Article Text
PDF
Abstract
Background Patients with refractory or high-risk myasthenia gravis (MG) respond poorly to conventional immunosuppressive therapy, requiring rescue therapies and often experiencing treatment toxicity. Rescue and injectable therapies do not induce remission and require repetitive administration leading to significant constraints on patients and the healthcare system. This long-term follow-up study demonstrates cyclophosphamide as a rapidly effective and safe treatment in patients with refractory or high-risk MG.
Methods Retrospective cohort study of MG patients treated with cyclophosphamide between January 2000 and June 2022 conducted at a quaternary neuroimmunology clinic in New South Wales, Australia.
Results 31 patients were treated: mean age of 64 years; median follow-up 3.6 years (5 months to 11 years); 94% seropositive to acetylcholine receptor (AChR) antibodies and 45% had thymoma. A reduced intensity cyclophosphamide induction protocol followed by oral antiproliferative maintenance is described.
Median myasthenia gravis composite scores reduced by >50% after the third cycle of cyclophosphamide. Complete cessation of prednisolone was possible in 11 patients while 20 remained on prednisolone with a median daily dose of 5 mg. Plasma exchange was ceased in 62% of patients and intravenous immunoglobulin ceased in 55%. Cyclophosphamide was generally well tolerated with mild cytopenias. There were no malignancies or cases of haemorrhagic cystitis.
Conclusion We describe a large cohort of high-risk MG patients treated with cyclophosphamide in a retrospective single-clinic cohort. We suggest cyclophosphamide should be considered for rapid remission induction, corticosteroid reduction and long-term freedom from recurrent injectable therapies in selected patients, typically those with AChR antibodies.
- MYASTHENIA
- NEUROIMMUNOLOGY
- IMMUNOLOGY
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
The existing literature on cyclophosphamide (CYC) in myasthenia gravis (MG) is limited as it is often overlooked due to concerns of toxicities and typically involves younger patients.
WHAT THIS STUDY ADDS
Reduced intensity CYC protocol is well tolerated in older patients without compromising efficacy, leading to steroid and rescue therapy reduction.
This is applicable even in patients with prior thymoma, typically a difficult-to-treat cohort with a low remission rate and often excluded from clinical trials.
Data on the long-term safety profile of CYC in refractory and high-risk MG.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The findings are applicable to both cost-restrained and high-income healthcare systems as CYC treatment offers genuine remission induction potential for patients with more severe acetylcholine receptor-antibody-associated MG.
Introduction
Myasthenia gravis (MG) is an autoimmune disorder of the postsynaptic neuromuscular junction characterised by fluctuating and fatigable weakness. MG prevalence has major geographical variations but has increased worldwide over the past seven decades.1 Prompt diagnosis and treatment initiation with conventional immunosuppressive therapies (ISTs) such as azathioprine (AZT), methotrexate (MTX) and mycophenolate mofetil (MMF) have improved prognosis.2 As a result, treatment strategies have evolved from reducing mortality to achieving complete, stable disease remission while minimising therapeutic adverse events.
A proportion of patients, however, experience refractory MG variably defined as failure to respond to conventional IST with corticosteroid and one or two IST. This group may require frequent injectable rescue therapies such as intravenous immunoglobulin (IVIg), plasma exchange (PLEX) or in some settings complement or neonatal Fc receptor (FcRn) inhibitors. These agents are collectively referred to in this paper as damage blocking therapies (DBTs) as they block or remove the effect of pathogenic antibodies without inducing remission, unlike immunosuppressive disease-modifying therapies (DMTs). High-risk MG patients may experience dose-limiting treatment toxicities, especially to corticosteroids, or have comorbid conditions that restrict the use of conventional immunosuppressive agents. Patients may also have severe or high-risk MG typically requiring intensive care unit admissions, often in association with thymoma, with slow or inadequate response to high and/or toxic doses of corticosteroids, in which case, it is known that the steroid-sparing benefit from DMTs such as AZA is likely to take 15 months or more3 and a more rapid control of disease is required. In addition, the physician must also tailor treatment requirements to the patient, incorporating differing treatment priorities and functional goals.
