Abstract

This article is comment on: Vakrakou et al. Clinico-radiologic features and therapeutic strategies in tumefactive demyelination: a retrospective analysis of 50 consecutive cases. Ther Adv Neurol Disord 2021; 14: https://journals.sagepub.com/doi/10.1177/17562864211006503.
We read with interest the recent article by Vakrakou et al. 1 about a retrospective analysis of clinico-radiological characteristics of patients with tumefactive demyelinating lesions (TDLs) and the seven-group stratification proposal. For group A, consistent with Marburg-like TDL, the authors proposed cyclophosphamide as a valid therapeutic option and reopened the debate about the treatment of aggressive demyelinating disorders. We would like to support induction therapy strategies in highly active disease at onset. In July 2019, we observed a case of aggressive form of Marburg-like multiple sclerosis in a 28-year-old woman, who woke up with a right-sided weakness and speech difficulties 10 days after delivery. Magnetic resonance imaging (MRI) showed a 3.5-cm lesion of the left corona radiata suggestive of TDL (Figure 1(a)–(e)) with a nodular contrast enhancement (CE) pattern. There were high values of diffusion restriction on diffusion-weighted imaging (DWI) sequence with a contextual hypointense apparent diffusion coefficient (ADC) map. Neither signs of oedema nor mass effect on adjacent structures were detected. A few small T2 and fluid-attenuated inversion recovery (FLAIR) hyperintense areas of altered signal of the contralateral white matter, the left temporal lobe and the left cerebellar hemisphere were also documented. There were no pathological signal alterations of the spinal cord or the brain stem. Cerebrospinal fluid (CSF) examination showed mild elevated proteins (45 mg/dl) and leucocytes (16 cells/mm3). Extensive chemical, microbiological and cytological CSF and serum examination ruled out rheumatological or infectious diseases. Oligoclonal bands (OCB), anti-MOG and anti-AQP4 antibodies were negative.

Brain magnetic resonance imaging (MRI). (a–e) MRI at symptom onset: (a) axial T1-weighted image showing an area of mild myelin rarefaction (hypointense signal) in the left corona radiata (orange arrowhead), (b) T1-weighted image after administration of gadolinium depicting a nodular contrast enhancement pattern, (c) axial fluid-attenuated inversion recovery (FLAIR) image illustrating the left tumefactive demyelinating lesion (TDL) and a small hyperintense area of the contralateral white matter, (d) TDL exhibiting restricted diffusion at diffusion-weighted imaging (DWI) and (e) TDL was markedly hypointense on apparent diffusion coefficient map. (f–i) Follow-up MRI (1 week after symptoms onset): (f) axial FLAIR image showing extension of TDL with areas of demyelination also involving the right hemisphere, (g) axial DWI image, (h) sagittal T2-weighted image and (i) axial T1-weighted image showing a linear-marginal contrast enhancement pattern. (j) Follow-up MRI (4 weeks after symptom onset): coronal FLAIR image depicting massive bilateral involvement of subcortical structures. (k) Summary of disease course.
Within 24 h from symptom onset, the patient developed a right-sided hemiplegia, global aphasia and left-sided hemineglect. Expanded Disability Status Scale (EDSS) score was 9. A computed tomography performed the day after evidenced that the initial left hemispheric TDL quickly extended to form large areas of demyelination, showing also initial signs of contralateral hypodensity configuring a highly active disease. Treatment with methylprednisolone (1000 mg/daily) for 7 days was undertaken, without any clinical improvement. At follow-up MRI (Figure 1(f)–(i)), there was a massive extension of the initial lesion to the periventricular and subcortical white matter without cortex invasion, with modification of the CE pattern (from intralesional to linear-marginal). She underwent 5 days of plasma exchange (1 apheresis/daily), but her neurological status further deteriorated. Seventeen days after disease onset, treatment with intravenous cyclophosphamide (1 g) was started. The patient showed minimal improvement regarding language comprehension and underwent an intensive hospital neurorehabilitation treatment. Three months later, she presented a mild residual right-sided hemiparesis and a minimal deficit of word retrieval. The patient continued cyclophosphamide (1 g every 4–5 weeks) for a total dose of 12 g. After 1 year, no new neurological symptoms occurred, but follow-up MRI showed two new T2-hyperintense lesions with CE. Cyclophosphamide infusions were discontinued and a steroid pulse therapy was performed. The patient was switched to rituximab with a dosing schedule of 1 g at days 1 and 15, with clinical and radiological control of disease. Another cycle of rituximab was performed 6 months later. After 24 months from disease onset, the patient had mild right hemiparesis (EDSS score 2) and showed no evidence of disease activity on brain MRI.
