Black lines indicate time from initial myasthenia gravis (MG) diagnosis to last follow-up, with thickened portions representing time of MG-related therapy. Dashed line indicates time of autologous HSCT. ICU indicates intensive care unit.
eTable. Autologous Stem Cell Collection and Autologous HSCT Conditioning Regimens
Customize your JAMA Network experience by selecting one or more topics from the list below.
Bryant A, Atkins H, Pringle CE, et al. Myasthenia Gravis Treated With Autologous Hematopoietic Stem Cell Transplantation. JAMA Neurol. 2016;73(6):652–658. doi:10.1001/jamaneurol.2016.0113
Some patients with myasthenia gravis (MG) do not respond to conventional treatment and have severe or life-threatening symptoms. Alternate and emerging therapies have not yet proved consistently or durably effective. Autologous hematopoietic stem cell transplant (HSCT) has been effective in treating other severe autoimmune neurologic conditions and may have similar application in MG.
To report 7 cases of severe MG treated with autologous HSCT in which consistent, durable, symptom-free, and treatment-free remission was achieved.
Design, Setting, and Participants
This retrospective cohort study reports outcomes at The Ottawa Hospital, a large, Canadian, tertiary care referral center with expertise in neurology and HSCT, from January 1, 2001, through December 31, 2014, with a median follow-up of 40 months (range, 29-149 months). Data collection and analysis were performed from February 1 through August 31, 2015. All patients with MG treated with autologous HSCT at The Ottawa Hospital were included. All had persistent severe or life-threatening MG-related symptoms despite continued use of intensive immunosuppressive therapies.
Autologous hematopoietic stem cell grafts were mobilized with cyclophosphamide and granulocyte colony-stimulating factor, collected by peripheral blood leukapheresis, and purified away from contaminating lymphocytes using CD34 immunomagnetic selection. Patients were treated with intensive conditioning chemotherapy regimens to destroy the autoreactive immune system followed by graft reinfusion for blood and immune reconstitution.
Main Outcomes and Measures
The primary outcome was MG disease activity after autologous HSCT measured by frequency of emergency department visits and hospitalizations and Myasthenia Gravis Foundation of America (MGFA) clinical classification, MGFA therapy status, and MGFA postintervention status. Safety outcomes included all severe autologous HSCT–related complications.
Seven patients underwent autologous HSCT, 6 for MG and 1 for follicular lymphoma with coincident active MG. Mean (SD) ages at MG diagnosis and at autologous HSCT were 37 (11) and 44 (10) years, respectively. Five patients (71%) had concurrent autoimmune or lymphoproliferative illnesses related to immune dysregulation. All patients had distinct clinical and electromyographic evidence of MG (MGFA clinical classification IIIb-V). All patients achieved durable MGFA complete stable remission with no residual MG symptoms and freedom from any ongoing MG therapy (MGFA postintervention status of complete stable remission). Three patients (43%) experienced transient viral reactivations, and 1 (14%) developed a secondary autoimmune disease after autologous HSCT, all of which resolved or stabilized with treatment. There were no treatment- or MG-related deaths.
Conclusions and Relevance
Autologous HSCT results in long-term symptom- and treatment-free remission in patients with severe MG. The application of autologous HSCT for this and other autoimmune neurologic conditions warrants prospective study.
Quiz Ref IDMyasthenia gravis (MG) is an antibody-mediated disease that affects the neuromuscular junction. Despite advances in immune-targeted therapies, a subset of patients have refractory disease with severe or life-threatening symptoms.1Quiz Ref ID Autologous hematopoietic stem cell transplant (HSCT) has been used to treat a variety of autoimmune neurologic conditions, including multiple sclerosis, chronic inflammatory demyelinating polyneuropathy, neuromyelitis optica, stiff person syndrome, and others.2-5 We report our center’s experience using autologous HSCT for 7 patients with severe MG.
Question Is immune system ablation with high-dose chemotherapy followed by autologous hematopoietic stem cell transplantation effective treatment for patients with severe myasthenia gravis?
Findings In this study of 7 patients with a severe myasthenia gravis that persisted despite multiple immunosuppressive treatments all patients achieved complete, durable, symptom- and treatment-free remission after high-dose chemotherapy, antithymocyte globulin, and CD34-selected autologous hematopoietic stem cell transplantation.
