Mexican Childhood Acute Lymphoblastic Leukemia: A Pilot Study of the MDR1 and MTHFR Gene Polymorphisms and Their Associations with Clinical Outcomes

Introduction

Acute lymphoblastic leukemia (ALL) is the most common malignancy diagnosed during childhood. In Mexico, 691 cases of leukemia in patients <20 years old were reported between 2005 and 2010, of which 562 (81%) were diagnosed with ALL (Perfil epidemiologico del Cáncer en niños y adolescentes en Mexico 2001). The current chemotherapy protocols cure almost 80% of patients, but side effects, relapse, resistance, and death remain a problem. It has been reported that genetic polymorphisms in patients with ALL can alter drug-metabolizing enzymes, transporters, and targets and, therefore, can influence both the efficacy and toxicity of chemotherapeutic agents (Bailey and Gregory, 1999).

The transport P-glycoprotein (P-gp) is a plasma membrane protein that is encoded by the ATP-binding cassette B1 (ABCB1) gene, which is also known as the multidrug resistance gene (MDR1) (Horio et al., 1988). P-gp is expressed in normal tissues on peripheral blood mononuclear cells, including natural killer cells, CD8⁺, and CD4⁺ cells, in the small and large intestine, kidneys, liver, brain (as part of the blood–brain barrier), testis, muscle, placenta, and adrenal glands (Thiebaut et al., 1987). Its physiological function is to protect cells from xenobiotics by functioning as an efflux pump. The P-gp substrates that are clinically relevant to oncology include the following: daunorubicin, doxorubicin, vincristine, and methotrexate (Hamidovic et al., 2010).

The MDR1 gene is located on chromosome 7q21 and comprises 28 exons (Lin and Yamazaki, 2003). Several polymorphisms have been identified in MDR1, including silent C3435T polymorphisms (rs1045642) in exon 26 at position 3435, which are associated with the altered expression and function of P-gp (Hoffmeyer et al., 2000). Previous research investigating the frequency of the 3435 C > T polymorphism found that this frequency varies between ethnicities; its association with clinical responses to some drugs has led to conflicting results. Some of the principal studies using digoxin as a P-gp substrate showed a positive association between the polymorphism 3435 TT and a high digoxin plasma level (Hoffmeyer et al., 2000). Likewise, an association between oral mucositis and the MDR1 3435 CT genotype has been reported in a Turkish population (Bektaş-Kayhan et al., 2012).

Methylenetetrahydrofolate reductase EC 1.5.1.20 (MTHFR) catalyzes the conversion of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate in the folic acid cycle and is blocked by methotrexate. The MTHFR gene is located on chromosome 1p36.6 and comprises 11 exons (Goyette et al., 1998). Another major polymorphism is a single-nucleotide polymorphism at nucleotide C677T (rs1801133), in which the replacement of cytosine with thymine results in an amino acid substitution (alanine to valine) at the 22 protein position. A lower enzymatic activity has been reported in homozygous 677TT individuals (Rozen, 1997; Ulrich et al., 2000; Chiusolo et al., 2002; Toffoli et al., 2003).

Due to such differences in activity, it has been hypothesized that genetic variants may modify the therapeutic response to antifolate chemotherapeutic agents, such as methotrexate (MTX). In support of this, an association between the T677 variant and gastrointestinal and hepatic toxicity after low-dose methotrexate therapy has been reported in leukemia patients (Chiusolo et al., 2002).

Due to the importance of the transport P-glycoprotein and methylenetetrahydrofolate reductase in the metabolism of chemotherapeutic agents, we genotyped the MDR1 rs1045642 and MTHFR rs1801133 polymorphisms and determined their associations with clinical outcomes in Mexican pediatric ALL patients.

