Rituximab in Refractory Myasthenia Gravis – Challenges and Lessons Learnt

Abstract

Nearly 10%–15% patients with generalized myasthenia gravis (MG) have refractory disease and are candidates for newer nonconventional immunotherapies. Rituximab has emerged as an attractive option in them. We describe the efficacy and safety of rituximab in 12 patients with refractory MG treated over a period of 4 years in a tertiary care center. We had a female-predominant cohort (10 females) in the age range 10–67 years. There were eight seropositive patients (acetylcholine receptor, muscle-specific kinase, or both antibodies) and eight had undergone thymectomy. Median number of crises was 2 before treatment with rituximab. Sustained improvement at 12 months was seen in seven patients and steroid dose could be reduced in them. Four worsened on therapy (including one death following prolonged ventilation and septicemia). One patient who had no response to therapy was re-evaluated and diagnosed with congenital myasthenic syndrome related to biallelic CHRNE mutation. In summary, nearly two-thirds (7/11, 63.6%) of autoimmune refractory MG in this cohort showed a good response to rituximab.

Introduction

Refractory myasthenia gravis (MG) refers to a severe and poorly responsive subset which constitutes approximately 10%–15% of generalized MG.[1,2] It is defined as moderate or severe weakness with maximal dose of corticosteroids and one or more immunosuppressive therapies at adequate dose and duration and/or inability to reduce immunosuppressive therapy and/or frequent crises even while on therapy.[2] These patients experience frequent relapses and require plasma exchange or intravenous immunoglobulin (IVIg), entailing significant disability and financial burden. Patients with muscle-specific kinase (MuSK) receptor antibody positivity have been observed more commonly to be treatment refractory compared to those with antibodies to acetyl choline receptor (AChR) or double-seronegative patients.[2]

Many novel drugs have emerged for refractory MG, but in the setting of a low–middle-income country such as India, rituximab (RTX), an anti-CD20 agent, has gained popularity as an accessible and relatively affordable agent. RTX causes destruction of B cells via macrophage- and neutrophil-induced phagocytosis, complement-dependent cytotoxicity, or antibody-dependent cellular cytotoxicity involving natural killer cells.[3] Many observational studies have shown the efficacy of RTX in MG, including refractory MG.[4,5,6,7]

In this series, we report our experience with the use of RTX in 12 patients with refractory MG. We studied the efficacy of RTX in terms of Myasthenia Gravis Foundation America-Post Intervention Status (MGFA-PIS) scores and reduction of steroid dose post-intervention. We additionally report the response to therapy, safety, and other specific observations in this cohort.

Methodology

We screened the records of patients with generalized MG and identified those who fulfilled the criteria for refractory MG (as per Mantegazza and Antozzi)[2] and were treated with RTX between January 2019 and January 2023 with at least 12 months follow-up. Details regarding the dosage and frequency of RTX infusion were noted. Clinical examination details, Myasthenia Gravis Foundation America (MGFA) scores, and use of immunosuppressants, steroid, and anticholinesterase medications were noted pre- and post-RTX. MGFA-PIS scores were determined pre- and post-RTX. Adverse events during treatment with RTX were also recorded. Consent waiver was approved by the Institutional Ethics Committee.

Results

We identified 27 patients with refractory MG over 4 years. among whom 12 patients treated with RTX were included for analysis. Patient characteristics have been summarized in Table 1. The median age of disease onset was 32 (range 9–47) years, with a marked female predominance (10 females). The median duration of MG at the onset of treatment with RTX was 42 months. Five were seropositive for AChR antibody, two were MuSK positive, and four were double seronegative. One patient had dual (AChR and MuSK) antibody positivity. Our patients had varying frequency of myasthenic crisis ranging from one to eight. Thymectomy was performed in eight patients. Both MuSK patients had undergone thymectomy due to poor disease control before antibody testing was available, and one seronegative patient had thymectomy for thymic cyst. One patient with a small thymoma decided against surgery [Table 1]. The mean dose of steroid before treatment with RTX was 45.4 (±15.4) mg/day. All patients were on concomitant steroid-sparing immunosuppressive therapies (seven on azathioprine, three on mycophenolate mofetil, and two on both). Two patients were given more than one immunosuppressant medication before switching to RTX. Significant steroid-related complications were observed in four patients.

