|Year : 2019 | Volume
| Issue : 1 | Page : 28-33
Clinical–radiological features of methotrexate-induced subacute leukoencephalopathy in patients with acute lymphoblastic leukemia: 'Panda eye sign' on diffusion weighted-magnetic resonance imaging
Abhishek Mahajan1, Hasmukh Jain2, Vijai Simha2, Tanvi Vaidya1, GV Santhosh Kumar1, Anurag Gupta1, Bhausaheb Bagal2, Manju Sengar2
1 Departments of Radiodiagnosis and Imaging, Tata Memorial Centre, Tata Memorial Hospital, Mumbai, Maharashtra, India
2 Department of Medical Oncology, Tata Memorial Centre, Tata Memorial Hospital, Mumbai, Maharashtra, India
|Date of Web Publication||9-Sep-2019|
Department of Radiodiagnosis and Imaging, Tata Memorial Hospital, Dr E Borges Road, Parel, Mumbai - 400 012, Maharashtra
Source of Support: None, Conflict of Interest: None
Background and Purpose: Subacute leukoencephalopathy in ALL is a rare complication after high dose methotrexate (HDMTX) administration and recognizing this self-remitting entity has important therapeutic implications. We did a retrospective study to evaluate the role of magnetic resonance imaging (MRI) in diagnosing this entity and asses the incremental value of diffusion weighted MR (DW-MRI).
Materials and Methods: We did a retrospective analysis of adult ALL patients being treated with the modified Berlin-Frankfurt-Münster (BFM)-90 protocol at our center between January 2014 and January 2015 and identified those who developed neurotoxicity after HDMTX. All these patients underwent a contrast enhanced CT and contrast enhanced MRI of brain with DW-MRI within 48 hours of presentation.
Results: Eleven patients were identified from a cohort of 239 patients (~5%). They presented with focal neurological deficits within ~14 days after HDMTX that resolved completely with conservative measures. The CT scans were normal in all these patients. A consistent finding seen in all these cases was the occurrence of restricted diffusion in the region of the centrum semiovale on DW-MRI. On diffusion maps, symmetrical areas of hyperintensity resembled Panda eyes and mean apparent diffusion coefficient cut-off of our series was 0.000453 x 10–3 +/- 0.000120 cm2/sec.
Conclusion: CT brain and conventional MR imaging have no significant role to play in diagnosing this entity however restricted diffusion in the centrum semiovale is a consistent imaging finding and the 'panda eye sign' as seen on DW imaging can be considered diagnostic for methotrexate induced subacute leukoencephalopathy and this sign can help in timely establishment of the diagnosis and appropriate management.
Keywords: Apparent diffusion coefficient, diffusion-weighted imaging, magnetic resonance imaging, methotrexate-induced neurotoxicity, quantitative diffusion imaging
|How to cite this article:|
Mahajan A, Jain H, Simha V, Vaidya T, Santhosh Kumar G V, Gupta A, Bagal B, Sengar M. Clinical–radiological features of methotrexate-induced subacute leukoencephalopathy in patients with acute lymphoblastic leukemia: 'Panda eye sign' on diffusion weighted-magnetic resonance imaging. Cancer Res Stat Treat 2019;2:28-33
|How to cite this URL:|
Mahajan A, Jain H, Simha V, Vaidya T, Santhosh Kumar G V, Gupta A, Bagal B, Sengar M. Clinical–radiological features of methotrexate-induced subacute leukoencephalopathy in patients with acute lymphoblastic leukemia: 'Panda eye sign' on diffusion weighted-magnetic resonance imaging. Cancer Res Stat Treat [serial online] 2019 [cited 2019 Nov 13];2:28-33. Available from: http://www.crstonline.com/text.asp?2019/2/1/28/266444
| Introduction|| |
Subacute leukoencephalopathy in acute lymphoblastic leukemia (ALL) is a rare complication after high-dose methotrexate (HDMTX) administration, and recognizing this self-remitting entity has important therapeutic implications. Methotrexate-induced neurotoxicity (MIN) can occur at different time intervals particularly in those being treated with intrathecal and high-dose regimens. The most common form has an acute onset that presents with features of meningitis including somnolence, confusion, headache, vomiting, and seizures. The existing literature is predominantly about the occurrence of acute toxicity and mainly in the pediatric age group. A less common entity is the occurrence of subacute toxicity occurring after the 1st week of methotrexate manifesting with mental status changes with or without focal neurological deficits similar to a cerebrovascular accident (CVA)., The last type of toxicity which is the chronic form occurs months to years later and is characterized by cognitive and behavioral abnormalities. Methotrexate-induced neurotoxicity is postulated to be caused by the disruption of the folate metabolism in the central nervous system (CNS) and direct neuronal damage.,, Risk factors for methotrexate toxicity include higher age (age >10 years) and high plasma methotrexate levels.