Materials and Methods

Patients and samples

One hundred nine pediatric patients with a confirmed diagnosis of ALL participated in this study. The inclusion criteria were an age less than 19 years, absence of other active malignancies, and a lack of infection with human immunodeficiency virus-1. All patients were diagnosed, treated, and followed up at the Hemato-Oncology Department at the Hospital Infantil de Mexico Federico Gomez, SSA in Mexico City. All diagnoses were based on the criteria developed by the World Health Organization (WHO) and conformed to the FAB (French–American–British) classification system.

The diagnosis of ALL was based on bone marrow morphology and immunophenotyping. According to the National Cancer Institute risk classification, patients were classified as having a standard risk (with ages ranging from 1 to 9.99 years and an initial white blood cell count (WBC) <50 × 10⁹/L) or a high risk (age <1 or ≥10 years and/or an initial WBC ≥50 × 10⁹/L).

Treatment was a modification of the St. Jude Total XIII protocol. In brief, remission induction therapy for all patients was started 7 days after the initiation of up-front therapy with dexamethasone (60 mg/m² per day for 28 days), vincristine (2.0 mg/m² thrice a week for 9 doses), daunorubicin (25 mg/m² on days 1 and 8), l-asparaginase (10,000 U/m² on day 12), and etoposide (300 mg/m²) plus cytarabine (300 mg/m²) on days 22, 25, and 29. Upon attaining complete remission, all patients received consolidation therapy with weekly high-dose methotrexate (2 g/m² over 2 hours for low-risk patients and 5 g/m² for high-risk patients), followed by leucovorin rescue. Postremission therapy (120 weeks) for lower-risk cases consisted of daily mercaptopurine (75 mg/m²) and weekly intramuscular methotrexate (50 mg/m²) and for high-risk patients, a total XV-based maintenance protocol.

Two milliliters of blood were collected in tubes containing dipotassium EDTA as an anticoagulant. Genomic DNA was isolated from the peripheral whole blood using the QIAmp DNA kit (Qiagen). The study was approved by the Institutional Research, Ethics (informed consent was obtained from the parents and/or patients), and Biosafety Committee of Hospital Infantil de Mexico “Federico Gomez,” under protocol number HIM-2008-014.

Clinical evaluation

The evaluation consisted of a physical examination and a laboratory test (including a complete blood cell count). Serum methotrexate concentrations were evaluated at 42 hours after the start of methotrexate infusion. Serum samples (20 μL) were centrifuged at 13,000 rpm for 5 minutes. To quantify MTX, TDx instrumentation (Abbott 10365 series) using FPIA (fluorescence polarization immunoassay) technology was used; this is an immunoassay based on the principles of antigen/antibody reactions. According to the MTX plasma pharmacogenetics on high-dose infusion, MTX plasma levels >1 μM was selected as the cutoff value. Mucositis was determined by oral examination. A single episode of mucositis was required to define its presence. Relapse was defined as the medullar reappearance of more than 20% blast cells or extramedullar leukemic infiltrates. Leukopenia grades were established according to the National Cancer Institute Common Toxicity Criteria, version 4.0. Patient characteristics and clinical data were collected retrospectively.

Genotyping

MDR1

Genotyping of MDR1 rs1045642 (3435 C>T) was performed using a modification of the method described by Tang et al. (2002). A stretch of 249 bp on exon 26 of the MDR1 gene was amplified by polymerase chain reaction (PCR) using 5′ GCT CCC AGG CTG TTT ATT TG 3′ as the forward primer and 5′ TGT TTT CAG CTG CTT GAT GG 3′ as the reverse primer. The 50-μL reaction contained 200 ng of genomic DNA, 1 × buffer with 1.5 mmol/L MgCl₂, 0.16 mmol/L of each of the four deoxynucleotide triphosphates, 0.4 mmol/L of each primer, and 2 U of Taq polymerase (New England BioLabs). The amplification program consisted of an initial denaturation step of 95°C for 5 minutes, followed by 35 cycles of 95°C for 1 minutes, 53°C for 1 minutes, and 72°C for 1 minutes. A final extension was performed at 72°C for 7 minutes.