Meanings Autologous hematopoietic stem cell transplantation can be considered for select patients at experienced institutions with severe myasthenia gravis for whom the risks of this procedure are outweighed by its potential benefits.
Cases were identified retrospectively from The Ottawa Hospital Bone Marrow Transplant Programme Database. All patients with MG who underwent autologous HSCT from January 1, 2001, through December 31, 2014, were included in the study. Data collection and analysis were performed from February 1 through August 31, 2015. Myasthenia gravis was diagnosed by clinical assessment, electromyography, and autoantibody testing. Acetylcholine receptor (AChR) antibody testing was performed by radioimmunoassay analysis at the Hamilton Regional Laboratory Medicine Program, with a reference range of less than 0.20 nmol/L. The primary outcome assessed was MG disease activity after autologous HSCT measured by (1) frequency of emergency department visits and hospitalizations and (2) Myasthenia Gravis Foundation of America (MGFA) clinical classification, therapy status, and postintervention status.6 Safety outcomes included all autologous HSCT–related complications of grade 3 or greater, defined by the Common Terminology Criteria for Adverse Events (CTCAE).7 For clarity and conciseness, mild or moderate autologous HSCT complications were not tabulated. The Ottawa Hospital Research Ethics Board reviewed and approved this project, including assembly of deidentified data for description and analysis.
Quiz Ref IDSeven patients underwent autologous HSCT from January 1, 2001, through December 31, 2014. Mean (SD) ages at MG diagnosis and at autologous HSCT were 37 (11) and 44 (10) years, respectively. Six patients underwent autologous HSCT for MG, and 1 patient underwent autologous HSCT for follicular lymphoma with coincident active MG (Table 1). Five patients (71%) had concurrent autoimmune or lymphoproliferative illnesses related to immune dysregulation. All patients had distinct clinical and electromyographic evidence of MG. The patients’ serologic status is given in Table 1. Although muscle-specific kinase antibody testing is now requested at diagnosis for AChR antibody–seronegative patients at our institution, this test was not performed on the 2 seronegative patients in this cohort because the test was not readily accessible in Canada at the time of their diagnosis. Before autologous HSCT, MG severity was graded as moderate (grade III) to life-threatening (grade V) by MGFA clinical classification6 and was refractory to treatment regimens that included pyridostigmine, corticosteroids, additional immunomodulators, and plasma exchange or intravenous immunoglobulin (Table 1). Of the 4 patients who underwent thymectomy, none had a thymoma. Six patients (86%) had at least 1 MG-related emergency department visit or hospitalization before autologous HSCT (Figure). Three patients (43%) required intensive care unit (ICU) admission, and 2 patients (29%) required intubation at least once.
The details of autologous HSCT are summarized in Table 2 and the eTable in the Supplement. All patients underwent stem cell mobilization with cyclophosphamide and granulocyte colony-stimulating factor. Stem cell grafts were harvested by peripheral blood leukapheresis and purified by positive selection of hematopoietic stem cells using an immunomagnetic anti-CD34 monoclonal antibody (CliniMACS CD34 Reagent System, Miltenyi Biotec Inc). Intensive conditioning regimens were used to achieve immune ablation before stem cell graft infusion. The first 2 patients received a conditioning regimen adapted from HSCT for aplastic anemia, an autoimmune hematologic illness. Patients 3 to 6 received a regimen that has been used for autologous HSCT of patients with multiple sclerosis.8 The patient with MG secondary to follicular lymphoma received a conditioning regimen used for the treatment of lymphoma.
Median follow-up was 40 months (range, 29-149 months) after autologous HSCT. At last follow-up, all patients were classified as being in complete stable remission (CSR) by MGFA criteria,6 indicating the absence of MG symptoms without needing MG-directed therapy (Table 1). At 8 months after autologous HSCT, all patients had discontinued immunosuppressive therapies. Six patients (86%) had discontinued MG therapy altogether, whereas 1 continued low-dose pyridostigmine therapy for 5 years (Figure). Six patients (86%) had no further hospitalizations or emergency department visits. One patient required ICU admissions during the first 18 months after autologous HSCT for airway sequelae from prior repeated intubations. This patient has not been hospitalized for MG for more than 11 years (Figure).