On magnetic resonance imaging (MRI), the subacute form is seen as bilateral symmetrical white hyperintensities on T2-weighted (T2W) images and fluid-attenuated inversion recovery (FLAIR) images. In severe cases, there may be extensive white matter signal abnormalities and areas of enhancement; this is termed as disseminated necrotizing leukoencephalopathy (DNL). Risk of DNL occurs when chemotherapy is administered concurrently with radiotherapy., Indeed, before the advent of advanced imaging techniques of MRI, the subacute form of methotrexate toxicity was attributed (possibly erroneously) to ischemic changes in the white matter,,, and it has only in the previous decade been recognized to be a metabolic insult., Although references exist in the literature regarding the changes in the water diffusion in the matter,,, the same has not been quantified in terms of the exact apparent diffusion coefficient (ADC) values on diffusion-weighted imaging (DWI) magnetic resonance (MR) and the topography of areas of diffusion restriction, which is a more useful way of referring to diffusion restriction on MRI and ruling out potential alternate diagnoses. The literature is limited on incremental colored diffusion maps and mean ADC (mADC) values and their role in diagnosing MIN. We report a case series of adolescent and young adult patients who manifested with CVA-like features after HDMTX and the incremental value of colored diffusion maps and ADC value in the diagnosis of MIN.
| Materials and Methods|| |
A retrospective review of the leukemia clinic database was performed for adolescent and adult (aged >14 years) ALL patients who were treated in the adult hematolymphoid unit with the modified Berlin–Frankfurt–Münster (BFM-90 protocol). 438 patients were registered between the years 2014 and 2015, of which 239 patients were treated at our center using the BFM-90 protocol. This protocol comprises an induction phase that lasts for 64 days (vincristine, steroids, daunomycin, l-asparaginase, 6-mercaptopurine, and cyclophosphamide), CNS phase with HDMTX (3 g/m 2 and 6-mercaptopurine), and reinduction phase (using the same drugs as induction, except that doxorubicin is used instead of daunomycin) followed by 2 years of maintenance. Patients who were treated with HDMTX (>1 g/m 2) and presented with new onset of neurological disturbances were identified. Eleven patients aged >14 years with ALL who developed acute-onset neurological symptoms within 2 weeks (14 days) after administration of HDMTX and underwent computed tomography (CT) and MR imaging with DW MRI (within 48 hours of presentation) were analyzed. These patients were not known to have any prior documented neurologic deficits or seizures and did not have lymphoblasts in the cerebrospinal fluid at diagnosis at baseline [Flow chart: [Figure 1].
|Figure 1: Flow chart of patient selection, treatment, and analysis (acute lymphoblastic leukemia who were selected to undergo radical treatment according to Berlin–Frankfurt–Münster-90 protocol)|
Click here to view
Imaging protocol and image analysis
Plain and postcontrast CT brain were performed in all patients. All the CT studies were performed on 16-slice multidetector CT Siemens CT machine. All the patients underwent contrast-enhanced MRI of brain with DW MRI within 48 hours of the onset of neurological deficits. MRI was performed on 1.5-T systems (HDxt GE) using 8 channel high-resolution dedicated brain coil. MRI protocol included spin-echo T1-weighted, fast spin-echo T2-weighted, FLAIR, DW, and gradient echo sequences (GRE) imaging. Postgadolinium injection, contrast-enhanced spin-echo T1-weighted imaging was also performed in all patients. The intravenous bolus dose of gadolinium contrast media was given at 0.1 mL/kg (0.1 mmol/kg) body weight. MR angiography was performed in five patients as these patients presented as stroke mimics.
