- Email: [email protected]
ARID5B, CEBPE and PIP4K2A Germline Genetic Polymorphisms and Risk of Childhood Acute Lymphoblastic Leukemia in Mexican Patients: A MIGICCL Study
Archives of Medical Research 47 (2016) 623e628
ORIGINAL ARTICLE
ARID5B, CEBPE and PIP4K2A Germline Genetic Polymorphisms and Risk of Childhood Acute Lymphoblastic Leukemia in Mexican Patients: A MIGICCL Study Vilma Carolina Bekker-Mendez,a Juan Carlos N u~ nez-Enrıquez,b Jose Luis Torres Escalante,c Enrique Alvarez-Olmos,d Pablo Miguel Gonzalez-Montalvoc,e Elva Jimenez-Hernandez,f Aurora Medina Sans on,g Yelda A. Leal,h Marıa Teresa Ramos-Cervantes,a Francisco Xavier Guerra-Castillo,a M onica Patricia Ortiz-Maganda,a Janet Flores-Lujano,b b Maria Luisa Perez-Saldivar, Martha Margarita Velazquez-Avi~ na,i Victoria Bolea-Murga,j Jose Refugio Torres-Nava,k Raquel Amador-Sanchez,l Karina Anastacia Solis-Labastida,m Julian Ramirez-Bello,n Jose Manuel Fragoso,o Juan Manuel Mejıa-Arangure,b,p and MIGICCLq a
Unidad de Investigacion Medica en Inmunologıa e Infectologıa, Hospital de Infectologıa ‘‘Dr. Daniel Mendez Hernandez’’, ‘‘La Raza’’, IMSS, Mexico City, Mexico b Unidad de Investigaci on Medica en Epidemiologıa Clınica, UMAE Hospital de Pediatrıa, Centro Medico Nacional (CMN) ‘‘Siglo XXI’’, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico c Servicio de Pediatrıa de la UMAE, Instituto Mexicano del Seguro Social (IMSS); Yucatan, Mexico d Laboratorio de Genomica del Cancer, Instituto Nacional de Medicina Genomica (INMEGEN), Mexico City, Mexico e Servicio de Oncologıa Pediatrica del Hospital O’Horan, SS, Yucatan, Mexico f Servicio de Hematologıa Pediatrica, Hospital General ‘‘Gaudencio Gonzalez Garza’’, Centro Medico Nacional (CMN) ‘‘La Raza’’, IMSS, Mexico City, Mexico g Servicio de Hemato-Oncologia, Hospital Infantil de Mexico Federico Gomez, Secretaria de Salud (SSa), Mexico City, Mexico h Unidad de Investigacion Medica Yucatan, Registro de Cancer, Unidad Medica de Alta Especialidad IMSS Merida, Yucatan, Mexico i Servicio de Onco-Pediatria, Hospital Juarez de Mexico, Secretaria de Salud (SSa), Mexico City, Mexico j Servicio de Hematologıa Pediatrica, Hospital General de Mexico, Secretaria de Salud (SSa), Mexico City, Mexico k Servicio de Oncologıa, Hospital Pediatrico de Moctezuma, Secretarıa de Salud del D.F., Mexico City, Mexico l Hospital General Regional No. 1 ‘‘Carlos McGregor Sanchez Navarro’’, IMSS, Mexico City, Mexico m Servicio de Hematologıa Pediatrica, UMAE Hospital de Pediatrıa, Centro Medico Nacional (CMN) ‘‘Siglo XXI’’, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico n Laboratorio de la Unidad de Investigacion en Enfermedades Metabolicas y Endocrinas del Hospital Juarez de Mexico, Mexico City, Mexico o Departamento de Biologıa Molecular, Instituto Nacional de Cardiologıa, Ignacio Chavez, Ciudad de Mexico, Mexico p Coordinacion de Investigacion en Salud, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico q Mexican Inter-Institutional Group for the Identification of the Causes of Childhood Leukaemia, Instituto Mexicano del Seguro Social, Instituto de Seguridad Social al Servicio de los Trabajadores del Estado, Secretarıa de Salud, Secretarıa de Salud del Gobierno del Distrito Federal, Mexico City, Mexico Received for publication September 29, 2016; accepted November 24, 2016 (ARCMED-D-16-00585).
Background and Aims. Childhood acute lymphoblastic leukemia (ALL) is the leading cause of childhood cancer-related deaths worldwide. Multiples studies have shown that ALL seems to be originated by an interaction between environmental and genetic susceptibility factors. The ARID5B polymorphisms are among the most reproducible ALL associated-risk alleles in different populations. The aim of the present study was to examine the contribution of ARID5B, CEBPE, and PIP4K2 risk alleles for the development of ALL in children from Mexico City and Yucatan, Mexico.
Address reprint request to: Dr. Vilma Carolina Bekker-Mendez, Unidad de Investigaci on en Inmunologıa e Infectologıa. Hospital de Infectologıa CMN ‘‘La Raza’’ IMSS. Tel/Fax: þ55 5724 5900 ext. 24322; 24321 Mexico, DF, Mexico; E-mail: bekkermendez@yahoo. com; Dr. Juan Manuel Mejıa-Arangure, Unidad de Investigacion Medica en Epidemiologıa Clınica, UMAE Hospital de Pediatrıa, Centro Medico
Nacional (CMN) ‘‘Siglo XXI’’, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico and Coordinacion de Investigacion en Salud, Instituto Mexicano del Seguro Social (IMSS), Torre Academia Nacional de Medicina 4to piso, Av. Cuauhtemoc 330, 06720 Mexico, DF, Mexico. Phone number: þ52155 56276900 ext. 21294. Fax number: þ52 155 57610841; E-mail: [email protected], [email protected].
0188-4409/$ - see front matter. Copyright Ó 2016 IMSS. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.arcmed.2016.12.003
624
Bekker-Mendez et al./ Archives of Medical Research 47 (2016) 623e628
Methods. A study was conducted with a total of 761 unrelated subjects. Two hundred eighty five ALL cases (111 from Yucatan and 174 from Mexico City) and 476 healthy subjects. Genotyping included the rs7088318 (PIP4K2A), rs10821936 (ARID5B), rs7089424 (ARID5B) and rs2239633 (CEBPE ) polymorphisms. Results. Associations between ALL and rs10821936 and rs7089424 ARID5B SNPs were found (OR 5 1.9, 95% CI (1.5e2.4) and OR 5 2.0, 95% CI (1.6e2.5), respectively). Moreover, a higher risk was observed in the homozygous risk genotypes of carriers from Mexico City (OR 5 3.1, 95% CI (2.0e4.9) and OR 3.1, CI 95% (2.0e4.8), respectively). Otherwise, the rs7088318 (PIP4K2A) and rs2239633 (CEBPE ) polymorphisms were not associated with ALL risk. Conclusions. Our analysis suggests that ARID5B confers risk for childhood ALL in a Mexican population. Ó 2016 IMSS. Published by Elsevier Inc. Key Words: Association study, ARID5B, PIP4K2A, CEBPE, Childhood acute lymphoblastic leukemia, Mexican.
