Genetic and immunophenotypic diversity of acute leukemias in children

Acute leukemia (AL) is the most frequently diagnosed childhood cancer in Poland; these account for about a third of all cancers in this age group. Acute lymphoblastic leukemia (ALL) accounts for approximately 80% of all leukemia cases in the population under 18 years of age, while the remaining 15–20% include acute myeloid leukemia (AML) [1]. The standardized incidence rate of leukemia is 35.4/1 million people in Poland. The incidence of each type of leukemia depends on age. Most ALL cases occur between the ages of two and five.

For AML in children, the peak incidence rate occurs in the first year of life and then gradually declines until age four, and then remains relatively constant throughout childhood [2]. Literature data indicate that children respond well to treatment. With the improvement of diagnosis and treatment criteria, the cure rate for ALL is close to 90%, and for AML it is about 60–70% [3].

Acute leukemias are a heterogeneous group of hematopoietic malignancies that occur as a result of clonal growth and accumulation of immature cells of the lymphoid or myeloid lineage in the bone marrow. Despite great interest from researchers, the pathogenesis of most childhood acute leukemias remains unknown. It is known, however, that as a result of disturbance of the mechanisms controlling the maturation process of normal cells, neoplastic transformation begins. This process is multistage, it results in the displacement of normal cell lines, and it is initiated as a result of various external factors. Established and suspected risk factors for leukemia can be classified as familial, genetic, environmental (benzene exposure, high doses of ionizing radiation, chemotherapy, electromagnetic fields), and lifestyle (smoking, obesity) [4]. Moreover, endogenous factors have an impact, including primary and secondary immune deficiencies and chronic stimulation of the immune system with its own antigens in the course of autoimmune diseases [5, 6].

Growing evidence from epidemiological studies strongly suggests that the increased incidence of leukemia is possibly associated with an abnormal immune response to infections in the early period of life. Research shows that the development of leukemia requires two factors: genetic and environmental. The first stage involves a genetic mutation that occurs before a fetus is born and which predisposes babies to leukemia. The second factor may be a childhood illness, as a result of exposure to one or more common infections [7, 8]. Recent experiments supporting the hypothesis of “delayed infection” as the cause of childhood B-cell lymphoblastic leukemia have shown that transcription factors determine the proper course of lymphocyte differentiation, while deregulation of factors such as ETV6, PAX5, IKZF1 significantly influences the susceptibility to ALL [8, 9].

The complex process of neoplastic transformation in acute leukemias requires at least two events resulting in damage to the cell's DNA. The abnormalities may concern both point mutations and double-stranded DNA breaks and result in chromosomal translocations, deletions, duplication, and inversion. Chromosomal translocations result in a fusion gene, most often acting as a transcription factor or signaling molecule. The fusion (chimeric) gene expression product is a fusion protein with oncogenic potential for hematopoietic cells. In the process of leukemogenesis, the altered cell loses the ability to differentiate and becomes insensitive to regulatory signals, including those that initiate the processes of apoptosis; however, the constant division and proliferation favor the formation of further genetic aberrations. The direct cause of leukemia manifestation are secondary mutations, including activating mutations of transcription factors or kinases, inactivating mutations, and deletions of genes encoding factors that inhibit the neoplasm growth [10].

