Chronic immunotherapy for myasthenia gravis

INTRODUCTION — Myasthenia gravis (MG) is an autoimmune neuromuscular disorder characterized by fluctuating motor weakness involving ocular, bulbar, limb, and/or respiratory muscles. The weakness is due to an antibody-mediated, immunologic attack directed at proteins in the postsynaptic membrane of the neuromuscular junction (acetylcholine receptors or receptor-associated proteins). MG is the most common disorder of neuromuscular transmission.

MG can be managed effectively with therapies that include anticholinesterase agents, immunosuppressive and immunomodulatory therapies, and thymectomy. While anticholinesterase agents and intravenous immune globulin (IVIG) are effective for rapid and short-term control of symptoms, most patients with MG require chronic immunotherapy at some point in their illness, if not indefinitely, to maintain disease stability.

The chronic immunosuppressive and immunomodulatory therapies used in the treatment of MG will be reviewed here. The general approach to the treatment of MG and to common clinical situations in patients with MG is discussed separately. (See "Overview of the treatment of myasthenia gravis" and "Myasthenic crisis".)

Other aspects of myasthenia gravis are discussed elsewhere.

●(See "Pathogenesis of myasthenia gravis".)

●(See "Clinical manifestations of myasthenia gravis".)

●(See "Diagnosis of myasthenia gravis".)

●(See "Differential diagnosis of myasthenia gravis".)

●(See "Management of myasthenia gravis in pregnancy".)

GLUCOCORTICOIDS

Indications and rationale — Immunotherapy is usually required for patients who remain significantly symptomatic on anticholinesterase therapy (pyridostigmine), or who become symptomatic again after a temporary response to pyridostigmine. Due to their rapid onset of effect, glucocorticoids are the most commonly used initial immunosuppressive therapy in this setting (table 1).

Observational studies and clinical experience support the efficacy of glucocorticoids in the treatment of MG [1]. Limited evidence from randomized, controlled trials likewise suggests that glucocorticoid treatment offers significant short-term benefit in MG compared with placebo [2]. The administration of moderate or high (1 mg/kg per day of prednisone) doses of glucocorticoids produces remission in approximately 30 percent of patients and marked improvement in another 50 percent [3,4]. The onset of benefit generally begins within two to three weeks and peaks after a mean of 5.5 months.

There are several options for patients who are at increased risk for glucocorticoid complications in order to minimize the duration of the initial prednisone course or to avoid glucocorticoids completely, including early introduction of a nonsteroidal immunosuppressive therapy (eg, azathioprine or mycophenolate) and bridge therapy with antibody-based biologics, intravenous immune globulin (IVIG), or therapeutic plasma exchange. The strategy should be individualized depending on the severity of disease and the nature of the comorbid risk factor. (See 'Glucocorticoid-sparing therapy' below and 'Weaning from glucocorticosteroids' below.)

Risk of transient worsening — When high-dose glucocorticoids are started, up to 50 percent of patients develop a transient deterioration that can be serious [4,5], and up to 10 percent have respiratory failure requiring mechanical ventilation. The transient worsening usually occurs 5 to 10 days after the initiation of glucocorticoids and lasts approximately five or six days.

For this reason, glucocorticoids are most often started in high doses only in hospitalized patients who are receiving concurrent plasma exchange or IVIG for myasthenic crisis. Plasma exchange and IVIG have a quick onset of action that precludes the transient worsening of MG that would otherwise occur due to the glucocorticoids. (See "Myasthenic crisis", section on 'Rapid therapies'.)

In the outpatient setting, a gradual dose escalation of oral prednisone is advised to reduce the risk of transient worsening. (See 'Initial dose escalation' below.)

Administration of prednisone

Initial dose escalation — To avoid the transient worsening that can occur early after starting high-dose glucocorticoids, an outpatient dose escalation regimen works quite effectively [6]. This is the preferred strategy for patients who do not have rapidly progressive symptoms, or those who do not have significant bulbar or respiratory symptoms. Patient with more severe exacerbations should be treated with plasma exchange or IVIG before starting glucocorticoids. (See "Myasthenic crisis", section on 'Rapid therapies'.)

We typically start prednisone at 20 mg daily and then increase by 5 mg every three to five days to a usual target dose of 60 mg per day (or 1 mg/kg per day, maximum 80 mg daily). This often takes four to eight weeks. By that time, it is generally clear if there is a response to glucocorticoids. Tapering of the dose can then begin after a month or so at this dose.

The maximum daily dose can also be individualized. If the patient has responded completely with resolution of the symptoms at a lower dose, then the dose does not need to go any higher. However, it is probably wise to hold the dose at that level for a month or so to be certain that the response is sustained, before beginning a tapering schedule.

Approach to tapering — Once an effective response is obtained, it is important not to taper the glucocorticoids too quickly, or else relapse is likely. Also, it should be remembered that worsening due to a reduction in dose typically takes at least two weeks.

Glucocorticoid tapering can be done with the final goal of achieving either a daily or alternate-day regimen. We use daily dosing of glucocorticoids in MG. However, there are no data to support the benefits of one over the other in myasthenia. Which of these regimens to choose is determined by the individual nature of the patient's medical health and the myasthenia. Daily dosing may be preferred for younger patients, those with little in the way of side effects, and patients with diabetes (to get a more even daily effect on blood sugars).

With either regimen, the daily dose may be reduced by 5 to 10 mg each month until below 30 mg, then more slowly thereafter (eg, 5 mg per month or slower). As an example, with the every-other-day regimen, the taper from 80 mg daily would begin with 80 mg alternating with 60 mg, then a month later to 80 mg alternating with 40 mg, and eventually to 80 mg alternating with 0 mg. Tapering of the high-dose day would then commence. It may be necessary to go back up to a higher dose by at least 10 to 20 mg if symptoms recur during the taper.

Many patients who respond to glucocorticoids can be effectively tapered down to approximately 10 mg per day (or 20 mg every other day) as a maintenance dose that keeps them minimally symptomatic. Young patients, in particular, tolerate such doses well with few of the side effects of these drugs. To avoid relapse, patients may need to stay on this maintenance dose for years, if not indefinitely. The relapse rate is not well defined.

Other immunotherapies are often needed in combination with glucocorticoids, either because of insufficient response, inability to taper glucocorticoids without recurrent symptoms, or development of toxicities of chronic steroid use. (See 'Glucocorticoid-sparing therapy' below.)

Side effects and monitoring — The major disadvantages with chronic glucocorticoid use are their well-recognized side effects. Most are not specific to myasthenia and are dose and duration of treatment dependent. These commonly include weight gain, Cushingoid facies, easy bruising and skin fragility, cataracts, aseptic necrosis of the femoral or humeral heads, hypertension, diabetes, and osteoporosis. (See "Major adverse effects of systemic glucocorticoids".)