Acute lymphoblastic leukemia (ALL) is the most frequent subtype of childhood cancer worldwide (1e3). The etiology of childhood ALL has not been established; however, many studies support the hypothesis that genetic polymorphisms influence racial differences in ALL incidence, either when the frequency of an ALL-predisposing genetic polymorphism differs by ethnicity or when these variants are associated with ALL in a race-specific manner (4e7). Genomewide association studies (GWAS) have identified ALLsusceptibility loci at 7p12.2, 9p21.3, 10p12.2, 10q21.2 and 14q11.2 (8,9). The most consistent across studies is the 10q21.2 locus, which harbors the AT-rich interactive domain 5B (ARID5B) gene. ARID5B, which is involved in regulation of embryonic development, cell growth and B-lymphocyte progenitor differentiation, bears germline single nucleotide polymorphisms (SNP) located in intron 3 that are strongly associated with risk of pediatric ALL including rs10821936 and rs7089424. Although its specific role in the pathogenesis of ALL remains unknown, one of the highlighted findings is the association between ARID5B SNPs with ALL susceptibility in multiple investigated populations. Interestingly, differences in the frequencies of the risk alleles among Caucasian, African, Asian and Hispanic derived populations have been found, being the highest in Hispanic descendent populations (9e12). As an example, the frequency of the rs10821936C risk allele was lower in control populations of African ancestry (0.13 in ASW and 0.2 in YRI) than in populations of European ancestry (0.33) (6). Likewise, the CCAAT/enhancer-binding protein epsilon (CEBPE ) and phosphatidylinositol-5-phosphate 4-kinase, type II, alpha (PIP4K2A) genes have been added to the list of risk genes associated with the development of childhood ALL in other populations (11e14). It is interesting to note that Hispanics have shown a worse prognosis, often need more intensive ALL treatment regimens, show older ages at diagnosis, and exhibit higher frequencies of the ARID5B ALL-risk variants (also linked to poorer outcomes) than non-Hispanic Whites (5,15). Mexican as well as Hispanic populations are characterized by a very complex genetic
background due to the different contributions from Native American, Caucasian and African populations during the conquest (16). Nowadays, the Mexican population is comprised by Mestizos (94%) and Amerindians (6%) (17,18). ALL has higher incidence rates in Hispanics than in other ethnic groups and is the leading cause of mortality for cancer in Mexican children (3). It is possible that differences in genetic susceptibility loci could explain the higher observed ALL incidence rates in Mexican population. The aim of the present study was to examine the contribution of ARID5B, CEBPE, and PIP4K2 risk alleles for the development of ALL in children from Mexico City and Yucatan, Mexico.
Methods Subjects We conducted a study in a sample of 761 unrelated Mexican subjects recruited in Merida, Yucatan and Mexico City. Cases were !18 years of age diagnosed with ALL, recruited from two hospitals from Merida, Yucatan and three tertiary level hospitals in Mexico City. Diagnosis of ALL was based on morphology and immunophenotype studies. ALL patients were classified as standard risk (age 1e9.99 years and initial white blood cell count (WBC) !50 109/L) or as high risk (NCI HR) (age ! 1 year or $10 years and initial WBC $50 109/L) according to the National Cancer Institute (NCI) Risk Classification (19). Controls corresponded to healthy adults from Mexico City recruited in the General Hospital of Mexico and Hospital Juarez of Mexico City. Children from Yucatan were included from the Unidad Medica de Alta Especialidad (UMAE)-IMSS, Merida, Yucatan (third-level hospital). Only individuals self-reported as having not been diagnosed with autoimmune, inflammatory, hematological diseases
ARID5B, CEBPE and PIP4K2A in Childhood Acute Lymphoblastic Leukemia
or cancer were included as controls. Institutional review committees from the participating institutions approved this study. Written informed consent was obtained from parents of children with ALL and from the control group. When possible, the child’s assent was also obtained. DNA Extraction We included bone marrow or saliva samples from ALL cases and peripheral blood from controls. Genomic deoxyribonucleic acid (DNA) from bone marrow and peripheral blood samples were extracted using the standard salting out technique (QIAGEN systems, Inc., Valencia, CA) and the genomic DNA extraction from the saliva was done using the ORAGENE Purification Kit (DNA Genotek Inc., ON, Canada) according to the manufacturer’s instructions. DNA purity and concentration was determined by spectrophotometry. Genotyping Analysis We selected four SNPs based on previous association studies in ALL, including two highly replicated ARID5B (rs10821936, rs7089424) variants (Table 1). Genotyping of the PIP4K2A (rs7088318), ARID5B (rs10821936, rs7089424) and CEBPE (rs2239633) was performed with the ABI 7900HT system using 10 ng of input DNA and Genotyping Master Mix and TaqMan Pre-Designed-SNPGenotyping-Assay (Life Technologies, Foster City, CA). PCR mix consisted of 10 ng of genomic DNA, 0.1 mM of each probe, 2.5 ml of TaqMan master mix (Life Technologies) and ddH2O up to a final volume of 5 ml. The amplification protocol included denaturing at 95 C for 10 min followed by 45 cycles of denaturing at 95 C for 15 sec and annealing and extension at 60 C for 1 min. The genotype of each sample was assigned automatically by measuring the allele-specific fluorescence using SDS 2.2.3 software for allelic discrimination (Applied Biosystems, Foster City, CA). A subset of 100 random samples was genotyped in duplicate and the reproducibility was 99%. To validate the TaqMan results, three random samples of each genotype were sequenced using the forward and reverse primers. PCR was performed with the following conditions: an initial denaturation at 95 C for 5 min, 35 cycles at 95 C for 30 s, 58 C for 30 s, and 68 C for 30 s, and a final extension at 72 C for 7 min. Sequencing was performed using an automated ABI PRISM 3100 DNA sequencer (Applied Biosystems). Statistical Analysis We used the FINNETI algorithm (http://ihg.gsf.de/cgi-bin/ hw/hwa1.pl) to test the Hardy-Weinberg equilibrium (HWE) for genotype distributions in ALL cases and controls. To assess the differences of the genotype and allele frequencies among populations and ethnic groups and
625
between genders we used the c2 test implemented in the Stat-Calc program (EPI INFO 7 v 7.201; Centers for Disease Control and Prevention, Atlanta, GA). We calculated the odds ratio (OR) and 95% confidence intervals (95% CI); p values !0.05 were considered statistically significant after adjusting for multiple comparisons base on the Bonferroni correction test ( p !0.01). Power calculations were performed using the Quanto Software, accounting for ALL prevalence in Mexico City (19), both the MAF and OR of each studied SNP.
Results Subjects This study included 761 unrelated subjects. A total of 285 ALL children fulfilled the ALL diagnostic criteria, from which 111 subjects (39.0%) were from Merida, Yucatan and 174 (61.0%) were from Mexico City, with a male predominance (55.0%). Patients’ age ranged from 0e17 years with a mean (SD) age of 5.7 2.8 years and 8.7 4.94 years in each population, respectively. There were no statistical differences between patients from Yucatan and Mexico City regarding gender and children’s age; however, patients from Yucatan showed an earlier-onset of ALL than others. The healthy group consisted of 476 subjects, 225 (47%) were male and 251 (53%) were female. We included healthy adults as controls because the risk alleles investigated did not seem to be associated with lifespan. Association Analysis No evidence from HWE was observed for any evaluated SNPs. Case-control association analysis including the 285 cases and the 476 controls did not show statistically significant differences in the distribution of the rs2239633 (CEBPE ) and rs7088318 (PIP4K2A) alleles between ALL patients and healthy subjects. Nevertheless, significant associations were found among both ARID5B SNPs (rs10821936 and rs7089424) and ALL. The rs10821936C and rs7089424G alleles were detected in 66.5% and 66.7% of patients, respectively, whereas they were carried by 50.6 and 49.2% of the healthy chromosomes, respectively, exhibiting both as risk alleles for leukemia (rs10821936C: OR 5 1.9, 95% CI (1.5e2.4), p !0.01 and rs7089424G: OR 5 2.0, 95% CI (1.6e2.5), p !0.01) (Table 1). To describe the observed ARID5B associations and capture associations under various genetic models, we measured the corresponding genotype OR in all samples. A strong allele dose-dependent effect on risk was observed at loci rs10821936 and rs7089424. Carriers of a homozygous risk genotype (rs10821936-CC and rs7089424-GG) homozygote mutant allele status meant increased risk (OR 5 3.1, 95% CI (2.0e4.9) p !0.01 and OR 5 3, 95% CI (2.0e4.8) p ! 0.01 respectively) to ALL (Table 1).