Regardless of the etiology, the first step in the process of neoplastic transformation is the occurrence of chromosomal instability (CIN), resulting in a disturbance of the regulation of oncogene expression and/or loss of neoplastic function of suppressor genes. CIN—that is, the accumulation of numerical and structural aberrations of chromosomes—is one of the characteristic features of neoplastic cells. Numerical and structural CINs arise during errors in chromosome segregation during mitosis. Abnormal duplication or segregation of chromosomes can promote whole chromosome gain or loss and/or changes in chromosome structure, including but not limited to translocations and deletions. Moreover, both numerical and structural changes predispose to the accumulation of successive changes in chromosomes, thus increasing CIN and changing the properties of the cell in an unpredictable manner, for example by increasing the rate of cell proliferation, increasing neoplasm malignancy or greater drug resistance. CIN is a source of genetic variation that promotes the adaptation of neoplasms to stressful environments or to cytotoxic antineoplastic drugs [11]. The phenomenon of CIN promotes carcinogenesis as it increases the probability of oncogene activation and inactivation of suppressor genes, and consequently allows the escaping out of control of regulatory mechanisms. Unfortunately, the mechanisms underlying CIN are not yet fully recognized. The abnormal course of the cell cycle leads to disturbances in the functioning of the cell life cycle checkpoints, errors in the attachment of the kinetochore to microtubules, abnormal sister chromatid cohesion, abnormal centrosome replication, abrasion of telomeres, and abnormalities in the checkpoint of the karyokinetic spindle syndrome (SAC) [12, 13]. In rapidly dividing cells, the number of chromosomes may be multiplied (polyploidy) or an incorrect number may be created by adding one or more chromosomes (aneuploidy) [14]. The appearance of aneuploidy may both promote and inhibit the formation of neoplasms [12].

In immunological, genetic, and molecular terms, acute leukemias are a heterogeneous group of neoplasms characterized by the clonal proliferation of immature cells of various hematopoietic lines. The proliferation of immature cells of the lymphoid lineage leads to ALL, which can be derived from B-lymphocyte precursor cells (B-cell precursor acute lymphoblastic leukemia: BCP-ALL) or T-lymphocytes (acute lymphoblastic leukemia: T-ALL). In contrast, clonal transformation of cells of the myeloid, monoid, erythroid, or megakaryocytic lineage leads to acute myeloid leukemia (AML) [10].

The most popular classification from 1976 was the FAB (French-American-British) classification, which includes the division of ALL into three subtypes – L1, L2, and L3 – and AML into eight subtypes – M0–M7 – distinguished on the basis of morphological cell features and cytochemical tests [15]. In clinical practice, however, the FAB classification turned out to be insufficient in the case of low-differentiated acute myeloid leukemias and acute lymphoblastic leukemias, where it is particularly important to determine the degree of maturity and the lymphocyte lineage from which cell proliferation occurs. The FAB classification is supplemented by the determination of the immunopheno-type of the tested cells: that is, determination of the expression of specific antigens of cluster differentiation (CD) occurring both on the cell surface and in the cytoplasm and in the cell nucleus. In 1995, the classification of acute lymphoblastic leukemias was introduced to the European Group for Immunological Classification of Leukemia (EGIL), whose criteria are based entirely on the blast cell immunophenotype [16]. The immunological classification of ALL is presented in Tables 1 and 2.

In 2016, the World Health Organization (WHO) introduced a new, currently binding classification that is the basis for the classification of hematopoietic neoplasms. This classification complements the FAB classification, and in addition to the morphological and cytochemical features of blast cells, it also includes immunophenotypic and cytogenetic and molecular characteristics, on the basis of which it is possible to initially determine the prognosis. The WHO classification has identified new AL subtypes or modified some of the existing ones. For example, among ALLs, the T-ALL subtype from early T-cell precursor acute lymphoblastic leukemia (ETP-ALL) has been distinguished, characterized by CD7 expression, lack of CD1a antigen, CD3 antigen surface expression, and lack of or low CD5 expression, with possible expression of myeloid antigens and frequent occurrence of specific mutations [17, 18]. In turn, among AML types, the previous FAB M3 AML subtype is now referred to as acute promyelocytic leukemia (APL) based on the presence of t(15; 17)(q22; q12) rearrangement or the PML/RARA translocation product detected by appropriate methods.

The classification of AML and ALL according to WHO is presented in Tables 3 and 4, respectively. Currently, acute leukemia is diagnosed when the percentage of blasts in the bone marrow is above 20% [18]. Genetic and molecular tests are becoming necessary in selecting the appropriate treatment regimen, including targeted therapy. Therefore, the WHO classification is updated on an ongoing basis based on new research results [19].

