New and Emerging Biological Therapies for Myasthenia Gravis: A Focussed Review for Clinical Decision-Making

Abstract

Myasthenia gravis (MG) is a rare autoimmune disease characterised by exertion-induced muscle weakness that can lead to potentially life-threatening myasthenic crises. Detectable antibodies are directed against specific postsynaptic structures of the neuromuscular junction. MG is a chronic condition that can be improved through therapies, but to date, not cured. Standard treatment has been unchanged for decades and includes symptomatic treatment with acetylcholine-esterase inhibitors and disease-modifying treatment with steroids, steroid-sparing immunosuppressants and thymectomy. Overall, a relevant proportion of patients does not achieve a satisfactory clinical improvement under standard treatment. Additionally, long-term therapy with steroids can cause significant side effects and latency to clinical improvement with standard steroid-sparing immunosuppressants and after thymectomy can take months to years. In recent years, treatment of MG has changed fundamentally due to improved evidence from phase 3 trials and the regulatory approval of complement inhibitors and FcRn inhibitors as add-on treatment options. This provides new optimism for substantially more patients reaching minimal manifestation status and has led to a shift in treatment strategy with more targeted therapies being employed early in the course of the disease, especially in patients with high disease activity. In this focussed review, we provide an overview of the diagnosis, classification and standard treatment of MG, followed by data from randomised controlled trials on the modern drugs already available for therapy and those still in the final stages of clinical development. In the second part, we provide an overview of real-world data for already approved therapies and outline how the availability of new biologicals is changing both clinical decision-making and patient journey.

Key Points

Standard treatment of myasthenia gravis does not achieve a satisfactory clinical improvement and stability in 30–50% of patients. Therefore, more effective and safe drugs are needed.

After decades of therapeutic stagnation, a new era has begun with the regulatory approval of complement inhibitors and FcRn inhibitors as add-on therapy. This enables a shift from pure broad-spectrum immunosuppressants to a combination with targeted immunotherapies, which should be used in patients with high disease activity, even in the early course of the disease. Additionally, several new biologicals, including B- and T-cell targeting and cell-based strategies such as CAR-T cell therapies, are being investigated.

For clinical decision-making, and to achieve the best possible treatment for the individual patient, further insights from real-world data and comparative studies as well as predictive biomarkers are needed to enable tailored and individualised therapy, to obtain guidance on discontinuation or switching between drugs and to provide information on possible synergistic or antagonistic effects.

Introduction

Myasthenia gravis (MG) is a rare, chronic neurological disorder characterised by specific autoantibodies directed against the neuromuscular junction. Symptoms present as fluctuating and exertion-induced muscle weakness and can include potentially life-threatening myasthenic crises. MG is a chronic condition that can be improved through therapies, but to date, not cured. Because of its well understood pathophysiology, MG serves as an exemplary disease for antibody-driven disorders in general and as a proof-of-principle condition for new therapeutic substances.

MG can be classified according to the detectable autoantibodies (ab), age at onset, thymic pathology, clinical manifestation (ocular versus generalised) and disease severity. More recently, the stratification into mild-to-moderate versus high disease activity (including treatment-refractory status) has gained importance, as treatment strategies are being adapted in the wake of new therapeutic options.

After decades of therapeutic stagnation and lack of evidence from phase 3 trials, a new era has begun with the regulatory approval of several monoclonal antibodies (namely complement inhibitors and neonatal fragment crystallisable receptor [FcRn] inhibitors) as add-on treatment regimens for generalised MG (gMG). Nevertheless, the burden of disease for those affected remains high and the need for more specific and more effective drugs remains relevant.

A Short Look into History—The Path from Lying in Bed to Monoclonal Antibodies

Historically, from its first description as myasthenia gravis pseudoparalytica in 1895 by Friedrich Jolly [1], ‘treatment’ consisted of lying in bed for most of the day, as symptoms are partially improved after rest. Around 1900, it was assumed that the mortality of severe cases was almost 100%. From the 1920s onwards, the development of modern intensive care medicine with the possibilities of mechanical ventilation (specifically the ‘iron lung’, originally developed for people suffering from polio) and the development of antibiotics significantly increased survival chances for patients with a myasthenic crisis [2]. In 1934 Mary Broadfoot Walker established the concept of acetylcholine-esterase inhibition as symptomatic treatment [3]. In the early 1960s, John Simpson recognised that MG is an autoimmune disease [4], and in the mid-1970s, it became clear that the disease is mediated by autoantibodies. These realisations led to two important milestones in the treatment of gMG: the introduction of plasmapheresis as a therapeutic procedure and the use of high-dose oral steroid treatment and subsequently steroid-sparing immunosuppressants such as azathioprine and ciclosporine [2, 5–7]. Mortality decreased substantially, but the proportion of patients with long-term remission remained low [2].

The first monoclonal antibody that was used to treat MG was the chimeric anti-CD20-antibody rituximab. However, due to conflicting evidence [8–11] it remained a treatment for few patients and is mainly used for refractory cases and muscle-specific tyrosine kinase (MuSK)-ab positive MG. Finally, a new era began in the treatment of MG with the regulatory approval of the first complement inhibitor eculizumab in 2017 [12].

New Drugs and Improved Evidence: the Era of Randomised Controlled Trials in Myasthenia Gravis

In recent years, the MG landscape has been influenced by an ever-increasing number of randomised controlled trials for numerous new therapies, mainly monoclonal antibodies. Early on, in 2000, the MG Foundation of America (MGFA) Taskforce on Clinical Research Standards proposed a set of guidelines to improve recording and reporting of clinical trials in MG, including standardisation of clinical outcome measures [13]. The guidelines were updated in 2012 [14]. In recent clinical trials, usually a combination of an examiner-determined physical examination scores, such as the quantitative myasthenia gravis (QMG) [15] score or the myasthenia gravis composite (MGC) [16], and a patient-reported outcome measure, such as myasthenia gravis activities of daily living (MG-ADL) [17] and myasthenia gravis quality of life (MG-QoL15 or MG-QoL15r) [18, 19] score, are being employed. For these outcome measures, a 3-point change in QMG [20], a 3-point change in MGC [14] and a 2-point change in MG-ADL [21] are regarded as a clinically meaningful improvement. Of note, most of the recent phase 3 trials utilised the patient-reported outcome measure MG-ADL as the primary endpoint. However, in most cases, the regulatory approval of new drugs for MG was based on response and improvement in the more objective secondary measure, for example, QMG.

While the clinically meaningful improvement in standardised scores is an important measure to evaluate new treatment options, the treatment goal for an individual patient should either be minimal symptom expression or remission status. The score that correlates best with minimal symptom expression is the Patient Acceptable Symptom State (PASS) [22, 23].

The standardisation of clinical outcome measures has led to improved evidence for treatment decisions but also influenced how disease course is being monitored. Therefore, clinical outcome measures used in MG trials not only play an important role in the research setting, but also in routine clinical and telemedical care to assess disease severity and activity and ultimately to guide therapy. A recent European expert consensus statement [24] recommends the consistent use of the MG-ADL across clinical settings, followed by a patient-reported outcome measure that determines patient satisfaction with disease status (e.g. PASS). In a second step, the use of the QMG and/or MG-QoL is recommended when the MG-ADL indicates disease worsening.

Aim of the Review

In this focussed review, we provide an overview of the new biologicals and discuss their use in the context of standard treatment and medical need in MG care. To this end, we present the available trial data for the approved biologicals, as well as those expected to come to market soon, and discuss the emerging real-world data for already approved drugs. Finally, we propose a pathway for clinical decision-making that is based on MG classification, the available data and patient preference.

Diagnosis and Classification of Myasthenia Gravis

Diagnosis of MG

The diagnosis of the disease is primarily based on the typical clinical appearance of the disease with exercise-induced muscle weakness, which can at least partially disappear after rest. This is often accompanied by a time-of-day dependency with more severe symptoms in the evening than in the morning. Most patients with MG present at first with ocular symptoms only. However, approximately 80% of cases subsequently develop gMG [2]. The disease can be confirmed by detecting the above-mentioned specific antibodies: approximately 75–85% of cases have detectable acetylcholinreceptor (AChR)-ab, 5–8% have detectable MuSK-ab and 1–3% have low-density lipoprotein receptor-related protein 4 (LRP4)-ab. However, approximately 5–15% of cases remain seronegative [25–28]. Serological testing should be accompanied by electrophysiological examinations: repetitive nerve stimulation (showing a decrement) detects gMG with around 75% sensitivity (ranging from 32% to 98% sensitivity) and single fibre electromyography (showing prolonged jitter or blocking of transmission) with around 75–98% of sensitivity but much lower specificity (≤ 70%) [29, 30]. Further, pharmacological testing can be performed with edrophonium, a reversible acetylcholinesterase inhibitor with rapid but short-lasting action. Similar to symptomatic treatment with longer lasting acetylcholinesterase inhibitors (such as pyridostigmine), it increases the concentration of acetylcholine in the synaptic cleft, thus improving neuromuscular transmission and reducing the symptoms.

In unclear diagnostic situations with seemingly clear clinical picture and high disease activity, but seronegative and electrophysiology negative status, further specialised procedures might be employed, such as intercostal muscle biopsy and histopathological examination of the neuromuscular junction [31].

As part of the initial diagnosis of myasthenia gravis, thymus gland imaging must always be carried out to clarify the presence of a thymoma [7].

Classification of MG

MG is a heterogenous disease, and some have gone this far as to say that MG is merely an umbrella term encompassing several distinct entities. Clinical subtypes exist on the basis of age at onset, autoantibody profile, clinical manifestation, gender distribution and thymic pathology. Disease severity is assessed with MGFA class (I–V) and has recently been extended with stratification into mild-to-moderate versus high disease activity (including treatment-refractory status) (Fig. 1). While most of the classification criteria remain individually unchanged during the course of the disease, it is in the nature of this chronic autoimmune disease that disease activity can change spontaneously as well as in response to therapy.

Fig. 1.

