Lack of Response to Pulse Cyclophosphamide in Neuromyelitis Optica: Evaluation of 7 Patients

The effect of immunosuppression with corticosteroids, azathioprine, mycophenolate, or rituximab for preventing relapses in patients with neuromyelitis optica (NMO) has been demonstrated in case series and retrospective studies.^1-4 These drugs are currently considered the mainstay treatment for preventing neurologic worsening in NMO.⁵ Herein, we describe our experience with pulse cyclophosphamide for treating patients with NMO, which was used prior to azathioprine and had its use interrupted owing to lack of efficacy.

We reviewed our previously reported series of patients with relapsing NMO followed up from February 1994 to August 2007 at the Federal University of São Paulo for those treated with intravenous cyclophosphamide for a thorough analysis of this drug's effect in preventing relapses and disability accumulation in patients with NMO. Inclusion criteria and data analysis are identical to those previously reported.¹

We identified 7 patients who received cyclophosphamide in pulse doses of 1 g associated with methylprednisolone, 1 g every 2 months (500-700 mg/m², according to each patient's body surface area) as a first-line treatment, within a mean of 17 months from the first demyelinating event (optic neuritis, myelitis, or both) (Table). During cyclophosphamide therapy, 5 patients continued relapsing and/or worsening, 1 patient died owing to a severe NMO relapse, and 1 abandoned follow-up at our center (Table); only 1 patient remained clinically stable. The remaining 5 patients on follow-up were switched to azathioprine associated with prednisone after clinical judgment of cyclophosphamide inefficiency. After treatment modification, there was a decrease in annualized relapse rate and/or progression index, with some patients showing improvement in their Expanded Disability Status Scale scores (Table). Patient 1 further showed an increase in annualized relapse rate and received rescue therapy with intravenous immunoglobulin every 2 months.

Neuromyelitis optica is an autoimmune demyelinating disease of the central nervous system characterized by severe relapses of optic neuritis and myelitis, which are the main factors associated with neurologic disability.⁶ Therefore, preventing relapses should be the main goal of treatments prescribed for patients with NMO.

In this series of Brazilian patients with relapsing NMO followed up at Federal University of São Paulo, treatment with cyclophosphamide was not able to halt relapses or neurologic disability progression, even being an immunosuppressant with similar mechanism of action to azathioprine (both broadly block DNA and RNA synthesis). Although some relapses were also observed in patients after switching to azathioprine, they were less severe, suggesting that less nervous tissue damage might have occurred, thus allowing better neurologic recovery. The association of NMO-IgG positivity and treatment response could not be established for these patients as this biomarker was only made available in Brazil after 2007 and the relationship between this antibody titer, disease severity, and treatment response is still under research.⁵

The reason for cyclophosphamide treatment failure could not be identified solely on clinical information, and we speculate that it might have occurred owing to the doses we used or dosing interval—even though 5 patients presented with a lymphocyte count below 1.000/μL (to convert to ×10⁹ per liter, multiply by 0.001) while taking the therapy—or even be related to a specific drug mechanism of action, which would be better understood in an experimental setting. Nevertheless, considering the known response of azathioprine,¹ mycophenolate,³ and rituximab,² the results presented here should discourage the use of cyclophosphamide in pulse doses for the treatment of NMO.

Correspondence: Denis B. Bichuetti, MD, PhD, Disciplina de Neurologia, Universidade Federal de São Paulo, Rua Botucatu, 740, São Paulo, SP, Brazil, 04023-900 (denisbichuetti@globo.com).

Author Contributions: Dr Bichuetti had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Bichuetti and Oliveira. Acquisition of data: Bichuetti and de Castro Boulos. Analysis and interpretation of data: Bichuetti, de Castro Boulos, and Gabbai. Drafting of the manuscript: Bichuetti and de Castro Boulos. Critical revision of the manuscript for important intellectual content: Oliveira and Gabbai. Statistical analysis: Bichuetti. Administrative, technical, and material support: de Castro Boulos. Study supervision: Oliveira and Gabbai.

Funding/Support: This study was approved by the internal review board of the Federal University of São Paulo.