626
Bekker-Mendez et al./ Archives of Medical Research 47 (2016) 623e628
Table 1. Association analysis among five ALL risk alleles and Mexican patients ALL children Gen PIP4K2A
ARID5B
ID SNP
Genotypes
rs7088318
AA AC CC A C AA vs. CC TT TC CC T C TT vs. CC TT TG GG T G TT vs. GG rs2239633 GA AA G A AA vs. GG
rs10821936
rs7089424 n total
CEBPE
Healthy individuals n [ 285 n (%)
n [ 476 n (%)
211 68 6 490 80
(74.0) (23.9) (2.1) (86.0) (14)
341 124 11 805 147
(71.6) (26.1) (2.3) (84.6) (15.4)
40 111 134 191 379
(14) (38.9) (47.0) (33.5) (66.5)
38 114 133 190 380
(13.3) (40.0) (46.7) (33.3) (66.7)
111 248 117 470 482 3.1 107 249 120 483 469 3.1 122 245 64 580 372 0.75
(23.3) (52.1) (24.6) (49.4) (50.6) (2.0e4.9) (22.5) (52.3) (25.2) (50.7) (49.2) (2.0e4.8) (42.8) (51.5) (13.4) (60.9) (39.1) (0.47e1.20)
GG 128 35 372 198
(44.9) (12.3) (65.3) (34.7)
OR (95% CI)
p
0.89 (0.66e1.20)
NS
0.88 (0.3e2.4)
NS
1.9 (1.5e2.4)
!0.01
!0.01
2.0 (1.6e2.5)
!0.01
!0.01 167 (35.1)
0.83 (0.6e1.0)
NS NS
OR, odds ratio; ALL, acute lymphoblastic leukemia. Number of subjects. NS indicates no significant differences. Differences in allele frequencies were measured by c2 test.
Genotypes and Allele Frequencies After we performed an origin matched case-control analysis, we failed to find statistically significant differences among the rs7088318 (PIP4K2A), rs2239633 (CEBPE ) and ALL in either the Maya or Mestizo population. The observed allele frequencies are summarized in Table 2. rs10821936C and rs7089424G alleles were detected in 66.8 and 67.1% of patients, respectively, whereas they were carried by 51.3 and 55% of the healthy controls from Yucatan, respectively, exhibiting both as risk alleles for leukemia (rs10821936C: OR 5 1.8, 95% CI (1.2e2.7), p !0.01 and rs7089424G: OR 5 1.6, 95% CI (1.0e2.5), p !0.01) (Table 2). ARID5B (rs10821936-CC) exhibits statistically significant differences (OR 5 2.5, 95% CI (1.1e5.7) p !0.05), whereas ARID5B (rs7089424-GG) failed to show a statistically significant difference (OR 5 2.1, 95% CI (0.9e4.9) p 5 NS) between ALL patients and healthy subjects from Yucatan. These SNPs were also associated with ALL and Mestizo populations. Significant differences were observed in both ARID5B homozygous risk genotypes (rs10821936-CC and rs7089424-GG) [OR 5 3.6, 95% CI (2.0e6.3) p !0.01 and OR 5 3.5, 95% CI (2.0e6.2) p !0.01, respectively]. ARID5B rs10821936 T/C and rs7089424 T/G were found in high-linkage disequilibrium
(LD) (r2 5 0.89) and haplotype analysis was performed. Statistical significance remained after multiple comparisons (100,000 permutations).
Discussion This study is one of the first to examine the association of ALL risk alleles located in the ARID5B, CEBPE, and PIP4k2A genes and the susceptibility to ALL in children from two regions from Mexico. Combining overall ALL children vs. healthy subjects, the ARID5B, SNPs exhibited differential distribution between groups. Both ARID5B polymorphisms (rs10821936 and rs7089424) were associated with susceptibility to ALL. OR significantly increased for homozygote risk allele carriers of both SNPs. ARID5B encodes the AT-rich interactive domain 5B, a member of the ARID family of proteins, which are highly conserved and play a key role in adipogenesis and in liver development. This gene is also involved in the proliferation of early B-cell progenitors and its expression correlates with genes implicated in several hematopoietic diseases (20). Our data provide evidence of the ARID5B contribution to pediatric ALL risk in Mexican
ARID5B, CEBPE and PIP4K2A in Childhood Acute Lymphoblastic Leukemia
627
Table 2. Genotypes and alleles frequencies of the ARID5B polymorphism in ALL and healthy subjects stratified by ethnicity Yucatan
SNP
Genotypes
rs10821936
TT TC CC T C TT vs. CC TT TG GG T G TT vs. GG
rs7089424
ALL n [ 111 19 38 54 76 146
(17.1) (34.2) (48.6) (34.2) (65.8)
17 39 55 73 149
(15.3) (35.1) (49.5) (32.9) (67.1)
Healthy individuals n [ 78 n (%) 20 36 22 76 80 2.5 16 38 24 70 86 2.1
(25.6) (46.2) (28.2) (48.7) (51.3) (1.1e5.7) (20.5) (48.7) (30.8) (45) (55) (0.9e4.9)
Mexico City
OR (95% CI) n (%)
1.8 (1.2e2.7)
p
!0.01
!0.05
1.6 (1.0e2.5)
!0.05
ALL n [ 174 21 73 80 115 233 3.6 21 75 78 117 231
(12.1) (42.0) (46.0) (33.0) (67.0) (2.0e6.3) (12.1) (43.1) (44.8) (33.6) (66.4)
NS
Healthy individuals n [ 398 n (%) 91 (22.9) 212 (53.3) 95 (23.9) 394 (49.5) 402 (50.5) !0.01 91 (22.9) 211 (53.0) 96 (24.1) 392 (49.2) 404 (50.8)
OR (95% CI) n (%)
p
1.9 (1.5.e2.5)
!0.01
1.9 (1.4e2.4)
!0.01
3.5 (2.0e6.2)
!0.01
n, total number of subjects; OR, odds ratio. NS indicates no significant differences. Differences in allele frequencies were measured by c2 test.
patients as has been observed in Caucasian-, Asian-, African- and Hispanic- derived populations (5,6,21). The observed risk for these loci in our cases vs. most populations was demonstrated to be similar to previously reported effects (22e24); however, the estimated risk for C (rs10821936) and G (rs7089424) allele homozygote carriers was higher than data frequently reported for B-lineage ALL. More interestingly, the frequencies of both risk-alleles documented in the present study seem to be the highest reported to date. To note, in a recent metaanalysis, the SNP rs7089424 showed significant associations among Caucasians but not among Asians in all genetic models. Further, the GG genotype frequencies in non-ALL controls display a decreasing tendency from Caucasian to Asian subgroups. Although we did not observe statistically significant associations of the germline SNPs in CEBPE and PIP4K2A, they have been reported to be associated with ALL susceptibility in diverse studies (11,13). On the other hand, Hsu et al. showed that CEBPE were associated with increased ALL risk in the California Hispanic population. Although our study had a relatively small sample size, which did not have sufficient power to detect slight effects of CEBPE and PIP4K2A on ALL, the consistency of our data in ARID5B with those published either for smaller or larger sample sizes (9,25) support the role of ARID5B in the susceptibility of this hematological malignancy and may possibly be an explanation of the higher frequency of ALL among Hispanic and Mexican children (5). Several studies have reported that the cancer incidence and mortality rates differ significantly among ethnic and racial groups (2,4,7). In order to assess these questions in our population, we need a larger sample size, well-characterized patients (including chimeric genes, hyperdiploidy, etc.), and
case-control matching by ancestry informative markers. In our study, the next step is the identification of these alleles in the survival of children with ALL. In conclusion, our results suggest that ARID5B polymorphisms are associated with the susceptibility to childhood ALL in Mexican population. On the other hand, PIPK4, CEBPE were not associated. Further association studies with larger samples and detailed ethnicity evaluation are required to investigate whether associations hold true.