Classification of MG. Overview of the different subtypes of MG: approximately 80–85% have AChR-ab. AChR-ab+ MG is classified according to age at onset into EOMG, LOMG and if associated with a thymoma, as TAMG. EOMG with female predominance, LOMG with male predominance. TAMG is evenly distributed between sexes. Approximately 1–3% have LRP4-ab; main age at onset of this subtype is 40–70 years. AChR- and LRP4-ab are of the immunoglobulin (Ig)G1–3 subtype. Approximately 5–15% remain seronegative; main age at onset is 30–60 years. Approximately 5–8% have MuSK-ab, which are of the IgG4 subtype and cause a distinct clinical phenotype. Main age at onset for MuSK-ab+ MG is between 20 and 60 years. jMG can be defined on the basis of the age-related classification, but should not be considered as a homogenous sub-group of MG. For example, jMG differs in some aspects from adult MG, for example, patients with jMG are more likely to be seronegative. According to the clinical manifestation, MG is categorised into ocular and generalised MG. Approximately 80% of patients with ocular presentation show a generalisation within the first 2 years after diagnosis. Generalised MG can be stratified according to MGFA IIa–V and according to disease activity into mild/moderate or high disease activity. In recent classifications, high disease activity encompasses treatment-refractory status. AChR-ab acetylcholine-receptor-antibody, EOMG early onset myasthenia gravis, IgG immunoglobulin G, jMG juvenile MG, LOMG late onset myasthenia gravis, LRP4-ab low-density lipoprotein receptor-related protein 4 antibody, MG myasthenia gravis, MGFA myasthenia gravis Foundation of America Classification, MuSK-ab muscle-specific tyrosine kinase antibody, TAMG thymoma-associated myasthenia gravis. All figures in this review have been created with biorender.com.

Appropriate recognition of these clinical subtypes is highly important because treatment strategies vary accordingly, for example, indication for thymectomy and biologicals depends on the subtype. We here give a focussed overview of the different subtypes and their (potential) consequences for treatment choices.

Firstly, the detected antibody determines different treatment strategies: AChR-ab and LRP4-ab belong to the IgG1 (–3) subtype and can therefore activate the complement system. In these subtypes, complement inhibition is a feasible treatment strategy. MuSK-ab in contrast belong to the IgG4 subtype and do not activate the complement system. This might in part explain the clinically distinct entity of MuSK-ab positive (MuSK-ab+) gMG with predominant weakness in bulbar and respiratory muscles.

AChR-ab positive (AChR-ab+) gMG is classified according to age at onset into early onset MG (onset < 50 years; EOMG) and late onset MG (onset > 50 years; LOMG). Together with the thymic pathology (thymoma-associated MG; TAMG versus non-thymoma-associated MG with either normal thymus, thymus hyperplasia or thymitis) age at onset is an important factor for prognosis of disease course and indication of thymectomy. If the disease starts in childhood or puberty, it is classified as juvenile MG (onset < 18 years; jMG). jMG has many features distinct from MG in adults, and there are sub-classifications for jMG which are discussed elsewhere [32].

The group of seronegative patients probably consists of different sub-groups and some patients might have still unknown antibodies. A study in 13 seronegative patients showing complement deposition at the neuromuscular junction indicates that this group might benefit at least in part from complement inhibition as treatment strategy [31].

In Japan these categories are summarised in the o-gELTMuN classification: first, patients with MG are separated into ocular and generalised forms. Then, generalised MG is divided into the following five subsets: AChR-ab+ early-onset MG, AChR-ab+ late-onset MG, AChR-ab+ thymoma-associated MG, MuSK-ab positive (MuSK-ab+) MG and seronegative MG [33]. While both the o-gELTMuN and the MGFA classification differentiate between ocular and generalised myasthenia, the MGFA classification focusses on disease severity, while the o-gELTMuN classification focusses on antibody-status and sub-groups within MG that necessitate different treatment approaches.

As in most autoimmune diseases, more women are affected by MG than men. AChR-ab+ EOMG, LRP4-ab+, MuSK-ab+ and seronegative MG have a clear female predominance, while AChR-ab+ LOMG has a male predominance. Thymus hyperplasia is more often found in female patients with MG while gender distribution in TAMG is more or less equal [25, 34]. Studies indicating a higher disease burden and lower quality of life in women with MG indicate that female gender might be an independent risk factor for high-disease activity and/or treatment-refractory status [26, 35–37].

Finally, disease course (clinical manifestation as ocular or generalised MG and disease severity according to MGFA class) and disease activity (mild-to-moderate versus high disease activity including treatment-refractory status) influence the recommended first-line therapy and the use of new biologicals.

Defining the Term ‘Treatment-Refractory’ MG

After the regulatory approval of eculizumab for ‘treatment-refractory’ gMG in the EU, a discussion was started as to how the term ‘treatment-refractory’ should be defined. Operational definitions were adopted and the following five criteria mainly emerged: failure to respond adequately to conventional treatment, inability to reduce immunosuppressive therapy without clinical relapse or need for ongoing rescue therapy (e.g. intravenous immunoglobulins [IVIg] and/or plasma exchange [PLEX] or immunoadsorption [IA]), severe or intolerable adverse effects from immunosuppressive or symptomatic therapy, comorbid conditions restricting use of conventional therapies and frequent myasthenic crises even while on immunosuppressive and symptomatic therapy [38, 39]. In most recent classifications, the term ‘treatment-refractory’ is being encompassed within the group of high diseases activity [40].

Standard Treatment of Myasthenia Gravis

Standard treatment of MG consists of symptomatic treatment with acetylcholine-esterase-inhibitors and disease-modifying treatment with steroids and steroid-sparing immunosuppressants for long-term management [41]. Patients fulfilling criteria for prognostic importance of thymectomy should undergo thymectomy within the first 2 years after diagnosis. An overview of standard treatment and newly approved agents for MG according to severity of adverse effects (AEs) and latency of clinical improvement is provided in Fig. 2.

Fig. 2.

Treatment of MG: latency of clinical improvement and magnitude of adverse effects for different therapies. Overview of standard treatment and newly approved agents for MG according to potential severity of adverse effects (AEs) and latency of clinical improvement. Severity of adverse effects is often dose dependent (especially for steroids) and latencies of clinical improvement should be seen as estimates. AChE-I acetylcholine esterase inhibitors, AZA azathioprine, C5-I complement factor 5 inhibitor, CSA cyclosporin A, FcRn-I neonatal fragment crystallisable receptor inhibitor, GCS glucocorticosteroids, IA immunoadsorption, IVIg intravenous immunoglobulins, MMF mycophenolate mofetil, MTX methotrexate, PLEX plasmapheresis, RTX rituximab, TAC tacrolimus. All figures in this review have been created with biorender.com.

Symptomatic Treatment

Symptomatic treatment with pyridostigmine is part of the initial treatment in most patients with MG. Individual tolerability in terms of dose-dependent side effects varies and pyridostigmine dose should be adjusted on the basis of symptom severity and in a flexible manner. To achieve this, patient education is essential. A reduction of pyridostigmine dose can indicate that the disease has stabilised and should guide the tapering of other therapies [41, 42].

Treatment of Exacerbations and Crises

Treatment of exacerbations and crises remain essentially unchanged and is carried out with intravenous immunoglobulins (IVIg) and/or plasma exchange (PLEX) or immunoadsorption (IA). Improvement from IVIg usually starts within a few weeks after and lasts 4–8 weeks. PLEX rapidly lowers antibody levels and removes pro-inflammatory factors contained in the plasma. The effect lasts for about 3 weeks. Because of its rapid onset of efficacy, PLEX is the recommended treatment for life-threatening myasthenic crises. However, in a situation with a severe ongoing infection, IVIg might be the safer approach albeit onset of improvement is slower than with PLEX. Overall efficacy of both strategies is similar. To maintain long-term minimal symptom expression or even pharmacological remission, additional immunosuppressants are usually needed [41].

Both therapies are also being used as maintenance strategies for patients with refractory MG, but evidence and recommendations for this approach vary. A clear recommendation exists only in Japan, where the guidelines for treatment of MG recommend early fast-acting treatment (EFT)—either IVIg or PLEX often combined with high-dose intravenous methylprednisolone (IVMP)—in patients in whom a reduction of steroid-dosage to 5 mg is otherwise not possible [43, 44]. Of note, Japanese guidelines recommend EFT before employment of more targeted biologicals such as C5 or FcRn inhibitors.

Disease-Modifying Treatment

Steroids

Since their introduction into MG therapy in the 1960s, steroids still play an essential role in disease-modifying treatment due to their fast onset of action. Long-term treatment with steroids is limited by relevant systemic side-effects such as, but not limited to, weight gain, diabetes, high blood pressure, cataract and osteoporosis. Therefore, steroid dosage should be tapered as soon as possible below 5 mg and steroid-sparing immunosuppressants should be started early in the disease course [40, 44].

Steroid-Sparing Immunosuppressants

Long-term immunosuppressants that are being used for treating MG are azathioprine, cyclosporine A (CSA), methotrexate (MTX), mycophenolate-mofetil (MMF) and tacrolimus. Treatment approaches as to which drug is first or second choice vary between countries due to a lack of comparative studies. However, azathioprine appears to be used most. Clinical effects can be expected to start after 12 months of treatment [5]. In the long-term, azathioprine is relatively safe and well tolerated. Side effects include liver toxicity, gastrointestinal, haematological, dermatological and infectious events. Rarely, severe bone marrow toxicity occurs. MMF is mostly used as second-line immunosuppressant after azathioprine and clinical effects can be expected after 6–12 months. Recent data suggest very similar efficacy of MMF compared with azathioprine [45]. The safety profile is similar to azathioprine with one important difference: MMF is teratogenic and should therefore be discontinued several months before planning a pregnancy. CSA is another second-line immunosuppressant with a much faster clinical response which can start as early as after 2 months. The safety profile of CSA includes an elevated risk of nephrotoxicity. Further, hypertension, paraesthesia, gum sensitivity, altered taste, increased hair growth, headache, muscle cramps and diarrhoea have been reported. Tacrolimus is mainly used in Japan and China. It also induces a fast clinical response which can start as early as after 2 months. The safety profile is comparable to CSA. MTX is mostly considered a third-line choice in patients who do not tolerate other immunosuppressants. Clinical response can be expected after 3–6 months [41, 42].

In patients with MuSK-ab+ MG, rituximab should be considered earlier in the treatment pathway than for patients with AChR-ab+ MG [41]. Rituximab is being discussed in detail in Sect. 4 of this review.

Thymectomy

The phase 3 MGTX trial, a randomised, single-blind trial, showed efficacy of thymectomy plus prednisolone (versus prednisolone only) for AChR-ab+ MG patients aged 18 to 65 years. Patients in the thymectomy group showed a reduction of prednisolone dose and an improvement in QMG over the 3-year follow-up period [46]. Treatment guidelines recommend thymectomy in patients with a thymoma or thymic enlargement on chest imaging. In non-thymoma-associated MG, thymectomy is recommended in patients with a AChR-ab+ gMG who are younger than 65 years and are within the first five (better two) years after diagnosis [40, 41, 47].

Overall, between 50% and 70% of patients with gMG achieve minimal manifestation, pharmacological remission or complete stable remission under standard treatment [45, 48]. The use of steroids is limited by significant adverse effects and the use of standard steroid-sparing immunosuppressants as well as thymectomy show a relevant clinical response only after months or years [42, 48]. Taken together, this highlights the need for better tolerated, more targeted therapies to treat gMG.