Table 1.

Clinical characteristics of the patient cohort with refractory MG

Patients (P)	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12
Age (in years)/gender	37/F	67/F	42/M	33/F	10/F	43/F	39/F	35/F	42/F	38/F	46/M	60/F
Age of onset of MG (in years)	15	47	25	29	9	39	38	34	30	38	12	47
Type of myasthenia	G	G	G	G	G	G	G	G	G	G	G	G
Antibody receptor status	MuSK	AChR, MuSK	AChR	Seroneg	Seroneg	AChR	Seroneg	Seroneg	AChR	AChR	MuSK	AChR
Episodes of worsening	7	6	13	4	4	1	1	2	4	2	5	2
No. of crisis	3	3	8	2	2	2	1	1	2	2	3	1
Acute therapies used	PLEX	PLEX, IVIg	IVIg, PLEX	PLEX	PLEX, IVIg	IVIg	PLEX	IVIg	PLEX	PLEX	PLEX	PLEX
Thymoma	N	N	Y	N	N	Y	N	N	Y	Y	N	Y
Thymectomy	Y	Y	Y	N	N	N	Y	N	Y	Y	Y	Y
Prednisolone dose pre-RTX (mg)	25	20	50	40	40	60	60	60	60	60	50	20
Immunosuppressants used before RTX	Aza	Aza	Aza, MMF	Mtx, Aza, MMF	MMF	Aza	Aza	MMF	Aza	MMF	Aza	Aza

AChR: anti-acetylcholine receptor antibody, Aza: azathioprine, F: female, G: generalized, IVIg: intravenous immunoglobulin, M: male, MG: myasthenia gravis, MMF: mycophenolate mofetil, Mtx: methotrexate, MuSK: Anti-muscle-specific tyrosine kinase antibody, N: no, RTX: rituximab, Seroneg: double seronegative, Y: yes, PLEX: plasma exchange

Among the 12 patients, the frequency of RTX administration ranged from one to six doses. The induction dose was 1 g biweekly, which was given for 11 patients. One patient received 500 mg every week for 4 weeks. The mean duration of follow-up was 21.9 (±12.4) months. All patients had a minimum follow-up of 12 months. Redosing was done based on CD19 count for six patients and by disease severity for two (500 or 1000 mg). Four were given only a single dose of RTX till the end of follow-up.

Improvement following RTX was observed in seven (58.3%) patients, who achieved minimal manifestation (MM) status and sustained improvement till the last visit [Table 2]. This included one MuSK-positive patient and three each with AChR-positive and double-seronegative antibody status. Three of the 7 patients required additional immunosuppressants. The mean dosage of steroids after RTX infusion was 19 (±13.7) mg/day. Corticosteroid doses could be reduced in seven and stopped in one following RTX.

Table 2.

Patient outcomes after RTX therapy

Patients (P)	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12
Antibody receptor status (final)	MuSK	AChR, MuSK	AChR	Seroneg	Seroneg	AChR	Seroneg	Seroneg	AChR	AChR	MuSK/CMS	AChR
Duration of myasthenia at which RTX was initiated (months)	240	240	192	36	12	48	14	8	144	6	24	144
Number of courses	3	1	2	6	2	6	2	2	3	3	3	2
MGFA class (before RTX)	IIIB	V	V	V	V	IVB	IVB	IVB	IVB	IVB	IIIA	IVB
MGFA post-RTX	IA	Expired	IIIA	I	IIA	IIIA	IIA	IIIA	IIA	IIA	IIIA	IIA
MGFA-PIS status	MM2	-	E	MM1	MM3	E	MM3	E	MM3	MM3	U	MM3
Steroid dose post-RTX (mg)	15	-	35	0	5	20	10	50	20	20	Nil	15
Duration of follow-up (months)	16	-	13	30	19	53	17	12	30	12	26	13
Relapses while on RTX	N	-	Y	N	N	Y	N	Y	N	N	N	N
Immunomodulation in addition to RTX	Nil	-	MMF, IVIg monthly	Nil	MMF, IVIg monthly	Nil	Nil	Aza	MMF	MMF	Nil	Nil