Acute leukemia is a heterogeneous group of diseases, varied in terms of a course, response to a treatment, and the occurrence of cytogenetic and molecular changes. In recent years, attempts have been made to identify the prognostic factors that predict the course of the disease at the time of diagnosis of AL. In new treatment protocols used in Poland—that is, AIEOP-BFM 2017 in ALL and AML-BFM 2019 in AML—stratification of patients into risk groups (high, intermediate, standard) is based on the response to treatment with a specific level of minimal residual disease (MRD) and the presence or absence of specific genetic disorders [21]. In designated risk groups, the intensity of therapy is varied, therefore the early identification of risk factors is important. Thanks to advances in chemotherapy and hematopoietic cell transplantation (HSCT), the long-term survival of ALL patients in children now exceeds 90% [22]. However, approximately 15–20% of children with ALL experience a recurrence (relapse). The relapse rate in patients with the first ALL recurrence is 71–93%, depending on the time and location of the relapse. Early relapse, up to 36 months after initial diagnosis, has a worse prognosis than isolated extramedullary or late relapse occurring more than 36 months after initial diagnosis. Although clinical remission can be achieved in the majority of ALL recurrences, long-term survival rates range from 40% to 50% [23, 24]. With regard to AML, the overall survival rate is approximately 70% [1, 25]. Recurrent AML is found in approximately 30% of patients, and the longterm survival rate in this group is approximately 40% [25].

In recent years, the progress in research on acute leukemia in children has contributed to the discovery of numerous anomalies related to, among other things, mutations of genes involved in the maturation of lymphocytes and in the regulation of the cell cycle. Some of these chromosomal aberrations, such as aneuploidy or chromosomal translocations, are already of broadly recognized diagnostic and prognostic importance, are included in modern therapeutic protocols, and contribute to the stratification of patients into specific risk groups.

In ALL, structural rearrangements are important, mostly including translocations. In BCP-ALL, the most common aberration is the t (12; 21) (p13; q22) translocation resulting in the ETV6-RUNX1 (TEL-AML1) fusion gene, which is a favorable prognostic factor. The t (12; 21) (p13; q22) translocation occurs in nearly 25% of children and its detection enables the stratification of patients into a group with a favorable prognosis. The disease-free survival rate is above 90%. On the other hand, patients with t (9; 22) (q34; q11) translocation with the formation of the BCR-ABL1 fusion gene and the so-called Philadelphia chromosome and patients with rearrangements involving the KMT2A (MLL) gene are classified to more intensive chemotherapy. KMT2A (MLL) rearrangements occur most frequently in infant ALL, however the occurrence of KMT2A-AFF1 t (4; 11) (q21; q23) translocation is associated with an unfavorable prognosis in all age groups. Sometimes, as in the case of the CDKN2A deletion, no influence of the aberration on the clinical course of the disease is observed [10, 26]. A common phenomenon in BCP-ALL is hyperdiploidy. High hyperdiploidy with a chromosome number above 51 is a favorable prognostic factor if the MRD level is not taken into account [27, 28]. On the contrary, hypodiploidy with the number of chromosomes below 44 is an unfavorable prognostic factor [29, 30, 31]. The most common genetic abnormalities are presented in Table 5.

In T-ALL, the most common chromosomal structural rearrangements include loci of genes encoding T-cell receptor (TCR) subunits and causing activation of oncogenes (TLX1, TLX3, TAL1, LMO1, LMO2, HOXA). Depending on the oncogene, the incidence of these genetic aberrations is 40–50%. The prognosis is favorable for t(10;14) (q24;q11) with TLX1 overexpression, and unfavorable for t(5;14)(q35;q32) with TLX3 overexpression. Activating mutations in the NOTCH1 gene, observed in various subtypes of pediatric T-ALL, have a direct impact on the increase in the self-renewal capacity of hematopoietic cells. The mutation has no unequivocally confirmed prognostic value, but a favorable prognosis is observed in some therapeutic protocols. Frequent rearrangements and fusion gene formation also take place within the 11q23 region of the KMT2A (MLL) gene, which has a negative impact on the expression of HOX genes, which are involved in normal hematopoiesis [10, 26]. The most common genetic abnormalities in childhood T-ALL are summarized in Table 6.