Patients were hospitalized for a median of 30 days (range, 13-43 days) (Table 2). Acute complications were transient and included CTCAE grade 3 mucositis in 2 patients (29%) and febrile neutropenia in 3 patients (43%) (Table 2). Absolute neutrophil count exceeded 500/μL (to convert to ×109/L, multiply by 0.001) a median of 11 days after autologous HSCT. No patient was admitted to the ICU for treatment-related toxic effects, and there were no regimen-related deaths. In the 2 months after autologous HSCT, viral reactivation occurred in 3 patients (43%) (Table 2). All reactivation resolved with antiviral treatment. One patient developed acquired amegakaryocytic thrombocytopenia 2 years after autologous HSCT. In 1 patient with coincident MG and lymphoma, lymphoma recurred and progressed 12 months after autologous HSCT and led to death 17 months later. In this patient, MG achieved CSR after autologous HSCT and remained so at the time of death.
Quiz Ref IDWe describe 7 patients with severe MG in whom autologous HSCT was followed by prolonged and complete symptom- and treatment-free remission. An immunoablative conditioning regimen that consisted of high-dose chemotherapy with or without total-body irradiation and antilymphocyte antibodies was administered to eliminate the established autoreactive immune system. The bone marrow and immune system were reconstituted by infusion of a previously harvested autologous stem cell graft depleted of residual mature immune cells by CD34 immunomagnetic selection. This treatment aims to wholly replace an autoreactive immune system with one that is protective and self-tolerant.
Compared with other MG treatments, autologous HSCT requires hospital admission. Quiz Ref IDThe conditioning regimen and the brief but profound immunosuppression it causes can lead to short-term complications, including opportunistic infections and rarer regimen-specific cardiac, renal, or other organ toxic effects. Late complications may involve endocrine dysfunction, including gonadal failure, infertility, and thyroid dysfunction. Immune dysregulation can lead to viral reactivations, secondary malignant tumors, and secondary autoimmune disease.9 In our cohort, the autologous HSCT procedure was tolerable, and unexpected acute toxic effects were not observed. The intense immune depletion resulted in viral reactivation in the first 2 months after autologous HSCT (Table 2). Acquired amegakaryocytic thrombocytopenia, a rare autoimmune hematopoietic condition, was observed in 1 patient and was likely transplant related: secondary autoimmune disease has been reported in 10% of patients after autologous HSCT for autoimmune disease.10 All autologous HSCT complications resolved or stabilized with treatment.
Although there was no transplant-related mortality (TRM) in our cohort, large registry data have reported TRM rates of 6% to 8% for recipients of autologous HSCT for autoimmune diseases.2,3 Smaller studies9,11,12 report lower TRM estimates of 1% to 5% owing to improved supportive care, patient selection, and increasing center-specific experience.
Prevalence of refractory MG has been estimated at 10% of patients with generalized disease, and patients in this cohort similarly exhibited infirmity despite treatment. Refractory MG is reported more frequently in female patients and those with muscle-specific kinase antibodies.13,14 Our population had a similar female preponderance (6 of 7 patients) but had predominantly AChR antibodies (5 of 7 patients). There were 2 AChR antibody–seronegative patients, representing 28% of our small cohort. This rate is somewhat higher than prevalence estimates of 10% to 15%15 but may reflect unidentified muscle-specific kinase antibody–seropositive patients. The AChR antibodies were not measured after autologous HSCT in our cohort. The presence of autoantibodies has not predicted response in other autoimmune disease after autologous HSCT. Anti-GADD antibodies did not disappear after autologous HSCT of 2 patients with stiff-person syndrome4 despite clinical evidence of disease remission, and cerebrospinal fluid oligoclonal bands did not disappear after autologous HSCT in patients with multiple sclerosis who entered remission.16
Other therapies have been used to treat patients with refractory MG, including thymectomy, cyclophosphamide, and monoclonal antibodies, such as rituximab and eculizumab.1 Thymectomy might lead to some clinical improvement, but complete remission is rare, and any advantages of additional procedures may be restricted to patients with limited initial thymectomy.1,17,18 Monthly pulses of cyclophosphamide or a single cycle of high-dose cyclophosphamide without autologous HSCT resulted in clinical improvement in most patients in a small observational study and a small randomized clinical trial19,20; however, complete remission and CSR were achieved in only a few. In a large cohort of patients experiencing neurologic and nonneurologic autoimmune disease, high-dose cyclophosphamide without autologous HSCT has demonstrated overall response rates up to 95%, but only 20% of patients remain event free after 5 years.21,22 Rituximab has demonstrated promising results in retrospective observational studies23-27 in which complete remission was variably achieved (12%-78%) at median follow-up times of 18, 32, and 39 months. Eculizumab has not been extensively studied, but a statistically significant improvement in quality-of-life measures was reported in 6 of 7 treated patients in a 14-patient, phase 2, randomized, multicenter crossover study.28
