Abstract
Keywords
Introduction
Aggressive multiple sclerosis (MS) causes early, rapid accumulation of disability with frequent and severe relapses, incomplete recovery from relapses, and/or high T2 lesion burden.1,2 While high-efficacy disease-modifying therapies (DMTs) reduce risk of relapse, acute treatment options for patients with a severe or catastrophic course remain limited. Severe relapses are typically treated with high-dose IV corticosteroids followed by plasmapheresis.3,4 Patients who do not respond to initial treatment lack universally accepted second-line options. Several alternatives studied in the past, including IVIG 5 and natalizumab, 6 have failed to demonstrate efficacy.
Cyclophosphamide is a DNA alkylating and chemotherapeutic agent used in clinical practice for severe MS relapses refractory to initial therapy. Despite its clinical use, there is no conclusive data demonstrating its efficacy. 7 Blinded randomized controlled studies investigating the efficacy of cyclophosphamide were conducted before MRI was routinely used in clinical care and showed inconsistent results.8,9 This study sought to describe the UBC MS clinic experience of using cyclophosphamide as a rescue therapy in MS patients during a treatment epoch where high-efficacy therapy and MRI were both available.
Methods
This study was approved by the University of British Columbia (UBC) research ethics board (#H24-02207) and the Vancouver Coastal Health Research Institute (#V24-02207).
Patients
We reviewed the UBC MS Clinic pharmacy database to identify patients followed at the clinic that received cyclophosphamide between June 2009 and June 2024. Patients meeting 2017 McDonald diagnostic criteria for MS were included. Most patients in this series were treated with cyclophosphamide due to insufficient recovery from a relapse following initial treatment with 1 g methylprednisolone IV daily for 3 to 5 days and/or plasmapheresis. Four patients received cyclophosphamide for progression independent of relapse activity. Patients with insufficient clinical information to determine extent of disability before and after treatment were excluded.
Data collection
Clinical and radiologic data were collected from available medical records. Neuroimaging reports and images were reviewed to assess for the development of new T2 hyperintense lesions and gadolinium enhancing lesions prior to and after cyclophosphamide treatment.
Treatment protocol
Patients were treated with pulse monthly therapy of cyclophosphamide. Doses given were 600 to 1000 mg/m2 for a duration of 1 to 10 months based on tolerability and perceived efficacy.
Clinical response
Clinical response was assessed based on the Expanded Disability Status Scale (EDSS) 10 scores at nadir preceding cyclophosphamide treatment and at last follow-up within 12 months of completing cyclophosphamide treatment. Brain and spinal cord MRIs obtained prior to cyclophosphamide treatment and up to 12 months after completion of cyclophosphamide treatment were reviewed for identification of new T2 hyperintense lesions and gadolinium enhancing lesions.
Results
Patient selection
Fifty-one patients who received cyclophosphamide while under the care of a treating neurologist at the UBC MS clinic were identified (Figure 1). Twenty-two patients were excluded for receiving cyclophosphamide for an indication other than MS. Six patients were excluded due to a lack of clinical information. The data of 23 patients were analyzed.
Patient selection flowchart.
Patient characteristics
Patient characteristics and response to cyclophosphamide were recorded (Table 1). The median age at initiation of cyclophosphamide treatment was 39 years (range 25–66) with a median EDSS score prior to cyclophosphamide therapy of 6 (range 2.5–9). The median disease duration was 0 years (range 0–28) with most patients having an MS severity score in the 10th decile. Fifteen out of 23 patients had a purely relapsing course at the time of cyclophosphamide initiation while 8/23 patients had a progressive or progressive relapsing course. Two patients had rebound disease activity, in one case (no. 5) following discontinuation of fingolimod in an attempted transition to natalizumab, and in the other (no. 8) due to a delayed natalizumab dose following influenza. Seven out of 23 patients were on maintenance DMT prior to cyclophosphamide, with most patients off therapy because the index event led to the initial diagnosis of MS. The localization of disability was spinal cord lesions in 16/23, infratentorial lesions in 8/23, and optic nerve lesions in 3/23. Patients received a median of 5 cyclophosphamide courses (range 1–10). Of the patients who had a relapse prior to cyclophosphamide initiation, 15/18 had initiated cyclophosphamide within 3 months of the relapse (range 1–5). The mean interval between completion of plasma exchange and cyclophosphamide initiation was 1.3 months (range 0–4).
