Subacute methotrexate neurotoxicity (MTX-NT) may occur days to weeks after systemic or intrathecal (IT) MTX administration and is often manifest by stroke-like symptoms. The pathogenesis of MTX-NT has mainly been associated with cerebral folate homeostasis, but the specific mechanism leading to the development of this complication is mostly unknown and is likely to be multifactorial. Most of studies aimed to determine putative genetic determinants of this syndrome have been focused on the methylenetetrahydrofolate reductase (MTHFR) C677T single nucleotide polymorphism (SNP). However, there are other functional polymorphisms that have also been identified in enzymes and transporters related to MTX and folate homeostasis. In this context, we carried out an extensive genetic analysis through the screening of 21 SNPs in 11 relevant genes in a five-year-old girl with acute lymphoblastic leukemia (ALL) who developed MTX-NT. The analysis revealed the presence of numerous genetic variants that may have accounted for the neurotoxicity observed. We discuss the putative role of MTX pharmacogenetics in the pathogenesis of MTX-NT.

MTX is a valuable drug in the treatment of childhood malignancy and is included in both antileukemic and antisolid tumor protocols. However, the drug also has a significant toxic effect on the central nervous system (CNS) and can potentially lead to severe neurologic morbidity [1].

Subacute methotrexate neurotoxicity is an uncommon complication of MTX therapy that generally develops within 5–14 days after the administration of intrathecal or high-dose intravenous (HD-IV) MTX and is manifest by abrupt onset of transitory focal neurological deficits, such as aphasia or hemiparesis. Current knowledge on MTX-NT is based upon spinal fluid analysis and neuroimaging techniques. At the onset of the symptoms, conventional CT scans, T1 or T2 weighted magnetic resonance (MR) imaging and angiography typically show no abnormalities, whereas diffusion-weighted imaging is able to show restricted diffusion of water in the brain that clears after resolution of the clinical symptoms [2]. Follow-up MR imaging shows variable abnormal T2 and FLAIR signal intensity in the deep white matter, with no detectable neurological sequelae in most patients [2].

We present the case of a five-year-old girl with acute lymphoblastic leukemia (ALL) who developed MTX-NT and was found to carry numerous allelic variants in relevant genes of the folate pathways.

A five-year-old Chinese girl presented with a two-month history of fatigue, pallor, and ecchymosis. Work-up revealed pancytopenia, hepatosplenomegaly and more than 90% lymphoblasts in the bone marrow. She was diagnosed with high risk B-precursor ALL, with aberrant expression of myeloid markers and pathological karyotype t (7,17) (q32,q21). Pre-treatment study revealed no CNS affectation by leukemia and thrombophilia tests were negative. In accordance with the protocol of the Spanish Society of Pediatric Hematology and Oncology [3], the patient was treated with induction therapy that included prednisone, daunorubicin, vincristine, cyclophosphamide, asparaginase and triple IT therapy (TIT) (12 mg MTX, 30 mg cytarabine, and 20 mg hydrocortisone on days +1, +8, and +15). The patient tolerated induction therapy without complications and was a rapid early responder, without evidence of residual disease at day +36.

One week later (day +43), consolidation treatment was started with oral mercaptopurine (40 mg/m2 daily), TIT and HD-IV MTX (3 g/m2) over 24-hr continuous infusion, followed 36 hr later by 15 mg/m2 of folinic acid every 6 hr. No other drugs were administered during this time. MTX clearance was normal and the treatment was well tolerated except for mild thrombopenia. However, 10 days after the first course of HD-IV MTX (day +53), the patient developed two intermittent episodes of moderate dysarthria without aphasia, followed 12 hr later by left hemiparesis and paresthesias that lasted less than 2 hr with subsequent complete recovery. On physical examination, blood pressure was normal; the patient had no fever and was awake and alert with left hemiparesis and speech disability. Ocular movements and fundoscopic and sensitive examinations were normal. An emergency CT scan was negative and the analyses of CSF revealed normal protein and glucose and no lymphoblasts. Microbiologic analyses and PCR testing for neurotropic virus were also negative. Hemogram, coagulation studies, acute phase reactants, electrolytes, and viral serology were all normal. Plasma homocysteine (Hcy) concentration was within normal range (5.94 μM/L). Prophylactic treatment with low-molecular-weight heparin and dexamethasone was implemented.