A single report29 of 1 patient with MG undergoing allogeneic HSCT describes a treatment-free remission with minor residual symptoms. The role of allogeneic HSCT in MG is limited, however, by its high mortality, with TRM rates reported as high as 38% in HSCT for other autoimmune disease and the occurrence of secondary MG as an infrequent manifestation of graft-vs-host disease.3,30-32
Compared with these alternative therapies, responses after autologous HSCT were more consistent (7 of 7 patients), complete (CSR), and sustained (29-149 months). We postulate that autologous HSCT is sufficiently rigorous to eliminate autoreactivity and that it subsequently reestablishes a long-lasting immune system that is functional and self-tolerant, obviating the need for maintenance immunosuppression or additional treatment. In comparison, the global immunosuppression produced by alternative therapies insufficiently suppresses autoreactivity and compromises normal protective immune function, leading to opportunistic infections and cancer in the long term.
The patients in this cohort were heterogeneous in regard to the MG therapy received before autologous HSCT. Notably, none of these patients had been treated with rituximab, eculizumab, cyclophosphamide, or prednisone doses greater than 60 mg/d. It is uncertain whether these alternate therapies or greater dose intensities might have improved control of MG in these patients. The optimal treatment and the specific role of autologous HSCT remain uncertain for patients with MG that remains active after several immunosuppressive agents have failed. In the absence of clear evidence, the use of autologous HSCT needs to be made on a case-by-case basis, balancing the potentially durable treatment-free remission achieved after autologous HSCT against its comparatively higher toxicity and resource needs and against the outcomes achievable by alternative treatments.
Our study was constrained by the retrospective nature of our analysis, which limited the breadth of available baseline and follow-up data. The small sample size and single-institution study environment may also limit the generalizability of our findings. The successful care of patients undergoing autologous HSCT requires significant institutional infrastructure and medical expertise. Rates of autologous HSCT complications and TRM are consistently higher in less experienced institutions.11 We would caution against the use of autologous HSCT for MG outside specialized centers with experience in HSCT and the care of patients with severe MG.
The ability to control autoimmunity by autologous HSCT has been demonstrated in other treatment-refractory autoimmune conditions, including neurologic diseases.2-5 We report the effectiveness of autologous HSCT for inducing long-term, treatment-free MG control for patients with severe disease that has not responded to other treatments. The role of autologous HSCT for MG warrants further exploration with prospective testing.
Accepted for Publication: January 12, 2016.
Corresponding Author: Harold Atkins, MD, FRCPC, The Bone Marrow Transplant Programme, University of Ottawa, The Ottawa Hospital, 501 Smyth Rd, PO Box 926, Ottawa, ON K1H 8L6, Canada (email@example.com).
Published Online: April 4, 2016. doi:10.1001/jamaneurol.2016.0113.
Author Contributions: Drs Bryant and Atkins had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Bryant, Atkins, Bredeson.
Acquisition, analysis, or interpretation of data: Bryant, Atkins, Pringle, Allan, Anstee, Bence-Bruckler, Hamelin, Hodgins, Hopkins, Huebsch, McDiarmid, Sabloff, Sheppard, Tay.
Drafting of the manuscript: Bryant, Atkins,
Critical revision of the manuscript for important intellectual content: All authors.
Administrative, technical, or material support: Bryant, Allan, Anstee, Hamelin, Hopkins, Bredeson.
Study supervision: Atkins, Bredeson.
Conflict of Interest Disclosures: None reported.
Previous Presentations: Poster abstracts of this work were presented at the American Society of Hematology Annual Meeting and Exposition; December 8, 2014; San Francisco, CA; and at the Canadian Blood and Marrow Transplant Group Annual Conference; May 14, 2015; Montreal, Quebec, Canada.
Create a personal account or sign in to: