2026 年 13 巻 p. 523-528
Methotrexate, a folate-analogue antimetabolite, has widely been used for the treatment of rheumatoid arthritis, and its association with the development of lymphoma has been recognized since the early 1990s. Although methotrexate-associated lymphoproliferative disorder shows diverse clinical manifestations, central nervous system involvement seems to be rare. The great majority of central nervous system methotrexate-associated lymphoproliferative disorder cases reported in the literature had an intraparenchymal lesion, and cases with dural lesion alone have been extremely rare. A 77-year-old woman with a 20-year history of oral methotrexate therapy for rheumatoid arthritis was referred for evaluation of progressive dural thickening. Contrast-enhanced T1-weighted magnetic resonance imaging of the brain showed strong enhancement of the thickened frontal dura mater. Dural biopsy via a small craniotomy revealed mucosa-associated lymphoid tissue lymphoma, leading to the diagnosis of methotrexate-associated lymphoproliferative disorder. Two months after discontinuation of methotrexate, follow-up magnetic resonance imaging showed regression of not only the dural thickening but also of the pulmonary lesions. Methotrexate-associated lymphoproliferative disorder presenting with dural thickening is rare. Considering that methotrexate-associated lymphoproliferative disorder often regresses following methotrexate cessation alone, early recognition of this adverse effect of methotrexate and prompt methotrexate discontinuation are crucial. In this case, imaging findings alone were insufficient for diagnosis, and dural biopsy was critical in establishing the diagnosis of methotrexate-associated lymphoproliferative disorder.
Methotrexate (MTX), classified as a synthetic disease-modifying antirheumatic drug, is one of the most commonly used drugs for the treatment of rheumatoid arthritis (RA).1,2) The relationship between MTX therapy and secondary lymphoma was first suggested in several case reports published in the early 1990s.3,4) Since then, lymphoproliferative disorders (LPDs) that occur in the setting of MTX therapy have become widely known to physicians. MTX-associated LPD (MTX-LPD) was first included in the World Health Organization (WHO) guidelines in 2008 under "other iatrogenic immunodeficiency-associated lymphoproliferative disorders," enhancing the recognition of this disease.5) Nevertheless, the patho-mechanism of MTX-LPD has not been fully understood, and it is still challenging to diagnose MTX-LPD or to differentiate it from primary malignancies or metastatic tumors.6-8) Regression of LPD following MTX discontinuation is the core concept of MTX-LPD and is regarded as strong evidence for the carcinogenic potential of MTX.8,9) Therefore, MTX-LPD should be considered in the differential diagnosis of tumors under the setting of MTX therapy. Furthermore, MTX-LPD involving central nervous system (CNS) is particularly rare.10,11) Here, we present a case of MTX-LPD manifesting as dural thickening without intraparenchymal lesions.
A 77-year-old woman with a 20-year history of MTX therapy for RA visited a local neurosurgery clinic with a complaint of dizziness. Brain magnetic resonance (MR) fluid attenuated inversion recovery (FLAIR) imaging revealed thickening of the bilateral frontal dura mater (Figure 1A). Contrast material had not been used at that imaging session. The dural lesions had been considered to be unrelated to her symptom, and thereafter, she had been followed up regularly at the clinic. Although she remained asymptomatic, follow-up MR FLAIR image obtained 1 year after the first MR imaging exhibited progressive dural thickening with de novo appearance of high-signal intensity area in the right frontal lobe (Figure 1B), and she was referred to our institution for further evaluation. The high-signal intensity lesion observed on the FLAIR imaging was depicted as an iso-intensity area on diffusion-weighted imaging (DWI) (Figure 1C). During the follow-up period, ground-glass opacity in the right lower lung field was incidentally noted on her chest X-ray. A subsequent chest computed tomography (CT) without contrast revealed infiltrates and a nodular lesion in the right lung. She was also referred to a pulmonologist of our institution with a clinical suspicion of lung cancer.
Non-contrast brain magnetic resonance (MR) fluid attenuated inversion recovery (FLAIR) image showing thickening of bilateral frontal dura mater (A). Follow-up MR FLAIR image obtained 1 year after the first MR imaging study exhibiting progressive dural thickening with de novo appearance of high signal intensity area in the right frontal lobe (B), which was depicted as an iso-intensity area on the diffusion-weighted imaging (C).
At the time of presentation to our institution, the patient exhibited no neurologic deficits, and her vital signs were all within normal limits. Her past medical history included a 30-year history of RA, hypertension, and recently diagnosed Sjögren syndrome confirmed by salivary gland scintigraphy. The patient's medications included oral MTX, 8 mg/week and supplemental folate, 5 mg/week. Her MTX dose had been increased from 6 mg/week 1 month prior to the presentation. Laboratory tests revealed normal white blood cell and platelet counts. Hemoglobin level was at the lower limit of normal (11.6 g/dL). Serum C-reactive protein was 0.27 mg/dL, and serum tumor marker levels, including neuron-specific enolase, cytokeratin 19 fragment, pro-gastrin-releasing peptide, carcinoembryonic antigen, and sialylated Lex-I antigen, were within normal ranges, whereas serum level of carbohydrate antigen 19-9 was slightly elevated. Serological tests for hepatitis B surface antigen, hepatitis C antibody, syphilis, and Treponema pallidum were all negative. Additional tests for interferon-γ, aspergillus-specific immunoglobulin G (IgG), cryptococcal antibody, and serum antibodies against Mycobacterium avium complex were also negative.