MR imaging, including angiography and diffusion-weighted sequences, was performed four days after the neurological events (day +57) showing no abnormalities, which could be attributable to the patient being asymptomatic at that time [4]. However, electroencephalogram (EEG) revealed δ-waves activity on right parieto-temporal region without epileptiform discharges. In any case, the patient was scheduled to receive prophylactic antiepileptic therapy with levetiracetam for one year. MR images were repeated six weeks after the clinical event revealing higher signal intensity on FLAIR images in the deep white matter (see Fig. 1). These abnormalities disappeared three months later, which is consistent with previous reports [5]. At that time the EEG had become normal as well.

thumbnail image

Figure 1. MR imaging findings six weeks after neurologic event. Axial T2 (A) and coronal FLAIR (B) sequences show deep and periventricular white matter hyperintensity, especially in the right hemisphere.

Download figure to PowerPoint

After these neurological events, the administration of MTX (IV or IT) was terminated and the consolidation phase was resumed with HD-IV cytarabine. The patient completed the rest of the therapy for ALL, which included TIT without MTX and oral MTX. Oral doses were adjusted to 7 mg/m2 because severe neutropenia was observed when higher doses of MTX were administered. Presently, she remains in complete remission, asymptomatic and with normal neurological examination.

The neurological syndrome experienced by this patient is consistent with MTX-NT. Indeed, the stroke-like symptoms occurred 10 days after MTX administration and resolved completely within 24 hr. In addition, FLAIR images performed six weeks later showed abnormal hyperintense signals involving right subcortical white matter in a distribution according to the neurological deficits [2]. Moreover, further studies ruled out additional processes that could have explained the symptoms, such as CNS infiltration as well as metabolic, infectious or vascular events.

The pathogenesis of MTX neurotoxicity is poorly understood and no specific risk factors for the development of this complication have been identified to date. Pharmacokinetic data in our patient did not support the occurrence of neurotoxicity, although MTX-NT is not necessarily associated with pharmacokinetic parameters [6]. In search of a mechanistic explanation for the adverse effects observed, we focused on MTX pharmacogenetics.

There is a growing body of evidence indicating that MTX toxicity in ALL and other patients, including CNS affectation, can be related to the presence of SNPs [7]. However, most of the available studies are focused in one or, less commonly, a few significant SNPs. In the present work, we have aimed to identify a wider genetic signature that could have accounted for the observed toxicity.

The patient was found to carry functional allelic variants in most of the genes involved in MTX and folate pathways (Table I). On the basis of data from previous association studies, three detected genotypes could have played a major role in the described complications. Most importantly, the patient was homozygous for the methylenetetrahydrofolate reductase (MTHFR) C677T SNP, a genotype present in 10–20% of Chinese subjects. This genotype has been consistently associated with MTX adverse effects and suggested to induce MTX-NT in young ALL patients [8, 9]. Second, the patient carried the wildtype CC genotype in position 1420 of the serine hydroxymethyltransferase (SHMT1) gene, which codes for a pivotal enzyme of the folate pathway. This genotype has been linked to increased citotoxicity in leukemic cells of children with ALL [10] and to MTX-induced neurotoxicity in rheumatoid arthritis (RA) patients [11].

Table I. Single Nucleotide Polymorphisms (SNPs) Analyzed in the Patient
PolymorphismSNP effectPatient's genotype (%)aCorrelation genotype-toxicity
  • The patient's genotype for each SNP, its frequency in the Asian (Chinese) population and the reported correlation with MTX toxicity is also shown. NA, frequency not available in Asians; NS, no significant association with toxicity reported for this polymorphism; -, Association with toxicity reported for the SNP but not for the patient's genotype; +, Association with toxicity reported for homozygous but not heterozygous carriers; ++, Association with toxicity reported for the patient's genotype; +++, Severe neurotoxicity reported for the patient's genotype.

  • a

    Frequency of the genotype carried by the patient in Asian (Chinese) populations. Data have been retrieved from the HapMap database ( or from population studies when available.

  • b

    Frequency variability higher than 50% between studies

  • c

    Evidences from studies carried out in diseases other than leukemia

  • d

    In vitro data.

 MTHFR C677TReduced enzyme activityTT (10-20)+++ [8, 9]
 MTHFR A1298CReduced enzyme activityAC (25-30)Controversial [15, 26].
 DHFR ins/del 19bpIncreased transcription ratedel/del (NA)++ [17]
 DHFR C-1610G/T, C-680A, A-317GDecreased transcription rateCC/AA/GG (NA)NS, association with ALL outcome [27]
 TS 2R/3RIncreased transcription rate2R/3R (30)++b [16]
 TS UTR-6bp ins/delLower mRNA expressionins/del (45)++b [16]
 CCND1 A870GIncreased expressionAG (65)+ [28]
 MTRR A66GReduced enzyme activityAG (35)++ [15]
 MS A2756GReduced enzyme activityAA (60-90)b [29]
 SHMT C1420TReduced enzyme activityCC (85)+++c,d [10, 11]
 RFC1 G80ADecreased MTX uptakeGA (b)++ [20]
 ABCG2 C421ADecreased transport activityAA (8)+++ [14]
 ABCB1 C3435TDecreased expressionCT (40-60)+ [14]
 ABCB1 G2677T/A?GT (25)NS
 ABCB1 C1236T?CT (45)++c [22]
 ABCC2 C-24T?CC (60)c [30]
 ABCC2 IVS 23+56 T>C?CC (NA)++c [22]
 ABCC2 G1058A?GG (100)c [22]
 ABCC2 G1249A?GA (15)+c [22]

