Background
Acute lymphoblastic leukemia (ALL), an aggressive lymphoid cancer, affecting red, white blood cells and platelets. It is reported as the most prevalent hematopoietic malignancy in children worldwide [
1,
2]. ALL represent about 25% of cancers that affect people before their twenties; with 2 to 5 years being the peak age. In Tunisia, 25 new children were diagnosed with ALL every year. New treatment strategies have increased the cure rate to 80% [
3,
4] and the five-year survival is estimated to 60% nowadays [
2]. ALL is characterized by infiltration to monoclonal immature cell medullar and extra medullar areas [
5,
6]. ALL is a multifactorial disease that includes the interaction between environmental and genetic factors [
7,
8]. Genetic etiology dominates in younger due to lower exposure to environmental factors compared to adults [
8,
9].
Recent evidence revealed that acquired somatic mutations could contribute to ALL [
10]. These mutations increase the proliferation and survival of progenitor cells [
10] and impair further cellular differentiation [
10]. Although these alterations contribute to the diagnosis and prognosis [
10], they are insufficient to understand the physiopathology ALL processes. The introduction of novel diagnostic tools, such as high-resolution karyotyping, CGH array and genome-wide molecular analysis [
6,
11] may reveal additional somatic mutations not captured by standard cytogenetic analysis. The heritable susceptibility to ALL is proven by the finding of recent candidate-gene and genome-wide association studies (GWAS). Some authors suggested that ALL risk is associated with multiple germline variants [
12]; others disprove these hypotheses suggesting that the genetic susceptibility to ALL is mediated by a small number of genes with a strong effect rather than the cumulative effect of small genes [
13]. In fact, the first GWAS performed by Trevino and colleagues, identified eighteen germline single nucleotide polymorphisms (SNPs) of 307,944 subjects with allele frequencies are higher in pediatric ALL population that in non-ALL controls (
p < 1 × 10–5) [
9]. Four loci particular SNPs located in 7p12.2 (
IKZF1), 9p12 (
CDKN2A/CDKN2B), 10q21.2 (
ARID5B) and 14q11.2 (
CEBPE) are reported by Sherborne which are associated with ALL risk: RR (1.5–1.6) and more specifically with the B-cell precursor (BCP) ALL [
14,
15].
IKAROS one of five members of large Zinc finger proteins (IKAROS, Helios, Aiolos, Eos and Pegasus), and is an important activator in the lymphopoiesis [
16]. IKAROS is a potential regulator of lymphocyte commitment and differentiation [
17], and immune system development [
18]. IKAROS is an important transcription factor expressed in a hematopoietic stem cell that regulates cell proliferation and differentiation during hematopoiesis, particularly in lymphoid cell lineage, and the adaptive immune system [
19,
20].
IKZF1 gene is located at chromosome 7 encodes IKAROS protein interacting with the deoxynucleotidyl transferase inducing its transcription. It especially targets the zone of dead centromeric chromosomes [
21]. It has been reported to have a key role in IKAROS protein dysfunction upon mutations in hematological malignancies, especially in lymphoblastic leukemia [
17,
22,
23]. Numerous studies documented the association of gene deletion and ALL [
24]. Few studies on African and Asian population investigated the association of single polymorphisms of
IKZF1 with infant ALL. Here we explored the association of the
IKZF1 gene variants, rs4132601 located on chromosome 7 (position 50,438,098 between rs6964823 and rs6944602, and strongly associated with ALL), and rs11978267 [
25], with ALL risk among Tunisian children.
Discussion
Genetic studies contributed to the understanding of the physiopathological mechanisms of malignant diseases such as leukemia. In fact, the genetic alterations are a major etiology of children leukemia which helps to explain the potential risk for ALL. Xu and colleagues hypothesized that the genetic susceptibility to ALL is mediated by a small number of genes with strong effect rather than the cumulative effect of small genes [
13].
Genome association studies have identified the
IKZF-1 gene polymorphism frequency through the deletion and single mutations in the children with ALL as compared to the controls [
24‐
26].
We examined the association between rs4132601 and ALL risk in 170 Tunisian pediatric patients. Our results (Table
3) confirm those of Papaemmanuil and colleagues, who reported the association of the
IKZF1 gene polymorphism rs4132601, localized in 3′ region of the IKAROS family zinc finger 1 (
IKZF1) gene with a pediatric ALL risk [
25]. It is in LD with rs6964823 and rs6944602 that form a block between 3′ UTR and intron 7 of
IKZF1 [
25]. However, previous studies that have examined the association between
IKZF1 rs4132601 polymorphism and acute lymphoblastic leukemia (ALL) risk reached controversial results. Li et al. analyzed 15 case-control studies with 8333 cases and 36,036 controls in a meta-analysis [
26]. The results suggested that rs4132601 was associated with an increased ALL risk. Significant associations were found among Caucasians and Hispanics [
27,
28], but not among Asians [
29]. On the other hand, when data are stratified by disease type, rs4132601 increased significantly risks in B-cell ALL and B hyperdiploid ALL, but not among T-cell ALL and AML subgroups [
26‐
30]. Our results showed that the G allele of rs4132601 is more frequent in patients than the control, which increased the ALL risk in Tunisian children. However, Dai and colleagues suggested in their meta-analysis that Europeans but not Asian G allele carriers children had a markedly increased risk of ALL [
30,
31].