New and Emerging Biological Therapies

MG is considered a prototypic autoimmune disease with a B-cell-mediated T-cell-dependent pathogenesis with known autoantibodies and IgG subtypes in the majority of cases with a multitude of different targeted mode of actions as potential therapeutic strategies. Figure 3 and Table 1 give an overview of all therapeutic agents, their biological target structures and route of administration.

Fig. 3.

Modern therapeutic agents for myasthenia gravis. Overview of the different modern therapeutics for MG according to their biological targets and including their route of administration: fast acting are complement inhibitors and FcRn inhibitors, while slower action onset is displayed by B- and T-cell targeting, cytokine-targeting biologicals and cell-based therapies. The coloured box behind each therapy indicates the level of evidence and whether regulatory approval exists. AChR acetylcholine receptor, BAFF B-cell activating factor, BAFF-R BAFF-receptor, BCMA B-cell maturation antigen, BCR B-cell receptor, BlyS B-lymphocyte stimulator, BTK Bruton’s tyrosine kinase, C5 complement factor 5, CAR chimeric antigen receptor, FcR fragment crystallisable receptor, FcRn neonatal fragment crystallisable receptor, ICOS inducible T-cell costimulatory, ICOS-L ICOS-ligand, IL6-R interleukin 6 receptor, TCR T-cell receptor. All figures in this review have been created with biorender.com.

Table 1.

Overview of new and emerging biological therapies in myasthenia gravis

Drug	Type of drug	Target	Effect/mechanism	Evidence/study in MG	Type of MG in study/case	Approval for MG
B-cell and plasma cell targeting
In clinical use (off-label)
Rituximab	Chimeric mouse/human monoclonal IgG1 ab	CD20	Depletion of CD 20 cells	Phase 2 study (Beat MG); NCT02110706; [10]	AChR and MuSK	Off-label-use in AChR-ab+ and MuSK-ab+ gMG
				MuSK: Class IV evidence (multicentre blinded prospective review; NCT02110706; [9]	Only MuSK
In clinical use in single cases
Obinutuzumab	Humanised IgG1κ monoclonal ab	CD20	Cell death of CD 20 cells	Case report [89]	Case: AChR-ab+ MG + CLL	–
Ofatumumab	Human IgG1 monoclonal ab	CD20	Depletion of CD 20 cells	Case report [90]	Case: AChR-ab+	–
Daratumumab	Human monoclonal IgG1κ ab	CD38	Reduction of plasma cells, plasmablasts, regulatory T- and NK- cells	Case series (1 MG case) [93]	Case: seronegative	Off-label-use in single cases
Bortezomib	Proteasome inhibitor	Proteasome	Inhibition of plasma cell and long-lived B cell proliferation, depletion of short- and long-lived B-cells	Case reports [98, 99]	Cases: MuSK-ab+; MuSK-ab+ and MM	Off-label-use in single cases
				Phase 2a study terminated due to recruitment difficulties; NCT02102594 [100]	–
In trials
Inebilizumab	Humanised IgGκ monoclonal ab	CD19	Depletion of CD 19 cells	Phase 3 study ongoing; NCT04524273	AChR and MuSK
Belimumab	Human monoclonal IgG1λ	BAFF/BLyS	Reduction of B-cell differentiation and maturation, reduction of CD 20-cells	Phase 2 study published with negative results; NCT01480596 [109]	AChR and MuSK
Telitacicept (RC18)	TACI-Ig fusion protein	BAFF/BLyS and APRIL	Reduction of mature B-cells	Phase 2 study published [111] NCT04302103	AChR and MuSK
				Phase 3 trial starting in 2024; NCT06456580	AChR and MuSK
Tolebrutinib	Small molecule	BTK	Irreversibly binds and inhibits BTK and modulates B-cell maturation, survival and activation and reduces inflammation	Phase 3 trial in MG terminated for strategic reasons; NCT05132569	AChR, MuSK, seronegative
Mezagitamab (Tak-079)	Human monoclonal IgG1 ab	CD 38	Reduction of plasma cells, T and NK cells	Phase 2 study completed, negative results posted on NCT; NCT04159805	AChR and MuSK
Complement inhibition
Regulatory approval
Eculizumab	Humanised monoclonal IgG2/4κ ab	C5	Inhibition of terminal complement/MAC activation	Phase 3 study published; REGAIN trial, [12]	AChR	In-label as add-on (USA) and for refractory (EU) AChR-ab+ gMG
Ravulizumab	Humanised monoclonal IgG2/4κ ab	C5	Inhibition of terminal complement/MAC activation	Phase 3 study published; CHAMPION MG trial [58]	AChR	In-label (add-on) for AChR-ab+ gMG
Zilucoplan	Short 35 kDa macrocyclic peptide	C5/C5b	Inhibition of terminal complement/MAC activation	Phase 3 study published; RAISE trial [62]	AChR	In-label (add-on) for AChR-ab+ gMG
In trials
Gefurulimab (ALXN1720)	Bispecific VHH antibody	C5/human serum albumin	Inhibition of terminal complement/MAC activation	Phase 3 study ongoing; NCT05556096	AChR	–
Pozelimab + Cemdisiran	Human monoclonal IgG4P antibody/ siRNA	C5/ C5-RNA	Inhibition of terminal complement/MAC activation/reduction of hepatic C5 level	Phase 3 study ongoing; NCT05070858	AChR and LRP4	–
Iptacopan (LNP023)	Small molecule	Complement factor B	Inhibition of alternative complement pathway	Phase 3 study ongoing; NCT06517758	AChR	–
Vemircopan (ALXN2050)	Small molecule	Complement factor D	Inhibition of alternative complement pathway	Phase 2 study in MG terminated by sponsor (NCT05218096); reasons unavailable at time of writing	AChR	–
Neonatal Fc-receptor inhibition
Regulatory approval
Efgartigimod	Anti-FcRn-IgG1 Fc fragment	FcRn	Inhibition of IgG3 > IgG4 autoantibody recycling, reduction of auto ab levels	Phase 3 study published; ADAPT [73] Phase 3 study, comparing SC and IV regimens in MG; completed, not published; NCT04735432	AChR, MuSK, seronegative	In-label (add-on) for AChR-ab+ gMG
Rozanolixizumab	Human anti-FcRn IgG4 ab	FcRn	Reduction of auto ab levels	Phase 3 study published; MycarinGstudy; [76]	AChR and MuSK	In-label (add-on) for AChR-ab+ and MuSK-ab+ gMG
In trials
Batoclimab	Humanised IgG1 monoclonal ab	FcRn	Reduction of auto ab levels	Phase 3 study ongoing (USA and Europe), study protocol published [81], NCT05403541; Phase 3 study published (China) [79], NCT05039190	AChR and MuSK
Nipocalimab (M281)	Human aglycosylated IgG1 anti-FcRn monoclonal ab	FcRn	Reduction of auto ab levels	Phase 2 study published, Vivacity-MG, NCT03772587; [82]	AChR and MuSK
				Phase 3 study completed, not published; NCT04951622	AChR, MuSK, LRP4, seronegative
B- and T-cell targeting
Cladribine	Purine analogue	B and T cells	Reduction of B and T cells	Pilot open-label study published [117];	AChR
				Phase 3 trial ongoing; NCT06463587	AChR, MuSK, LRP4, seronegative
T-cell targeting
Iscalimab (CFZ533)	Monoclonal ab	CD40	Non-depleting, blocking of CD40, blocking of primary and recall T-cell-dependent antibody responses	Phase 2 trial published; [120]; NCT02565576	AChR and MuSK
Cytokine targeting
Satralizumab	Humanised IgG2 monoclonal ab	IL-6R	IL-6R signalling blockade, inhibition of T-cell activation and differentiation	Phase 3 study completed, abstract on baseline data available; LUMINESCENCE; NCT04963270	AChR, MuSK, LRP4
Tocilizumab	Humanised monoclonal ab	IL-6R	IL-6R signalling blockade, inhibition of T-cell activation and differentiation	Retrospective case series; [124] 2 prospective, open-label-studies [125, 126]	AChR AChR
Cell-based therapies
Autologous hematopoietic stem cell transplantation	–	–	Immune ablation and repopulation	15 cases reported, [127–132]	AChR, MuSK, and seronegative	Ultima ratio for single cases
				Phase 1 study terminated; NCT00424489	–
				Phase 2 ongoing (Fred Hutchinson Cancer Center); NCT00716066	Not MG specific
In trials
Descartes-08, rCAR T Cells	anti-BCMA rCAR-T cells	BCMA (B-cell maturation antigen	RNA Car T-Cells directed against BCMA leading to elimination of plasma cells expressing BCMA	Phase 1b/2a study published [135]; phase 2b ongoing; NCT04146051	AChR, MuSK, seronegative
Anti-CD19 CAR T cells	Fully human autologous anti-CD19 CAR T cells	CD19	DNA CAR-T cells directed against CD19 leading to depletion of CD19 positive B cells	Case reports in MG [136] Phase 2, open-label study starting in 2024; (KYV-101; NCT06193889) Phase 1/2 open-label study starting in 2024; (CABA-201; NCT06359041)	AChR and 2 cases co-diagnosis AChR-ab+ gMG & VGCC-ab+ LEMS

ab antibody, AChR acetylcholine receptor, APRIL a proliferating-inducing ligand, BAFF B-cell activating factor, BCMA B-cell maturation antigen, BlyS B-lymphocyte stimulator, BTK Bruton’s tyrosine kinase, CAR chimeric antigen receptor, gMG generalised myasthenia gravis, IgG immunoglobulin G, IgG4P IgG4 with a proline substitution, IL-6R interleukin 6 receptor, LEMS Lambert–Eaton myasthenic syndrome, LRP4 low-density lipoprotein receptor-related protein 4, MG myasthenia gravis, MuSK muscle-specific tyrosine kinase, TNF tumour necrosis factor, VGCC voltage-gated calcium channel, VHH heavy-chain variable region antigen-binding fragment

Complement Inhibition

Approximately 80% of patients with MG have detectable AChR-ab, which belong to the IgG1 (–3) subtype. Their pathophysiological role is mediated through several different processes: blockade, accelerated endocytosis and degradation of acetylcholine receptors, but also complement-mediated membrane damage and inflammation. By targeting the central complement component C5, the terminal part of the complement cascade is inhibited, leading to reduced formation of the membrane attack complex (MAC or sC5b-9) [49]. Alternate strategies being investigated for treatment of MG are the inhibition of complement factor D [50] and complement factor B. The main risk factor for complement-inhibitor-therapy is the elevated risk of meningococcal infection, which must be mitigated by vaccination.