Atypical demyelinating syndromes 2 are rare and difficult to diagnose as they might mimic a cerebral vascular event or a tumour at onset. We would like to point out some cardinal aspects that helped us to face this aggressive form of inflammatory demyelinating disorder. The extremely rapid clinical deterioration with the accumulation of maximal disability within 24 h from symptom onset configures a multifocal cognitive syndrome rather than an altered state of consciousness. This aspect might be a clue for the differential diagnosis with acute disseminated encephalomyelitis (sudden onset of cortical syndromes versus diffused encephalopathy). The first symptomatic lesion of our patient differs from classic TDL, which usually presents with oedema and mass effect, T2-hypintense rim in the same area of ring enhancement, peripheral hypointensity on DWI sequences, frequent presence of OCB, response to steroids or plasma exchange. 3 Atypical characteristics in our case also included the nodular pattern of CE and the infiltrative lesion morphology, which have already been linked to a poorer clinical outcome by Wallner-Blazek et al. 4 DWI/ADC sequences depicting remarkable cytotoxic oedema, which involved in our case the whole lesion, might be also considered prognostically unfavourable.
Marburg disease typically does not respond to therapies usually effective for multiple sclerosis so that recognition of this syndrome becomes crucial to establish a prompt treatment.
Neuropathological findings of Marburg multiple sclerosis encompass large plaques with extensive areas of myelinolysis with macrophagic infiltrates, B- and T-lymphocyte infiltrates, necrosis and variable axonal loss.5,6 Notably, our case did not respond to methylprednisolone and plasma exchange in line with the patient described by Suzuki et al. 6 Cyclophosphamide has the peculiarity to act both on cell-mediated and humoral immune response through its effects on T (naïve) and B lymphocytes. 7 Furthermore, it has been proven to enter the central nervous system and to effectively penetrate the blood–brain barrier and the brain parenchyma. Side effects include haemorrhagic cystitis (due to acrolein toxicity to the bladder epithelium), increased bladder cancer risk and gonadal toxicity. Cyclophosphamide is therefore an effective drug to ablate the immune system, especially the autoreactive cells, leaving the hematopoietic precursors intact. Of course, further studies with larger populations on long-term efficacy and sequelae are needed.
There is emerging evidence which favours induction strategies over escalation therapies in case of highly aggressive inflammatory demyelinating disorders to obtain rapid disease control and to increase the neurological reserve. The principle of induction therapies is to perform a pharmacological ablation of the bone marrow resetting the immune system.
By facing a non-typical TDL with highly active disease, which does not fully respond to steroid therapy and plasma exchange, neurologists should rapidly consider the use of intravenous therapy with cyclophosphamide.8–10 In case of disease relapse or drug toxicity under cyclophosphamide therapy, anti-CD20 monoclonal antibodies might be a valid therapeutic option.
Footnotes
Acknowledgements
We thank Professor Alex Rovira, Department of Radiology, Vall d’Hebron University Hospital, Barcelona, for the support in the neuroimaging diagnosis.
Author contributions
T. Nicoletti and A. Bianco contributed equally to this study.
Conflict of interest statement
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
Ethics statement
The study was carried out in compliance with the Helsinki Declaration and with the guidelines of the Ethical Committee of our Institution. The patient has given her written informed consent for publication of her clinical data and medical images.