Several studies have been conducted to examine the possible interaction between
IKZF1 G allele of rs4132601 and environmental ALL risk factors. Rudant and colleagues reported a negative correlation of
IKZF1 allele risk with maternal use of home insecticides during pregnancy and positive interactions with early immune modulation, breastfeeding, and history of repeated early common infections, but there was no risk with paternal preconception smoking [
32]. Rudant and Linabery noted that some selected demographic factors such as birth weight and maternal age don’t change the risk of some SNPs of
ARID5B and
IKZF1 [
32,
33]. Also, Linabery and Evans in their independent studies showed an evident interaction between rs4132601 risk and paternal preconception smoking or maternal use of foli
IKZF1c acid [
33,
34].
The rs4132601 is a substitution polymorphism (T > G) located in 3′UTR to the IKAROS Family zinc finger 1 (
IKZF1) gene (located on 7p12.2). IKAROS proteins are major regulators of lymphocyte maturation and CD4, CD8 T-cell lineage differentiation [
35]. The animal model revealed that mice without IKAROS protein develop aggressive lymphoblastic leukemia [
36,
37]. However, the role of the rs4132601 in the leukemogenesis is unclear; Górniak and colleagues hypothesized that the G allele decreases the stability and level of the mRNA expression; which delays progression through B-cell maturation and increases the risk of ALL [
25,
38]. Lautner-Csorba and colleagues reported that the rs4132601 strongly increases the probability (0.97) of ALL and showed that the lower expression associated with the rs4132601 might contribute to the increased vulnerability to the disease [
39]. Our results had disapproved of this hypothesis. The Pearson test did not demonstrate any correlation between rs4132601 and current state (death, relapse) (Table
5).
The rs11197867
IKZF1 gene variant, the second investigated polymorphism, is an intronic polymorphism located at the position 50,433,798 of chromosome 7. Our data showed the lack of association of rs11197867with ALL, irrespective of the genetic model used. To our knowledge, few studies have replicated this polymorphism. Dai and colleagues in their meta-analysis of 20 studies with 4960 patients and 28,034 controls reported that the rs11978267 polymorphism in the
IKZF1 gene increased significantly AL risk [
31], especially among Europeans [
28,
33,
40] but not among African and mixed populations [
13]. When stratified by types of ALL, a significant correlation was found in the BCP-ALL sub-group in the allelic and ALL genetic models. Moreover, ethnicity appears to change ALL risk. In fact, Europeans with a G allele had a markedly increased risk of ALL, which was not observed among Asians [
30].
Mixed Lineage Leukemia (MLL) gene rearrangements assigns a poor prognostic [
41] in both ALL and AML infant cases. In fact, MLL status markedly influences the five-year survival rates for infant ALL. It’s reported that 80% of ALL infants with negative MLL rearrangement live up to 5 years or more. The risk death increase with ALL/MLL+. However, this influence is less pronounced in infant AML. Further, ALL/MLL+ infants are usually diagnosed as early as the age of 6 months while those with ALL/MLL− are often depicted in older infants [
41]. Burkhardt asserted that the combined cytogenetic MLL with a homozygote variant
IKZF1 allele (rs1197867) increases infant leukemia risk. The risk is similarly increased among young patients with AML/MLL+ and AML/MLL- but only among young patients with ALL/MLL+ [
42].
Papaemmanuil and colleagues showed a functional effect for rs4132601, which is in complete linkage disequilibrium with rs11978267, since each additional copy of the variant allele resulted in greater attenuation of
IKZF1 mRNA expression, and consequently induced leukemia [
25].
The
IKZF1 gene is polymorphic. Churchman and al. identified 28 germline
IKZF1 variants in children with ALL, mostly in B-ALL [
43]. The Iranian (Persian) study including four polymorphisms of the
IKZF1 gene rs4132601 T > G, rs11980379 T > C, and rs10272724 T > C, rs11978267 A > G showed that the first three are associated with risk ALL under the dominant model and that rs11978267 A / G is not associated [
44]. Moreover, Oris et al. reported that the four polymorphisms studied rs696823, rs11978267, rs4132601 and rs6944602 are risk factors for ALL [
28]. Yemeni study has noted the association between rs10235796 and rs6964969 variants and ALL, with a borderline association denoted for
IKZF1 rs4132601 T/G variant and a lack of association between rs11978267, rs7789635 and ALL [
45].
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.