Cyclophosphamide (CYC) is a DNA alkylating agent that interferes in transcription processes and DNA replication. It has been widely used for autoimmune diseases4 and has been demonstrated to be effective in the management of refractory MG.5–8 However, its clinical utility has been limited by concerns about side effects, unfamiliarity with treatment protocols and the lack of a standardised post-treatment regimen leading to a lack of sustained clinical benefit.6 7
This retrospective review aims to evaluate the long-term clinical outcomes of refractory MG patients treated with CYC therapy, followed by DMT and the incidence of any adverse events. Six patients in this study were included from our initial study of induction CYC in refractory MG, two cases were excluded due to inadequate follow-up records.8
We hypothesise that the lack of a standardised post-treatment approach to initiating a maintenance corticosteroid-sparing agent in MG contributes to the lack of sustained clinical benefit from CYC therapy.
Methods
A retrospective cohort study conducted at the neuroimmunology clinic at Concord Hospital, New South Wales, Australia. The clinic serves as a quaternary neuroimmunology referral centre with a focus on MG patients requiring complex therapy.
Study population and clinical variables
We reviewed our clinical database for MG patients treated with CYC between January 2000 and June 2022. Demographic and clinical data were extracted from electronic medical records. Disease severity was documented with the myasthenia gravis composite (MGC) scale, using data collected during routine consultations in this clinical service. MGC was recorded at the following time points, pre-CYC, at 10–14 days after each CYC treatment and during the last clinic follow-up. White cell count, neutrophil and lymphocyte profile were recorded at nadir, defined as 10–14 days after each CYC treatment schedule. Adverse events were classified according to Common Terminology Criteria for Adverse Events (CTCAE), V.5.0 guidelines.9
Statistical analysis
Standard descriptive statistics were used for the presented data. Continuous variables were reported as either mean (SD) or median (range) based on their distribution and presented as both in certain figures. Categorical variables were reported as percentage.
Treatment schedule
All patients underwent immunosuppression screening prior to commencement of CYC and gave informed consent after risks and benefits were discussed.10 Where appropriate, fertility counselling and oocyte/embryo/sperm preservation were performed prior to CYC. Of note, a satisfactory neutrophil count of >1.5×109/L was required in the fortnight prior to each infusion. CYC was infused via peripheral IV access in the outpatient setting unless MG severity necessitated hospitalisation.
CYC was administered every 4 weeks for six cycles, at a dose of 0.75 g/m2 with a maximum dose of 1500 mg per cycle. The dose was reduced to 0.5 g/m2 in renal impairment (<30% of normal glomerular filtration rate) with smaller dose reductions if the neutrophil count at the 10–14 day nadir fell below 1.5×109/L. Prophylactic anti-emetics were concurrently administered with each infusion. Patients received Pneumocystis jirovecii pneumonia prophylaxis with cotrimoxazole during the treatment period and for 1 month after the last CYC cycle.
Maintenance immunotherapy with MMF, AZT or MTX was commenced 4–6 weeks after the final CYC cycle.
Results
Patient and disease characteristics
34 patients with treatment refractory (n=14) or severe generalised MG (n=17) were identified, 3 patients were excluded due to insufficient data. Treatment refractory MG within this cohort was defined as poor disease control despite corticosteroids, disease blocking therapy (IVIg or PLEX) and a previous trial of at least one IST. High-risk patients included were generalised MG patients with poor disease control despite corticosteroids and frequent disease blocking therapy (IVIg or PLEX), complicated by adverse events to corticosteroids and/or at risk of or requiring critical care admission. The diagnosis of MG was confirmed in each patient based on typical clinical symptoms and examination findings, acetylcholine receptor (AChR) and/or muscle-specific kinase (MuSK) antibody profile and/or confirmatory electrophysiological studies. A total of 31 patients were included with 12 females and 19 males (table 1). The mean age was 64 years (range 32–83 years). 29 patients were seropositive for AChR antibodies, 1 was seropositive for both AChR and MuSK antibodies and 1 was seronegative to AChR, MuSK and low-density lipoprotein receptor-related protein 4 (LRP4) antibodies. 14 patients (45.2%) had thymoma; 17 patients (54.8%) underwent thymectomy with their thymic histology summarised in table 1. Various therapies trialled before CYC included AZT (n=6), MMF (n=11), MTX (n=1), rituximab (RTX, n=4) therapy, plus pyridostigmine (n=23). 21 (68%) of patients required maintenance IVIg and 20 (64.5%) required maintenance PLEX. All patients were treated with prednisolone pre-CYC with a median daily dose of 25 mg (range 1.25–75 mg).
