether these risk loci contribute to the manuscript. increased ALL incidence observed in Hispanics relative to populations of European or African ancestry (Table 2). Interestingly, the risk KM Walsh1,2,8, AP Chokkalingam3,8, L-I Hsu3, C Metayer3, allele of rs3731217 in CDKN2A has an allele frequency of 100% in 4 5 6 7 2 3 Native Americans. Despite the absence of the minor (protective) AJ de Smith ,DIJacobs,GVDahl,MLLoh,IVSmirnov, K Bartley , XMa5,JKWiencke2,LFBarcellos3, JL Wiemels4 and PA Buffler3 allele in this population, this SNP explains only a small proportion of 1 the increased B-cell ALL risk observed in Hispanics compared with Program in Cancer Genetics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA, USA; European or African-ancestry populations. 2 Previously identified risk alleles in CEBPE, PIP4K2A and ARID5B Division of Neuroepidemiology, Department of Neurological are also more common in Native American and Hispanic Surgery, University of California, San Francisco, CA, USA; 3School of Public Health, University of California, Berkeley, CA, USA; populations than in Europeans. SNP rs2239633 in CEBPE 4 accounted for a 1.03-fold increased risk of B-cell ALL in Hispanics Department of Epidemiology and Biostatistics, University of California, San Francisco, CA, USA; versus Caucasians (95% CI: 1.002–1.067). In addition, rs7089424 in 5 ARID5B accounted for a 1.11-fold increased risk of B-cell ALL in Yale School of Public Health, Yale School of Medicine, New Haven, CT, USA; Hispanics versus Caucasians (95% CI: 1.005–1.212) (Table 2). As this 6 SNP is more strongly associated with hyperdiploid B-cell ALL than Division of Pediatric Hematology/Oncology, Department of Pediatrics, Stanford School of Medicine, Stanford, CA, USA and with other subtypes, it can explain an even larger proportion of 7 the differences observed across populations in the incidence of Department of Pediatrics, Benioff Children’s Hospital, University of this ALL subtype (Supplementary Table S3). California, San Francisco, CA, USA E-mail: [email protected] Our findings suggest that the increased risk of B-cell ALL 8 observed in Hispanic populations is due, at least in part, to an The first two authors should be regarded as joint first authors. effect of Native American ancestry. In our sample, each 20% increase in the proportion of an individual’s genome that is of Native American origin conferred a 1.20-fold increased risk of REFERENCES B-cell ALL. Because increased Native American ancestry was also 1 Yamamoto JF, Goodman MT. Patterns of leukemia incidence in the United States associated with known ALL risk alleles, even among controls, we by subtype and demographic characteristics, 1997-2002. Cancer Causes Control believe the increased risk of ALL associated with increased Native 2008; 19: 379–390. American ancestry is not easily attributed to potential confound- 2 Trevino LR, Yang W, French D, Hunger SP, Carroll WL, Devidas M et al. Germline genomic variants associated with childhood acute lymphoblastic leukemia. Nat ing factors. Genet 2009; 41: 1001–1005. Taken together, the risk alleles in CDKN2A, PIP4K2A, CEBPE and 3 Papaemmanuil E, Hosking FJ, Vijayakrishnan J, Price A, Olver B, Sheridan E et al. ARID5B may account for an important proportion of the ALL Loci on 7p12.2, 10q21.2 and 14q11.2 are associated with risk of childhood acute incidence differences observed across ethnicities. Although lymphoblastic leukemia. Nat Genet 2009; 41: 1006–1010. these variants are associated with ALL risk in numerous 4 Sherborne AL, Hosking FJ, Prasad RB, Kumar R, Koehler R, Vijayakrishnan J et al. populations,5,12–14 their increased frequency in populations with Variation in CDKN2A at 9p21.3 influences childhood acute lymphoblastic Native American ancestry may result from a founder effect leukemia risk. Nat Genet 2010; 42: 492–494. occurring during migration to the New World and genetic drift 5 Xu H, Yang W, Perez-Andreu V, Devidas M, Fan Y, Cheng C et al. Novel Sus- during subsequent population expansion. ceptibility Variants at 10p12.31-12.2 for Childhood Acute Lymphoblastic Leukemia in Ethnically Diverse Populations. J Natl Cancer Inst 2013; e-pub ahead of print As a corollary to the positive association between Native 9 March 2013. American ancestry and ALL risk, increased European ancestry is 6 Yang JJ, Cheng C, Devidas M, Cao X, Fan Y, Campana D et al. Ancestry and associated with decreased B-cell ALL risk in this Hispanic sample. pharmacogenomics of relapse in acute lymphoblastic leukemia. Nat Genet 2012; However, were European ancestry protective, both Hispanic 43: 237–241.