AChR: anti-acetylcholine receptor antibody, Aza: azathioprine, CMS: congenital myasthenic syndrome, E: exacerbated, IVIg: intravenous immunoglobulin, MGFA: Myasthenia Gravis Foundation of America, MM: minimal manifestation, MMF: mycophenolate mofetil, MuSK: anti-muscle-specific kinase antibody, N: no, PIS: Post-Intervention Status, RTX: rituximab, Seroneg: double seronegative, U: unchanged, Y: yes

Three patients experienced relapse while on RTX therapy. Two patients who developed relapse while on RTX achieved MM status with monthly IVIg administration and additional immunosuppressant. Relapse in the third patient was due to poor drug compliance. One patient with multiple comorbidities developed cardiac failure and progressed to myasthenic crisis, which was controlled with plasma exchange followed by RTX. She developed cardiopulmonary complications, had prolonged ventilation with hospital-acquired infections and expired. The remaining 11 patients did not have any major side effects related to the infusion.

One 46-year-old male patient (P11) in our study cohort was MuSK antibody positive, but had no response RTX or to other immunosuppressants. His disease onset was at 12 years with multiple relapses and three episodes of myasthenic crisis, which apparently improved with plasma exchange. He had undergone thymectomy due to poor response before antibody test was performed. His childhood photographs confirmed ptosis in the first decade. Genetic testing yielded homozygous pathogenic variant in CHRNE, confirming the diagnosis of a congenital myasthenic syndrome (CMS) and a possible false antibody positivity. He was slowly weaned off all immunotherapy without any marked worsening. Excluding this patient, the final RTX responsiveness was 60% in AChR-positive patients and 75% in double-seronegative patients, while the one true MuSK-positive patient also showed improvement.

Discussion

RTX is being increasingly used in the treatment of refractory myasthenia, as has been supported by previous series demonstrating its efficacy.[8,9,10,11] A recent meta-analysis showed that 64% of patients achieved MM status following RTX and 81% of patients stopped taking immunosuppressants at the last visit.[4] In our cohort, the response to therapy was similar (63.6%, excluding the CMS patient), but most of the patients required additional immunotherapy. Though the dose of corticosteroid could be reduced in the responders, three of them required 20 mg/day of prednisolone even after 1 year of RTX therapy. Real-world data on the newly approved immunotherapies (eculizumab and efgartigimod) have shown responder rates between 62% and 70%[12,13] and even lower rates of MM status, highlighting the significant challenges posed by refractory MG.

Adverse effects with RTX have been reported in 4%–26% in various studies.[14] Risk of infections and hypogammaglobulinemia are of particular concern in MG patients with simultaneous poly-immunotherapy and older age. One of our patients who had cardiopulmonary comorbidities and prolonged ventilatory requirement succumbed to septicemia. Deaths and progressive multifocal leukoencephalopathy have been reported in other cohorts as well.[9,14]

The patient number in this series was too limited to explore the impact of antibody status. Interestingly, we had three patients with double-seronegative status who responded well to RTX. There are only limited studies favoring the use of RTX in refractory seronegative patients.[8,15] Our case series also highlights an important clinical aspect that needs to be considered in all “refractory” cases of myasthenia. One of the MuSK antibody-positive patients who failed to respond to immunomodulation turned out to have CMS. This underlines the need for re-evaluation of diagnosis before labeling patients as “treatment-refractory MG.”

This series of refractory MG patients treated with RTX shows that majority of the patients achieved sustained improvement in terms of post-intervention status, reduction in number of relapses, steroid dose, and anticholinesterase drug dose. We also want to highlight the challenges of incomplete improvement, need for concomitant immunotherapy including steroids, and sepsis, as have been seen in other series. Limitations of this series include the small sample size, heterogeneity of antibody status, and retrospective design of the study. The long-term outcomes and remission could not be assessed in this study. Our experience suggests that RTX is an attractive option when conventional therapies fail, but is far from being the panacea for all. Realistic expectations and patient risk profile should guide the choice of therapy.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.