Imaging parameters used for various sequences were as follows: (a) T1: repetition time (TR) ms/echo time (TE) ms, TR/TE: 500/20; flip angle, 90°; and matrix, 256 × 256 for spin-echo; (b) Fast spin echo T2-weighted imaging: TR/TE: 3000/90; flip angle, 90°; and matrix, 256 × 256 for; (c) FLAIR: 8000/80; 90°; and matrix, 256 × 256 for FLAIR imaging. The other parameters were as follows: 5-mm slice thickness with a 1-mm slice gap.
DW imaging at b = 0 and 1000 s/mm 2 (TR/TE 6000/20) in three orthogonal directions was performed in the axial plane before contrast material injection and combined into a trace image. DW images were both analyzed qualitatively and quantitatively. Colored maps were analyzed qualitatively and were classified as hyperintense, isointense, and hypointense as compared to normal white matter. For quantitative assessment, the region of interest was marked on both abnormal and normal brain parenchyma, and the mADC values of the two were compared. The mean mADC values (10−3 cm 2/s) were calculated on a voxel-by-voxel basis using FuncTool software (GE Healthcare) on an Advantage Windows (ADW) 4.4 workstation (Sun Microsystems, Santa Clara, CA, USA).
All statistical analyses were performed using SPSS (Statistical Package for the Social Sciences) for Windows, version 22.0 (SPSS, Chicago, IL, USA), and descriptive statistical methods were used. Mean and standard deviation were calculated for mADC in normal and abnormal brain parenchyma. P < 0.05 indicated a significant statistical difference.
| Results|| |
Patient characteristics and clinical findings
The total number of patients screened are described in [Figure 1]. Eleven patients (6 males and 5 female) aged >14 years with ALL (mean age at presentation, 24.5 years) with acute onset of neurological symptoms within 2 weeks (mean time to presentation, 10.6 days) after administration of HDMTX (1–3 g/m 2) were identified. The patients were not known to have any prior documented neurologic deficits or seizures. The patients were noted to have no lymphoblasts in the cerebrospinal fluid at diagnosis at baseline. Hemiparesis was the most common symptom seen in 6 patients (>50%) followed by dysarthria/aphasia (33%), followed by facial palsy and altered sensorium in 40%. They were treated with antiplatelet medications and supportive care measures. Three patients received dextromethorphan. All the patients had a complete recovery within 2 weeks and were subsequently followed up for an average of 17.1 months and did not show any signs of recurrence or progression of neurological deficit. None of the patients received any further HDMTX. The patient characteristics and clinical details are summarized in [Table 1].
All the patients underwent CT brain at the time of presentation and revealed no significant abnormality. MRI with MR angiogram (MRA) and DWI was performed in all the cases within 48 hours of presentation [Table 2]. The most common and consistent imaging findings on MR were focal and symmetrical regions of restricted diffusion on DWI images and colored ADC maps, mimicking the appearance of the eyes of a panda bear, involving both centrum semiovale and splenium of corpus callosum. This finding was seen in seven patients (63.7%) [Figure 2] and [Figure 3] and matched the appearance of 'panda eye sign' [Figure 4]. Asymmetric/unilateral restricted diffusion was the second most common MR finding seen in the rest (n = 4, 36.3%) of the cases. The mADC cutoff of our series was 0.000453 × 10−3 ± 0.000120 cm 2/s. This finding was not associated with any significant altered signal abnormality on conventional MR sequences. In two patients, mild hyperintense signal intensity was on T2 weighted, and in one patient, mild hyperintensity in the corresponding area was seen on FLAIR images [Case 2; [Figure 3]. After gadolinium injection, all except one case showed no significant abnormal enhancement of these lesions. The MRA showed focal decreased flow in the peripheral branches of middle cerebral artery in two patients (18.1%).