Acknowledgments This work was supported by the Consejo Nacional de la Ciencia y la Tecnologıa (CONACYT) through its program, Fondo Sectorial de Investigaci on en Salud y Seguridad Social (SALUD 2007-171223/FIS/IMSS/PROT/592; SALUD 2010-1-141026/FIS/IMSS/ PROT/895; SALUD 2015-1- 262190); by the Fondo Sectorial de Investigaci on para la Educaci on (CB-2007-1-83949/FIS/ IMSS/PROT/616 and CB-2015-1-258042); by the Convocatoria de Proyectos de Investigaci on para Atender Problemas Nacionale2013 (PDCPN2013-01-215726, FIS/IMSS/PROT/1364); and by the Instituto Mexicano del Seguro Social (FIS/IMSS/ PROT/PRIO/14/031, FIS/IMSS/PROT/G10/846, FIS/IMSS/PROT/ G12/1134 and FIS/IMSS/PROT/G15/1407).
Conflict of interests The authors declare that there is no conflict of interests regarding the publication of this paper. Mexican Inter-institutional Group for the Identification of the Causes of Childhood Leukaemia: Francisco PantojaGuillen, Eric Israel Gutierrez-Juarez, Elva Jimenez-Hernandez, Francisco Xavier Guerra-Castillo, Cecilia Rosel-Pech, Monica Patricia Ortiz-Maganda, Emiliano Tesoro-Cruz,
628
Bekker-Mendez et al./ Archives of Medical Research 47 (2016) 623e628
Marıa Teresa Ramos-Cervantes, Marıa del Carmen Rodrıguez-Zepeda, Nancy Nu~ nez Villegas, Aurora MedinaSans on, Yelda A. Leal, Galileo Escobedo-Gonzalez, Juan Manuel Fragoso-Luna, Arturo Fajardo-Gutierrez, Janet Flores-Lujano, Marıa Luisa Perez-Saldivar, Jose Gabriel Pe~naloza-Gonzalez, Martha Margarita Velazquez-Avi~ na, Victoria Bolea-Murga, Nancy Carolina Reyes-Zepeda, Rosa Martha Espinosa-Elizondo, Jose Refugio Torres-Nava, Raquel Amador-Sanchez, Cesar Ra ul Gonzalez-Bonilla, Luz Victoria Flores-Villegas, Karina Anastacia Solis-Labastida, Gabriela Bibiana Martınez-Morales, Dulce L opez-Sanchez, Haydee Rosas-Vargas, and Julian Ramırez-Bello.
References 1. Kaatsch P. Epidemiology of childhood cancer. Cancer Treat Rev 2010; 36:277e285. 2. Ward E, DeSantis C, Robbins A, et al. Childhood and adolescent cancer statistics, 2014. CA Cancer J Clin 2014;64:83e103. 3. Fajardo-Gutierrez A, Juarez-Oca~na S, Gonzalez-Miranda G, et al. Incidence of cancer in children residing in ten jurisdictions of the Mexican Republic: importance of the cancer registry (a populationbased study). BMC Cancer 2007;7:68. 4. Metayer C, Milne E, Clavel J, et al. The Childhood Leukemia International Consortium. Cancer Epidemiol 2013;37:336e347. 5. Xu H, Cheng C, Devidas M, et al. ARID5B genetic polymorphisms contribute to racial disparities in the incidence and treatment outcome of childhood acute lymphoblastic leukemia. J Clin Oncol 2012;30:751e757. 6. Yang W, Trevi~ no LR, Yang JJ, et al. ARID5B SNP rs10821936 is associated with risk of childhood acute lymphoblastic leukemia in blacks and contributes to racial differences in leukemia incidence. Leukemia 2010;24:894e896. 7. Valery PC, Moore SP, Meiklejohn J, et al. International variations in childhood cancer in indigenous populations: a systematic review. Lancet Oncol 2014;15:e90ee103. 8. Migliorini G, Fiege B, Hosking FJ, et al. Variation at 10p12.2 and 10p14 influences risk of childhood B-cell acute lymphoblastic leukemia and phenotype. Blood 2013;122:3298e3307. 9. Xu H, Yang W, Perez-Andreu V, et al. Novel susceptibility variants at 10p12.31-12.2 for childhood acute lymphoblastic leukemia in ethnically diverse populations. J Natl Cancer Inst 2013;105:33e42. 10. Healy J, Richer C, Bourgey M, et al. Replication analysis confirms the association of ARID5B with childhood B-cell acute lymphoblastic leukemia. Haematologica 2010;95:1608e1611. 11. Burmeister T, Bartels G, Gr€oger D, et al. Germline variants in IKZF1, ARID5B, and CEBPE as risk factors for adult-onset acute lymphoblastic leukemia: an analysis from the GMALL study group. Haematologica 2014;99:e23ee25.
12. Evans TJ, Milne E, Anderson D, et al. Confirmation of childhood acute lymphoblastic leukemia variants, ARID5B and IKZF1, and interaction with parental environmental exposures. PLoS One 2014; 9:e110255. 13. Peyrouze P, Guihard S, Grardel N, et al. Genetic polymorphisms in ARID5B, CEBPE, IKZF1 and CDKN2A in relation with risk of acute lymphoblastic leukaemia in adults: a Group for Research on Adult Acute Lymphoblastic Leukaemia (GRAALL) study. Br J Haematol 2012;159:599e602. 14. Walsh KM, de Smith AJ, Chokkalingam AP, et al. Novel childhood ALL susceptibility locus BMI1-PIP4K2A is specifically associated with the hyperdiploid subtype. Blood 2013;121: 4808e4809. 15. Chokkalingam AP, Hsu LI, Metayer C, et al. Genetic variants in ARID5B and CEBPE are childhood ALL susceptibility loci in Hispanics. Cancer Causes Control 2013;24:1789e1795. 16. Moreno-Estrada A, Gignoux CR, Fernandez-Lopez JC, et al. The genetics of Mexico recapitulates Native American substructure and affects biomedical traits. Science 2014;344:1280e1285. 17. Instituto Nacional de Geografıa y Estadıstica. Available at: http:// www3.inegi.org.mx/sistemas/biinegi/?ind 5 1005000039. Accessed on July 10, 2014. 18. Ruiz-Linares A, Adhikari K, Acu~na-Alonzo V, et al. Admixture in Latin America: geographic structure, phenotypic diversity and selfperception of ancestry based on 7,342 individuals. PLoS Genet 2014;10:e1004572. 19. Perez-Saldivar ML, Fajardo-Gutierrez A, Bernaldez-Rıos R, et al. Childhood acute leukemias are frequent in Mexico City: descriptive epidemiology. BMC Cancer 2011;11:355. 20. Walsh KM, de Smith AJ, Welch TC, et al. Genomic ancestry and somatic alterations correlate with age at diagnosis in Hispanic children with B-cell acute lymphoblastic leukemia. Am J Hematol 2014;89: 721e725. 21. Hsu LI, Chokkalingam AP, Briggs FB, et al. Association of genetic variation in IKZF1, ARID5B, and CEBPE and surrogates for earlylife infections with the risk of acute lymphoblastic leukemia in Hispanic children. Cancer Causes Control 2015;26:609e619. 22. Ross JA, Linabery AM, Blommer CN, et al. Genetic variants modify susceptibility to leukemia in infants: a Children’s Oncology Group report. Pediatr Blood Cancer 2013;60:31e34. 23. Guo LM, Xi JS, Ma Y, et al. ARID5B gene rs10821936 polymorphism is associated with childhood acute lymphoblastic leukemia: a metaanalysis based on 39,116 subjects. Tumour Biol 2014;35:709e713. 24. Zeng H, Wang XB, Cui NH, et al. Associations between AT-rich interactive domain 5B gene polymorphisms and risk of childhood acute lymphoblastic leukemia: a meta-analysis. Asian Pac J Cancer Prev 2014;15:6211e6217. 25. Lin CY, Li MJ, Chang JG, et al. High-resolution melting analyses for genetic variants in ARID5B and IKZF1 with childhood acute lymphoblastic leukemia susceptibility loci in Taiwan. Blood Cells Mol Dis 2014;52:140e145.