Complement Inhibitors in Clinical Use

Eculizumab (Anti-C5)

In 2017, the monoclonal antibody eculizumab became the first C5 inhibitor to gain regulatory approval for the therapy of treatment-refractory gMG on the basis of the REGAIN trial results. Although eculizumab did not achieve the primary endpoint compared with placebo (26 weeks of follow-up, 63 eculizumab and 62 placebo patients; change from baseline to week 26 in MG-ADL, measured by worst-rank analysis of covariance [ANCOVA]), relevant secondary endpoints and responder analyses (including MG-ADL, QMG and MG-QoL15) were positive [12]. In subsequent phase 3 gMG trials, the critical worst-rank ANCOVA approach of REGAIN was no longer used as the primary endpoint, but the secondary endpoints including the statistical analyses were. Data from the open-label extension period (OLE) showed a significant reduction in rate of exacerbations, rescue therapies and hospitalisation [51]. Further analysis of the REGAIN and REGAIN OLE also showed a greater reduction of fatigue for the eculizumab group that correlated with clinical improvement of MG and was sustained through the 52 weeks OLE period [52]. Additionally, case reports have shown efficacy of eculizumab in patients with ventilator-dependent MG [53–55]. Recently, data from a phase 3, open-label study on eculizumab in jMG showed efficacy in both primary endpoints (QMG and MG-ADL) [56]. This led to regulatory approval of eculizumab for jMG in Europe in 2023. Eculizumab has shown a good safety profile (most frequent AEs: infections, headache, diarrhoea and nausea) and is administered intravenously every 2 weeks (after an induction period of four weekly doses).

Ravulizumab (Anti-C5)

Ravulizumab is a monoclonal antibody that has been developed by introducing changes to the structure of eculizumab with the goal of a prolonged half-life. The prolonged half-life is achieved through enhanced FcRn binding and recycling of drug but makes the binding of C5 slightly weaker and pH dependent [57]. Ravulizumab showed efficacy versus placebo in the primary endpoints (MG-ADL and QMG) of the CHAMPION MG trial (26 weeks, 86 ravulizumab, 89 placebo) and subsequently gained regulatory approval in 2022 [58]. A post hoc sub-group analysis found no difference in clinical response to ravulizumab comparing disease duration of ≤ 2 years versus > 2 years before start of treatment [59]. A further post hoc analysis examined the time from starting treatment to a relevant clinical response. According to this analysis, the median time to minimal clinically important differences, as measured by MG-ADL (≥ 3 points), was approximately 2 weeks, and as measured by the QMG (≥ 5 points), was approximately 4 weeks [60]. Ravulizumab has shown a good safety profile (most frequent AEs: headache, diarrhoea and nausea) and is administered intravenously every 8 weeks (after induction doses at week 0 and 2).

Zilucoplan (Anti-C5/C5b)

Zilucoplan is a short macrocyclic peptide that targets C5 and additionally C5b and sterically hinders binding of C5b to C6, therefore preventing formation of the membrane attack complex in two ways [61]. Zilucoplan showed efficacy versus placebo in the primary endpoint (MG-ADL) and secondary endpoints (QMG) of the RAISE trial (12 weeks, 86 zilucoplan, 88 placebo) [62]. Zilucopan has shown a good safety profile (most frequent AEs: injection-site reactions, headache and diarrhoea) and is self-administered as a once-daily subcutaneous injection. Data from an interim-analysis of the OLE of the RAISE trial (NCT04225871) after a median of 1.2 years of zilucoplan treatment continue to show a similar safety profile and overall good efficacy with rapid improvement for the group switching from placebo to zilucoplan [63]. Additionally, data from a post hoc analysis of the RAISE trial and the RAISE OLE data showed that zilucoplan improved the fatigue score over 12 weeks compared with placebo and these improvements continued over the 60 week OLE period, with 65% of patients achieving only mild or no fatigue [64].

Complement Inhibitors with Emerging Evidence

Gefurulimab (Anti-C5)

Gefurulimab (ALXN1720) is a bispecific anti-C5 and anti-human serum albumin heavy-chain variable region antigen-binding fragment (VHH) antibody fragment. Its main advantage over existing C5-inhibitors could be the subcutaneous injection paired with a longer half-life (injections once weekly) [65]. A randomised controlled phase 3 trial is currently ongoing (NCT05556096).

Pozelimab with Cemdisiran (Anti-C5/C5-RNA)

The combination of pozelimab with cemdisiran explores a different concept by co-administering a C5-inhibitor (pozelimab) with a small synthetic interfering ribonucleic acid (siRNA) (cemdisiran). Cemdisiran suppresses hepatic production of C5. Both drugs are administered as subcutaneous injections. Pozelimab has gained regulatory approval for treatment of CD55-deficient protein-losing enteropathy (CHAPLE disease) and the combination of pozelimab with cemdisiran is being investigated for PNH and MG [66, 67]. A randomised controlled phase 3 trial in MG is currently ongoing (NCT05070858).

Iptacopan (Anti-Factor B)

Iptacopan (LNP023) is a complement factor-B inhibitor and targets the alternative pathway. More specifically, the inhibition of factor B blocks the activity of the alternative pathway-related C3 convertase [68, 69]. Iptacopan is a small molecule that is administered orally. It has received regulatory approval for IgA-Nepropathy. A randomised controlled phase 3 trial in MG is currently ongoing (NCT06517758).

Vemircopan (Anti-Factor D)

Vemircopan (ALXN2050) is a complement factor-D inhibitor and thus targets the alternative pathway and specifically the amplification loop of the alternative pathway [70]. It is a small molecule that is administered orally. The phase 2 trial for MG (NCT05218096) was stopped at the sponsor’s decision after an interim analysis and no results are available at the time of writing.

FcRn Inhibition

FcRn is an MHC class I ­like molecule expressed in many cell types. Its importance as a target for immunomodulating therapies is derived from the role it plays in IgG recycling. FcRn transports IgG (and albumin) across endothelial cells and thus prolongs their half-life substantially. The drawback of this important mechanism is the increase in the half-life of pathogenic autoantibodies in IgG-mediated diseases. Drugs targeting FcRn lead to increased degradation of endogenous IgG, including pathogenic autoantibodies [71].

FcRn Inhibitors in Clinical Use

Efgartigimod (Anti-FcRn)

Efgartigimod is a Fc-antibody fragment that targets FcRn and reduces IgG levels by about 60–70%. Unlike full monoclonal antibodies targeting FcRn, efgartigimod retains the pH-dependent binding that allows recycling of efgartigimod and prolongs its half-life [72]. Efgartigimod showed efficacy versus placebo in the primary endpoint (≥ 2-point MG-ADL improvement sustained for ≥ 4 weeks in the first treatment cycle) of the ADAPT MG trial (26 weeks, 84 efgartigimod, 83 placebo) [73]. A total of 77% of patients were AChR-ab+, 4% were MuSK-ab+ and 19% were seronegative. In Europe and the USA, efgartigimod was approved only for AChR-ab+ MG, whereas in Japan, the approval includes MuSK-ab+ and seronegative MG. The subcutaneous formulation showed similar efficacy to the intravenous formulation in a non-inferiority trial (ADAPT-SC) [74]. Efgartigimod has shown a good safety profile (most frequent AEs: infections, headache and nausea; subcutaneous administration: additionally, injection-site reactions). It can be administered intravenously or subcutaneously in a cyclic manner (4 weekly administration followed by a drug-free interval of at least 4 weeks) [73–75]. Further trials to investigate other administration schemes (such as continuous administration every 2 weeks without an interval) are underway (NCT04980495; ADAPT-NXT).

Rozanolixizumab (Anti-FcRn)

Rozanolixizumab is a humanised monoclonal antibody that targets the IgG binding region of FcRn and reduces IgG by about 60–70%. Rozanolixizumab showed efficacy in both treatment arms versus placebo in the primary endpoint (change in MG-ADL from baseline to day 43) of the MycarinG trial (14 weeks consisting of 6 treatment weeks and 8 observation weeks, 66 rozanolixizumab 7 mg/kg, 67 rozanolixizumab 10 mg/kg, 67 placebo) [76]. In total, 90% of patients were AChR-ab positive and 11% were MuSK-ab positive. Rozanolixizumab has gained regulatory approval for both AChR- and MuSK-ab+ gMG in Europe and the USA, and is therefore the first monoclonal antibody with regulatory approval for MuSK-ab+ gMG. Rozanolixizumab has shown an acceptable safety profile (most frequent AEs: headache, diarrhoea, pyrexia and nausea) [76, 77]. A recent meta-analysis comparing trial data of 4 FcRn-inhibitors, however, reported a signal that rozanolixizumab showed a higher risk of mild and moderate AEs than placebo [78]. It is administered subcutaneously by pump infusion in a cyclic manner (six weekly administration followed by a variable drug-free interval). Due to the short duration of the phase 3 trial and the reported signal of higher risk of mild-to-moderate AEs in the meta-analysis, more data on the safety of rozanolixizumab is needed and could be provided by the OLE trials currently underway (NCT04124965 completed, and most participants rolled over to NCT04650854).

FcRn Inhibitors with Emerging Evidence

Batoclimab (Anti-FcRn)

Batoclimab is a fully human monoclonal antibody that targets FcRn and is being developed as a low-volume subcutaneous (SC) injection. It has been shown to reduce IgG levels by about 70% [79, 80]. Results from a phase 3 trial conducted at 27 centres only in China (NCT05039190) are already available (67 batoclimab and 65 placebo). Batoclimab showed efficacy versus placebo in the primary outcome (sustained MG-ADL improvement, defined as ≥ 3-point reduction in MG-ADL score from baseline for at least 4 consecutive weeks, in the first cycle). The trial consisted of a 6-week treatment period with weekly dosing followed by a 4-week observation period and an optional second treatment cycle (which was required by 88% of patients). Most frequently reported AEs were infections, peripheral oedema, diarrhoea, MG worsening and hypercholesterolemia [79]. The currently ongoing randomised controlled phase 3 FLEX trial in the USA, Europe and Japan has a duration of 24 weeks (12-week placebo- controlled induction period and 12-week maintenance period) and is assigning patients to three groups: batoclimab 680 mg once weekly, batoclimab 340 mg once weekly or placebo once weekly (NCT05403541). It is planned that two long-term extension periods over 52 weeks are to follow [81].