|Table 2: Computed tomography, conventional magnetic resonance, diffusion-weighed magnetic resonance findings|
Click here to view
|Figure 2: A 16-year-old boy with preB acute lymphoblastic leukemia. Ten days after the 2nd high-dose methotrexate dose, the patient developed dysarthria and dysphagia. The patient had complete resolution of symptoms spontaneously within 24 hours. His computed tomography and magnetic resonance brain was normal. T1, T2 (a), T2-fluid-attenuated inversion recovery, and postcontrast magnetic resonance imaging did not reveal any significant abnormality. The diffusion-weighted imaging showed bilateral restriction in diffusion (b and c and colored [d and e] diffusion-weighted maps) with mean apparent diffusion coefficient of the affected area 0.000416 cm2/s. Further high-dose methotrexate was stopped, and I2A multicenter protocol (MCP) protocol was initiated (12A phase of MCP protocol contains high dose AraC given at 2 g/m2 twice a day for 2 days in addition to 6MP and cyclophosphamide)|
Click here to view
|Figure 3: A 24-year-old male with T-acute lymphoblastic leukemia and 9p21 deletion on cytogenetics presented with aphasia and blurring of vision post 2nd high-dose methotrexate dose which recovered spontaneously in a period of 7 days. With conservative management, there was full resolution of symptoms. Magnetic resonance imaging showed no abnormality on T1W image (a). Mild hyperintensity is seen on fluid-attenuated inversion recovery (b) and T2 (c) images. Postcontrast magnetic resonance image revealed hypoenhancement of the corresponding T2 abnormal area (d). Grey scale (e) and colored (f) diffusion and apparent diffusion coefficient map showed areas of restricted diffusion in bilateral cerebral hemispheres with apparent diffusion coefficient value of 0.000230|
Click here to view
|Figure 4: (a-c) showing areas of restricted diffusion on colored mean apparent diffusion coefficient maps consistently localized to the centrum semiovale seen in our series, resembling the eye of a panda bear (d) and the proposed “Panda eye sign”|
Click here to view
| Discussion|| |
Subacute leukoencephalopathy in acute ALL is a rare complication after HDMTX administration, and recognizing this self-remitting entity has important therapeutic implications. Subacute methotrexate-induced toxicity occurs days to weeks after methotrexate and manifests as abrupt onset of neurological deficits presenting similar to a cerebrovascular event. This type of presentation has also been observed with 5-fluorouracil  and its derivative, carmofur, and with capecitabine, and it has been called 'delayed leukoencephalopathy with stroke like presentation' by Baehring and Fulbright.
The recognition of this nonvasculogenic entity is important because the treatment varies accordingly. The occurrence of a sudden onset of neurological change in a patient on chemotherapy makes a cerebrovascular event a likely possibility, and therefore, all the patients are likely to have a plain and contrast CT of the brain as the first imaging modality. The CT brain was normal in all the eleven patients in our series. MRI with DWI is the modality of choice in identifying and characterizing the pathological process in this scenario. This is the first and largest reported case series of MIN studied with DW MRI, and ours is the first report of the mean ADC cutoff for diagnosing MIN., Sandoval et al. reported a 13-year-old female patient with MIN that was diagnosed during the consolidation phase for preB-cell ALL. During the initial work-up of the patient, T2 images were normal, and DW images showed restricted diffusion within the white matter. On the follow-up MRI, these white matter abnormalities were seen to correspond to the areas of restricted diffusion. In Fisher et al.'s series of six patients, DWI in all cases showed restricted diffusion in the deep periventricular area that preceded abnormalities in the same region by FLAIR imaging. Another series of MIN reported by Rollins et al. found restricted diffusion that was limited to the white matter in five of six cases and involving the cortex in one patient. In our series, all the patients subsequently underwent MRI wherein all the eleven patients had diffusion restriction and lower mADC values on MRI in the centrum semiovale (deep white matter). Interestingly, two patients in our series had abnormal MRAs and showed arterial abnormalities on the affected side; however, none of them showed any T2W or FLAIR abnormality. The conventional MRI did not reveal a significant finding in our series.
In general, toxic and acquired metabolic disorders produce a widespread, symmetric pattern of injury that often involves the deep gray nuclei and cerebral cortex. However, in our series of methotrexate-induced encephalopathy, we observed predominant involvement of the centrum semiovale and corpus callosum, and in all the eleven patients, no other area of the white matter was involved.,, The centrum semiovale is the area of the brain beneath the cerebral cortex which contains axons of the pyramidal tract which are considered the longest axons in the CNS and the presence of fibers which are densely packed, which may be the reason why there is overt manifestation of toxic insult to this particular area of the brain. However, DW-MRI changes in other areas of white matter have been seen like the frontoparietal white matter and corpus callosum., There are multiple postulated mechanisms explaining methotrexate-induced neurotoxicity, including chronic folate depletion in the brain tissue, relative homocysteine excess  with increased excitatory amino acids, and alterations of biopterin and adenosine metabolisms that lead to decreased neurotransmitter synthesis. However, this is a transient neurotoxic insult that is likely to resolve spontaneously.