Ten microliters of PCR products were digested with DpnII enzyme (New England BioLabs) at 37°C for 4 hours. The DNA fragments were then separated and visualized by electrophoresis on 2% agarose gels. The wild-type gene (rs1045642 CC) showed a single band at 249 bp, and the heterozygous condition (rs1045642 CT) showed three bands of 249, 162, and 87 bp. The homozygous condition (rs1045642 TT) had two bands of 162 and 87 bp.

MTHFR

A modification of the method described by Fross et al. (1995) was used for the detection of MTHFR rs1801133 (677 C>T). A stretch of 198 bp on exon 4 of the MTHFR gene was amplified by PCR using 5′ TGA AGG AGA AGG TGT CTG CGG GA 3′ as the forward primer and 5′AGG ACG GTG CGG TGA GAG TG 3′ as the reverse primer. The C to T polymorphism at codon 677 introduces a restriction site for the HinfI enzyme. Each 50-μL reaction contained 200 ng of genomic DNA, 1× buffer with 1.5 mmol/L MgCl₂, 0.16 mmol/L of each of the four deoxynucleotide triphosphates, 0.4 mmol/L of each primer, and 2 U of Taq polymerase (New England BioLabs). The amplification program consisted of an initial denaturation step of 95°C for 5 minutes, followed by 30 cycles of 95°C for 1 minutes, 58°C for 1 minutes, and 72°C for 1 minutes. A final extension was performed at 72°C for 7 minutes.

Ten microliters of PCR products were digested with HinfI enzymes (New England BioLabs) at 37°C for 4 hours. The DNA fragments were then separated and visualized by electrophoresis on 3% agarose/2% SFR agarose gels. The wild-type gene (rs1801133 CC) showed a single band at 198 bp, and the heterozygous condition (rs1801133 CT) showed three bands of 198, 175, and 23 bp. The homozygous condition (rs1801133 TT) had two bands of 175 and 23 bp.

Statistical analysis

The statistical analysis was conducted using SPSS 14.0. The results are expressed as the mean ± standard deviation (SD) of the measured parameters. The leukocyte counts (total WBC) were grouped according to the common toxicity criteria of the NCI. The genotype frequency was determined by direct counting. The odds ratios (ORs) and 95% confidence intervals (CIs) were estimated by logistic regression as a measure of the association between the genotype and the clinical variables. The regression model includes each genotype as an independent variable; mucositis, relapse, MTX plasma concentration, and leukocyte count were included as dependent variables. All models were adjusted for age and sex. An alpha level of 5% (p = 0.05) was considered significant.

Results

This study recruited 140 children with ALL, 15 patients were excluded due to loss of any of clinical data follow-up and 16 had insufficient blood samples or loss of DNA integrity. Only 109 patients completed the study protocol. The general characteristics of the subjects and their genotypic distribution are presented in Table 1. The mean age was 8.1 years (SD = 4.24; range = 1–18.), and most of the patients presented with ALL L2 (55%). The “CC” genotype was the most common genotype for MDR1 rs1045642 (46.8%), and the “CT” genotype was the most common genotype for MTHFR rs1801133 (40.4%).

The multivariate logistic regression showed a significant genetic association with MTX plasma levels, mucositis, and leukocyte count (Table 2). MDR1 rs1045642 CC patients showed a higher risk of presenting with MTX plasma levels >1 μM (OR = 2.79; 95% CI = 1.11–7.01) and grade I leukopenia (OR = 3.53; 95% CI = 1.45–8.56). MDR1 rs1045642 TT carriers had a lower risk of MTX plasma levels >1 μM (OR = 0.30; 95% CI = 1.25–0.97) and grade I leukopenia (OR = 0.343; 95% CI = 1.21–0.97). MTHFR rs1801133 CC carriers showed a higher risk of mucositis development (OR = 3.49; 95% CI = 1.36–8.95) than MTHFR rs1801133 CC and TT carriers. No association was found between relapse and MDR1 rs1045642 or MTHFR rs1801133.