ORIGINAL ARTICLE
ARID5B, CEBPE and PIP4K2A Germline Genetic Polymorphisms and Risk of Childhood Acute Lymphoblastic Leukemia in Mexican Patients: A MIGICCL Study Vilma Carolina Bekker-Mendez,a Juan Carlos N u~ nez-Enrıquez,b Jose Luis Torres Escalante,c Enrique Alvarez-Olmos,d Pablo Miguel Gonzalez-Montalvoc,e Elva Jimenez-Hernandez,f Aurora Medina Sans on,g Yelda A. Leal,h Marıa Teresa Ramos-Cervantes,a Francisco Xavier Guerra-Castillo,a M onica Patricia Ortiz-Maganda,a Janet Flores-Lujano,b b Maria Luisa Perez-Saldivar, Martha Margarita Velazquez-Avi~ na,i Victoria Bolea-Murga,j Jose Refugio Torres-Nava,k Raquel Amador-Sanchez,l Karina Anastacia Solis-Labastida,m Julian Ramirez-Bello,n Jose Manuel Fragoso,o Juan Manuel Mejıa-Arangure,b,p and MIGICCLq a
Unidad de Investigacion Medica en Inmunologıa e Infectologıa, Hospital de Infectologıa ‘‘Dr. Daniel Mendez Hernandez’’, ‘‘La Raza’’, IMSS, Mexico City, Mexico b Unidad de Investigaci on Medica en Epidemiologıa Clınica, UMAE Hospital de Pediatrıa, Centro Medico Nacional (CMN) ‘‘Siglo XXI’’, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico c Servicio de Pediatrıa de la UMAE, Instituto Mexicano del Seguro Social (IMSS); Yucatan, Mexico d Laboratorio de Genomica del Cancer, Instituto Nacional de Medicina Genomica (INMEGEN), Mexico City, Mexico e Servicio de Oncologıa Pediatrica del Hospital O’Horan, SS, Yucatan, Mexico f Servicio de Hematologıa Pediatrica, Hospital General ‘‘Gaudencio Gonzalez Garza’’, Centro Medico Nacional (CMN) ‘‘La Raza’’, IMSS, Mexico City, Mexico g Servicio de Hemato-Oncologia, Hospital Infantil de Mexico Federico Gomez, Secretaria de Salud (SSa), Mexico City, Mexico h Unidad de Investigacion Medica Yucatan, Registro de Cancer, Unidad Medica de Alta Especialidad IMSS Merida, Yucatan, Mexico i Servicio de Onco-Pediatria, Hospital Juarez de Mexico, Secretaria de Salud (SSa), Mexico City, Mexico j Servicio de Hematologıa Pediatrica, Hospital General de Mexico, Secretaria de Salud (SSa), Mexico City, Mexico k Servicio de Oncologıa, Hospital Pediatrico de Moctezuma, Secretarıa de Salud del D.F., Mexico City, Mexico l Hospital General Regional No. 1 ‘‘Carlos McGregor Sanchez Navarro’’, IMSS, Mexico City, Mexico m Servicio de Hematologıa Pediatrica, UMAE Hospital de Pediatrıa, Centro Medico Nacional (CMN) ‘‘Siglo XXI’’, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico n Laboratorio de la Unidad de Investigacion en Enfermedades Metabolicas y Endocrinas del Hospital Juarez de Mexico, Mexico City, Mexico o Departamento de Biologıa Molecular, Instituto Nacional de Cardiologıa, Ignacio Chavez, Ciudad de Mexico, Mexico p Coordinacion de Investigacion en Salud, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico q Mexican Inter-Institutional Group for the Identification of the Causes of Childhood Leukaemia, Instituto Mexicano del Seguro Social, Instituto de Seguridad Social al Servicio de los Trabajadores del Estado, Secretarıa de Salud, Secretarıa de Salud del Gobierno del Distrito Federal, Mexico City, Mexico Received for publication September 29, 2016; accepted November 24, 2016 (ARCMED-D-16-00585).
Background and Aims. Childhood acute lymphoblastic leukemia (ALL) is the leading cause of childhood cancer-related deaths worldwide. Multiples studies have shown that ALL seems to be originated by an interaction between environmental and genetic susceptibility factors. The ARID5B polymorphisms are among the most reproducible ALL associated-risk alleles in different populations. The aim of the present study was to examine the contribution of ARID5B, CEBPE, and PIP4K2 risk alleles for the development of ALL in children from Mexico City and Yucatan, Mexico.
Address reprint request to: Dr. Vilma Carolina Bekker-Mendez, Unidad de Investigaci on en Inmunologıa e Infectologıa. Hospital de Infectologıa CMN ‘‘La Raza’’ IMSS. Tel/Fax: þ55 5724 5900 ext. 24322; 24321 Mexico, DF, Mexico; E-mail: bekkermendez@yahoo. com; Dr. Juan Manuel Mejıa-Arangure, Unidad de Investigacion Medica en Epidemiologıa Clınica, UMAE Hospital de Pediatrıa, Centro Medico
Nacional (CMN) ‘‘Siglo XXI’’, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico and Coordinacion de Investigacion en Salud, Instituto Mexicano del Seguro Social (IMSS), Torre Academia Nacional de Medicina 4to piso, Av. Cuauhtemoc 330, 06720 Mexico, DF, Mexico. Phone number: þ52155 56276900 ext. 21294. Fax number: þ52 155 57610841; E-mail: [email protected], [email protected].
0188-4409/$ - see front matter. Copyright Ó 2016 IMSS. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.arcmed.2016.12.003
624
Bekker-Mendez et al./ Archives of Medical Research 47 (2016) 623e628
Methods. A study was conducted with a total of 761 unrelated subjects. Two hundred eighty five ALL cases (111 from Yucatan and 174 from Mexico City) and 476 healthy subjects. Genotyping included the rs7088318 (PIP4K2A), rs10821936 (ARID5B), rs7089424 (ARID5B) and rs2239633 (CEBPE ) polymorphisms. Results. Associations between ALL and rs10821936 and rs7089424 ARID5B SNPs were found (OR 5 1.9, 95% CI (1.5e2.4) and OR 5 2.0, 95% CI (1.6e2.5), respectively). Moreover, a higher risk was observed in the homozygous risk genotypes of carriers from Mexico City (OR 5 3.1, 95% CI (2.0e4.9) and OR 3.1, CI 95% (2.0e4.8), respectively). Otherwise, the rs7088318 (PIP4K2A) and rs2239633 (CEBPE ) polymorphisms were not associated with ALL risk. Conclusions. Our analysis suggests that ARID5B confers risk for childhood ALL in a Mexican population. Ó 2016 IMSS. Published by Elsevier Inc. Key Words: Association study, ARID5B, PIP4K2A, CEBPE, Childhood acute lymphoblastic leukemia, Mexican.