Third, the patient was homozygous for the ATP-binding cassette (ABC)G2 421A minor allele, a genotype carried by less than 10% of Chinese individuals that decreases the ability to pump substrates such as MTX out of the cell [12, 13]. Indeed, the 421AA genotype has been related to the occurrence of MTX adverse effects in children with ALL [14]. Moreover, the combined presence of the ABCG2 421A and ABCB1 3435T minor alleles detected in the patient has also been related with higher risk of encephalopathy [14].

The patient carried other genotypes that could also account for MTX-induced toxicity, although probably playing a less important role. For instance, the methionine synthase reductase (MTRR) 66AG heterozygous genotype has been recently related to increased risk of developing toxicity in children with ALL treated with HD MTX [15]. In addition, the patient was heterozygous for the two SNPs analyzed in the thymidylate synthase (TS) gene, which codes for a pharmacological target of MTX. While there is as yet no evidence of a link between TS SNPs and toxicity in the leukemia setting, a study in psoriasis patients did find an association with increased occurrence of MTX side effects [16]. Interestingly, the authors showed that even heterozygous individuals were also at higher risk for toxicity [16].

The patient was also homozygous for three of the four polymorphisms analyzed in the dihydrofolate reductase (DHFR) gene, which codes for another major MTX target enzyme. Of these SNPs, the 19bp del/del genotype has been associated with increased toxicity in leukemia patients treated with MTX [17].

With regard to SNPs in genes involved in MTX or folate transport, the patient was heterozygous for the G80A polymorphism in the reduced folate carrier 1 (RFC1, SLC19A1). The 80A minor allele has been associated to increased toxicity in children with ALL treated with MTX [18, 19], even in heterozigosity [20], although it should be remarked that the functional role of this SNP is still controversial [21].

In addition, data from the RA setting regarding polymorphisms in efflux transporters indicate that two minor alleles detected in the patient (ABCC2 IVS 23 + 56C and 1249A) might lead to higher risk of MTX toxicity [22]. Furthermore, the observed heterozygous ABCB1 1236GT and ABCC2 IVS 23 + 56TT genotypes have also been associated with increased MTX side effects in the same population [22].

It should be noted that large studies on the association between MTX toxicity and genetics in childhood ALL are still scarce. Because of this, some of the conclusions for the SNPs analyzed (especially regarding ABC transporters) were extrapolated from other diseases, particularly RA, and therefore should be taken cautiously.

According to the scheme depicted in Fig. 2, the excess of MTX and the decreased functionality of folate enzymes caused by the described genetic alterations could lead to increased Hcy intracellular levels [23]. Toxic effects of Hcy may be mediated by cerebrovascular ischemia via oxidative stress [24], which is consistent with the stroke-like symptoms developed by our patient. This underlying mechanism has also been proposed for patients with MTX-NT carrying the MTHFR 677T variant [8, 9]. It should be noted that Hcy plasma concentrations in the patient were normal. In fact, it has been suggested that other biological markers in the CSF could correlate better with the MTX-NT [25].

thumbnail image

Figure 2. Overview of methotrexate (MTX) mechanisms of action in the folate metabolic pathway. Blank arrows mark the location of the genotypes carried by the patient that most likely could have contributed to the reported complications. Solid arrows mark the predicted consequences of these mutations: Three mechanisms underlying the onset of neurotoxicity are proposed.

Download figure to PowerPoint

In summary, the genetic analysis shows that the patient carried numerous genetic variants previously associated with MTX side effects, including neurotoxicity, which may have accounted for some of the symptoms described. However, the current scenario of MTX pharmacogenetics is extremely complex. The wide array of applications of this drug has probably hampered the drawing of definitive conclusions with respect to the clinical impact of genetic polymorphisms, as research efforts have been divided into study designs with greatly different populations, therapies, etc. Larger, prospective studies are therefore needed to fully elucidate the genetic determinants of MTX-induced toxicity in childhood ALL patients.