Patients should be advised of the potential weight gain and provided an appropriate diet if necessary. Calcium (1500 mg per day) and vitamin D (400 to 800 international units per day) supplementation should be used to reduce bone mineral loss, especially postmenopausal patients. For those most at risk for osteoporosis, bone density should be measured at the start of treatment and then periodically. If bone mineral loss has occurred, patients are generally given a bisphosphonate or an alternative agent. (See "Prevention and treatment of glucocorticoid-induced osteoporosis".)

A histamine-2 blocker does not need to be used routinely, but it should be given in patients with a history of peptic ulcer disease or those who develop symptoms of gastritis. Blood pressure and serum glucose levels should be routinely monitored. Yearly checks for cataracts and glaucoma are also recommended.

Pneumocystis pneumonia prophylaxis should be considered for patients who are treated with significant doses of glucocorticoids (eg, ≥20 mg of prednisone daily for one month or longer) in combination with a second immunosuppressive drug. (See "Treatment and prevention of Pneumocystis pneumonia in patients without HIV", section on 'Prophylaxis'.)

GLUCOCORTICOID-SPARING THERAPY — Nonsteroidal immunotherapy plays an important role in the management of MG for patients with an insufficient response to glucocorticoids, for those who cannot taper glucocorticoids below a reasonably acceptable level without return of symptoms, and as a steroid-sparing therapy in patients with toxicities of chronic glucocorticoid use.

Selection of therapy

AChR-positive and seronegative MG — Azathioprine and mycophenolate mofetil are the most commonly used glucocorticoid-sparing therapies in patients with acetylcholine receptor (AChR) antibody positive or seronegative MG. Alternative agents include cyclosporine, tacrolimus, efgartigimod alfa, rozanolixizumab, ravulizumab, and zilucoplan, and methotrexate is used infrequently.

None of these agents have been compared head to head in patients with MG, and they have variable side effect profiles, cost, availability, and time to maximal effect (table 1).

Absent data from comparative trials, azathioprine is supported as a first-line glucocorticoid-sparing agent by expert consensus, observational data, and limited randomized trial evidence [1]. Mycophenolate mofetil is equally reasonable and is widely used first line in many centers, although randomized trial evidence does not yet support its effectiveness. Each of the antibody-based biologic therapies may be used as a glucocorticoid-sparing therapy with a more rapid onset of effect than other available agents. Cyclosporine and tacrolimus have activity in MG and may achieve results faster than azathioprine or mycophenolate, but concerns about kidney toxicity and drug interactions limit their use as first-line steroid-sparing agents. In rare patients with contraindications to both drugs (eg, active liver disease for azathioprine and lymphopenia for either azathioprine or mycophenolate), cyclosporine or tacrolimus may be used as an initial oral glucocorticoid-sparing agent. (See 'Administration of specific drugs' below.)

MuSK-positive MG — International expert consensus guidelines emphasize growing clinical evidence that muscle-specific tyrosine kinase (MuSK)-positive MG has important differences with regard to treatment compared with AChR antibody-positive and seronegative MG [1]. Most patients with MuSK-positive disease are poorly responsive to anticholinesterase agents. While many patients respond to glucocorticoids, they are more likely to remain steroid dependent despite the addition of other immunotherapeutic agents such as azathioprine or mycophenolate mofetil.

Rituximab has emerged as a preferred early therapy in these patients, especially if there is an unsatisfactory response to initial glucocorticoids, based on accumulating referral center experience and observational data suggesting that MuSK-positive patients may respond particularly well to rituximab. If cost or access to rituximab is prohibitive, azathioprine and mycophenolate mofetil remain reasonable to use as first-line steroid-sparing therapies [7]. Rozanolixizumab was approved by the United States Food and Drug Administration (FDA) in 2023, including for patients with MuSK-positive MG, but its place in MG treatment has not yet been defined. (See 'Rituximab' below and 'Azathioprine' below and 'Mycophenolate' below and 'Rozanolixizumab' below.)

Pregnancy — Pyridostigmine is a first-line therapy for MG during pregnancy, and prednisone is the immunotherapy of choice in patients who are not adequately controlled by symptomatic therapy alone.

In patients who cannot be adequately controlled on prednisone, selective use of azathioprine during pregnancy is warranted when the benefits of immunosuppression with this agent appear to outweigh the risks. There is a consensus that azathioprine is safer in pregnancy than other immunosuppressant drugs, including mycophenolate mofetil, cyclosporine, and tacrolimus. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Azathioprine and 6-mercaptopurine'.)

As with azathioprine, selective use of cyclosporine or tacrolimus during pregnancy is warranted when the benefits of immunosuppression with these drugs appear to outweigh the risks. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Cyclosporine' and "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Tacrolimus'.)

Mycophenolate carries a moderate to high risk of fetal harm. Thus, its use for patients who may become pregnant should be reserved for those with severe disease who use reliable means of contraception. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Mycophenolate mofetil'.)

Prepregnancy planning and management of MG during pregnancy and the postpartum period are reviewed separately. (See "Management of myasthenia gravis in pregnancy".)

Older adults — We use glucocorticoids in older individuals who need a relatively quick response to immunotherapy and then try to add another agent (most often azathioprine or mycophenolate) to replace the glucocorticoids for the long term.

Other than a tendency to use glucocorticoids more sparingly in this group, the use of azathioprine and mycophenolate does not differ from younger patients. Due to the potential kidney side effects, we do not commonly use cyclosporine in these patients. (See 'AChR-positive and seronegative MG' above.)

There are few studies that look at the treatments in this particular age group. However, the usefulness of immunotherapy is supported by a study of outcome at one year or longer in 149 patients with disease onset after age 60 who were treated with azathioprine with or without prednisone [8]. Better outcomes and fewer side effects were observed when prednisone was avoided or was combined with azathioprine.

Administration of specific drugs

Azathioprine — Azathioprine is a purine analogue that inhibits the synthesis of nucleic acids. It interferes with T and B cell proliferation and lowers AChR antibodies in patients with MG.

The goal of azathioprine in most cases is to maintain disease control while avoiding the long-term consequences of glucocorticoid treatment. The more rapidly acting glucocorticoid (prednisone) is usually started first for short-term effect, and then azathioprine is initiated once prednisone has been tapered to a stable, relatively low dose. The ultimate goal is to slowly withdraw the steroids entirely. However, this can take longer than a year to achieve.