Nipocalimab (Anti-FcRn)

Nipocalimab is an aglycosylated monoclonal antibody that targets the IgG binding site of the FcRn and is being developed as an intravenous infusion every 2 weeks. A phase 2 trial (NCT03772587; Vivacity-MG) with 68 patients who were treated over 8 weeks used a five-arm dose and administration scheme finding design (placebo, nipocalimab 5 mg/kg every 4 weeks, nipocalimab 30 mg/kg every 4 weeks, nipocalimab 60 mg/kg once, nipocalimab 60 mg/kg every 2 weeks). No new safety concerns were identified; the most common adverse events were diarrhoea, headache and nasopharyngitis. Nipocalimab did not significantly improve MG-ADL at any individual dose when comparing each dosing arm with placebo. However, if all the dosing arms were combined and compared with placebo, there was a statistically significant improvement in MG-ADL versus the placebo group. The authors concluded that this warranted further evaluation of nipocalimab for gMG [82]. A meta-analysis comparing four FcRn-inhibitors has reviewed the available data and concluded that Nipocalimab failed to show efficacy over placebo [78]. A randomised controlled phase 3 trial (placebo versus nipocalimab 60 mg/kg every 2 weeks) has been completed (NCT04951622). Preliminary results have been presented at the Congress of the European Academy of Neurology in July 2024 showing that nipocalimab has reached its primary endpoint of significant MG-ADL improvement and significant improvement in QMG (secondary endpoint) compared with placebo. Peer-reviewed published results are expected soon.

B-Cell and Plasma-Cell Targeting

B cells and plasma cells play an important role in the pathogenesis of MG. They are involved in antigen presentation, autoantibody production and cytokine production. Therefore, therapeutic approaches targeting surface structures (CD 19/20 for B cells and CD 38 for plasma cells) and subsequently depleting B or plasma cells, as well as cytokine-mediated (BAFF/BlyS) approaches, are expected to be effective in MG.

B-Cell and Plasma-Cell Inhibitor in Clinical Use

Rituximab (Anti-CD20)

Rituximab is a chimeric mouse/human monoclonal antibody that targets the B cell surface antigen CD20, which is expressed on most B cells except early development pre-B cells and long-lived plasma cells. The latter is especially important, because long-lived plasma cells produce antibodies. Rituximab depletes CD20 positive B cells primarily by complement-mediated cytotoxicity. The effect lasts approximately 6–12 months. It was first used in treatment of non-Hodgkin’s lymphoma and has since been used successfully (but off label) for many autoimmune diseases [8]. Due to its chimeric nature, it carries an increased risk of allergic reactions and is therefore administered with an anti-allergic premedication. Further, the risk of reactivation of hepatitis-B-virus infection and rare reports of progressive multifocal leukoencephalopathy (PML), especially when combined with other immunosuppressants, have to be taken into account [83, 84].

Rituximab has been administered to patients with AChR-ab+ and MuSK-ab+ MG who did not reach substantial disease stability under standard treatment, including juvenile MG [8, 85, 86]. Especially in MuSK-ab+ MG, rituximab is an important treatment option as reported in a systematic review [8]. Additionally, a multicentre, blinded prospective review evaluating the treatment of 119 MuSK-ab+ patients provided class IV evidence that rituximab increased the probability of a favourable outcome [9]. A further meta-analysis reviewing 24 studies involving 417 patients found that overall, 64% of patients achieved minimal manifestation state or better with a higher proportion of MuSK-ab positive than AChR-positive patients [87]. However, data from a randomised controlled phase 2 trial (BeatMG) in AChR-ab+ MG over a period of 52 weeks (27 placebo, 24 rituximab) failed to show efficacy for rituximab in the primary endpoint (combined outcome of ≥ 75% reduction of daily prednisone dose with clinical improvement or stability defined as ≤ 2-point increase in MGC score) [10]. In a more pragmatic trial, the randomised, placebo-controlled RINOMAX trial, which compared rituximab with placebo in newly diagnosed AChR-ab+ gMG over 16 weeks (rituximab 25, placebo 22), a greater proportion in the rituximab group reached minimal disease manifestation (defined as QMG ≤ 4 and daily prednisolone ≤ 10 mg and no rescue treatment) [11]. The RINOMAX trial was based on retrospective data that showed that rituximab might be more efficient in early onset MG [88]. In summary, the evidence on rituximab therapy in MG remains heterogenous and large randomised controlled trials are lacking.

B-Cell and Plasma-Cell Inhibitors in Single Cases

Obinutuzumab and Ofatumumab (Anti-CD 20)

There are single case reports on the successful use of the more modern anti-CD 20 monoclonal antibodies obinutuzumab and ofatumumab in MG [89, 90]. Obinutuzumab has regulatory approval for chronic lymphatic leukaemia (CLL) and the reported case had concomitant MG and CLL. Ofatumumab has regulatory approval for multiple sclerosis but to date, no investigational trial has been started in MG.

Daratumumab (Anti-CD 38)

The plasma-cell-depleting anti-CD 38 antibody daratumumab has been used as off-label therapy in patients with different autoimmune diseases including MG [91–93]. CD 38 is expressed on plasma cells and plasmablasts, but also on natural killer (NK) cells and subpopulations of regulatory T cells [94], making it a promising target for treating autoimmune diseases. Daratumumab has regulatory approval for treatment of multiple myeloma. Its use in autoimmune diseases has only been considered in cases that were unresponsive to first- and second-line as well as escalation immunotherapies. The main reasoning behind its use is the fact that it targets plasma cells responsible for production of autoantibodies, including pathogenic antibodies. Supportive of this hypothesis are the findings of a retrospective case series in different autoimmune diseases which found that daratumumab therapy led to a marked reduction in pathogenic antibody titres [93]. No randomised controlled trials in MG have been registered.

Bortezomib (Proteasome Inhibitor)

Bortezomib is a proteasome inhibitor that leads to apoptosis in cells that have a high protein turnover. This mechanism of action means that not only tumour cells, but also plasma cells, that have a high protein turnover due to constant secretion of antibodies are sensitive to bortezomib. Bortezomib is approved as first-line therapy in multiple myeloma [95, 96]. After experimental evidence in MG models [97], single case reports in MG exist [98, 99]. Of note, both reported cases were MuSK-ab positive. However, this might be due to the limited on-label escalation treatment options for this MG sub-group. The unicentric, non-randomised, non-placebo controlled phase 2a (TAVAB) trial that planned to investigate the use of Bortezomib in antibody-mediated autoimmune diseases (SLE, RA and MG) was terminated due to recruitment difficulties (NCT02102594) [100]. No further investigational trials in MG have been registered.

B-Cell and Plasma-Cell Inhibitors with Emerging Evidence

Inebilizumab (Anti-CD19)

Inebilizumab is a humanised anti-CD19 monoclonal antibody that depletes CD19-expressing B cells. This includes a broad spectrum of B cells including CD19-positive subpopulations of (auto)antibody-producing plasma cells [101]. On the basis of the results of a randomised controlled phase 2/3 trial (N-MOmentum; NCT02200770) in neuromyelitis optica spectrum disorder (NMOSD), it has received regulatory approval for NMOSD in the USA and Japan in 2020 [102, 103]. Inebilizumab is contraindicated in patients with active hepatitis B virus (HBV) infection and active or untreated latent tuberculosis. A randomised controlled phase 3 trial in AChR-ab+ and MuSK-ab+ MG (MINT trial) has been completed (NCT0452427). In October 2024, at the American Association of Neuromuscular & Electrodiagnostic Medicine (AANEM) Annual Meeting, first positive results from the MINT trial were presented, demonstrating that, compared with placebo, inebilizumab significantly improved MG-ADL (primary endpoint) and QMG (secondary endpoint) over the 26-week study period in patients with gMG, including the AChR-ab+ and MuSK-ab+ sub-groups. No new safety concerns were identified. The study also included a protocol-driven steroid tapering to 5 mg prednisone per day by week 24. Published, peer-reviewed results are expected soon.

Targeting BAFF/BLyS and APRIL

The tumour necrosis factor (TNF) family ligands B-cell activating factor (BAFF) (also called B-lymphocyte stimulator [BLyS]) and a proliferating-inducing ligand (APRIL) play an important role in B-cell survival and regulation [104]. BAFF/BLyS and APRIL interact with three receptors: transmembrane activator and CAML interactor (TACI), B-cell maturation antigen (BCMA) and the BAFF-Receptor. The BAFF–APRIL subfamily of ligands is a promising target for the treatment of autoimmune diseases, since there is evidence in mice that certain (but not all) autoreactive B cells have a greater dependency on BAFF/BLyS for their survival than non-autoreactive B cells [105].

Belimumab (Anti-BAFF/BlyS)

Belimumab (formerly LymphoStat B) is a monoclonal antibody directed against soluble BAFF/BLyS and thus inhibits B-cell proliferation and maturation [106]. Belimumab is approved for SLE since 2011 on the basis of the results of two phase 3 trials (BLISS-52 with 865 participants and BLISS-76 with 819 participants) [105, 107, 108]. However, a randomised controlled phase 2 trial in MG with 40 participants (placebo 22, belimumab 18) failed to show efficacy of belimumab in the primary endpoint defined by improvement of QMG score [109].

Telitacicept/RC18 (Anti-BAFF/BlyS and APRIL)

Telitacicept (also known as RC18) is a fully human TACI-Fc fusion protein that targets BAFF/BLyS and APRIL, thereby inhibiting B-cell proliferation and maturation. Telitacicept has been approved for the treatment of SLE in China in 2021 [110]. Telitacicept was safe and well tolerated in a phase II study of 29 patients and reduced clinical severity (QMG) up to week 24 [111] (NCT04302103). No results were available at the time of writing. A randomised controlled phase 3 trial including AChR-ab+ and MuSK-ab+ patients is starting in 2024 (NCT06456580).

Tolebrutinib (Anti-BTK)

Bruton’s tyrosine kinase (BTK) inhibitors are being investigated for use in several autoimmune and mast cell driven diseases. BTK is an important link between cell-surface receptors and downstream signalling pathways and plays an important role in B‐cell development, migration and activation [112, 113]. Tolebrutinib is a small molecule that is being administered orally and was being investigated for use in MG. However, the phase 3 trial (NCT05132569) was placed on partial clinical hold in 2022 by the US Food and Drug Administration (FDA) on the basis of a limited number of cases of drug-induced liver injury in other tolebrutinib-trials. The study was then terminated by the sponsor for strategic reasons after enrolling only six patients, while other tolebrutinib-trials resumed activity after protocol amendments.

Mezagitamab/TAK-079 (Anti-CD 38)

Mezagitamab (also known as TAK-079) is human monoclonal antibody (mAb) that targets CD38, which is expressed on antibody-producing plasmablasts and plasma cells [94]. A phase 2 trial over 16 weeks (12 placebo, 12 mezagitamab 300 mg, 12 mezagitamab 600 mg) has been completed and negative results for MG-ADL and QMG have been posted on clinicaltrials.gov but not yet published (NCT04159805). The results have, however, been analysed within a Bayesian network analysis comparing 11 antibody-treatments for MG with data from randomised controlled trials. In this comparative analysis, mezagitamab ranked on the lower end for both efficacy outcomes and safety [114].