After stratification by risk group (Tables 3 and 4), the data showed a significant genetic association of high-risk patients, who were MDR1 rs1045642 CC carriers, with MTX plasma levels (OR = 3.20; 95% CI = 0.40–54.82), relapse (OR = 3.94; 95% CI = 1.45–10.69), and leukocyte count (OR = 5.26; 95% CI = 0.50–13.39). There was also an association between the standard risk patients, who were MTHFR rs1801133 CC carriers, and mucositis (OR = 24; 95% CI = 2.03–282.67).

Discussion

It has been shown that genetic polymorphisms in patients with ALL can alter drug-metabolizing enzymes, transporters, and targets and can, therefore, influence both the efficacy and toxicity of chemotherapeutic agents. For this reason, the main goal of this study was to investigate the association of the MDR1 rs1045642 and MTHFR rs1801133 polymorphisms with clinical outcomes (mucositis, relapse, MTX plasma levels, and total leukocyte count) in pediatric ALL patients in a pediatric public hospital in Mexico City.

A comparison of the gene frequency results obtained herein with those of other populations revealed some differences. We found that MDR1 rs1045642 CC was the most frequent genotype (46.8%), whereas in a Turkish population, the heterozygote MDR1 rs1045642 CT was the most frequent genotype (Bektaş-Kayhan et al., 2012); this genotype was found in our population at least 20% less frequently than in the Turkish population (18.3%). Comparisons with Indian and Brazilian populations revealed a difference in the frequency of the MTHFR rs1801133 CC genotype, which occurred at a frequency of 28.4% in our study, which was at least 50% less than the frequency in the Indian population and 25% less than the frequency in the Brazilian population (Sadananda Adiga et al., 2010; Vasconcelos de Deus et al., 2012).

Finally, our study found the MTHFR rs1801133 CC genotype (31.2%) at a frequency that was at least 20% higher than that in the Brazilian, Indian, and German populations (Schnakenberg et al., 2005; Sadananda Adiga et al., 2010; Vasconcelos de Deus et al., 2012). It is well known that ethnic differences may be accompanied by differences in genetic frequencies; our results confirm the importance of analyzing each population rather than generalizing the results obtained in other populations, especially in pharmacogenetic studies.

Our results demonstrate the importance of the association between transporter gene variability and drug plasma levels as carriers of MDR1 rs1045642 CC had a 3.20-fold higher risk of MTX plasma levels >1 μM than individuals in the high-risk group (Tables 3 and 4). However, our data are not in agreement with those studies that note the importance of the T allele in drug clearance, where it was reported that subjects that harbor the T allele contain more than fourfold higher plasma levels of digoxin than those with the CC genotype (Hoffmeyer et al., 2000), and carriers of the T allele with gastrointestinal stromal tumors or chronic myeloid leukemia have a higher imatinib clearance (Gurney et al., 2007).

We believe that these discrepancies are essentially due to differences in the methodologies, diseases, and population backgrounds of the above studies. The first study was performed by measuring MDR1 expression and the P-gp in vivo activity in eight samples of healthy Caucasian volunteers, whereas the second was an in vivo test of the effects of phenotype and genotype on imatinib clearance in 22 patients with gastrointestinal stromal tumors or chronic myeloid leukemia. Because the individual responses to a drug are a complex system, and to strengthen our finding, we recommend the analysis of additional variables, such as other polymorphisms, and organ function evaluation.

Oral mucositis is one of the most common complications of radiotherapy and chemotherapy treatment for cancer. It is a complex condition resulting from several factors such as the interaction between antineoplasic agents and epithelial cells, the oral microbiota, and the deficient immune status of the patient (Pereira Pinto et al., 2006).