Leukemia (2013) 2376 – 2424 & 2013 Macmillan Publishers Limited Letters to the Editor 2419 7 Ma X, Bufﬂer PA, Layefsky M, Does MB, Reynolds P. Control selection strategies in susceptibility to glioma: a novel candidate SNP approach. Front Genet case-control studies of childhood diseases. Am J Epidemiol 2004; 159: 915–921. 2012; 3: 203. 8 Aldrich MC, Zhang L, Wiemels JL, Ma X, Loh ML, Metayer C et al. Cytogenetics of 12 Vijayakrishnan J, Sherborne AL, Sawangpanich R, Hongeng S, Houlston RS, Hispanic and White children with acute lymphoblastic leukemia in California. Pakakasama S. Variation at 7p12.2 and 10q21.2 inﬂuences childhood acute lym- Cancer Epidemiol Biomarkers Prev 2006; 15: 578–581. phoblastic leukemia risk in the Thai population and may contribute to racial 9 Falush D, Stephens M, Pritchard JK. Inference of population structure using differences in leukemia incidence. Leuk Lymphoma 2010; 51: 1870–1874. multilocus genotype data: linked loci and correlated allele frequencies. Genetics 13 Yang W, Trevino LR, Yang JJ, Scheet P, Pui CH, Evans WE et al. ARID5B SNP rs10821936 2003; 164: 1567–1587. is associated with risk of childhood acute lymphoblastic leukemia in blacks 10 Li JZ, Absher DM, Tang H, Southwick AM, Casto AM, Ramachandran S et al. and contributes to racial differences in leukemia incidence. Leukemia 2010; 24: Worldwide human relationships inferred from genome-wide patterns of variation. 894–896. Science 2008; 319: 1100–1104. 14 Xu H, Cheng C, Devidas M, Pei D, Fan Y, Yang W et al. ARID5B genetic poly- 11 Jacobs DI, Walsh KM, Wrensch M, Wiencke J, Jenkins R, Houlston RS et al. morphisms contribute to racial disparities in the incidence and treatment out- Leveraging ethnic group incidence variation to investigate genetic come of childhood acute lymphoblastic leukemia. J Clin Oncol 2012; 30: 751–757.

Supplementary Information accompanies this paper on the Leukemia website (http://www.nature.com/leu)

Detection of PICALM-MLLT10 (CALM-AF10) and outcome in children with T-lineage acute lymphoblastic leukemia