Twenty-one SNPs affecting 11 genes involved in the MTX and folate intracellular pathways were analyzed in this study (Table I). SNPs were chosen on the basis of a previous reported association with MTX toxicity. In a few cases no direct evidence for this association was evident in the ALL setting but a relation with toxicity had been shown for other diseases (e.g., RA and psoriasis) (Table I). No direct association with MTX toxicity has been reported for four of the SNPs analyzed, namely ABCB1 G2677T/A and DHFR C-1610G/T, C-680A and A-317G. The ABCB1 G2677T/A SNP is a nonsynonimous polymorphism with the potential to alter the transporting ability of P-glycoprotein, a transmembrane protein likely involved in MTX transport. On the other hand, DHFR SNPs have been shown to decrease the transcription rate of the gene and to modify the outcome of ALL. For these reasons, a putative association with toxicity seems plausible and the SNPs were included in the analysis.

All polymorphisms were screened for by means of PCR-RFLP and direct sequencing.


  1. Top of page
  • 1
    Shuper A,Stark B,Kornreich L, et al. Methotrexate-related neurotoxicity in the treatment of childhood acute lymphoblastic leukemia. Isr Med Assoc J 2002; 4: 10501053.
  • 2
    Rollins N,Winick N,Bash R, et al. Acute methotrexate neurotoxicity: Findings on diffusion-weighted imaging and correlation with clinical outcome. AJNR Am J Neuroradiol 2004; 25: 16881695.
  • 3
    Badell I,Munoz A,Estella J, et al. Long-term results of two consecutive trials in childhood acute lymphoblastic leukaemia performed by the Spanish Cooperative Group for Childhood Acute Lymphoblastic Leukemia Group (SHOP) from 1989 to 1998. Clin Transl Oncol 2008; 10: 117124.
  • 4
    Kuker W,Bader P,Herrlinger U, et al. Transient encephalopathy after intrathekal methotrexate chemotherapy: Diffusion-weighted MRI. J Neurooncol 2005; 73: 4749.
  • 5
    Inaba H,Khan RB,Laningham FH, et al. Clinical and radiological characteristics of methotrexate-induced acute encephalopathy in pediatric patients with cancer. Ann Oncol 2008; 19: 178184.
  • 6
    Rubnitz JE,Relling MV,Harrison PL, et al. Transient encephalopathy following high-dose methotrexate treatment in childhood acute lymphoblastic leukemia. Leukemia 1998; 12: 11761181.
  • 7
    Gervasini G. Polymorphisms in methotrexate pathways: What is clinically relevant, what is not, and what is promising. Curr Drug Metab 2009; 10: 547566.
  • 8
    Mahadeo KM,Dhall G,Panigrahy A, et al. Subacute methotrexate neurotoxicity and cerebral venous sinus thrombosis in a 12-year old with acute lymphoblastic leukemia and methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism: Homocysteine-mediated methotrexate neurotoxicity via direct endothelial injury. Pediatr Hematol Oncol 2010; 27: 4652.
  • 9
    Strunk T,Gottschalk S,Goepel W, et al. Subacute leukencephalopathy after low-dose intrathecal methotrexate in an adolescent heterozygous for the MTHFR C677T polymorphism. Med Pediatr Oncol 2003; 40: 4850.
  • 10
    de Jonge R,Hooijberg JH,van Zelst BD, et al. Effect of polymorphisms in folate-related genes on in vitro methotrexate sensitivity in pediatric acute lymphoblastic leukemia. Blood 2005; 106: 717720.
  • 11
    Weisman MH,Furst DE,Park GS, et al. Risk genotypes in folate-dependent enzymes and their association with methotrexate-related side effects in rheumatoid arthritis. Arthritis Rheum 2006; 54: 607612.
  • 12
    Vlaming ML,van Esch A,Pala Z, et al. Abcc2 (Mrp2), Abcc3 (Mrp3), and Abcg2 (Bcrp1) are the main determinants for rapid elimination of methotrexate and its toxic metabolite 7-hydroxymethotrexate in vivo. Mol Cancer Ther 2009; 8: 33503359.
  • 13
    Mizuarai S,Aozasa N,Kotani H. Single nucleotide polymorphisms result in impaired membrane localization and reduced atpase activity in multidrug transporter ABCG2. Int J Cancer 2004; 109: 238246.
  • 14
    Erdilyi DJ,Kamory E,Csokay B, et al. Synergistic interaction of ABCB1 and ABCG2 polymorphisms predicts the prevalence of toxic encephalopathy during anticancer chemotherapy. Pharmacogenomics J 2008; 8: 321327.