Patient characteristics and response to cyclophosphamide.
Cyclophosphamide efficacy
Thirteen out of 23 patients had an improvement in EDSS, including 6/23 with a ≥3 EDSS numerical improvement (Figure 2). All 6 patients with ≥3 EDSS improvement had a solely relapsing course. Mean EDSS improved from 5.7 to 4.3, and median EDSS improved from 6 (range 2.5–9) to 4 (range 2–8). Zero out of 23 patients had relapses within 1 year following completion of cyclophosphamide treatment. Two out of 23 patients had new T2 lesions within 1 year following completion of cyclophosphamide treatment. Both patients had highly active disease with substantial improvement in EDSS following cyclophosphamide (nos. 7 and 22). Of the 8 patients with a progressive or progressive relapsing course, 1/8 had improvement in EDSS, and 4/8 had ongoing progression at follow-up within 1 year of completing cyclophosphamide treatment.
EDSS score prior to initiation of cyclophosphamide and 1 year after cyclophosphamide.
Cyclophosphamide safety
Adverse events other than leukopenia and nausea were recorded. One case each of sinus infection, dental infection, and epididymitis were reported, all of which resolved with antibiotic treatment. No serious adverse events were reported.
Illustrative cases
Patient no. 7 is a previously healthy 34-year-old man that presented with blurred vision in his right eye. He received 1 g methylprednisolone IV daily for 3 days for treatment of optic neuritis. Two weeks later his vision began to deteriorate in his left eye, and he developed lower extremity sensory changes and weakness that quickly progressed to paraplegia and urinary retention requiring catheterization with an EDSS of 9.0. MRI brain and cervical and thoracic spine showed numerous T2 hyperintense lesions affecting the bilateral optic nerves, within the brain in juxtacortical, periventricular, and infratentorial locations, and within the cervical and thoracic cord in a morphology and distribution consistent with MS (Figure 3). Two cerebral and one cervical cord lesion demonstrated gadolinium enhancement. Lumbar puncture demonstrated 15 × 106/L leukocytes with 93% lymphocytes, a protein of 1.12 g/L, and CSF-specific oligoclonal banding. He received an additional 1 g methylprednisolone IV daily for 3 days followed by 5 courses of plasmapheresis without improvement. He was then started on cyclophosphamide. One week later, he developed new symptoms in his arms with right greater than left sensorimotor dysfunction. He was treated with 1 g methylprednisolone IV daily for 5 days and continued on cyclophosphamide. Two weeks after his second course of cyclophosphamide, he began to ambulate unassisted without a gait aid. He also developed epididymitis and recovered after a 2-week course of antibiotics. Serum NMO-IgG and MOG-IgG were negative. Repeat MRI imaging at 3 and 6 months showed new cerebral and spinal cord T2 hyperintense lesions, including four lesions with gadolinium enhancement. He completed 6 courses of cyclophosphamide before transitioning to rituximab for maintenance therapy. He developed new T2 hyperintense lesions 3 months after transitioning to rituximab. Despite ongoing radiologic activity, he continued to improve clinically, with an EDSS of 3.0 6 months after transitioning to rituximab. Repeat imaging 9 months after transitioning to rituximab was stable. He was followed for 2 years thereafter with clinical and radiologic stability on rituximab.
Representative MRI images of patient no 7. (a) Brain sagittal T2 flair demonstrating T2 hyperintense lesions in the upper pons and at C1 in the cervical cord. (b) Thoracic spine sagittal T2 demonstrating short segment T2 hyperintensities at T4, T7, and T12.
Patient no. 19 is a 26-year-old woman with a history of type 1 diabetes that presented with blurred vision in her left eye. She initially described no vision inferiorly and diminished vision superiorly in the left eye that progressed to no motion perception over 2 days. Seven days after initial symptoms she was treated with 1 g methylprednisolone IV daily for 5 days. Two days after steroid initiation her vision deteriorated in her right eye. At nadir she had light perception in her left eye and motion perception in her right eye for an EDSS of 4.0. MRI brain and orbits with gadolinium showed increased T2 signal and gadolinium enhancement in the left greater than right prechiasmatic optic nerves, the corpus callosum, and in juxtacortical locations (Figure 4). Lumbar puncture showed 32 × 106/L leukocytes with 99% lymphocytes, a protein of 0.32 g/L, and CSF-specific oligoclonal banding. She was treated with five courses of plasmapheresis with no improvement. She was started on cyclophosphamide 3 weeks after symptom onset and began to notice visual recovery following her infusion. At 1 month follow-up after her first course of cyclophosphamide, visual acuity had improved to 20/100 OD and counting fingers OS. Serum NMO-IgG was negative and serum MOG-IgG was indeterminate. At 5-month follow-up after three total courses of cyclophosphamide she had substantial visual recovery with a visual acuity of 20/40 OD and 20/120 OS for an EDSS of 3.0. She was transitioned to rituximab for maintenance disease-modifying therapy.