Detecting genetic disorders is also a basic element of AML diagnostics. The detection of disorders such as the t(8;21), t(15;17) translocation, chromosome 16 inversion, or the finding of chromosome 11 abnormalities involving the KMT2A gene are the bases for the recognition of specific AML subtypes. Moreover, according to the latest WHO classification from 2016, for the diagnosis of leukemia, it is more important and sufficient to determine the presence of t(8;21), t(15;17) or inv(16) inversion regardless of the percentage of bone marrow blasts [7]. The most common genetic abnormalities found in pediatric AML are presented in Table 7.

According to the currently used AIEOP-BFM 2017 protocol, independent factors with confirmed prognostic significance in ALL are: response to therapy expressed as MRD level on day 15 of remission-inducing treatment assessed by flow cytometry and on day 33 and before the start of the M protocol assessed by PCR. MRD levels on day 15. ≥ 10% and absence of complete remission by day 33 are unfavorable prognostic factors for both BCP-ALL and T-ALL, as is a poor response to prednisone on day 8 of treatment. In addition to the response to treatment, equally important risk factors are repeated genetic aberrations, including the presence of KMT2A, TCF3-HLF rearrangements, hypodiploidy < 45 chromosomes [17]. The immunophenotype of leukemic blasts and the distinction between BCP-ALL and T-ALL are of less importance, although within BCP-ALL the pro-B-ALL CD10-phenotype is strongly correlated with KMT2A rearrangement. Moreover, a better prognosis is associated with the DNA index (DI) of leukemic cells higher than 1.16 [32, 33].

According to the guidelines of the AML-BFM 2019 protocol, factors with a good prognostic value in AML in children include: the presence of t(16;16), t(8;21), t(1;11), inversion of chromosome 16 (inv (16)) and presence of NPM1 gene mutation or CEBPA biallelic mutation, with normal karyotype. In turn, the unfavorable prognostic factors include abnormalities within the 12p chromo-some, the presence of t(2;12), t(4;11), t(5;11), t(6;11), t(10;11) translocations, t(6;9), t(7;12), t(9;22), t(16;21), chromosome 5 monosomy (or 5q arm), inversion of chromosome 3 (inv (3)), gene mutations WT1, intra-tandem FLT3 gene duplication (FLT3-ITD), or a complex karyotype with three or more aberrations, including at least one structural aberration [34].

This paper presents the current achievements in the field of diagnosis of acute leukemias in children and a review of methods useful in routine laboratory diagnostics. Cytogenetics, immunophenotyping using flow cytometry, and molecular techniques are important, complementary methods used in the diagnosis of acute leukemias, also helpful in identifying risk factors and in drawing conclusions about prognosis. As a result of the enormous progress made in laboratory diagnostics, especially in the field of molecular techniques, in recent years, the number of prognostic markers in acute leukemias has increased significantly. On their basis, patients are currently stratified into risk groups with varying intensity of treatment.

Clinical observations indicate that quantitative cytogenetic disturbances may also affect the effectiveness of ALL treatment, and cytometric analysis of DNA content in leukemic cells is a sensitive and appropriate method for the identification of hyper-diploid clones. Flow cytometry is particularly useful when cell ploidy cannot be determined by cytogenetic methods. Moreover, introduction of methods into diagnostics that allow shortening the waiting time for the result is important from a clinical point of view and results in reducing costs. The identification of biologically distinct leukemia subtypes not only enables patient stratification, but also offers the opportunity to introduce individualized treatment based on the detected molecular target. Examples of already implemented individual therapies are the use of tyrosine kinase inhibitors in ALL with the confirmed presence of the BCR-ABL1 fusion protein. Moreover, immunotherapies with the use of bispecific antibodies and T lymphocytes with chimeric receptors (CAR-T cells) are an effective method of treating leukemia, especially in cases resistant to conventional treatment [22].