Patient and disease characteristics
Median MGC scores pre-CYC treatment were 14/50 (range 4–35, MGC upper limit=50). All 31 patients had poor disease control as indications for CYC therapy with various toxicities from prednisolone therapy including weight gain, osteoporosis, hyperglycaemia and psychiatric concerns. Five patients had comorbid medical issues complicating the use of conventional immunosuppression or prednisolone use. These included two patients with infective complications, two with toxic complications with previous AZT use and one with concomitant tuberculosis and hepatitis B seropositivity (table 1).
One patient had CYC therapy for the combined management of refractory MG and gamma-aminobutyric acid (GABA) subtype A plus α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor double seropositive encephalitis associated with a resected thymoma.
Clinical response
Median MGC scores reduced by >50% after the third cycle of CYC (range 0–18) allowing for a concomitant prednisolone reduction (figure 1) with an overall improvement of pretreatment median MGC scores of 14 to post-treatment median MGC score of 1, assessed at the last follow-up clinic visit. This clinical benefit allowed a reduction in prednisolone, IVIg and PLEX administration. Remarkably, during the last follow-up, complete cessation of prednisolone was possible in 11 patients and 20 patients remained on prednisolone with a median daily dose of 5 mg (range 1.25–15 mg) (table 2). Characteristics of patients with complete cessation of prednisolone and achieving MGC of 0 during the last follow-up are included in online supplemental tables.
Supplemental material
Clinical response measured by (A) MGC and (B) prednisolone dose after each treatment cycle and at last follow-up. MGC, myasthenia gravis composite.
MGC and prednisolone outcomes
There was a notable decrease of 50% reduction in frequency of administration of IVIg. PLEX was ceased in 13/21 (62%) of patients and IVIG ceased in 11/20 patients (55%). Six patients pre-CYC were on sequential PLEX and IVIG combined.
Four (13%) were CYC non-responders with no improvement in MGC post-CYC treatment. One of these four patients with a resected thymoma, on reassessment, was found to be seropositive not only for AChR but also MuSK antibodies and achieved pharmacological remission after rituximab, with subsequent cessation of all rescue therapy. The remaining three patients had AChR antibodies with ongoing subjective symptoms but were difficult on assessment to establish objective weakness. All three were administered ongoing IVIG and/or PLEX.
Treatment tolerability and long-term outcomes
All patients underwent frequent clinical and laboratory monitoring for cytopenia during the 6-month CYC treatment period. Cytopenias were mild with no patients requiring support with blood products or granulocyte colony-stimulating factor (figure 2).
Mean and median values of (A) white cell count, (B) lymphocyte and (C) neutrophil profile at nadir of each treatment cycle. (D) Number of patients with neutrophil values
≤
2 and 
at nadir of each treatment cycle.
There were a total of 12 adverse events involving 9 patients. Most of the adverse events were grade 1 CTCAE with asymptomatic lymphopenia, nausea and renal impairment which required a dose reduction in one or more treatment cycles. There was one grade 2 CTCAE of cellulitis requiring oral antibiotics. There were three grade 3 CTCAE with infections requiring hospitalisation and intravenous antibiotics. Adverse effects are summarised in table 3. CYC cessation occurred in one patient who developed worsening MG during the fourth treatment cycle for thymoma-associated generalised MG and was subsequently found to have seropositivity to both AChR and MuSk antibodies then switched to rituximab as noted previously.
Cumulative adverse event profile
Median follow-up was 3.6 years (5 months to 11 years). There were no malignancies or cases of haemorrhagic cystitis.
One death was recorded 7 years post-CYC infusion with P. jirovecii pneumonia considered to be one contributing factor. This patient had an excellent response to CYC but was a frail elderly nursing home resident with multiple comorbid conditions on concomitant prednisolone therapy. There was no evidence of myelodysplasia or haematological malignancy, and the cause of death was considered unrelated to prior CYC treatment given the long interval.