Acute lymphoblastic leukemia (ALL) is the most frequent subtype of childhood cancer worldwide (1e3). The etiology of childhood ALL has not been established; however, many studies support the hypothesis that genetic polymorphisms influence racial differences in ALL incidence, either when the frequency of an ALL-predisposing genetic polymorphism differs by ethnicity or when these variants are associated with ALL in a race-specific manner (4e7). Genomewide association studies (GWAS) have identified ALLsusceptibility loci at 7p12.2, 9p21.3, 10p12.2, 10q21.2 and 14q11.2 (8,9). The most consistent across studies is the 10q21.2 locus, which harbors the AT-rich interactive domain 5B (ARID5B) gene. ARID5B, which is involved in regulation of embryonic development, cell growth and B-lymphocyte progenitor differentiation, bears germline single nucleotide polymorphisms (SNP) located in intron 3 that are strongly associated with risk of pediatric ALL including rs10821936 and rs7089424. Although its specific role in the pathogenesis of ALL remains unknown, one of the highlighted findings is the association between ARID5B SNPs with ALL susceptibility in multiple investigated populations. Interestingly, differences in the frequencies of the risk alleles among Caucasian, African, Asian and Hispanic derived populations have been found, being the highest in Hispanic descendent populations (9e12). As an example, the frequency of the rs10821936C risk allele was lower in control populations of African ancestry (0.13 in ASW and 0.2 in YRI) than in populations of European ancestry (0.33) (6). Likewise, the CCAAT/enhancer-binding protein epsilon (CEBPE ) and phosphatidylinositol-5-phosphate 4-kinase, type II, alpha (PIP4K2A) genes have been added to the list of risk genes associated with the development of childhood ALL in other populations (11e14). It is interesting to note that Hispanics have shown a worse prognosis, often need more intensive ALL treatment regimens, show older ages at diagnosis, and exhibit higher frequencies of the ARID5B ALL-risk variants (also linked to poorer outcomes) than non-Hispanic Whites (5,15). Mexican as well as Hispanic populations are characterized by a very complex genetic
background due to the different contributions from Native American, Caucasian and African populations during the conquest (16). Nowadays, the Mexican population is comprised by Mestizos (94%) and Amerindians (6%) (17,18). ALL has higher incidence rates in Hispanics than in other ethnic groups and is the leading cause of mortality for cancer in Mexican children (3). It is possible that differences in genetic susceptibility loci could explain the higher observed ALL incidence rates in Mexican population. The aim of the present study was to examine the contribution of ARID5B, CEBPE, and PIP4K2 risk alleles for the development of ALL in children from Mexico City and Yucatan, Mexico.
Methods Subjects We conducted a study in a sample of 761 unrelated Mexican subjects recruited in Merida, Yucatan and Mexico City. Cases were !18 years of age diagnosed with ALL, recruited from two hospitals from Merida, Yucatan and three tertiary level hospitals in Mexico City. Diagnosis of ALL was based on morphology and immunophenotype studies. ALL patients were classified as standard risk (age 1e9.99 years and initial white blood cell count (WBC) !50 109/L) or as high risk (NCI HR) (age ! 1 year or $10 years and initial WBC $50 109/L) according to the National Cancer Institute (NCI) Risk Classification (19). Controls corresponded to healthy adults from Mexico City recruited in the General Hospital of Mexico and Hospital Juarez of Mexico City. Children from Yucatan were included from the Unidad Medica de Alta Especialidad (UMAE)-IMSS, Merida, Yucatan (third-level hospital). Only individuals self-reported as having not been diagnosed with autoimmune, inflammatory, hematological diseases
ARID5B, CEBPE and PIP4K2A in Childhood Acute Lymphoblastic Leukemia
or cancer were included as controls. Institutional review committees from the participating institutions approved this study. Written informed consent was obtained from parents of children with ALL and from the control group. When possible, the child’s assent was also obtained. DNA Extraction We included bone marrow or saliva samples from ALL cases and peripheral blood from controls. Genomic deoxyribonucleic acid (DNA) from bone marrow and peripheral blood samples were extracted using the standard salting out technique (QIAGEN systems, Inc., Valencia, CA) and the genomic DNA extraction from the saliva was done using the ORAGENE Purification Kit (DNA Genotek Inc., ON, Canada) according to the manufacturer’s instructions. DNA purity and concentration was determined by spectrophotometry. Genotyping Analysis We selected four SNPs based on previous association studies in ALL, including two highly replicated ARID5B (rs10821936, rs7089424) variants (Table 1). Genotyping of the PIP4K2A (rs7088318), ARID5B (rs10821936, rs7089424) and CEBPE (rs2239633) was performed with the ABI 7900HT system using 10 ng of input DNA and Genotyping Master Mix and TaqMan Pre-Designed-SNPGenotyping-Assay (Life Technologies, Foster City, CA). PCR mix consisted of 10 ng of genomic DNA, 0.1 mM of each probe, 2.5 ml of TaqMan master mix (Life Technologies) and ddH2O up to a final volume of 5 ml. The amplification protocol included denaturing at 95 C for 10 min followed by 45 cycles of denaturing at 95 C for 15 sec and annealing and extension at 60 C for 1 min. The genotype of each sample was assigned automatically by measuring the allele-specific fluorescence using SDS 2.2.3 software for allelic discrimination (Applied Biosystems, Foster City, CA). A subset of 100 random samples was genotyped in duplicate and the reproducibility was 99%. To validate the TaqMan results, three random samples of each genotype were sequenced using the forward and reverse primers. PCR was performed with the following conditions: an initial denaturation at 95 C for 5 min, 35 cycles at 95 C for 30 s, 58 C for 30 s, and 68 C for 30 s, and a final extension at 72 C for 7 min. Sequencing was performed using an automated ABI PRISM 3100 DNA sequencer (Applied Biosystems). Statistical Analysis We used the FINNETI algorithm (http://ihg.gsf.de/cgi-bin/ hw/hwa1.pl) to test the Hardy-Weinberg equilibrium (HWE) for genotype distributions in ALL cases and controls. To assess the differences of the genotype and allele frequencies among populations and ethnic groups and
625
between genders we used the c2 test implemented in the Stat-Calc program (EPI INFO 7 v 7.201; Centers for Disease Control and Prevention, Atlanta, GA). We calculated the odds ratio (OR) and 95% confidence intervals (95% CI); p values !0.05 were considered statistically significant after adjusting for multiple comparisons base on the Bonferroni correction test ( p !0.01). Power calculations were performed using the Quanto Software, accounting for ALL prevalence in Mexico City (19), both the MAF and OR of each studied SNP.
Results Subjects This study included 761 unrelated subjects. A total of 285 ALL children fulfilled the ALL diagnostic criteria, from which 111 subjects (39.0%) were from Merida, Yucatan and 174 (61.0%) were from Mexico City, with a male predominance (55.0%). Patients’ age ranged from 0e17 years with a mean (SD) age of 5.7 2.8 years and 8.7 4.94 years in each population, respectively. There were no statistical differences between patients from Yucatan and Mexico City regarding gender and children’s age; however, patients from Yucatan showed an earlier-onset of ALL than others. The healthy group consisted of 476 subjects, 225 (47%) were male and 251 (53%) were female. We included healthy adults as controls because the risk alleles investigated did not seem to be associated with lifespan. Association Analysis No evidence from HWE was observed for any evaluated SNPs. Case-control association analysis including the 285 cases and the 476 controls did not show statistically significant differences in the distribution of the rs2239633 (CEBPE ) and rs7088318 (PIP4K2A) alleles between ALL patients and healthy subjects. Nevertheless, significant associations were found among both ARID5B SNPs (rs10821936 and rs7089424) and ALL. The rs10821936C and rs7089424G alleles were detected in 66.5% and 66.7% of patients, respectively, whereas they were carried by 50.6 and 49.2% of the healthy chromosomes, respectively, exhibiting both as risk alleles for leukemia (rs10821936C: OR 5 1.9, 95% CI (1.5e2.4), p !0.01 and rs7089424G: OR 5 2.0, 95% CI (1.6e2.5), p !0.01) (Table 1). To describe the observed ARID5B associations and capture associations under various genetic models, we measured the corresponding genotype OR in all samples. A strong allele dose-dependent effect on risk was observed at loci rs10821936 and rs7089424. Carriers of a homozygous risk genotype (rs10821936-CC and rs7089424-GG) homozygote mutant allele status meant increased risk (OR 5 3.1, 95% CI (2.0e4.9) p !0.01 and OR 5 3, 95% CI (2.0e4.8) p ! 0.01 respectively) to ALL (Table 1).