Methotrexate-induced subacute neurotoxicity must be therefore diagnosed when DWI shows areas of restricted diffusion across multiple vascular beds in the deep cerebral white matter particularly in the centrum semiovale after a course of HDMTX in the preceding few weeks. The appearance of 'panda eye sign' was consistently seen in seven of the eleven patients in our series, and we propose this as a characteristic diagnostic sign that will help radiologists and oncologists to make a timely diagnosis of this entity. This is particularly important in adults receiving HDMTX as the threshold for diagnosis of CVA in adults is much lower compared to the pediatric population. It is important to recognize this entity, as any major intervention is not called for in these patients. Dextromethorphan, a noncompetitive antagonist of the N-methyl-1-aspartate receptor, is the only recommended drug used in the treatment and has resulted in quick resolution of symptoms, and thus, the overall outcome is favorable.
Even though the study is a retrospective analysis, considering the prevalence and incidence of MIN, this is the largest series highlighting the role of DWI and colored ADC maps and mADC value in timely diagnosis. The management protocols and imaging protocol were uniform in all the patients. mADC value of the affected side was compared to the normal white matter to arrive at a cutoff mADC value.
To conclude, methotrexate-induced subacute toxicity is not related to the dose of methotrexate or delayed clearance of the drug after administration at high doses, which implies that further doses of methotrexate may be administered safely. CT brain and conventional MRI have no significant role to play in diagnosing this entity. Restricted diffusion in the centrum semiovale is the diagnostic sign, and 'panda eye sign' as seen on DW imaging can be considered as diagnostic sign for this entity.
| Conclusion|| |
CT brain and conventional MRI have no significant role to play in diagnosing this entity; however, restricted diffusion in the centrum semiovale is a consistent imaging finding and the 'panda eye sign' as seen on DWI can be considered diagnostic for methotrexate-induced subacute leukoencephalopathy, and this sign can help in timely establishment of the diagnosis and appropriate management.
Declaration of patient consent
The authors certify that they have obtained all appropriate consent forms. In the form, the patient has given his/her consent for his/her images and other clinical information to be reported in the journal. The patient understands that his/her name and other identifying information will not be published and due efforts will be made to conceal his/her identity, but that anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Tufekci O, Yilmaz S, Karapinar TH, Gozmen S, Cakmakci H, Hiz S, et al.
A rare complication of intrathecal methotrexate in a child with acute lymphoblastic leukemia. Pediatr Hematol Oncol 2011;28:517-22.
Bhojwani D, Sabin ND, Pei D, Yang JJ, Khan RB, Panetta JC, et al.
Methotrexate-induced neurotoxicity and leukoencephalopathy in childhood acute lymphoblastic leukemia. J Clin Oncol 2014;32:949-59.
Agarwal A, Vijay K, Thamburaj K, Ouyang T. Transient leukoencephalopathy after intrathecal methotrexate mimicking stroke. Emerg Radiol 2011;18:345-7.
Buizer AI, de Sonneville LM, van den Heuvel-Eibrink MM, Veerman AJ. Behavioral and educational limitations after chemotherapy for childhood acute lymphoblastic leukemia or wilms tumor. Cancer 2006;106:2067-75.
Cole PD, Beckwith KA, Vijayanathan V, Roychowdhury S, Smith AK, Kamen BA, et al.
Folate homeostasis in cerebrospinal fluid during therapy for acute lymphoblastic leukemia. Pediatr Neurol 2009;40:34-41.
Vezmar S, Becker A, Bode U, Jaehde U. Biochemical and clinical aspects of methotrexate neurotoxicity. Chemotherapy 2003;49:92-104.
Kim JY, Kim ST, Nam DH, Lee JI, Park K, Kong DS, et al.
Leukoencephalopathy and disseminated necrotizing leukoencephalopathy following intrathecal methotrexate chemotherapy and radiation therapy for central nerve system lymphoma or leukemia. J Korean Neurosurg Soc 2011;50:304-10.
Sakamaki H, Onozawa Y, Yano Y, Imai K, Sasaki T, Ibuka T, et al.