B- and T-Cell Targeting Drug with Emerging Evidence

Cladribine

Cladribine is a purine analogue that selectively depletes B and T cells. Cladribine has successfully been used in relapsing-remitting multiple sclerosis (RRMS) [115] and long-term observation studies have shown an acceptable efficacy/safety profile [116]. Results of a prospective open-label study in 13 difficult-to-treat cases of MG provided evidence of efficacy of subcutaneous cladribine: after 6 months of treatment 11 participants demonstrated significant clinical improvement, defined as a reduction of ≥ 3 points on the Myasthenia Gravis Composite scale [117]. A case report of a patient with RRMS treated with cladribine who subsequently developed concomitant AChR-ab+ MG with severe myasthenic crisis despite low B and T cell counts [118] should caution the use of cladribine in MG until further evidence can be provided. A phase 3 trial investigating the efficacy and safety of a new oral cladribine formulation in gMG in comparison with placebo has been launched in 2024 (NCT06463587).

T-Cell Targeting Drug with Emerging Evidence

To date, no T-cell targeting therapy has been approved in MG. One possible target structure is CD 40, a surface molecule expressed on T cells. The CD40 dependent pathway plays an important role in T-cell-dependent immune responses, antigen presenting cell function and development of humoral memory [119].

Iscalimab/CFZ533 (Anti-CD 40)

Iscalimab (formerly known as CFZ533) is a human monoclonal antibody that targets CD 40. It leads to a blocking of the CD 40 dependent pathway while it does not deplete CD40-positive cells. Iscalimab has been investigated in a randomised controlled phase 2 trial (22 placebo, 22 iscalimab) over 52 weeks, where it failed to show efficacy over placebo in the primary endpoint (improvement in QMG score at week 25). There was, however, a signal that it improved secondary outcomes (MG-ADL and MGC) in the sub-group of non-thymectomised patients (NCT02565576) [120]. Therefore, further randomised controlled trials are needed to evaluate this therapeutic strategy in MG.

Cytokine-Targeting Drugs with Emerging Evidence

Another strategy under investigation is targeting the interleukin-6 receptor (IL-6R). IL-6 is a pro-inflammatory cytokine produced by different cells including B and T cells and binds to membrane-bound IL-6R and soluble IL-6R. The IL-6 pathway is involved in several processes involved in disease pathogenesis in autoimmune diseases, such as activation and differentiation of naive T cells into pro-inflammatory helper T cells and differentiation of B cells into antibody-producing plasmablasts [121, 122]. In MG, treatment-naïve patients displayed elevated levels of IL-6, which decreased with immunosuppressive treatment and corresponded to disease activity [123].

Satralizumab (IL6-Receptor)

Satralizumab is a humanised monoclonal antibody that targets the interleukin 6 receptor (IL6-R) and inhibits T-cell activation and differentiation as well as differentiation of B cells into plasmablasts. It has been approved for the treatment of seropositive NMOSD [121]. In MG, a randomised controlled phase 3 trial (LUMINESCENCE) has been completed and baseline characteristics of the study are available as abstract (NCT04963270). The study has enrolled 186 participants and included patients with AChR-ab, MuSK-ab and LRP4-ab. The final results are expected soon, but the authors have received the information that the sponsor is probably not going to apply for regulatory approval in MG.

Tocilizumab (IL6-Receptor)

Tocilizumab is a humanised monoclonal antibody that targets IL-6R. It is approved for treatment of rheumatoid arthritis. In MG, mainly case series and small open-label studies exist, such as a retrospective case series including five patients with AChR-ab+ MG who were all older than 65 years at disease onset and reported a favourable efficacy and safety for tocilizumab in combination with low-dose oral steroids [124]. Further, a prospective, open-label, single-arm study in 14 patients with AChR-ab+ gMG over 48 weeks reported efficacy of tocilizumab in the primary endpoint (improvement in QMG score at week 12); 12 patients achieved minimal manifestation status at the end of the study period. All reported adverse events as mild to moderate, with elevated liver enzymes and infections being the most frequent [125]. Another prospective, open-label study with a control group receiving standard treatment (20 tocilizumab, 14 control group) reported greater reduction in MG- ADL score starting week 4 and sustained through week 24 for the tocilizumab group. The most frequently reported adverse event was thrombocytopenia. All adverse events were rated mild to moderate [126].

Cell-Based Treatments in Single Cases

Autologous Haematopoietic Stem Cell Transplantation

Autologous hematopoietic stem cell transplantation (HSCT) is the only treatment option that eradicates all autoreactive T and B cells leading to a reshaping of the immune system or so-called immune resetting. HSCT is being considered in carefully selected, severe and treatment-refractory autoimmune disorders including MG. Thus far, most experience with HSCT in autoimmune disorders has been gathered for multiple sclerosis, with close to 2000 patients treated until August 2022 [127]. A major concern with HSCT is the uncertain risk–benefit ratio including the aggressive conditioning regimen resulting in systemic toxicity and elevated risk of severe infections, as well as risk of death. Potential long-term complications include secondary autoimmunity, impairment of fertility and secondary neoplasms [127].

For MG, 15 cases are reported in the literature that were AChR-ab+ (9 cases), MuSK-ab+ (4 cases) or seronegative (2 cases) [127–132]. Follow-up ranged from 1.5 to 10 years and most patients achieved either complete stable remission or minimal manifestation status with most being able to discontinue all MG medications. However, there is also one case of secondary autoimmune disease manifesting as new onset MG after autologous HSCT for systemic sclerosis [133].

A phase 1 trial where 9 patients were enrolled has been terminated (NCT00424489) and partial results have been posted on clinicaltrials.gov but not published. According to the posted results, 3 deaths occurred (one due to high-grade lymphoma, one known diabetic on tight insulin control died of insulin overdose and one patient with prior splenectomy and thymectomy who died of sepsis bacteremia) but no further results are available at the time of writing. A phase 2 trial on autologous HSCT in 14 different autoimmune disorders, with MG being one, is currently ongoing (NCT00716066).

Cell-Based Treatments with Emerging Evidence

CAR-T Cell Therapy

Similar to most treatment strategies, the concept of chimeric antigen receptor (CAR) T-cell therapy emerged from oncology and is subsequently being adapted and investigated for treatment of autoimmune diseases. CAR-T cells are genetically modified T cells that express the CAR, a synthetic receptor that enables T cells to recognise target antigens without the involvement of antigen-presenting cells. The recognition of the target antigen leads to T-cell activation and subsequently to a targeted destruction of cells expressing the antigen [134]. Different B cell targets are in therapeutic use or being investigated in trials. In MG two targets are being investigated thus far: BCMA and CD19. Safety concerns include side effects of the lymphodepleting conditioning treatment (usually fludarabine and cyclophosphamide), elevated risk of infections due to B cell depletion and associated hypogammaglobulinemia, risk of cytokine-release syndrome (CRS) and immune-effector-cell-associated neurotoxicity syndrome (ICANS), potential malignant complications from long-term persistence of CAR-T cells and off-target toxicity. Most of these have been described in CAR-T cell therapy in autoimmune diseases but with a much lower frequency and often less severity than in cancer CAR-T cell therapy [134]. Of note, none of the reported cases in MG has reached a remission status, leaving the question open as to whether this therapy would need to be administered repeatedly.

Anti-BCMA rCAR-T Cells/Descartes-08

Descartes-08 is an approach using autologous RNA CAR-T cells (rCAR-T) that target BCMA. rCAR-T cells are engineered with RNA instead of DNA and were built upon knowledge from RNA-vaccines. By using RNA, which is temporary and non-replicable, a few issues that arise with DNA CAR-T cells are being addressed: instead of being integrated permanently into the T-cell genome and replicating with cell division, the CAR-encoding mRNA does not replicate together with the activated and proliferating rCAR T-cells. Additionally, since Descartes-08 uses an ex vivo approach, no lymphodepleting conditioning treatment is necessary prior to administration [135]. The technology was assessed in a first prospective, open-label, non-randomised phase 1b/2a study (NCT04146051) that enrolled 14 patients with MG (11 AChR-ab, 2 MuSK-ab, 1 seronegative) and tested several different dosing strategies. Median follow-up was 6 months, and most participants reached an improvement in MG-ADL and QMG, however, they were not symptom free. Reported adverse effects were mostly mild (headache, nausea, fever). Two severe adverse events occurred (urticaria and myocardial infarction, the latter being deemed unrelated to Descartes-08) [135]. A phase 2b trial has been completed (NCT04146051) and peer-reviewed published results are expected soon.

Anti-CD19 CAR-T Cells

Anti-CD19 CAR-T cells use the conventional DNA-based approach and are delivered following a conventional lymphodepleting regimen. Three cases with successful treatment of refractory MG have been published: the first case was a patient with severe, treatment refractory, AChR-ab+ gMG who underwent anti-CD19 CAR-T cell therapy using a second-generation anti-CD19 CAR-T construct comprising a fully human CD19 binding domain. One adverse event occurred (self-limiting and resolving grade 1 transaminitis). Reported follow-up was over 2 months, during which the patients substantially improved in arm-holding time, walking distance and QMG [136]. The two other cases had a concomitant diagnosis of AChR-ab+ gMG and VGCC-ab positive LEMS and were both regarded treatment refractory to multiple therapies. They underwent the same anti-CD19 CAR-T cell therapy regimen and achieved a substantial improvement in MG-ADL and QMG as well as walking distance over a follow-up of 2 and 6 months. Reported adverse effects were flu-like symptoms (fever, muscle and joint pain) consistent with CRS, and in addition one patient with mild arterial hypotension, graded as CRS grade 2. One patient developed minor neurological side effects (mild fatigue, dysarthria and dysgraphia) interpreted as ICANS grade 1. Side effects were successfully treated with tocilizumab, dexamethasone and additionally with anakinra in one patient. Circulating CD19+ B cells were effectively depleted in both patients, accompanied by substantial reduction in pathogenic AChR- and VGCC-ab [137].

A phase 2, open-label study (KYV-101; NCT06193889) as well as a phase 1/2 open-label study (CABA-201; NCT06359041) on autologous fully human anti-CD19 CAR-T cell therapy are about to start in 2024 (NCT06193889).

Other CAR-T Cell Strategies

Other CAR-T cell strategies that are being investigated include MuSK-CAAR-T cells. These chimeric autoantibody receptor (CAAR) T cells are directed against antigen-specific B cells (in this case B cells recognising MuSK) while sparing healthy B cells. An open-label phase 1 trial is starting in 2024 (NCT05451212).

Real-World Data and Meta-Analyses for New Biologicals

There is a rapidly growing body of real-world data (RWD) for the use of C5 inhibitors and FcRn inhibitors in MG; at the time of writing for eculizumab, ravulizumab and efgartigimod. RWD for rozanolixizumab and zilucoplan was not available yet and are only mentioned if they were part of a meta-analysis. Due to the high research activity in the field, this section is likely to be outdated soon.