626
Bekker-Mendez et al./ Archives of Medical Research 47 (2016) 623e628
Table 1. Association analysis among five ALL risk alleles and Mexican patients ALL children Gen PIP4K2A
ARID5B
ID SNP
Genotypes
rs7088318
AA AC CC A C AA vs. CC TT TC CC T C TT vs. CC TT TG GG T G TT vs. GG rs2239633 GA AA G A AA vs. GG
rs10821936
rs7089424 n total
CEBPE
Healthy individuals n [ 285 n (%)
n [ 476 n (%)
211 68 6 490 80
(74.0) (23.9) (2.1) (86.0) (14)
341 124 11 805 147
(71.6) (26.1) (2.3) (84.6) (15.4)
40 111 134 191 379
(14) (38.9) (47.0) (33.5) (66.5)
38 114 133 190 380
(13.3) (40.0) (46.7) (33.3) (66.7)
111 248 117 470 482 3.1 107 249 120 483 469 3.1 122 245 64 580 372 0.75
(23.3) (52.1) (24.6) (49.4) (50.6) (2.0e4.9) (22.5) (52.3) (25.2) (50.7) (49.2) (2.0e4.8) (42.8) (51.5) (13.4) (60.9) (39.1) (0.47e1.20)
GG 128 35 372 198
(44.9) (12.3) (65.3) (34.7)
OR (95% CI)
p
0.89 (0.66e1.20)
NS
0.88 (0.3e2.4)
NS
1.9 (1.5e2.4)
!0.01
!0.01
2.0 (1.6e2.5)
!0.01
!0.01 167 (35.1)
0.83 (0.6e1.0)
NS NS
OR, odds ratio; ALL, acute lymphoblastic leukemia. Number of subjects. NS indicates no significant differences. Differences in allele frequencies were measured by c2 test.
Genotypes and Allele Frequencies After we performed an origin matched case-control analysis, we failed to find statistically significant differences among the rs7088318 (PIP4K2A), rs2239633 (CEBPE ) and ALL in either the Maya or Mestizo population. The observed allele frequencies are summarized in Table 2. rs10821936C and rs7089424G alleles were detected in 66.8 and 67.1% of patients, respectively, whereas they were carried by 51.3 and 55% of the healthy controls from Yucatan, respectively, exhibiting both as risk alleles for leukemia (rs10821936C: OR 5 1.8, 95% CI (1.2e2.7), p !0.01 and rs7089424G: OR 5 1.6, 95% CI (1.0e2.5), p !0.01) (Table 2). ARID5B (rs10821936-CC) exhibits statistically significant differences (OR 5 2.5, 95% CI (1.1e5.7) p !0.05), whereas ARID5B (rs7089424-GG) failed to show a statistically significant difference (OR 5 2.1, 95% CI (0.9e4.9) p 5 NS) between ALL patients and healthy subjects from Yucatan. These SNPs were also associated with ALL and Mestizo populations. Significant differences were observed in both ARID5B homozygous risk genotypes (rs10821936-CC and rs7089424-GG) [OR 5 3.6, 95% CI (2.0e6.3) p !0.01 and OR 5 3.5, 95% CI (2.0e6.2) p !0.01, respectively]. ARID5B rs10821936 T/C and rs7089424 T/G were found in high-linkage disequilibrium
(LD) (r2 5 0.89) and haplotype analysis was performed. Statistical significance remained after multiple comparisons (100,000 permutations).
Discussion This study is one of the first to examine the association of ALL risk alleles located in the ARID5B, CEBPE, and PIP4k2A genes and the susceptibility to ALL in children from two regions from Mexico. Combining overall ALL children vs. healthy subjects, the ARID5B, SNPs exhibited differential distribution between groups. Both ARID5B polymorphisms (rs10821936 and rs7089424) were associated with susceptibility to ALL. OR significantly increased for homozygote risk allele carriers of both SNPs. ARID5B encodes the AT-rich interactive domain 5B, a member of the ARID family of proteins, which are highly conserved and play a key role in adipogenesis and in liver development. This gene is also involved in the proliferation of early B-cell progenitors and its expression correlates with genes implicated in several hematopoietic diseases (20). Our data provide evidence of the ARID5B contribution to pediatric ALL risk in Mexican
ARID5B, CEBPE and PIP4K2A in Childhood Acute Lymphoblastic Leukemia
627
Table 2. Genotypes and alleles frequencies of the ARID5B polymorphism in ALL and healthy subjects stratified by ethnicity Yucatan
SNP
Genotypes
rs10821936
TT TC CC T C TT vs. CC TT TG GG T G TT vs. GG
rs7089424
ALL n [ 111 19 38 54 76 146
(17.1) (34.2) (48.6) (34.2) (65.8)
17 39 55 73 149
(15.3) (35.1) (49.5) (32.9) (67.1)
Healthy individuals n [ 78 n (%) 20 36 22 76 80 2.5 16 38 24 70 86 2.1
(25.6) (46.2) (28.2) (48.7) (51.3) (1.1e5.7) (20.5) (48.7) (30.8) (45) (55) (0.9e4.9)
Mexico City
OR (95% CI) n (%)
1.8 (1.2e2.7)
p
!0.01
!0.05
1.6 (1.0e2.5)
!0.05
ALL n [ 174 21 73 80 115 233 3.6 21 75 78 117 231
(12.1) (42.0) (46.0) (33.0) (67.0) (2.0e6.3) (12.1) (43.1) (44.8) (33.6) (66.4)
NS
Healthy individuals n [ 398 n (%) 91 (22.9) 212 (53.3) 95 (23.9) 394 (49.5) 402 (50.5) !0.01 91 (22.9) 211 (53.0) 96 (24.1) 392 (49.2) 404 (50.8)
OR (95% CI) n (%)
p
1.9 (1.5.e2.5)
!0.01
1.9 (1.4e2.4)
!0.01
3.5 (2.0e6.2)
!0.01
n, total number of subjects; OR, odds ratio. NS indicates no significant differences. Differences in allele frequencies were measured by c2 test.
patients as has been observed in Caucasian-, Asian-, African- and Hispanic- derived populations (5,6,21). The observed risk for these loci in our cases vs. most populations was demonstrated to be similar to previously reported effects (22e24); however, the estimated risk for C (rs10821936) and G (rs7089424) allele homozygote carriers was higher than data frequently reported for B-lineage ALL. More interestingly, the frequencies of both risk-alleles documented in the present study seem to be the highest reported to date. To note, in a recent metaanalysis, the SNP rs7089424 showed significant associations among Caucasians but not among Asians in all genetic models. Further, the GG genotype frequencies in non-ALL controls display a decreasing tendency from Caucasian to Asian subgroups. Although we did not observe statistically significant associations of the germline SNPs in CEBPE and PIP4K2A, they have been reported to be associated with ALL susceptibility in diverse studies (11,13). On the other hand, Hsu et al. showed that CEBPE were associated with increased ALL risk in the California Hispanic population. Although our study had a relatively small sample size, which did not have sufficient power to detect slight effects of CEBPE and PIP4K2A on ALL, the consistency of our data in ARID5B with those published either for smaller or larger sample sizes (9,25) support the role of ARID5B in the susceptibility of this hematological malignancy and may possibly be an explanation of the higher frequency of ALL among Hispanic and Mexican children (5). Several studies have reported that the cancer incidence and mortality rates differ significantly among ethnic and racial groups (2,4,7). In order to assess these questions in our population, we need a larger sample size, well-characterized patients (including chimeric genes, hyperdiploidy, etc.), and
case-control matching by ancestry informative markers. In our study, the next step is the identification of these alleles in the survival of children with ALL. In conclusion, our results suggest that ARID5B polymorphisms are associated with the susceptibility to childhood ALL in Mexican population. On the other hand, PIPK4, CEBPE were not associated. Further association studies with larger samples and detailed ethnicity evaluation are required to investigate whether associations hold true.