Representative MRI images of patient no 19. (a) Brain sagittal T2 flair demonstrating two juxtacortical T2 hyperintense lesions. (b) Brain and orbit post-gadolinium T1 demonstrating left greater than right prechiasmatic optic nerve enhancement.
Discussion
This single center case series suggests that cyclophosphamide can be effective and relatively safe as a rescue therapy for severe MS relapses refractory to high-dose corticosteroids and/or plasmapheresis. Our study was limited by the retrospective design, the absence of a control group, and a small, heterogenous population.
In our cohort, cyclophosphamide was used in both relapsing and progressive patients. Six patients treated with cyclophosphamide exhibited substantial recovery with an improvement of ≥3 in EDSS. All six patients had a relapsing course and received cyclophosphamide within three months of the relapse causing disability. In eight patients with a progressive or progressive relapsing course, cyclophosphamide was used due to a perceived acceleration in the rate of decline as determined by the treating clinician. One patient had an EDSS improvement of 0.5, while the others had a stable EDSS. Four of eight patients had stabilization in their progression at last follow-up within 1 year following cyclophosphamide treatment. This is in keeping with prior studies suggesting greater efficacy of cyclophosphamide in treating MS relapses with limited efficacy in patients with a predominantly progressive course.7–9
Common side effects associated with cyclophosphamide include nausea, alopecia, infertility, infection, bladder toxicity such as hemorrhagic cystitis, and risk of secondary malignancy, predominantly in patients exposed to higher cumulative doses of cyclophosphamide. 7 Cyclophosphamide was well tolerated in our cohort. There were no serious adverse events. Three out of 23 patients had infections that resolved following antibiotic treatment without further complication. Patients were followed with annual urine cytology with no cases of bladder cancer detected.
Cyclophosphamide is a small molecule DNA alkylating agent that depletes both B and T cells. 11 In contrast to B-cell depleting monoclonal antibodies, it achieves superior penetration into the central nervous system and can be administered concurrently with plasmapheresis, properties that may enable more effective suppression of ongoing central nervous system inflammation. In addition, cyclophosphamide induces B and T cell depletion within 1 week, 11 and its rapid onset of action may promote greater recovery compared to MS immunotherapies that require more time to exert their full immunosuppressive effect. Furthermore, relative to relapse therapies such as high-dose methylprednisolone and plasmapheresis, its broader and more potent immunomodulatory effects may allow it to be effective in cases refractory to these established therapies.
Conclusion
In patients with aggressive MS who demonstrate an inadequate response to steroids and/or plasmapheresis, treatment with cyclophosphamide was associated with improvement in EDSS amongst those with a relapsing course. Despite the availability of a broad range of high-efficacy DMTs for relapse prevention, comparable advances in the treatment of acute relapses have not been realized. This gap represents a substantial unmet clinical need, and one that cyclophosphamide may be uniquely positioned to address, particularly for patients with aggressive forms of MS. For individuals who remain refractory to steroids and plasmapheresis, current practice typically entails empiric use of off-label therapies such as cyclophosphamide or the rapid initiation of a maintenance DMT, most commonly a B-cell depleting monoclonal antibody. In the setting of catastrophic MS attacks, cyclophosphamide may be an indispensable treatment option for patients who do not recover following corticosteroids and plasma exchange. A randomized controlled trial directly comparing cyclophosphamide with early initiation of B-cell–depleting monoclonal antibodies could inform optimal therapeutic strategies for the acute management of severe MS relapses.
Footnotes
Acknowledgments
This research was made possible through data collected by the UBC MS clinic.
Author contributions
VB contributed to study design, data analysis, and preparing the manuscript. AS and RC contributed to study design and manuscript revision. All authors contributed to maintaining records of cyclophosphamide use and clinical outcomes, as well as review and approval of the final manuscript.
Declaration of conflicting interests
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.