RWD for Eculizumab

Data from the OLE period of the REGAIN trial as well as data from clinical experience with eculizumab in MG and other diseases continue to show a good safety profile with few serious adverse effects mostly related to severe infections [51, 138–142]. Reported severe infections include severe pneumonia, meningococcal infections and disseminated cryptococcosis [139, 140, 143]. For long-term safety, there are rare reports of probable eculizumab-associated hepatotoxicity with more cases reported in children and none in MG thus far [144, 145]. In retrospective real-world studies, eculizumab continues to show efficacy in the treatment of gMG with clinically meaningful improvement in outcome measures in 70–100% of patients, reduction of exacerbations and hospitalisations and decreased concomitant drug requirement [138, 140–142, 146]. Moreover, eculizumab has been used successfully in several cases of ventilator-dependent MG with successful weaning and significant reduction in disease burden after prolonged myasthenic crisis [53–55, 147–150]. One retrospective cohort study comparing eculizumab with rituximab treatment in patients with treatment-refractory MG found that patients on eculizumab showed a greater reduction in clinical outcome measures, while there was no difference in the rate of myasthenic crisis [151]. Another retrospective cohort study comparing eculizumab with efgartigimod found similar efficacy for reduction in MG-ADL but a greater reduction of QMG score and a greater steroid-sparing effect in the eculizumab group [140].

RWD for Ravulizumab

Emerging RWD for ravulizumab for MG was still scarce at the time of writing. However, data from other indications (PNH and aHUS, pediatric and adult data) and first RWD from MG continues to show a good safety profile comparable to the one in the RCTs [152–156]. Of note, these studies include patients who switched from eculizumab to ravulizumab. Switching was performed at 2 weeks after the last eculizumab dose and was not associated with specific AEs [152–154, 156]. Two small retrospective real-world studies on the use of ravulizumab in MG have been published: one included ten C5-inhibitor-naïve patients and eight patients switching from eculizumab. From the ten C5-inhibitor-naïve patients, six (60%) achieved a clinical meaningful reduction of MG-ADL (improvement of > 2 point) and eight (80%) were able to reduce their steroid dosage. One C5-inhibitor-naïve patient discontinued due to MG worsening; two of the eight patients switching from eculizumab achieved a clinical meaningful reduction of MG-ADL and three were able to reduce their steroid dosage after switching to ravulizumab. None of the switchers discontinued over an observation period of 5 months [156]. The other included 36 patients treated with eculizumab, of whom 15 switched to ravulizumab when it became available (13 discontinued eculizumab, 8 remained on eculizumab). After switching to ravulizumab, 10 remained stable, 3 improved and 2 worsened; 14 of 15 patients were asked whether they preferred ravulizumab or eculizumab overall and regarding several sub-categories including convenience, efficacy and side effects. Notably, the majority preferred ravulizumab overall and for convenience, while for efficacy preferences where evenly distributed between the two medications. For side effects, all patients rated ravulizumab and eculizumab as having the same safety profile [142]. In another prospective cohort study the short-term effects of ravulizumab (41 patients over 10 weeks) and efgartigimod (21 patients over 8 weeks) on AChR-ab effector functions, including complement activation and clinical outcome measures, were compared. Clinically meaningful improvement was observed in 44% (MG-ADL) and 29% (QMG) of patients treated with ravulizumab compared with 35% (MG-ADL) and 20% (QMG) of patients treated with efgartigimod. This was substantially less than in the corresponding RCTs. The authors attributed this partly to the short follow-up period and to the higher proportion of patients with treatment-refractory MG in the cohort. Ravulizumab reduced systemic terminal complement activation. However, in the short follow-up period, neither treatment had significant effects on complement pathways proximal to C5 or on functional capacities of AChR-ab [157].

To date, there is one reported case on the successful use of ravulizumab in crisis-like severe exacerbation (MGFA IVb) [158]. Further RWD on ravulizumab comprising larger cohorts is expected to emerge soon.

RWD for Efgartigimod

Emerging RWD for efgartigimod for MG was still scarce at the time of writing. Safety data are mainly available from the OLE periods of the ADAPT and ADAPT-SC trials (ADAPT+; NCT03770403 and ADAPT-SC+; NCT04818671). In ADAPT+, data from 145 patients who received at least 1 cycle of efgartigimod and up to 17 cycles corresponding to 217 participant-years of exposure continue to show the same safety profile as in the RCT [159]. In ADAPT-SC+, data from 164 patients showed similar adverse events compared with the intravenous formulation with the exception of a rate of 42% of injection-site reactions. These were however, reported to be all mild to moderate and did not lead to treatment discontinuation [74]. One small prospective study from China including 14 patients over 8 weeks (one single cycle of efgartigimod) found that 12 (86%) patients achieved MG-ADL responder status (≥ 2-point improvement) [160]. A larger prospective cohort study from China reported on 61 patients (56 AChR-ab+, 4 MuSK-ab+, 1 seronegative) who received at least one cycle of efgartigimod. A second cycle was only started in 20% of patients with a mean interval length of 6 weeks. Notably, TAMG accounted for most cases (44%, 27 out of 61), and 97% of patients achieved clinically meaningful improvement within the first cycle of efgartigimod [161]. Further, five small retrospective studies with RWD have been published, with most of them pursuing the question of finding the best individual cycle interval. In total, they report on 89 patients receiving between one and six cycles of efgartigimod. Overall, between 86% and 100% of patients achieved clinically meaningful improvements. Efgartigimod effect mostly lasted between 4 and 8 weeks, with exceptional cases of longer intervals between 12 and up to 21 weeks, and follow-up cycles were administered accordingly [140, 162–165]. Reported AEs were mostly mild and similar to the RCTs, with the exception of one fatality due to respiratory insufficiency where a probable severe MG worsening combined with newly diagnosed asthma was treated too late [162].

Furthermore, one of the five retrospective studies compared eculizumab (32 patients) with efgartigimod (31 patients). The study reported similar efficacy for reduction in MG-ADL between efgartigimod and eculizumab but a lesser reduction of QMG score and a lesser steroid-sparing effect in the efgartigimod group compared with eculizumab. The study reported that patients on efgartigimod had a slightly higher probability to discontinue treatment, with the main reason being MG worsening [140]. Yet another one of the five retrospective studies reported specifically on 16 double-seronegative (AChR- and MuSK-ab negative) patients (up to six cycles of efgartigimod; median interval between cycles 6 weeks). The study found only a slight improvement in MG-ADL over the whole group and only a slight reduction in prednisolone dosage. However, mean steroid dosage was already very low at the beginning of efgartigimod treatment [164].

Moreover, 13 cases of successful use of efgartigimod after myasthenic crisis (MGFA V) and 4 cases in crisis-like severe exacerbation (MGFA IVb; including 2 cases with MuSK-ab+ MG) have been reported, with all patients weaning off artificial ventilation and achieving significant clinical improvement after the first cycle of efgartigimod [166–170]. Further, one retrospective study from China compared efgartigimod versus IVIG in the treatment of impending myasthenic crisis or crisis-like severe exacerbation (efgartigimod nine patients, IVIG ten patients). The study found the efgartigimod group had a greater reduction in the MG-ADL score, however, follow-up was only 4 weeks, and one upper respiratory tract infection was reported in the efgartigimod group [171].

In summary, the safety profile of efgartigimod continues to be good but identifying the best individual cycle interval continues to pose clinical challenges. The question of whether efgartigimod is a feasible treatment strategy in seronegative MG remains unanswered. Further, RWD on efgartigimod comprising larger cohorts is expected to emerge soon.

Analyses Comparing Trial Data of Different New Biologicals for MG

Several systematic reviews, meta-analyses and indirect trial data comparisons attempting to compare the available data from RCTs on new biologicals for MG have been published. In general, the evidence from these analyses should be viewed with caution as available data and definitions of endpoints in different trials show important variations which make direct comparison difficult.

An analysis comparing the RCT phase 3 data for efgartigimod (ADAPT trial) and ravulizumab (CHAMPION MG trial) concluded that efgartigimod had a significantly improved response in MG-QoL15r, MG-ADL and QMG at week 4 and at time of best response. However, if data were compared at 26 weeks, responses for MG-ADL and QMG did not differ between the two drugs [172]. A meta-analysis comparing available data on four FcRn-inhibitors (batoclimab, efgartigimod, nipocalimab, rozanolixizumab) concluded that taken together, FcRn inhibition was more efficacious than placebo. However, in the analyses for single drugs, batoclimab, efgartigimod and rozanolixizumab showed efficacy over placebo, while nipocalimab failed to show efficacy over placebo [78]. Of note, the included data on nipocalimab were from a phase 2 five-arm dose-finding trial and conclusions on efficacy of nipocalimab might not be valid. For safety, rozanolixizumab, was the only drug in the analysis that caused an increased incidence of mild and moderate adverse events [78].

A further meta-analysis comparing RCT data from six biologicals (eculizumab, ravulizumab, zilucoplan, efgartigimod, rozanolixizumab and rituximab) found no significant differences between complement inhibitors and FcRn inhibitors in their efficacy versus placebo. In this analysis, rituximab was the only drug that failed to show efficacy in MG-ADL and QMG [173].

Two Bayesian network meta-analyses attempt to compare 11 new biologicals (FcRn inhibitors batoclimab, efgartigimod, nipocalimab and rozanolixzumab; complement inhibitors eculizumab, ravulizumab and zilucoplan; B-cell targeting drugs belimumab, mezagitamab and rituximab; and T-cell targeting drug iscalimab) by using data from RCTs [114, 174]. Both analyses included data from 13 trials, however, for eculizumab, rozanolixizumab and zilucoplan, the selected studies differed and seemingly arbitrarily included only phase 2 or phase 3 data. Moreover, the compared data includes very heterogeneous numbers of patients exposed to each drug, ranging from 18 to 96, and the trial endpoints differ substantially. This probably explains why some of the reported CIs are very broad. In our view, these analyses do not allow valid and clinically relevant conclusions on the comparison of efficacy between the different drugs.

One further network meta-analysis attempted to estimate the number needed to treat (NNT), number needed to harm (NNH) and cost per improved outcome (CPIO) for efgartigimod, rozanolixizumab, ravulizumab and zilucoplan. In this analysis, the NNT was a ≥ 3-point improvement in QMG was best for efgartigimod (2.2), followed by zilucoplan (4.0), ravulizumab (4.4) and rozanolixizumab (5.4). For different outcomes (≥ 5-point improvement in QMG, ≥ 3- or ≥ 5-point improvement in MG-ADL), NNT for efgartigimod was always lowest, but the ranking of the other agents changed. The NNH for any adverse events were similar, but here, too, efgartigimod was best. The CPIO was also lowest for efgartigimod. However, this only applies for the USA since treatment costs depend on country-specific conditions of the healthcare systems [175].