Acknowledgments This work was supported by the Consejo Nacional de la Ciencia y la Tecnologıa (CONACYT) through its program, Fondo Sectorial de Investigaci on en Salud y Seguridad Social (SALUD 2007-171223/FIS/IMSS/PROT/592; SALUD 2010-1-141026/FIS/IMSS/ PROT/895; SALUD 2015-1- 262190); by the Fondo Sectorial de Investigaci on para la Educaci on (CB-2007-1-83949/FIS/ IMSS/PROT/616 and CB-2015-1-258042); by the Convocatoria de Proyectos de Investigaci on para Atender Problemas Nacionale2013 (PDCPN2013-01-215726, FIS/IMSS/PROT/1364); and by the Instituto Mexicano del Seguro Social (FIS/IMSS/ PROT/PRIO/14/031, FIS/IMSS/PROT/G10/846, FIS/IMSS/PROT/ G12/1134 and FIS/IMSS/PROT/G15/1407).
Conflict of interests The authors declare that there is no conflict of interests regarding the publication of this paper. Mexican Inter-institutional Group for the Identification of the Causes of Childhood Leukaemia: Francisco PantojaGuillen, Eric Israel Gutierrez-Juarez, Elva Jimenez-Hernandez, Francisco Xavier Guerra-Castillo, Cecilia Rosel-Pech, Monica Patricia Ortiz-Maganda, Emiliano Tesoro-Cruz,
628
Bekker-Mendez et al./ Archives of Medical Research 47 (2016) 623e628
Marıa Teresa Ramos-Cervantes, Marıa del Carmen Rodrıguez-Zepeda, Nancy Nu~ nez Villegas, Aurora MedinaSans on, Yelda A. Leal, Galileo Escobedo-Gonzalez, Juan Manuel Fragoso-Luna, Arturo Fajardo-Gutierrez, Janet Flores-Lujano, Marıa Luisa Perez-Saldivar, Jose Gabriel Pe~naloza-Gonzalez, Martha Margarita Velazquez-Avi~ na, Victoria Bolea-Murga, Nancy Carolina Reyes-Zepeda, Rosa Martha Espinosa-Elizondo, Jose Refugio Torres-Nava, Raquel Amador-Sanchez, Cesar Ra ul Gonzalez-Bonilla, Luz Victoria Flores-Villegas, Karina Anastacia Solis-Labastida, Gabriela Bibiana Martınez-Morales, Dulce L opez-Sanchez, Haydee Rosas-Vargas, and Julian Ramırez-Bello.
References 1. Kaatsch P. Epidemiology of childhood cancer. Cancer Treat Rev 2010; 36:277e285. 2. Ward E, DeSantis C, Robbins A, et al. Childhood and adolescent cancer statistics, 2014. CA Cancer J Clin 2014;64:83e103. 3. Fajardo-Gutierrez A, Juarez-Oca~na S, Gonzalez-Miranda G, et al. Incidence of cancer in children residing in ten jurisdictions of the Mexican Republic: importance of the cancer registry (a populationbased study). BMC Cancer 2007;7:68. 4. Metayer C, Milne E, Clavel J, et al. The Childhood Leukemia International Consortium. Cancer Epidemiol 2013;37:336e347. 5. Xu H, Cheng C, Devidas M, et al. ARID5B genetic polymorphisms contribute to racial disparities in the incidence and treatment outcome of childhood acute lymphoblastic leukemia. J Clin Oncol 2012;30:751e757. 6. Yang W, Trevi~ no LR, Yang JJ, et al. ARID5B SNP rs10821936 is associated with risk of childhood acute lymphoblastic leukemia in blacks and contributes to racial differences in leukemia incidence. Leukemia 2010;24:894e896. 7. Valery PC, Moore SP, Meiklejohn J, et al. International variations in childhood cancer in indigenous populations: a systematic review. Lancet Oncol 2014;15:e90ee103. 8. Migliorini G, Fiege B, Hosking FJ, et al. Variation at 10p12.2 and 10p14 influences risk of childhood B-cell acute lymphoblastic leukemia and phenotype. Blood 2013;122:3298e3307. 9. Xu H, Yang W, Perez-Andreu V, et al. Novel susceptibility variants at 10p12.31-12.2 for childhood acute lymphoblastic leukemia in ethnically diverse populations. J Natl Cancer Inst 2013;105:33e42. 10. Healy J, Richer C, Bourgey M, et al. Replication analysis confirms the association of ARID5B with childhood B-cell acute lymphoblastic leukemia. Haematologica 2010;95:1608e1611. 11. Burmeister T, Bartels G, Gr€oger D, et al. Germline variants in IKZF1, ARID5B, and CEBPE as risk factors for adult-onset acute lymphoblastic leukemia: an analysis from the GMALL study group. Haematologica 2014;99:e23ee25.
12. Evans TJ, Milne E, Anderson D, et al. Confirmation of childhood acute lymphoblastic leukemia variants, ARID5B and IKZF1, and interaction with parental environmental exposures. PLoS One 2014; 9:e110255. 13. Peyrouze P, Guihard S, Grardel N, et al. Genetic polymorphisms in ARID5B, CEBPE, IKZF1 and CDKN2A in relation with risk of acute lymphoblastic leukaemia in adults: a Group for Research on Adult Acute Lymphoblastic Leukaemia (GRAALL) study. Br J Haematol 2012;159:599e602. 14. Walsh KM, de Smith AJ, Chokkalingam AP, et al. Novel childhood ALL susceptibility locus BMI1-PIP4K2A is specifically associated with the hyperdiploid subtype. Blood 2013;121: 4808e4809. 15. Chokkalingam AP, Hsu LI, Metayer C, et al. Genetic variants in ARID5B and CEBPE are childhood ALL susceptibility loci in Hispanics. Cancer Causes Control 2013;24:1789e1795. 16. Moreno-Estrada A, Gignoux CR, Fernandez-Lopez JC, et al. The genetics of Mexico recapitulates Native American substructure and affects biomedical traits. Science 2014;344:1280e1285. 17. Instituto Nacional de Geografıa y Estadıstica. Available at: http:// www3.inegi.org.mx/sistemas/biinegi/?ind 5 1005000039. Accessed on July 10, 2014. 18. Ruiz-Linares A, Adhikari K, Acu~na-Alonzo V, et al. Admixture in Latin America: geographic structure, phenotypic diversity and selfperception of ancestry based on 7,342 individuals. PLoS Genet 2014;10:e1004572. 19. Perez-Saldivar ML, Fajardo-Gutierrez A, Bernaldez-Rıos R, et al. Childhood acute leukemias are frequent in Mexico City: descriptive epidemiology. BMC Cancer 2011;11:355. 20. Walsh KM, de Smith AJ, Welch TC, et al. Genomic ancestry and somatic alterations correlate with age at diagnosis in Hispanic children with B-cell acute lymphoblastic leukemia. Am J Hematol 2014;89: 721e725. 21. Hsu LI, Chokkalingam AP, Briggs FB, et al. Association of genetic variation in IKZF1, ARID5B, and CEBPE and surrogates for earlylife infections with the risk of acute lymphoblastic leukemia in Hispanic children. Cancer Causes Control 2015;26:609e619. 22. Ross JA, Linabery AM, Blommer CN, et al. Genetic variants modify susceptibility to leukemia in infants: a Children’s Oncology Group report. Pediatr Blood Cancer 2013;60:31e34. 23. Guo LM, Xi JS, Ma Y, et al. ARID5B gene rs10821936 polymorphism is associated with childhood acute lymphoblastic leukemia: a metaanalysis based on 39,116 subjects. Tumour Biol 2014;35:709e713. 24. Zeng H, Wang XB, Cui NH, et al. Associations between AT-rich interactive domain 5B gene polymorphisms and risk of childhood acute lymphoblastic leukemia: a meta-analysis. Asian Pac J Cancer Prev 2014;15:6211e6217. 25. Lin CY, Li MJ, Chang JG, et al. High-resolution melting analyses for genetic variants in ARID5B and IKZF1 with childhood acute lymphoblastic leukemia susceptibility loci in Taiwan. Blood Cells Mol Dis 2014;52:140e145.