Discussion
We describe a large series of refractory and or severe MG patients treated with IV CYC. This study is an extension of our initial case series, with extended follow-up of up to 11 years8 and provides class IV evidence for the treatment of high-risk MG patients of which there is a paucity of evidence.
87% of the MG patients in our cohort demonstrated dramatic clinical improvement with IV CYC every 4 weeks for six cycles. This is a remarkable effect within this patient cohort given all patients had moderate-to-severe disease prior to CYC treatment despite various immunosuppressive and immunomodulatory (IVIg/PLEX) agents. The therapeutic response allowed a reduction in the median prednisolone dose by 80%, a clinical benefit that is relevant to minimise the long-term sequelae of corticosteroid use.
There is widespread hesitation about using CYC due to concerns of toxicities. In the acute phase, marrow suppression and consequential life-threatening infections are valid safety concerns as with previous case series using high-dose CYC (200 mg/kg).11 However, our case series reflects the safety of pulse low-dose CYC, with a maximum total treatment dose of 9 g, administered via a peripheral line in an outpatient setting. There were no serious infections requiring cessation of CYC and no patients required premature cessation of CYC therapy or haematological support. The safety profile of similar low-dose regimens is further supported by other studies6 7 with good tolerability.
Fertility can be negatively affected by CYC, but this tends to be of greater concern for younger patients, in whom the corticosteroid offset advantages of CYC may be less relevant. In such patients, therapeutic thymectomy with early-onset MG appears to have a greater benefit.12 13 Second, a standard corticosteroid-sparing agent or tacrolimus±recurrent injectable therapies may be more appropriate and better tolerated in younger patients.14 More generally in refractory MG, it may also be the case that newer DBTs such as the FcRn inhibitors are able to suppress MG sufficiently that longer trials of standard DMTs while minimising corticosteroid toxicity can be embarked on, without resorting to potent therapies such as CYC.
In selected patients, long-term moderate to high-dose corticosteroids may pose a greater risk to health than a course of CYC. This includes patients with thymoma, where high doses are required, older patients or patients with vascular risk factors, and where corticosteroid-specific side effects are present such as osteoporosis with fractures or psychosis. Among 5011 patients with giant cell arteritis with a mean age of 72 years who were treated with corticosteroids, there was an increased risk of diabetes, osteoporosis, fractures, glaucoma, serious infections requiring hospitalisation and even death.15 Our data with 11/31 patients ceasing corticosteroids and a median daily dose of 5 mg in the remainder suggests that CYC treatment averted potentially substantial steroid-related morbidity.
Transient lymphopenia was noted (<0.5×109/L) in eight patients with four persistently lymphopenic at the end of treatment. Lymphopenia has been identified as a risk factor for serious infection in other settings16 but these were not observed in this setting. We note routine use of prophylaxis with trimethoprim/sulfamethoxazole may have mitigated against serious infections in our protocols. The overall white cell count, neutrophil and lymphocyte profile are presented in figure 2.
The rapid onset of clinical improvement (median of 3 treatment cycles) makes CYC a valuable treatment tool in severe disease necessitating rapid clinical responses. This is in comparison to the standard protocol of corticosteroid and conventional ISTs which demonstrate a clinical benefit after 6–15 months of treatment3 17 resulting in prolonged symptoms and disability compounded by high doses of total corticosteroid exposure. Repeated adjuvant DBT use such as IVIg or PLEX in this cohort of patients results in substantial healthcare costs, productivity loss with frequent hospital visits and incidence of adverse effects related to infusions/exchanges.18 19 PLEX has not been suggested to have remission induction potential and chronic use of IVIG has been shown to not have a corticosteroid-sparing effect.20 Some countries have recently introduced repeated injectable therapies such as complement inhibitors and FcRn inhibitors which reduce severity, but these do not have long-term benefits or remission induction potential and are very high cost.21 Long-term use of DBT also confines a patient to chronic healthcare system engagement rather than disease freedom.