Disseminated necrotizing leukoencephalopathy following irradiation and methotrexate therapy for central nervous system infiltration of leukemia and lymphoma. Radiat Med 1993;11:146-53.
Ebner F, Ranner G, Slavc I, Urban C, Kleinert R, Radner H, et al.
MR findings in methotrexate-induced CNS abnormalities. AJNR Am J Neuroradiol 1989;10:959-64.
Packer RJ, Grossman RI, Belasco JB. High dose systemic methotrexate-associated acute neurologic dysfunction. Med Pediatr Oncol 1983;11:159-61.
Yim YS, Mahoney DH Jr., Oshman DG. Hemiparesis and ischemic changes of the white matter after intrathecal therapy for children with acute lymphocytic leukemia. Cancer 1991;67:2058-61.
Haykin M, Gorman M, van Hoff J, Fulbright R, Baehring J. Diffusion-weighted MRI correlates of subacute methotrexate-related neurotoxicity. J Neurooncol 2006;76:153-7.
Küker W, Bader P, Herrlinger U, Heckl S, Nägele T. Transient encephalopathy after intrathekal methotrexate chemotherapy: Diffusion-weighted MRI. J Neurooncol 2005;73:47-9.
Fisher MJ, Khademian ZP, Simon EM, Zimmerman RA, Bilaniuk LT. Diffusion-weighted MR imaging of early methotrexate-related neurotoxicity in children. AJNR Am J Neuroradiol 2005;26:1686-9.
Rollins N, Winick N, Bash R, Booth T. Acute methotrexate neurotoxicity: Findings on diffusion-weighted imaging and correlation with clinical outcome. AJNR Am J Neuroradiol 2004;25:1688-95.
Valle DA, Kakehasi FM, Melo RM, Siqueira CM, Soares TF, Rodrigues KE, et al.
Stroke-like encephalopathy following high-dose intravenous methotrexate in an adolescent with osteosarcoma: A case report. Rev Bras Hematol Hemoter 2016;38:364-7.
Khaddar S, Patil V, Noronha V, Joshi A, Menon N, Prabhash K. Delirium and seizures in a patient with head-and-neck squamous cell carcinoma on docetaxel, cisplatin, and 5-fluorouracil. Cancer Res Stat Treat 2019;2:105-7. [Full text]
Baehring JM, Fulbright RK. Delayed leukoencephalopathy with stroke-like presentation in chemotherapy recipients. J Neurol Neurosurg Psychiatry 2008;79:535-9.
Sandoval C, Kutscher M, Jayabose S, Tenner M. Neurotoxicity of intrathecal methotrexate: MR imaging findings. AJNR Am J Neuroradiol 2003;24:1887-90.
Cheung YT, Sabin ND, Reddick WE, Bhojwani D, Liu W, Brinkman TM, et al.
Leukoencephalopathy and long-term neurobehavioural, neurocognitive, and brain imaging outcomes in survivors of childhood acute lymphoblastic leukaemia treated with chemotherapy: A longitudinal analysis. Lancet Haematol 2016;3:e456-66.
Eichler AF, Batchelor TT, Henson JW. Diffusion and perfusion imaging in subacute neurotoxicity following high-dose intravenous methotrexate. Neuro Oncol 2007;9:373-7.
Davidson A, Payne G, Leach MO, McVicar D, Britton JM, Watson M, et al.
Proton magnetic resonance spectroscopy ((1) H-MRS) of the brain following high-dose methotrexate treatment for childhood cancer. Med Pediatr Oncol 2000;35:28-34.
Quinn CT, Griener JC, Bottiglieri T, Hyland K, Farrow A, Kamen BA, et al.
Elevation of homocysteine and excitatory amino acid neurotransmitters in the CSF of children who receive methotrexate for the treatment of cancer. J Clin Oncol 1997;15:2800-6.
Millot F, Dhondt JL, Mazingue F, Mechinaud F, Ingrand P, Guilhot F, et al.
Changes of cerebral biopterin and biogenic amine metabolism in leukemic children receiving 5 g/m2 intravenous methotrexate. Pediatr Res 1995;37:151-4.
Drachtman RA, Cole PD, Golden CB, James SJ, Melnyk S, Aisner J, et al.
Dextromethorphan is effective in the treatment of subacute methotrexate neurotoxicity. Pediatr Hematol Oncol 2002;19:319-27.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2]