In this study, we found no significant association between MDR1 rs1045642 and the development of mucositis. However, in a Turkish population, there was a positive association between MDR1 3435 CT and mucositis (OR = 2.80) (Frosst et al., 1995). These contradictory results can be attributed to differences between the two studies, especially in terms of the sample size and the frequency of mucositis. On the other hand, in this study, we report that carriers of MTHFR rs1801133 CC have a threefold higher risk of developing mucositis (OR = 3.49; 95% CI = 1.36–8.95) than those with the MTHFR rs1801133 CT or TT genotype (Table 2). However, a previous Mexican study found no association between the genotype and mucositis (Ruiz-Argüelles et al., 2007), possibly because it analyzed only 28 adult samples.

To determine whether the association between MTHFR rs1801133 and mucositis varies with age, an additional study will be required. Furthermore, in the stratified analysis, we found an increased risk in the standard group of patients (Tables 3 and 4) with MTHFR rs1801133 CC; however, because this group was small, these conclusions should be interpreted with caution. The association between MTHFR rs1801133 CC and the development of mucositis could be partly explained by the relation MTHFR-genotype/enzymatic activity (Frosst et al., 1995). It is possible that those patient carriers of MTHFR rs1801133CC (encodes to a high activity enzyme) under MTX-treatment have more capability to inhibit DNA synthesis, causing cell division alteration and death and/or damage oral epithelial cells.

Further studies are needed to determine mechanisms by which the MTHFR polymorphism is involved in the pathogenesis of mucositis. Because lesions of oral mucositis are often very painful, can compromise nutrition and oral hygiene, and increase the risk of local and systemic infection, and given the economic impact of mucositis management, it is important to consider MTHFR rs1801133 genotyping in the management of ALL pediatric patients.

Regarding the relapse variable, bivariate analyses have not found any significant associations; however, when analyzed by risk group, we found a significant association between relapse and the high-risk group carriers of MDR1 rs1045642 CC (OR = 3.94; Tables 3 and 4).

Leukopenia is a typical result of chemotherapy treatment. Our analysis of the total leukocyte count (adverse event grade I) in MDR1 rs1045642 patients revealed that the MDR1 rs1045642 CC carriers had a 3.53-fold risk of developing leukopenia, which increases to 5.26-fold among those classified as ALL high-risk group patients. On the other hand, MDR1 rs1045642 TT was associated with a significant lower risk of leukopenia.

In these contexts, increased values in some clinical variables (WBC, blast%, and mean LDH levels) have been reported for all ALL MDR1 C3435T genotype carriers, suggesting that an association with the CC genotype is a poor prognostic factor. These results support the hypothesis that the CC genotype (high expression) may be eliminating antileukemic drugs that are P-gp substrates, leading to lower intracellular drug concentrations and a poor prognosis (Rao et al., 2010). In addition, it has been suggested that a possible genetic predisposition to neutropenia exists in patients with solid tumor variants at the 3435 locus (1.5-fold higher risk) (Sissung et al., 2006), and an association between neutropenia and variant allele carriers of ABCB1 rs1045642 (OR = 5.174; 95% CI: 1.674–15.989) has been described in Lebanese children with ALL (Zgheib et al., 2014).

These results support the importance of pharmacogenetic clinical studies, but the association with immune response cells has not been documented in detail; thus, further studies are required.

Genetic polymorphisms in patients with ALL can alter drug-metabolizing enzymes, transporters, and targets; therefore, they can influence both the efficacy and the toxicity of chemotherapeutic agents. Our results suggest that carriers of MDR1 rs1045642 CC have a higher risk of MTX plasma levels >1 μM and leukopenia grade I, and patients with the MTHFR rs1801133 CC genotype have a higher risk of developing mucositis as a result of ALL treatment. MDR1 rs1045642 and MTHFR rs1801133 can be considered candidates for the identification of pediatric patients with a high risk of suffering adverse events such as mucositis and MTX toxicity during ALL treatment. A pilot clinical-intervention study should be conducted, and additional polymorphisms should be analyzed.

Acknowledgment

This study was partially supported by an HIM-2008-014 grant.

Author Disclosure Statement

No competing financial interests exist.