The remission rate in our AChR cohort with a high-risk population for MG (male, thymomatous) is better than the expected remission rate from prior studies3 and is much higher than the reported 30% in a rituximab series for AChR positive MG patients.22 This is supported by the finding that CYC demonstrated good efficacy compared with conventional ISTs in a recent meta-analysis.23
Our cohort consisted of primarily AChR patients and this is attributed to our clinical understanding that RTX is preferred as a therapeutic option in patients with MuSK antibodies.24 25 Given our single patient with double seropositive experience, we recommend checking both antibodies prior to treatment with CYC, even with thymoma where AChR antibodies are expected.
The adverse effects of prolonged corticosteroid use are far more consequential for the older patient compared with long-term risks of CYC of malignancy and fertility issues. The older population may be accepting of these CYC risks in the longer term in return for a rapid return of quality of life and minimising corticosteroid exposure, which should be factored into therapeutic decisions for different age groups. We have found that the reduced intensity protocol is tolerated in older patients without compromising efficacy. Prior CYC series have cohorts in which the majority of patients are less than 65 years6 26 but half (15) of our patients were over the age of 65 years. We recognise the limitations of a retrospective single clinic series but note that in our tertiary referral clinic with a high thymoma rate, CYC is only used as a treatment in a minority of patients.
The maintenance steroid-sparing agent of choice following CYC in our clinical service was MMF due to the cohort-specific reason of high rates of non-melanomatous skin cancers in the Australian Caucasian population. The choice of the most appropriate corticosteroid-sparing maintenance agent should be personalised to the individual MG patient, with considerations of childbearing potential, comorbidities and local practices.
In conclusion, we propose CYC with subsequent routine initiation of a corticosteroid-sparing maintenance agent should be considered as an efficacious therapy to achieve remission in a considered cohort of AChR MG patients. The economic benefit of CYC is particularly applicable to cost-restrained healthcare systems, especially compared with recent therapies which are primarily not remission inducing and require continuous repetition. Even in a high-income healthcare system, we suggest CYC should be considered and discussed with patients for rapid remission induction, corticosteroid reduction and long-term freedom from recurrent injectable therapies in selected patients.
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
Ethics statements
Patient consent for publication
Ethics approval
Ethics approval was granted by the Sydney Local Health District human research ethics committee (2021/ETH00927) and the study was conducted in accordance with the Declaration of Helsinki.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
X @darshi_r
Contributors The authors confirm their contribution to the paper as follows: study conception and design: DSR and SWR; data collection: FC; analysis and interpretation of results: FC, DSR and SWR; draft manuscript preparation: all authors. All authors reviewed the results and approved the final version of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests SR has received research funding from the National Health and Medical Research Council (NHMRC, Australia), the Petre Foundation, the Brain Foundation (Australia), the Royal Australasian College of Physicians, and the University of Sydney. She is supported by an NHMRC Investigator Grant (GNT2008339). She serves as a consultant on an advisory board for UCB and Limbic Neurology. She has been an invited speaker for Biogen, Excemed, Alexion, Novartis and Limbic Neurology for research and education meetings. She is a non-remunerated member of the medical advisory board of The MOG Project and The Sumaira Foundation. SWR declares funds over the last 5 years including but not limited to travel support, honoraria, trial payments, research and clinical support to the neurology department or academic projects of which I am a member have been received from bodies and charities: NHMRC, MRFF, NBA, Myasthenia Alliance Australia, Lambert Initiative, Beeren foundation, anonymous donors; and from pharmaceutical/biological companies: Alexion, Biogen, CSL, Genzyme, Grifols, Merck, Novartis, Roche, Sandoz, Sanofi, UCB. Additional interests and potential conflicts of interest include:Co-founder/shareholder of RxPx health, National IVIG Governance Advisory Council & Specialist Working Group Australia (Neurology) (paid), Australian Medical Services Advisory Committee ad hoc subcommittee on IVIG (paid), Australian Technical Advisory Group on Immunisation Varicella Zoster working party (unpaid), Public Salary as a staff specialist neurologist from Concord Hospital Sydney Local Health District (paid), Private billings from patients and medicare Australia reimbursement as a private practice neurologist (paid), Medical advisor (unpaid) to various patient and advocacy groups. All other authors declare no competing interests for content.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Linked Articles
- Editorial commentary