●Pretreatment screening – Before starting azathioprine, it is prudent to screen patients for variants in the thiopurine methyltransferase (TPMT) gene that causes TPMT deficiency. One in 300 individuals is homozygous for a genetic variant and has very low or absent enzyme levels. Such patients should not receive azathioprine because they cannot metabolize the drug and may develop life-threatening bone marrow suppression. Patients who are heterozygous for a genetic variant generally have low enzyme activity but can tolerate azathioprine at lower than usual doses. Specific dosing recommendations for thiopurines according to TPMT genotype are available from the Clinical Pharmacogenetics Implementation Consortium [9]. The guidelines are updated and available online at PharmGKB. (See "Overview of pharmacogenomics", section on 'Thiopurines and polymorphisms in TPMT and NUDT15'.)

Either commercially available genotype and/or phenotype (enzyme levels) testing for TPMT deficiency may be done. The TPMT erythrocyte enzyme level assay in serum is preferred because of lower cost and ready availability.

●Dosing – Azathioprine, whether used alone or in combination with glucocorticoids, is begun at a dose of 50 mg daily for two to four weeks. If it is tolerated without systemic side effects, it can be increased by 50 mg increments every two to four weeks to a maintenance dose of 2 to 3 mg/kg total body weight. In most patients, this is typically 150 to 200 mg each day.

●Adverse effects and monitoring – Monthly monitoring of complete blood counts (CBCs) and liver function studies is warranted for the first six months and then less frequently if stable.

The most common side effect encountered with azathioprine is a flu-like illness with fever, nausea, vomiting, and malaise in 10 percent and then less frequent hematologic, gastrointestinal, or liver problems. The initial flu-like illness is idiosyncratic and develops within the first few weeks of treatment. The drug should be discontinued in the patient, and the symptoms quickly abate. (See "Pharmacology and side effects of azathioprine when used in rheumatic diseases".)

Hepatotoxicity, suppression of the white blood count, and pancreatitis are less common but serious potential adverse effects. For these reasons, it is useful to monitor with a CBC and liver function tests monthly for at least six months. The azathioprine dose should be decreased and the CBC monitored more frequently if the white blood count falls below 4000 cells/mm³. Azathioprine should be discontinued if the white blood cell count falls below 3000 cells/mm³. Azathioprine should also be discontinued if the liver function studies are more than mildly abnormal or continue to rise. Serum amylase should be performed if symptoms develop that are consistent with pancreatitis, and azathioprine should be discontinued if the amylase is elevated in this setting.

Azathioprine causes a macrocytosis that increases the mean corpuscular volume (MCV). This is of little concern, but it can be used as a means to assess dosing and compliance [10].

Although it is quite uncommon in patients with MG, the risk of malignancy with long-term azathioprine use may be increased for non-Hodgkin lymphomas [11] and nonmelanotic skin cancer. The malignancy risk is likely relatively low and analogous to the experience seen with its use in rheumatoid arthritis. (See "Pharmacology and side effects of azathioprine when used in rheumatic diseases", section on 'Malignancy risk'.)

There is an important drug interaction between azathioprine and the xanthine oxidase inhibitors allopurinol and febuxostat. Xanthine oxidase inhibitors interfere with the metabolism of 6-mercaptopurine, the active metabolite of azathioprine. Therefore, xanthine oxidase inhibitors should not be used in combination with azathioprine if possible.

In observational studies, azathioprine improves MG symptoms in 70 to 90 percent of patients, but its major drawback is that the onset of any beneficial effect is delayed 6 to 12 months, with a maximal effect often not until one to two years of treatment [12-15]. This was demonstrated in a three-year randomized trial of prednisolone plus azathioprine versus prednisolone plus placebo in 34 patients with generalized MG [16]. At one year, the median prednisolone dose required to maintain remission was similar between groups (37.5 versus 45 mg prednisolone every other day for azathioprine and placebo groups, respectively). By three years, however, azathioprine lowered the median prednisolone dose (0 versus 40 mg every other day), the proportion of patients still requiring prednisolone (20 versus 63 percent), and the number of relapses and treatment failures. Confidence in these results is limited by patient attrition, in addition to the small sample size.

Mycophenolate — Mycophenolate mofetil blocks purine synthesis in lymphocytes and inhibits their proliferation. Like other steroid-sparing agents, it is typically started after patients achieve stability on low doses of prednisone. It may have a slightly faster onset of effect compared with azathioprine.

●Dosing – The usual dose of mycophenolate is 2000 mg (1000 mg twice a day). After starting at 500 mg twice a day, the dose can be increased after one to four weeks to 1000 mg twice a day. Monthly CBCs are usually suggested for the first six months and periodically thereafter.

●Adverse effects and monitoring – Side effects are generally few with mycophenolate. The most common adverse side effects are gastrointestinal, mostly nausea or diarrhea. Leukopenia can occur, but rarely to the degree that warrants discontinuation of mycophenolate, particularly at doses (2000 mg daily or less) used in the treatment of MG. Nonetheless, monthly CBCs during the first six months of mycophenolate therapy, and then less frequently, are suggested. (See "Mycophenolate: Overview of use and adverse effects in the treatment of rheumatic diseases".)

An unanswered question is the long-term safety of mycophenolate. There is no long-term experience with this agent in MG, as there is with azathioprine. In addition, it is hard to extrapolate mycophenolate risks from transplant recipients, where much higher mycophenolate doses and multiple other immunomodulating drugs are used. Most patients who develop lymphoid malignancies on mycophenolate are taking a combination of immunosuppressive agents; therefore, the risk attributable to mycophenolate as a single agent is uncertain. There are conflicting data regarding whether the risk of late malignancy, particularly lymphomas, is increased or not following therapy with mycophenolate [17-20].

Accumulated experience and expert opinion also support use of mycophenolate mofetil in MG, and it is widely used as a first-line agent in many centers [1]. Some feel that it has a slightly faster onset of effect compared with azathioprine and is better tolerated on average. As with azathioprine, relatively short-term trials have failed to show a glucocorticoid-sparing effect within the first year of mycophenolate use compared with placebo [21-23]. In the largest trial, 172 patients with generalized MG taking prednisone ≥20 mg daily for at least four weeks were randomly assigned to mycophenolate (2000 mg daily) or placebo [21]. At 36 weeks, there were no statistically significant differences in any outcomes, including a composite primary outcome (achievement of minimal disease symptoms or pharmacologic remission; 44 versus 39 percent for mycophenolate versus placebo), myasthenia symptoms, and median glucocorticoid dose.

Observational studies both before and after this trial have suggested a benefit of mycophenolate with longer-term follow-up, however [24-27]. One retrospective study from two centers included 102 patients with acetylcholine receptor (AChR) antibody-positive generalized disease who were treated with mycophenolate alone or mycophenolate and prednisone [24]. Eighty percent of those followed for more than 24 months had a desirable outcome (minimal clinical manifestations or better), and the percentage of those who achieved this target began to rise as early as six months into treatment. The prednisone dose required began to fall after 12 months. After 24 months of mycophenolate treatment, 55 percent of the patients were able to come off prednisone entirely and 75 percent were on less than 7.5 mg daily.