  • 15
    Huang L,Tissing WJ,de Jonge R, et al. Polymorphisms in folate-related genes: Association with side effects of high-dose methotrexate in childhood acute lymphoblastic leukemia. Leukemia 2008; 22: 17981800.
  • 16
    Campalani E,Arenas M,Marinaki AM, et al. Polymorphisms in folate, pyrimidine, and purine metabolism are associated with efficacy and toxicity of methotrexate in psoriasis. J Invest Dermatol 2007; 127: 18601867.
  • 17
    Ongaro A,De Mattei M,Della Porta MG, et al. Gene polymorphisms in folate metabolizing enzymes in adult acute lymphoblastic leukemia: Effects on methotrexate-related toxicity and survival. Haematologica 2009; 94: 13911398.
  • 18
    Imanishi H,Okamura N,Yagi M, et al. Genetic polymorphisms associated with adverse events and elimination of methotrexate in childhood acute lymphoblastic leukemia and malignant lymphoma. J Hum Genet 2007; 52: 166171.
  • 19
    Shimasaki N,Mori T,Samejima H, et al. Effects of methylenetetrahydrofolate reductase and reduced folate carrier 1 polymorphisms on high-dose methotrexate-induced toxicities in children with acute lymphoblastic leukemia or lymphoma. J Pediatr Hematol Oncol 2006; 28: 6468.
  • 20
    Kishi S,Cheng C,French D, et al. Ancestry and pharmacogenetics of antileukemic drug toxicity. Blood 2007; 109: 41514157.
  • 21
    Whetstine JR,Gifford AJ,Witt T, et al. Single nucleotide polymorphisms in the human reduced folate carrier: Characterization of a high-frequency G/A variant at position 80 and transport properties of the His(27) and Arg(27) carriers. Clin Cancer Res 2001; 7: 34163422.
  • 22
    Ranganathan P,Culverhouse R,Marsh S, et al. Methotrexate (MTX) pathway gene polymorphisms and their effects on MTX toxicity in Caucasian and African American patients with rheumatoid arthritis. J Rheumatol 2008; 35: 572579.
  • 23
    Gellekink H,Blom HJ,van der Linden IJ, et al. Molecular genetic analysis of the human dihydrofolate reductase gene: Relation with plasma total homocysteine, serum and red blood cell folate levels. Eur J Hum Genet 2007; 15: 103109.
  • 24
    Stamler JS,Osborne JA,Jaraki O, et al. Adverse vascular effects of homocysteine are modulated by endothelium-derived relaxing factor and related oxides of nitrogen. J Clin Invest 1993; 91: 308318.
  • 25
    Vezmar S,Schusseler P,Becker A, et al. Methotrexate-associated alterations of the folate and methyl-transfer pathway in the CSF of ALL patients with and without symptoms of neurotoxicity. Pediatr Blood Cancer 2009; 52: 2632.
  • 26
    Pakakasama S,Kanchanakamhaeng K,Kajanachumpol S, et al. Genetic polymorphisms of folate metabolic enzymes and toxicities of high dose methotrexate in children with acute lymphoblastic leukemia. Ann Hematol 2007; 86: 609611.
  • 27
    Dulucq S,St-Onge G,Gagne V, et al. DNA variants in the dihydrofolate reductase gene and outcome in childhood ALL. Blood 2008; 111: 36923700.
  • 28
    Costea I,Moghrabi A,Laverdiere C, et al. Folate cycle gene variants and chemotherapy toxicity in pediatric patients with acute lymphoblastic leukemia. Haematologica 2006; 91: 11131116.
  • 29
    Dervieux T,Greenstein N,Kremer J. Pharmacogenomic and metabolic biomarkers in the folate pathway and their association with methotrexate effects during dosage escalation in rheumatoid arthritis. Arthritis Rheum 2006; 54: 30953103.
  • 30
    Rau T,Erney B,Gores R, et al. High-dose methotrexate in pediatric acute lymphoblastic leukemia: Impact of ABCC2 polymorphisms on plasma concentrations. Clin Pharmacol Ther 2006; 80: 468476.

Jose Manuel Vagace*, Cristina Caceres-Marzal†, Mercedes Jimenez‡, Maria Soledad Casado*, Silvia Gonzalez de Murillo‡, Guillermo Gervasini‡, * Department of Pediatric Hematology, University Hospital Infanta Cristina, Badajoz, Spain, † Pediatric Neurology, University Hospital Infanta Cristina, Badajoz, Spain, ‡ Department of Pharmacology, University of Extremadura, Badajoz Spain.