The varying outcomes when different statistical methods are employed highlight the difficulties in deriving reliable conclusions from retrospective and indirect analyses that attempt to compare different study designs and sample sizes. Therefore, it appears questionable whether meaningful conclusions should be drawn from these kind of analyses for clinical decision-making.

Clinical Decision-Making and Patient Journey with New Biologicals

In this part of the review, we propose a new treatment paradigm and a decision pathway. As evidence for these clinical considerations is mostly lacking, this should be regarded as an expert recommendation from a tertiary care centre treating a large cohort of patients with MG. However, different opinions might exist in the field, and with further evidence these might need to be refined in the near future.

The regulatory approval of several new monoclonal antibodies for MG has led to a shift from broad-spectrum immunosuppressants alone to add-on target-specific and selective immunological agents. Generally, MG treatment is becoming more personalised, considering MG subtype and personal patient preferences. Additionally, therapy should now be guided by disease activity with a shift towards a strategy similar to the ‘hit hard and early’ concept from multiple sclerosis therapy. The overarching goal is fast and effective disease control. Evidence for this strategy comes from a systematic review that associated early start of therapy after disease onset with a higher probability of remission and is corroborated for rituximab by the results of a Swedish cohort study and the RINOMAX trial and for EFT consisting of IVIG or PLEX often combined with high-dose intravenous methylprednisolone by Japanese registry data [11, 43, 88, 176]. Direct evidence for a higher efficacy of C5 and FcRn inhibitors if employed early is currently lacking. However, the new agents have two main advantages: onset of action is fast and adverse effects are fewer. Importantly, these therapies are very expensive and therefore economic reasoning also plays a role in most countries.

Figure 4 gives an overview of the patient journey for gMG under the new treatment paradigm. The main novelty is that patients with high disease activity should be considered for add-on treatment with C5 inhibitors or FcRn inhibitors (if AChR-ab+) or add-on treatment with the FcRn-inhibitor rozanolixizumab (if MuSK-ab+) early in the course of their disease. Of note, the Japanese guidelines recommend to first use EFT (IVIG or PLEX in combination with high-dose steroids) before more targeted biologicals such as C5 inhibitors are employed [44].

Fig. 4.

Patient journey for MG–New treatment paradigm. After diagnosis, most patients receive symptomatic treatment with acetylcholinesterase inhibitors and are started on steroids. Steroids should be tapered as fast as clinically possible (goal: ≤ 5 mg/day) and steroid-sparing immunosuppressants (or here NSIST: non-steroidal immunosuppressant therapies) should be considered early. Thymectomy is recommended in AChR-ab+ patients who fulfil criteria for prognostic importance of thymectomy and should be performed within the first 2 years after diagnosis. In LRP4-ab+ MG and seronegative MG, thymectomy might be considered in individual cases. Rescue treatment with IVIg, PLEX or IA should be initiated at any point during the disease course if it is deemed clinically necessary. In patients with high disease activity, for AChR-ab+ MG add-on C5 inhibitors or FcRn inhibitors are available. For MuSK-ab+ MG, rituximab should be considered as steroid-sparing immunosuppressant. Additionally, for MuSK-ab+ MG, the FcRn-inhibitor rozanolixizumab has regulatory approval as add-on therapy. AChE acetylcholine esterase, AChR-ab acetylcholine-receptor-antibody, IA immunoadsorption, IVIg intravenous immunoglobulins, MG myasthenia gravis, MuSK-ab muscle-specific tyrosine kinase antibody, PLEX plasmapheresis. All figures in this review have been created with biorender.com.

We would like to emphasise that biologicals should be used as therapy enhancement and not instead of standard treatment: steroids and steroid-sparing immunosuppressants should be maintained and steroids should be tapered to the lowest effective dose but not withdrawn. The main reasoning behind this recommendation is the fact that in the RCTs of eculizumab, efgartigimod, ravulizumab, rozanolixizumab and zilucoplan, up to 90% of participants continued to receive steroids and/or non-steroidal immunosuppressant therapies (NSISTs) throughout the trial period.

High disease activity has been defined in the updated German treatment guidelines for myasthenic syndromes [40], including three possible scenarios:

(a) Moderate/high MGFA status (≥ MGFA IIb) and/or at least two recurrent severe exacerbations/myasthenic crises with the need for therapeutic intervention within 1 year after diagnosis despite adequate disease-modifying and symptomatic therapy.

or

(b) Persistent symptoms relevant to daily living (≥ MGFA IIa) and severe exacerbation/myasthenic crisis within the last calendar year despite adequate disease-modifying and symptomatic therapy.

or

(c) Persistent symptoms relevant to daily living, even of the mild/moderate course type (≥ MGFA IIa), for more than 2 years despite adequate disease-modifying and symptomatic therapy.

In the absence of comparative studies and biomarkers that go beyond the autoantibody to determine the MG subtype and indicate which monoclonal antibody is most effective in which sub-group of patients, the decision as to which biological is (best) suited for which patient is left to the clinician. Additionally, patient preference regarding frequency and route of administration should be considered. In Fig. 5 we propose a decision pathway for patients with high disease activity that takes antibody type, patients’ personal preference and required vaccinations into account. This decision pathway is based on the authors’ clinical experience, the regulatory approval for the different agents in Europe and the updated German treatment guidelines for myasthenic syndromes [40]. Importantly, standard treatment with steroids (goal: < 5 mg/day) and steroid-sparing immunosuppressants should be maintained, as these biologicals are add-on options to intensify therapy.

Fig. 5.

Proposed clinical decision pathway for patients with high disease activity: which biological for whom? In patients with high disease activity, add-on monoclonal antibodies should be considered. Standard treatment with steroids (goal: ≤ 5 mg/day) and steroid-sparing immunosuppressants should be maintained. To decide which biological for whom, first, the antibody subtype needs to be considered. If the patient is AChR-ab positive, C5 inhibitors and FcRn inhibitors have regulatory approval. To start a C5 inhibitor, complete meningococcal vaccination is mandatory. In Germany this comprises one dose of meningococcal ACWY vaccine and two or three doses of meningococcal B vaccine. To start a FcRn inhibitor, serum IgG needs to be above 5 g/l. Patient’s preference regarding application schemes (i.v. versus s.c. and daily [zilucoplan], cyclic [efgartigimod and rozanolixizumab] or interval-based [ravulizumab and eculizumab] application schemes) should be taken into account. To attempt a prediction of the clinical response to one of the two mechanisms, for FcRn inhibitors historic response to plasmapheresis and for complement inhibitors presence of complement disposition at the neuromuscular junction in a muscle biopsy might be considered. However, we would like to emphasise, that no data exist on these clinical considerations. A switch in mode of action should be considered if one of the two mechanisms shows inefficient clinical response. The precise timing is, however, unclear. At our centre, we try to hold the patient on one biological for a minimum of 4–6 months, unless side effects or a myasthenic crisis make an earlier switch necessary. Relevant side effects should prompt an early switch to a different agent. Of note, the proposed strategy for C5 inhibitors to first use ravulizumab or zilucoplan is based on the regulatory approval in Europe: ravulizumab and zilucoplan are approved as add-on treatment in AChR-ab+ gMG, whereas in Europe, eculizumab is strictly only approved for therapy-refractory AChR-ab+ gMG. In countries where regulatory approval of eculizumab does not differ from ravulizumab, it might be reasonable to start with eculizumab first (e.g. Japan). If the patient is MuSK-ab positive, either rituximab or rozanolixizumab are valid options. In Japan, efgartigimod has also received regulatory approval for MuSK-ab+ MG. Rituximab (off-label-use in most countries including Germany) necessitates a complete vaccination status, while rozanolixizumab/efgartigimod require a serum IgG level above 5 g/l. Patient preference for a low frequency intravenous application would favour rituximab, while a preference for cyclic subcutaneous pump infusions would favour rozanolixizumab. AChR acetylcholine-receptor, MuSK-ab muscle-specific tyrosine kinase antibody, s.c. subcutaneous. *In Japan, efgartigimod has received regulatory approval for AChR-ab+, MuSK-ab+ and seronegative gMG. All figures in this review have been created with biorender.com.

Conclusions and Outlook

After decades of therapeutic stagnation, MG treatment is finally becoming more personalised, considering MG subtype and patients’ personal preferences. The regulatory approval of complement inhibitors and FcRn inhibitors as add-on treatment regimens enables a shift from broad-spectrum immunosuppressants alone to add-on targeted and selective immune modulation. Additionally, several new biologicals including B and T cell targeting and cell-based strategies such as CAR-T cell therapies are being investigated.

The availability of more targeted and fast-acting biologicals has raised the hope among patients and clinicians that a substantially larger proportion of patients will achieve satisfactory clinical improvements or even remission status. The coming years will show whether the new agents can hold that promise.

While excitement in the community about scientific advances is high, important questions remain open: when should novel agents be initiated and when discontinued (apart from lack of efficacy or side effects)? What should be the precise intervals and reasons for inter-class and intra-class switching? Are there possible synergistic or antagonistic effects between the new biologicals? Are the new agents safe over long-term periods? Are they efficient to treat seronegative MG?

Predictive biomarkers to help decide which patient might profit most from which biological are lacking, despite an intensive scientific search. As the new agents are being employed early in patients with high disease activity, more data are needed to define what ‘early’ means, for example, whether the new biologicals could and should also be employed as first-line therapies in treatment-naïve patients. Another open discussion regards the question of whether and at what dosage standard treatment should be continued together with the new biologicals.

As new evidence will hopefully shed light on some of these questions, decision pathways and treatment guidelines will need to be updated.

Declarations

Funding

Open Access funding enabled and organised by Projekt DEAL.

Authors contributions

Concept and planning of the review: L.G. and A.M.; writing of first draft: L.G.; figures: P.D.; and critical revision of the manuscript: all authors. All authors read and approved the final manuscript.

Conflict of interest

L.G. reports no conflicts of interest. P.D. received speaker´s honoraria from Alexion and UCB. S.H. received speaker´s honoraria and honoraria for attendance at advisory boards from Alexion, argenx, Roche and UCB. A. M. received speaker or consultancy honoraria or financial research support (paid to his institution) from Alexion, Argenx, Axunio, Destin, Grifols, Hormosan Pharma, Janssen, Merck, Octapharma, UCB and Xcenda. He serves as member of the medical advisory board of the German Myasthenia Gravis Society.

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and material

No datasets were generated or analysed for this review article; data that support the findings referenced in this article can be found in their corresponding referenced original study.

Code availability

Not applicable.