Brain contrast-enhanced T1-weighted MR imaging showed marked dural thickening in the bilateral frontal dura (Figure 2A and B). Fluorodeoxyglucose positron emission tomography demonstrated increased uptake in the right lung (Figure 2C), but no increased uptake either in the brain (Figure 2D) or in the other organs was noted. Two weeks after the presentation to our institution, the patient underwent bronchoscopic lung biopsy. Histopathological examination of the lung specimen revealed lymphoid cell proliferation without atypical cells. Although atypical epithelioid cells had been present, the finding fell short of establishing the diagnosis of any malignancies.
Brain contrast-enhanced T1-weighted MR imaging showing marked dural thickening in the bilateral frontal dura mater. A: axial image; B: sagittal image. Note that the high signal intensity area in the right frontal lobe observed in the FLAIR imaging (Figure 1B) had no contrast enhancement. Fluorodeoxyglucose positron emission tomography demonstrating increased uptake in the right lung (C, white arrowhead) without increased uptake in the brain (D).
FLAIR: fluid attenuated inversion recovery; MR: magnetic resonance
Regarding the dural lesions, either hypertrophic pachymeningitis related to an autoimmune disease or intradural meningioma had initially been listed as differential diagnosis. Additional serological tests revealed elevated expression of speckled anti-nuclear antibody (1:160) and elevated serum levels of rheumatoid factor (198.3 IU/mL; normal <15 IU/mL). Serum levels of soluble interleukin-2 receptor (1,420 U/mL; normal <121 U/mL) and β-2 microglobulin (2.7 mg/L; normal <1.9 mg/L) were also substantially elevated, suggesting the possibility of lymphoma. Other laboratory data related to autoimmune diseases were within normal limits, including serum perinuclear anti-neutrophil cytoplasmic antibody and immunoglobulin G4 levels. To establish the definitive diagnosis, a dural biopsy was contemplated. She agreed to our treatment proposal and was admitted for the biopsy.
Under general anesthesia, the patient was placed in the supine position. A skin incision was made over the right forehead along a skin crease and a small craniotomy was performed in the frontal bone with an assistance of intraoperative navigation system. The thickened dura mater was exposed, a small portion of which was resected and submitted to a histopathological laboratory. The thickened dura was soft and bled easily during dissection and resection, which was inconsistent with the typical manifestation of hypertrophic pachymeningitis. Her postoperative course was uneventful, and she was discharged 2 days after the biopsy.
The pathohistological examination with hematoxylin-eosin staining disclosed marked proliferation of small lymphocytes (Figure 3A). Immunohistochemistry showed the dominance of B-cells: the cells were positive for CD20 (Figure 3B), but were negative for Bcl6, CD5, and CD10. The cells are positive for the λ immunoglobulin light chain (Figure 3C), but lack κ light chain expression (Figure 3D). Those findings, in conjunction, pointed to the diagnosis of mucosa-associated lymphoid tissue (MALT) lymphoma.12) Considering the long-term history of MTX therapy, the diagnosis of MTX-LPD was established and MTX was discontinued. One month later, exacerbation of RA was noted, and salazosulfapyridine was initiated. Follow-up contrast-enhanced MR imaging 6 months after the MTX cessation demonstrated significant regression of dural thickening (Figure 4A and B). The FLAIR high-signal intensity area in the right frontal lobe, which had been observed in the previous imaging study (Figure 1B), had disappeared (Figure 4C). The pulmonary lesions also substantially subsided on the CT scans. Over a 12-month follow-up, she remained asymptomatic and no recurrence of either the brain or pulmonary lesions was observed.
A pathohistological slide with hematoxylin-eosin staining showing marked proliferation of small lymphocytes (A). Immunohistochemistry staining showing the dominance of B-cells that were positive for CD20 (B). The cells were positive for the λ immunoglobulin light chain (C), but lacked κ light chain expression (D).
Brain contrast-enhanced T1-weighted MR imaging obtained 6 months after cessation of methotrexate showing marked regression of the dural thickening. A: axial image; B: sagittal image. The FLAIR high-signal intensity area in the right frontal lobe which had been observed in the previous MR imaging (Figure 1B) had disappeared (C).
FLAIR: fluid attenuated inversion recovery; MR: magnetic resonance
This case exhibited dural and pulmonary lesions concomitantly, both of which regressed promptly after MTX discontinuation. Although the histopathological examination of the lung sample did not reveal malignant cells, clinical course suggests that both lesions might have shared the same etiology, i.e., MTX-LPD. The progression of dural thickening observed on serial MR imaging coincided with the increase in MTX dose. Several authors have indicated that a higher dose of MTX is more likely to elicit MTX-LPD in patients with RA,13) and the recent increase in the MTX dosage might have contributed to the progression of MTX-LPD. Honda et al.14) reported that an MTX dose exceeding 8 mg/week was a significant risk factor for the development of MTX-LPD. In terms of the duration of MTX treatment, no definite association with MTX-LPD has been reported. Tanaka et al.15) reported that the duration from MTX initiation to LPD onset ranged widely, from 11.0 to 149.5 months (median: 74.8 months). Recently, a large epidemiological study in Japan found that 0.69% of patients with RA using MTX developed LPD during a 3-year observation period.16) MTX-LPD generally regresses spontaneously following cessation of MTX,10,11,17) although some life-threatening cases of MTX-LPD have been reported.18) Therefore, early diagnosis of MTX-LPD and prompt discontinuation of MTX are critical.
Although MTX-LPD often involves extranodal lesions,17) MTX-LPD with CNS involvement has rarely been reported. A recent systematic review reported by In et al.11) identified only 13 cases of CNS MTX-LPD. Most cases of CNS MTX-LPD in the literature manifested as solitary or multiple intraparenchymal lesions in the cerebrum or medulla. CNS MTX-LPD with dural involvement is particularly uncommon and may be even more difficult to diagnose. Our literature search retrieved only four cases of CNS MTX-LPD involving the intracranial or spinal dura mater (Table 1).19-21) All cases had received MTX at doses exceeding 8 mg/week, which may be regarded as a risk factor for MTX-LPD.14,16) In the other three cases, elevated anti–Epstein-Barr virus (EBV) IgG titers had been documented, suggesting a possible association with prior EBV infection (Table 1). In this case, either hypertrophic pachymeningitis or intradural meningioma had initially been suspected based on MR imaging findings. Therefore, EBV serological testing had not been performed in this case, despite its known association with MTX-LPD.7-10)
Summary of Reported Cases of MTX-Associated Lymphoproliferative Disorder Involving the Dura Mater
| No. (Ref) | Authors | Age/Sex | MTX dose (mg/week) | EBV infection | Symptoms | Lesion site | Other lesions | Histopathology | Response after MTX cessation |
|---|---|---|---|---|---|---|---|---|---|
| EBV: Epstein-Barr virus; F: female; IVLBCL: intravascular large B-cell lymphoma; M: male; MALT: mucosa-associated lymphoid tissue; MTX: methotrexate; Tx: thoracic | |||||||||
| 1 (19) | Tsukamoto et al. | 69/M | 10 | Yes | Bilateral leg weakness | Tx spine | No | Hodgkin’s lymphoma | Good |
| 2 (20) | Kamio et al. | 54/F | 10 | Yes | Bilateral leg numbness | Tx spine | No | Hodgkin’s lymphoma | Good |
| 3 (21) | Kikuchi et al. | 50/F | 8 | Yes | None | Cranium | Skin | IVLBCL | Good |
| 4 | The present case | 77/F | 8 | Unknown | None | Cranium | Lung | MALT lymphoma | Good |
Although histopathological findings in the aforementioned four cases had been heterogeneous, clinical outcomes were generally favorable. Notably, in the two cases with intracranial dural involvement, neurological manifestations were either absent or were nonspecific. This observation highlights the diagnostic difficulty of CNS MTX-LPD confined to the intracranial dura mater. Both cases exhibited dural thickening in the frontal convexity; however, this case was diagnosed as MTX-induced MALT lymphoma, with different pathological finding from the other case that had been diagnosed as intravascular large B-cell lymphoma. The diagnosis of intravascular large B-cell lymphoma had been established by skin biopsy in that case,21) indicating that histopathological diagnosis of the dural lesion had not been established. To the best of our knowledge, this case is the first MTX-LPD case with intracranial dural involvement confirmed histopathologically by biopsy. Though intraparenchymal lesions are more common in CNS MTX-LPD, it remains unclear why the lesion in this case had been confined to the dura mater. Multiple authors reported cases of MALT lymphoma arising from the dura.12,22-24) Considering that the dura had also been the origin of MALT lymphoma in this case, the dural cells may have some yet unexplained tendency to develop into MALT lymphoma.
The follow-up MR imaging obtained approximately a year after the initial MR imaging demonstrated de novo appearance of a high signal intensity area on FLAIR sequence in the right frontal lobe (Figure 1B). The lesion did not exhibit restricted diffusion on DWI (Figure 1C) and showed no contrast enhancement (Figure 2A). These imaging findings were not suggestive of malignant lymphoma. Although the exact etiology of this intraparenchymal lesion remains unclear, partial obstruction of the superior sagittal sinus by the thickened dura and subsequent venous congestion might have caused the reversible high-intensity lesion in the FLAIR imaging.
In conclusion, MTX-LPD may rarely manifest itself with intracranial dural thickening. Imaging studies alone may not be sufficient for the diagnosis, and biopsy is a reliable diagnostic method to be considered when a CNS lesion develops in patients receiving MTX.
All authors have no conflict of interest.
The consent from all the participants (the patient and her family) were obtained in written and oral forms.
The authors certify that they did not make any use of AI-assisted tools or technologies in the writing of a manuscript, in the production of images or graphical elements of the paper, or in the collection and analysis of data.