Leukemia (2013) 27, 2419–2421; doi:10.1038/leu.2013.149 protocol and statistical analyses are reported on Supplementary Materials. Characteristics of the 187 evaluable children with T-ALL are Approximately 10–15% of pediatric patients with acute lympho- listed in Table 1; of them, 14 children (7.5%) were positive blastic leukemia (ALL) have a T-cell phenotype. The prognosis of for PICALM-MLLT10 and their characteristics are listed in these patients has improved over the last years, owing to the Supplementary Table 1. use of more intensive treatment strategies.1 Characterization The PICALM-MLLT10 fusion transcript was detected more of molecular alterations with prognostic impact in T-ALL may be of frequently in males (12 out of 14). Patient’s median age was 7 great help for an early identification of patients at high risk (HR) of years (range 2–13). PICALM-MLLT10-positive cases were more likely failure in whom more intensive treatments, including allogeneic to present with HR features, including a white blood cell (WBC) hematopoietic stem cell transplantation, may be considered. count higher than 100 000/ml(n ¼ 8).1,8 PICALM-MLLT10 fusion PICALM (clathrin-assembly protein-like lymphoid myeloid transcript was significantly more frequent in patients with cortical leukemia gene)-MLLT10 (formerly CALM-AF10) results from a and mature T, than in those with early-T immunophenotype recurring t(10;11)(p13;q14-21) chromosomal translocation and is (P ¼ 0.02). Eight patients were ‘poor prednisone (PDN) responders’ the most frequent fusion transcript detected in patients with (PPR), which, by study design, were assigned to the HR group. T-ALL, at an overall rate of 10%, including both adults and Based on MRD levels, in the PICALM-MLLT10-positive group, eight children.2 This translocation fuses PICALM (or CALM) and MLLT10 children were presented with an intermediate risk pattern and six (also called AF10). The Groupe Français de Cytogeńe´tique children with an HR pattern (including three who also were PPR). He´matologique firstly observed this translocation in a patient Based on the type of steroid given in induction, nine children with histiocytic lymphoma.3 The immunophenotypic with PICALM-MLLT10-positive T-ALL were treated with PDN characterization of the leukemic cell bearing this fusion gene and five children were treated with dexamethasone (DXM). showed that mature PICALM-MLLT10-positive cases expressed Among the PICALM-MLLT10-negative patients, DXM was given CD5, T-cell receptor (TCR) g/d and CD4 or CD8 (or both), whereas to 53 (30%) out of 173 children treated on both AIEOP- Berlin- immature cases, which are currently defined as early-T, expressed Frankfurt-Munster ALL-2000 (n ¼ 38) and AIEOP R-2006 (n ¼ 15) few T-lineage markers other than CD5, TdT, cCD3 and CD7, and studies, respectively. Details on patients’ outcome are shown in were often positive for CD13, CD33 or CD34.4 The presence of Supplementary Table 2. PICALM-MLLT10 has been associated with a poor prognosis and The relapse rate was similar in both PICALM-MLLT10-positive several studies included very few children with T-ALL.4–7 and negative patients. In the group of PICALM-MLLT10-positive We analyzed a large cohort of children with T-ALL, enrolled patients, relapse occurred in four patients (21.5%) and involved in two subsequent AIEOP (Associazione Italiana Ematologia the central nervous system in three: this proportion appears Oncologia Pediatrica) protocols, with the aim of evaluating both to be higher than in the PICALM-MLLT10-negative counterpart incidence and prognostic impact of PICALM-MLLT10. From 1 (5.8%; P ¼ 0.02). September 2000 to 31 December 2007, a total of 309 patients with There was no statistically significant difference in terms of T-lineage ALL (11.8% of the total number of children diagnosed final outcome between PICALM-MLLT10-positive and -negative with ALL), aged between 1 and 18 years (infants o1 year of age patients, the 5-year event-free survival (EFS) being 71.4% (SE 12.1) were eligible for a separate protocol), were treated in Italy vs 62.5% (SE 3.8) (P ¼ 0.53), respectively (Figure 1a). Consistently, on the AIEOP- Berlin-Frankfurt-Munster (BFM) ALL-2000 (n ¼ 258 cumulative incidence (CI) of relapse was very similar, as shown in until 31 July 2006) or the subsequent AIEOP R-2006 (n ¼ 51) Figure 1b. Same results were obtained after adjusting for risk studies. Leftover biological material for analysis of molecular group in a Cox model (data not shown). alterations was available in 187 children. Follow-up was updated When the analysis was restricted to the subgroup of patients at December 2009 and median follow-up was 4.7 years. with cortical or mature immunophenotype, which included the Detailed information regarding diagnosis, methods, risk majority of PICALM-MLLT10-positive patients (n ¼ 12), there was stratification based on minimal residual disease (MRD), treatment no statistically significant difference between PICALM-MLLT10-

Accepted article preview 14 May 2013; advance online publication, 18 June 2013

& 2013 Macmillan Publishers Limited Leukemia (2013) 2376 – 2424