Cyclosporine — Cyclosporine selectively inhibits calcineurin, impairs the production of interleukin-2, inhibits the function of T helper cells, and dampens T lymphocyte-dependent immune responses. Kidney toxicity, hypertension, and drug interactions limit its use as a first-line steroid-sparing agent, however.

●Dosing – When used in patients with MG, cyclosporine is typically started at 2.5 mg/kg per day in divided doses. The dose can be increased in 0.5 mg/kg per day increments every four to eight weeks as tolerated to a maximum of 5 mg/kg per day. Clinical practice varies with regard to monitoring of cyclosporine levels; some have suggested aiming for trough levels <300 ng/mL.

Clinical benefit in MG can often be appreciated as early as one to two months after starting cyclosporine.

●Adverse effects and monitoring – Hypertension and nephrotoxicity are the most common limiting adverse effects of cyclosporine. Progressive nephrotoxicity occurs in up to 10 percent of patients. Serum creatinine and blood pressure must be monitored regularly, as well as serum trough cyclosporine levels. (See "Cyclosporine and tacrolimus nephrotoxicity".)

Other cyclosporine side effects include tremor, nausea, gingival hyperplasia, myalgias and flu-like symptoms, and hypertrichosis. The risk of malignancy, primarily squamous cell skin cancer and lymphoma, may also be increased with long-term use [28]. (See "Pharmacology of cyclosporine and tacrolimus", section on 'Side effects'.)

Through its metabolism via the cytochrome P450 system, cyclosporine interacts with many other medications as a substrate of CYP3A4 and an inhibitor of CYP2C8/9 and 3A4. Some drugs reduce the serum cyclosporine levels, whereas others raise the levels. It is important to check trough cyclosporine levels after any of these medications are added to or withdrawn from the patient's regimen.

Observational studies and small placebo-controlled trials have demonstrated the efficacy of cyclosporine both in newly treated patients with MG and in prednisone-dependent patients [28-32]. The maximal effect is achieved by seven months on average [28].

In the largest trial, 39 patients with prednisone-dependent MG were treated with either prednisone plus placebo or prednisone plus cyclosporine at a dose of 5 mg/kg per day (in two divided doses) [30]. At six months, the cyclosporine group showed greater improvement in quantitative strength testing. This group also had a nonsignificantly lower mean prednisone dose. Over the subsequent 18 months of open-label follow-up, progressive nephrotoxicity led to discontinuation of cyclosporine in 10 percent, and cumulative side effects led to discontinuation in a total of 35 percent.

Tacrolimus — Tacrolimus (FK506), most often used after organ transplantation, is an immunosuppressive macrolide molecule similar in action to cyclosporine. Like cyclosporine, it has a relatively rapid onset of effect but has a limited first-line role due to the potential for serious toxicities. (See 'AChR-positive and seronegative MG' above.)

●Dosing – The dosing of tacrolimus is not standardized. In a large observational series in patients with MG, tacrolimus was started at 0.1 mg/kg per day in two divided doses and then adjusted to achieve a plasma tacrolimus drug concentration of 7 to 8 ng/mL [33]. The mean tacrolimus dose was higher at study onset (5.5 mg daily) than at study end (3.5 mg/day). In two randomized trials in patients with MG, a fixed dose of 3 mg daily was used [34,35].

●Adverse effects – Tacrolimus is less nephrotoxic than cyclosporine, and this is especially true given its use in MG at doses markedly lower than that used early after organ transplantation. Nephrotoxicity has not been a limiting factor in the experience of treating MG to date [33-41].

Side effects of tacrolimus include hyperglycemia, hypomagnesemia, tremor, and paresthesias. These generally resolve after adjustment of the dose, if necessary, based on serum levels and magnesium replacement. In one study there was no increase in infections in the tacrolimus group [41]. However, there were three malignancies that developed (two with lung cancer and one with kidney cancer). These all developed within six months of therapy and were felt by the investigators to be unlikely related to the administration of tacrolimus.

Although well tolerated in small trials in patients with MG, tacrolimus has the potential for severe side effects. (See "Pharmacology of cyclosporine and tacrolimus", section on 'Side effects'.)

Small, randomized trials of tacrolimus have been mixed. In a pilot trial of 34 newly diagnosed patients with generalized MG, patients randomly assigned to low-dose tacrolimus (3 mg/day) required less prednisolone to maintain minimal manifestations of MG at one year compared with the placebo group, and there were no significant toxicities [34]. In a larger trial of 80 patients with stable MG on prednisolone, the mean prednisolone dose in the last 12 weeks of the study was nonsignificantly lower in the tacrolimus group compared with placebo (4.9 versus 6.5 mg, p = 0.08) [35]. However, the trial was limited by the stable, mildly affected patient population and the short duration (28 weeks) of follow-up.

In a number of uncontrolled studies, tacrolimus has been used successfully to treat MG at low doses (generally 3 to 8 mg/day) with a response rate of 67 to 87 percent [33,37-40]. As an example, one report evaluated 212 patients with MG treated with tacrolimus who were followed for a mean of 49 months [33]. Prednisone was withdrawn in 95 percent of patients. Pharmacologic remission (ie, no symptoms or signs of MG for at least one year while on some form of immunomodulating therapy for MG) or complete remission (ie, no symptoms or signs of MG for at least one year while on no therapy for MG) was achieved by ≥85 percent of patients.

Antibody-based biologic therapies — Biologic therapies for MG include those that target the Fc-receptor or complement proteins promote autoantibody degradation or inhibit autoimmune-mediated membrane destruction. Onset of effect appears to be faster than several oral agents, making them appealing options for early MG therapy. Efgartigimod alfa was approved by the United States FDA for use in MG in 2021, ravulizumab was approved in 2022, and rozanolixizumab and zilucoplan were approved in 2023, but their places in the spectrum of MG treatments have not yet been defined.

Efgartigimod — Efgartigimod alfa is an IgG antibody Fc-fragment molecule designed to promote degradation of IgG autoantibodies [42,43]. There are several scenarios where efgartigimod may be used. It may be considered early in the course of disease instead of glucocorticoids in diabetic patients or others where glucocorticoids may provide unacceptable risks or have intolerable side effects. It may also be used as a bridge therapy to slower-acting agents (eg, azathioprine or mycophenolate) or as chronic periodic maintenance therapy. Future data are warranted to assess the long-term effects of efgartigimod and to help identify optimal patients for this therapy.

●Dosing – Efgartigimod alfa is given either as a weekly intravenous (IV) infusion at 10 mg/kg or as a subcutaneous formulation (efgartigimod alfa-hyaluronidase) given by a health care provider as a fixed-dose injection [44,45]. The use of the subcutaneous formulation may be preferred in patients with limited venous access. Both formulations are administered by a health care professional and are given weekly for four weeks with observation of the treatment response afterward to determine the next course of treatment. Some patients may have a robust and prolonged response and may only need one or two treatment cycles a year. It may work like maintenance plasmapheresis by reducing antibody levels, without the risks of chronic venous line complications.

●Adverse effects – Common adverse effects with efgartigimod reported in the phase III clinical trial included headache, nasopharyngitis, and upper respiratory infection [46]. Other adverse effects include leukopenia and development of pathogenic serum antibodies. Adverse reactions reported in postmarketing surveillance include infusion-related reactions such as hypertension, chills, and abdominal pain and hypersensitivity reactions including hypotension and anaphylaxis [47].

Efgartigimod was found to be effective in the multicenter ADAPT trial that included 167 patients with generalized MG from North America, Europe, and Japan [46]. Most patients were taking glucocorticoids or other immunosuppressive medications (86 percent), and AChR antibodies were positive in 77 percent. Symptomatic improvement of at least two points in the MG Activities of Daily Living (MG-ADL) score was more common in patients assigned to weekly infusions of efgartigimod 10 mg/kg IV for four weeks than in those assigned to placebo (68 versus 30 percent). The rate of adverse effects was similar in treatment and placebo groups; headache, nasopharyngitis, and upper respiratory infection were reported most frequently. On the basis of these data, efgartigimod was approved by the FDA and in Japan for use in AchR antibody-positive patients with generalized MG. In a subsequent single-center retrospective review of 17 patients who were treated with efgartigimod, a >50 percent improvement in nonocular symptoms was reported by 16 patients (94 percent) at three-month follow-up [48]. Mean improvements in baseline MG-ADL score were 5.5 and 7.1 points at three and six months, respectively. Most patients were taking glucocorticoids, but only 41 percent had attempted a steroid-sparing agent. Efgartigimod was selected for most patients due to a poor response or contraindication to other therapies.

Rozanolixizumab — Rozanolixizumab is a monoclonal antibody that binds to the Fc receptor to promote lysosomal degradation of IgG autoantibodies [49].

As with other antibody-based biologic agents, there are multiple scenarios where rozanolixizumab may be used. It may be considered early in the course of disease instead of glucocorticoids, as a bridge therapy to slower-acting agents, or as chronic periodic maintenance therapy. Rozanolixizumab was approved by the US FDA for patients with AchR-positive or MuSK-positive MG [50]. Future data are warranted to assess the long-term effects of rozanolixizumab and to help identify optimal patients for this therapy.

●Dosing – Rozanolixizumab is given by a health care provider as a subcutaneous infusion weekly for six weeks according to patient body weight. The dose is 420 mg for patients who weigh <50 kg, 560 mg for patients who weigh 50 to 100 kg, and 840 mg for patients who weigh >100 kg [50]. Subsequent infusion cycles are determined after observation to assess the initial response to treatment.

●Adverse effects – Common adverse effects with rozanolixizumab include headache, infections, and diarrhea.

In a placebo-controlled trial of 200 patients with AchR-positive or MuSK-positive MG, improvement in baseline symptoms (MG-ADL scores) by day 43 was greater among those assigned either to rozanolixizumab 7 mg/kg or 10 mg/kg than patients assigned to placebo (least-squares mean difference [-3.37 and -3.4 versus -0.8]) [51]. The mean age of patients was 52 years old, and mean duration since MG diagnosis was 5.8 years; nearly two-thirds of patients were also taking glucocorticoids. Most adverse effects were mild, including headache, diarrhea, fever, and nausea.

Ravulizumab — Ravulizumab is a humanized monoclonal antibody that specifically binds with the terminal complement protein C5, preventing disruption of neuromuscular transmission, presumably by inhibiting membrane attack complex-mediated destruction of the postsynaptic membrane. As with efgartigimod, there are several scenarios where ravulizumab may be used. It may be considered early in the course of disease instead of glucocorticoids, as a bridge therapy to slower-acting agents, or as chronic periodic maintenance therapy. Future data are warranted to assess the long-term effects of ravulizumab and to help identify optimal patients for this therapy.

●Dosing – Ravulizumab is given as by a health care provider as a variable weight-based dose. After a loading dose, the maintenance dose is given two weeks later and then continued every eight weeks thereafter.

This medication is structurally similar to eculizumab but has been altered to maintain a concentration in the serum that requires less frequent dosing at eight-week intervals [52]. Although there are no data in patients with MG, patients with paroxysmal nocturnal hemoglobinuria have been safely and effectively switched from eculizumab administered every two weeks to ravulizumab administered every eight weeks [53].

●Adverse effects – In the phase III clinical trial, there were no treatment-limiting side effects [52]. Headache was the most frequently reported adverse effect and the proportion of patients who experienced adverse effects were similar between groups.

Patients should be immunized with meningococcal vaccines at least two weeks prior to administering the first dose of ravulizumab.

In a trial of 175 patients with MG on glucocorticoids, patients with AchR antibody MG who received ravulizumab infusions every eight weeks after initial loading doses had greater improvements in the MG-ADL score at 26 weeks than those assigned to placebo (-3.1 versus -1.4) [52]. Patients had mild to moderate symptoms (median MG-ADL score 9), and most were taking glucocorticoids or other immunosuppressants at baseline. The rate of adverse events was similar between groups and the most frequently reported were headache, diarrhea, and nausea. On the basis of these results, ravulizumab was approved by the United States FDA for use in AChR antibody-positive patients with generalized MG.

Zilucoplan — Zilucoplan is a macrocyclic peptide that is designed to inhibit cleavage of complement C5 and prevent the development membrane attack complex that contributes to destruction of the postsynaptic membrane.

The exact role of zilucoplan in the chronic treatment of MG is uncertain. It may be used as early therapy in place of glucocorticoids, as bridge therapy, or chronic maintenance therapy. Zilucoplan was approved by the United States FDA for use in patients with AChR-positive generalized MG [54].

●Dosing – Zilucoplan is given as a self-administered daily subcutaneous injection at prespecified doses of 16.6, 23, or 32.4 mg according to actual body weight.

●Adverse effects – The most common adverse reactions in clinical trials were injection site bruising, upper respiratory infections, and diarrhea [55,56]. Uncommon serious adverse reactions include meningococcal meningitis and pancreatitis.

Patients should be immunized with meningococcal vaccines at least two weeks prior to administering the first dose of zilucoplan.

In a clinical trial of 174 patients with generalized MG, improvement in baseline MG-ADL scores at 12 weeks was greater among patients assigned to zilucoplan at a dose of 0.3 mg/kg once daily than those assigned to placebo (least-squares mean difference -2.09, 95% CI -3.24 to -0.95) [55]. Adverse effects were generally mild and included injection site reactions, headache, and diarrhea. Approximately 50 percent of patients were considered treatment refractory at study entry, and more than 60 percent were taking glucocorticoids at baseline.

Methotrexate — Oral methotrexate may be used as a glucocorticoid-sparing therapy in patients with MG who have not responded to other agents based upon favorable results from an open-label study with a small number of patients [7,57]. However, a 12-month, double-blind, randomized controlled trial of 50 patients with MG found that methotrexate did not significantly reduce the use of prednisone (as measured by the area under the dose-time curve) or improve secondary outcome measures of MG compared with placebo [58].

Weaning from glucocorticosteroids — Azathioprine and other glucocorticoid-sparing therapies are most commonly started once prednisone has been tapered to the lowest dose that will maintain a reasonable clinical response.

●Tapering – After the minimum time to onset of clinical response for the added immunotherapeutic agent has passed (eg, as long as 12 months on azathioprine), we begin to slowly taper prednisone to as low a dose as can be achieved, preferably none at all.

Balancing the improvement of MG symptoms with the side effects of the immunotherapies is a constant challenge. Even when steroid-sparing therapies are used, it is common to continue the use of pyridostigmine in order to reduce the required doses of immunotherapeutic drugs in an effort to minimize adverse effects. (See "Overview of the treatment of myasthenia gravis", section on 'Initial symptomatic therapy'.)

After the glucocorticoids have been tapered, we then try to taper pyridostigmine. Many, but certainly not all, patients can be minimally symptomatic or in remission on one immunotherapeutic agent alone.

●Bridge therapy for patients intolerant to glucocorticoids – For patients with MG in whom it is especially desirable to avoid glucocorticoids (such as those with poorly controlled diabetes) or for those who are not successfully weaned to lower doses of prednisone, we often use immunotherapeutic agents with a quicker onset of effect until the more slowly acting immunotherapy takes effect (table 1). Options include:

•Intravenous immune globulin (IVIG)

•Plasmapheresis

•Efgartigimod

•Rozanolixizumab

•Ravulizumab

•Zilucoplan

•Eculizumab

The most common regimen has been monthly courses of IVIG. Monthly plasmapheresis is an alternative to IVIG but is frequently limited by venous access problems associated with frequent use [1]. Efgartigimod, rozanolixizumab, ravulizumab, zilucoplan, and eculizumab may play a similar role in these patients, but experience with these agents in this setting is limited. (See "Overview of the treatment of myasthenia gravis", section on 'Plasma exchange and IVIG as rescue or bridge therapies' and 'Antibody-based biologic therapies' above and 'Eculizumab' below.)

REFRACTORY DISEASE

Definition and general approach — Approximately 10 percent of patients with severe MG are refractory to, or are limited by the specific toxicities of, the first-line glucocorticoid-sparing oral therapies (azathioprine, mycophenolate, cyclosporine, and tacrolimus) [59,60]. Some require unacceptably high doses of glucocorticoids despite concurrent use of these agents. Refractory disease is generally considered to include patients with any of the following:

●Disease that is unchanged or worse after glucocorticoids and at least one other immunotherapeutic agent, used in adequate dose and duration, with persistent symptoms or side effects that limit function

●Need for ongoing rescue therapy with intravenous immune globulin (IVIG) or plasma exchange, or frequent myasthenic crises, while on immunotherapy

●Intolerable adverse reactions or the presence of comorbid illnesses that preclude use of conventional immunotherapies

There is variability in how the glucocorticoid-sparing therapies are used and ordered among experts. Most would require a minimum of glucocorticoids plus an adequate trial (ie, one year or more) of either azathioprine or mycophenolate (but not both), and perhaps efgartigimod, rozanolixizumab, ravulizumab, or zilucoplan before considering a patient to have refractory disease. In such patients, some would also consider a trial of cyclosporine or tacrolimus before moving on to the treatment options discussed below.

Although there are no evidence-based guidelines in patients with refractory disease, more aggressive treatment is generally warranted, and the approach is highly individualized [1,7]. Eculizumab may be an effective therapy and has been approved by the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) for this use in refractory acetylcholine receptor (AChR) antibody-positive patients [7]. Chronic, or maintenance, IVIG or plasma exchange (at regular intervals) can be used successfully in patients with refractory generalized MG of any serologic status. Alternative agents include rituximab or cyclophosphamide [1,7]. Rituximab is particularly efficacious in muscle-specific tyrosine kinase (MuSK)-positive MG and should be considered early in the course of the disease. Other biologic therapies may be used for select patients with refractory disease, although experience in this setting is limited.

Pharmacologic options

Eculizumab — Complement-mediated injury to the postsynaptic membrane likely plays a major role in the pathogenesis of MG. Eculizumab is a humanized monoclonal antibody that binds to C5 and inhibits the formation of C5b-induced membrane attack complex. Based on the available data and considering the potential for infectious complications and lack of comparative and long-term safety and efficacy data, eculizumab is most appropriate to consider in selected refractory patients with severe disease or dependence on maintenance IVIG or plasma exchange [7].

Eculizumab was first reported to be of possible benefit in a crossover trial of 14 patients with AChR-positive, refractory MG [61]. A larger multicenter trial of eculizumab (REGAIN) was then performed in 125 patients with generalized AChR-positive MG who had failed at least two immunosuppressive therapies, one of which could have been chronic IVIG or plasma exchange [62]. Although the trial failed to show a statistically significant difference in the MG Activities of Daily Living (MG-ADL) total score at 26 weeks (rank-based treatment difference -11.7 points favoring eculizumab, 95% CI -24.4 to 0.96), the majority of other prespecified endpoints favored eculizumab over placebo, including the proportion of patients with ≥5 point improvement in the quantitative MG scale score (45 versus 19 percent) and the rate of MG exacerbations (10 versus 24 percent). On the basis of these data, eculizumab was approved by the FDA and EMA for use in refractory AChR antibody-positive patients. In an open-label extension of the REGAIN trial in which 117 patients were treated with eculizumab 1200 mg every two weeks for up to three years (median 23 months), functional gains were maintained over time, and the exacerbation rate in the open-label extension period was 75 percent lower than the one-year pre-study baseline rate (25 versus 102 events per 100 patient-years) [63]. Other outcomes also supported a sustained benefit, including hospitalization rates and use of rescue therapies. Changes in glucocorticoid dosing and other immunosuppressive therapies were not reported.

Eculizumab is given by intravenous (IV) infusion starting at least two weeks after vaccination for meningococcal meningitis. The dose in MG consists of 900 mg infused weekly for four weeks, followed by 1200 mg a week later, and then 1200 mg every two weeks. The optimal duration of therapy in responding patients is not yet known. Some experts who have treated patients for longer than six months indicate that they try to lengthen the interval between infusions at a point when the patient appears to have maximally responded. In the REGAIN trial, clinical benefit appeared to occur rapidly in responders, and therefore eculizumab should probably be discontinued if no clinical improvement has occurred by three to four months [62-64].

Eculizumab increases the risk of life-threatening Neisserial infections, including N. meningitidis, and administration of eculizumab has been associated with a 1000-fold to 2000-fold increased incidence of meningococcal disease. Patients should be immunized with meningococcal vaccines (both quadrivalent ACYW serogroup conjugate and serogroup B) at least two weeks prior to administering the first dose of eculizumab. In addition, daily oral antibiotic prophylaxis should be considered, as described separately, because invasive meningococcal disease has occurred among patients receiving eculizumab despite receipt of meningococcal vaccine [65]. In the United States, eculizumab is available only through a restricted program under a Risk Evaluation and Mitigation Strategy (REMS). (See "Treatment and prevention of meningococcal infection", section on 'Patients receiving C5 inhibitors'.)

Otherwise, eculizumab infusions are generally well tolerated. Commonly reported adverse events in clinical trials include headache (16 to 38 percent), nasopharyngitis (15 to 32 percent), diarrhea (13 to 23 percent), upper respiratory infection, arthralgias, and nausea. Risk of infection, especially with encapsulated bacteria, is increased with eculizumab due to blockade of terminal complement activation. In the REGAIN open-label extension study with a median follow-up of two years, 19 percent of patients developed an infectious event of special interest, including five cases of sepsis, septic shock, or pseudomonas sepsis and one case each of aspergillus, cytomegalovirus, and pseudomonas infection; there was one case of meningococcal infection after the data cutoff that resolved with antibiotics [63].

Rituximab — Rituximab is a monoclonal antibody against B cell membrane marker CD20. Originally developed to treat B cell lymphomas, there is a good theoretical rationale for its use in MG, a B cell mediated disease. It has proven to be relatively safe in other autoimmune disorders.

A large and growing number of case reports and case series support the potential efficacy of rituximab in patients with refractory MG [66-70]. In a prospective study of 22 patients with refractory MG treated with rituximab, the mean time to relapse was 17 months. Among 14 patients taking prednisone, the mean daily dose decreased from 25 mg at baseline to 7 mg after treatment with a mean follow-up of 29 months [71]. In an observational study of 72 patients with new-onset or refractory generalized MG, those treated with low-dose rituximab had shorter time to remission, lower use of adjunctive treatments, and fewer adverse events than patients treated with conventional immunotherapy [72]. A retrospective study of 40 patients with MG found patients age ≥65 years old improved with rituximab at rates similar to younger patients [73].

Results of a randomized phase II trial of rituximab in patients with AChR antibody-positive, generalized MG did not show a significant steroid-sparing effect at one year compared with placebo [74]. However, rituximab may have benefit for some patients with early MG. In a small trial of patients with AChR antibody-positive MG ≤12 months duration, patients assigned to a single dose of rituximab 500 mg had a greater likelihood of minimal manifestations of the disease at 16 weeks and reduced need for rescue medications than those assigned to placebo [75]. Additional data are warranted to assess longer term benefit and help identify patients for early treatment with rituximab.

Rituximab appears to be particularly effective in patients with MuSK-positive MG, who often respond relatively poorly to first-line immunotherapies [66,76-80]. A 2017 systematic review identified 47 single case reports and cases series totalling 169 patients with MG treated with rituximab, of whom one-third (57 patients) were MuSK positive [66]. Compared with patients with AChR-positive disease, those with MuSK-positive MG were more likely to respond to treatment, with higher rates of minimal disease manifestations or better (72 versus 30 percent) as well as pharmacologic or complete sustained remission for one year or more (47 versus 16 percent). In a multicenter retrospective study that included 77 patients with MuSK-positive MG who were deemed prospectively by blinded reviewers to be reasonable candidates for a trial of rituximab versus placebo within the first year of therapy, the 24 patients who ultimately received rituximab were more likely to have a favorable outcome than the 31 patients who did not receive rituximab (58 versus 16 percent) [80]. Favorable outcome was defined using a novel scale combining postintervention disease status and number and dose of immunotherapies used. The percentage of patients on prednisone at last follow-up was also lower in patients who received rituximab (29 versus 74 percent).

The optimal dose and dosing interval of rituximab for MG has not been established. A common initial dose is 375 mg/m² weekly for four doses. Others administer 1000 mg as a fixed dose and repeat in two weeks for a total of 2000 mg. Whether and when to repeat rituximab dosing in patients who respond to the initial course is generally guided by clinical symptoms and lymphocyte recovery. Patients who do not respond to the initial regimen can be redosed at six months; such patients who fail to respond to a second course are probably best considered refractory to rituximab [80].

Rituximab is well tolerated by most patients. Both minor and serious infusion reactions may occur, and patients require both premedication and monitoring during and immediately after all infusions. (See "Infusion-related reactions to therapeutic monoclonal antibodies used for cancer therapy".)

Patients may be more susceptible to certain infections. Rare cases of progressive multifocal leukoencephalopathy (PML) have been reported in patients treated with rituximab, but it is unknown if rituximab increases the risk of PML [57,58]. (See "Progressive multifocal leukoencephalopathy (PML): Epidemiology, clinical manifestations, and diagnosis".)

Other biologic therapies — Other biologic therapies may also be used in patients with refractory MG. Options may include efgartigimod, rozanolixizumab, ravulizumab, and zilucoplan. However, clinical experience with these agents for patients with refractory MG is limited. (See 'Antibody-based biologic therapies' above.)

Maintenance IVIG or plasma exchange — Periodic administration of IVIG or plasma exchange is sometimes used to maintain remission in patients with MG that is not well controlled despite the use of chronic immunomodulating drugs. There are no studies comparing this strategy with other options for refractory disease, but in the author's experience, maintenance IVIG is necessary and useful in some patients when other strategies have failed. (See "Overview of the treatment of myasthenia gravis", section on 'Plasma exchange and IVIG as rescue or bridge therapies'.)

Cyclophosphamide — Cyclophosphamide is an alkylating agent that reduces proliferation of both B and T cells. It can be effective in MG, but it has several serious potential side effects. Since there are effective agents with less toxicity, as detailed above, cyclophosphamide is usually reserved for the small percent of cases that are refractory to the other immunotherapies.

Administration of monthly high-dose IV cyclophosphamide has demonstrated equally good efficacy and much lower toxicity than daily oral cyclophosphamide in other autoimmune disorders, such as lupus nephritis [81]. Although data are limited, the efficacy of pulsed cyclophosphamide in MG is suggested by the following reports:

●In a randomized, controlled trial of 23 patients with MG who were either poorly controlled or had significant glucocorticoid-related side effects, patients received their usual medications plus IV pulse cyclophosphamide (initial dose 500 mg/m² of body surface area) or placebo [82]. Subsequent doses of cyclophosphamide were titrated according to changes in muscle strength and side effects. The dose was not modified as long as clinical improvement occurred and no side effects were observed. However, the subsequent dose was increased by 25 percent of the previous dose if there was no clinical improvement or side effects or increased by 10 percent if the previous dose was higher than 1000 mg/m² body surface area. The maximum single dose was 2000 mg/m² body surface area. Regardless of muscle strength, the subsequent dose was decreased by 25 percent if there were side effects or if the white blood cell count was <3000/mm³ or the polymorphonuclear cell count was <2000 mm³. The cyclophosphamide treatments were given monthly for six months, then every other month, for a total of nine treatments. At one year, the following observations were reported:

•There was a significant reduction in the mean daily prednisone dose in the treatment group compared with the placebo group (14.4 mg versus 32.5 mg).

•Five of 12 cyclophosphamide-treated patients were able to get off glucocorticoids altogether, but none were able to do so in the placebo group.

•There was also a significant increase in muscle strength measures in patients in the treatment arm versus the placebo patients.

There were no dropouts in the study due to toxicity of treatment. The incidence of hematologic and infectious complications was the same in both groups. During the study and two subsequent years of follow-up, there were no cases of hemorrhagic cystitis, cardiomyopathy, or pneumonitis. One cyclophosphamide-treated patient developed bladder carcinoma two years after the therapy was administered.

●In a retrospective report, eight patients with MG refractory to conventional immunotherapy received pulsed IV cyclophosphamide (750 mg/m² every four weeks for six months) followed by oral immunosuppression with a conventional agent [83]. Six patients improved within three months of treatment, while two were nonresponders. At a mean follow-up of 31 months, four patients remained in clinical remission.

A novel approach to refractory patients with myasthenia involves the use of high-dose cyclophosphamide as immunoablative therapy [84,85]. This approach remains investigational.

The complications of oral daily cyclophosphamide use are myriad and include anorexia, nausea and vomiting, alopecia, persistent leukopenia, hemorrhagic cystitis, and increased risk of malignancies. As a result, oral daily therapy is seldom used. As noted in the studies above, monthly pulse cyclophosphamide is associated with fewer of these symptoms [81,82]. The risk of late malignancy, including bladder cancer, remains a concern. (See "General principles of the use of cyclophosphamide in rheumatic diseases".)

Investigational therapies — The available immunotherapies for MG are effective. However, their utility is limited by their broad effects on the immune system. Their general immunosuppressive properties also underlie many of their potential serious side effects. Improved understanding of the immune events in MG should allow more focused, immune-based therapies. Experimental immunologic approaches to target T cell signaling pathways, T cell–B cell interactions, B cells, Fc receptors, complement pathways, and proinflammatory cytokines are being considered in MG [7,86-89]. Chimeric antigen receptor T cell therapy (CAR-T) uses autologous lymphocytes combined with the extracellular target antibody domain to enhance T-cell-activated immune response. It is typically used for advanced hematologic cancers. Preliminary data suggest this approach may also have benefit for patients with MG [90].

Additional data are warranted to assess the long-term safety and benefit of these agents.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Myasthenia gravis" and "Society guideline links: Thymomas and thymic carcinomas".)

SUMMARY AND RECOMMENDATIONS

●Glucocorticoid initial therapy – For most patients with myasthenia gravis (MG) who remain significantly symptomatic on pyridostigmine or who become symptomatic after a temporary response, we recommend glucocorticoids as initial therapy (Grade 1B).

•Typical dosing – A typical regimen is oral prednisone, started at 20 mg daily, with gradual dose escalation by 5 mg every three to five days to a usual target dose of 60 mg per day (or 1 mg/kg per day, maximum 80 mg daily). Gradual dose escalation is advised to avoid the early transient worsening associated with high-dose glucocorticoids. (See 'Glucocorticoids' above.)

•Subsequent tapering – To reach the lowest dose necessary to maintain disease stability, glucocorticoids should be gradually tapered approximately one month after an effective clinical response has been achieved. We typically reduce prednisone by 5 to 10 mg per month until a daily dose of 30 mg is reached then even more slowly (eg, no more than 5 mg per month). (See 'Approach to tapering' above.)

•Monitoring – Patients should be monitored regularly for side effects and toxicities of glucocorticoids. Calcium and vitamin D supplementation should be used to reduce bone mineral loss, and Pneumocystis pneumonia prophylaxis should be considered in patients on prolonged courses of steroids, particularly at doses of prednisone ≥20 mg daily and other risk factors for immunosuppression. (See 'Side effects and monitoring' above.)

●Glucocorticoid-sparing therapy – For patients with an insufficient response to glucocorticoids, inability to taper steroids below a reasonably acceptable level without return of symptoms, or intolerance of chronic steroids, we recommend the addition of steroid-sparing immunotherapy (Grade 1B). Our preferred first-line agents vary based on individual patient and disease factors:

•Patients with AChR-positive or seronegative MG – For most patients with acetylcholine receptor (AChR) antibody-positive or seronegative MG, we suggest either azathioprine or mycophenolate mofetil as initial steroid-sparing therapy (Grade 2C). Alternative options include cyclosporine, tacrolimus, efgartigimod, rozanolixizumab, ravulizumab, or zilucoplan. (See 'AChR-positive and seronegative MG' above and 'Administration of specific drugs' above.)

•Patients with MuSK-positive MG – For most patients with muscle-specific tyrosine kinase (MuSK)-positive MG, we suggest rituximab first line (Grade 2C). If cost or access to rituximab is prohibitive, azathioprine, mycophenolate mofetil, or rozanolixizumab remain reasonable to use as first-line steroid-sparing therapies. (See 'MuSK-positive MG' above and 'Rituximab' above.)

•Weaning from glucocorticoids – A slow prednisone taper is typically started once the minimum time to onset of clinical response of the steroid-sparing agent has passed. The clinical benefit of immunotherapy may be delayed for 6 to 12 months, with a maximal effect often not seen until one to two years of treatment, depending on agent used (table 1). (See 'Administration of specific drugs' above and 'Weaning from glucocorticosteroids' above.)

●Options for refractory disease – Approximately 10 percent of patients with severe MG are refractory to, or are limited by the specific toxicities of, the first-line immunotherapies. In these refractory patients, treatment options include antibody-based biologic therapies, maintenance IVIG or plasma exchange, and cyclophosphamide. (See 'Definition and general approach' above.)