Key Points
Question
Can a rapid-tapering regimen of prednisone be used in patients with generalized myasthenia gravis?
Findings
In a single-blind, parallel, randomized clinical trial including 117 patients with generalized myasthenia gravis, the rapid-tapering regimen was associated with a significantly greater proportion of patients who reached the minimal manifestation status of generalized myasthenia gravis and did not require prednisone treatment at 12 and 15 months.
Meaning
The rapid tapering of prednisone therapy appears to be feasible, beneficial, and safe in patients with generalized myasthenia gravis and warrants testing in other autoimmune diseases.
Abstract
Importance
The tapering of prednisone therapy in generalized myasthenia gravis (MG) presents a therapeutic dilemma; however, the recommended regimen has not yet been validated.
Objective
To compare the efficacy of the standard slow-tapering regimen of prednisone therapy with a rapid-tapering regimen.
Design
From June 1, 2009, to July 31, 2013, a multicenter, parallel, single-blind randomized trial was conducted to compare 2 regimens of prednisone tapering. Data analysis was conducted from February 18, 2019, to January 23, 2020. A total of 2291 adults with a confirmed diagnosis of moderate to severe generalized MG at 7 specialized centers in France were assessed for eligibility.
Interventions
The slow-tapering arm included a gradual increase of the prednisone dose to 1.5 mg/kg every other day and a slow decrease once minimal manifestation status of MG was attained. The rapid-tapering arm consisted of immediate high-dose daily administration of prednisone, 0.75 mg/kg, followed by an earlier and rapid decrease once improved MG status was attained. Azathioprine, up to a maximum dose of 3 mg/kg/d, was prescribed for all participants.
Main Outcomes and Measures
The primary outcome was attainment of minimal manifestation status of MG without prednisone at 12 months and without clinical relapse at 15 months. Intention-to-treat analysis was conducted.
Results
Of the 2291 patients assessed, 2086 did not fulfill the inclusion criteria, 87 declined to participate, and 1 patient registered after trial closure. A total of 117 patients (58 in the slow-tapering arm and 59 in the rapid-tapering arm) were selected for inclusion by MG specialists and were randomized. The population included 62 men (53%); median age was 65 years (interquartile range, 35-69 years). The proportion of patients having met the primary outcome was higher in the rapid- vs slow-tapering arm (23 [39%] vs 5 [9%]), with a risk ratio of 3.61 (95% CI, 1.64-7.97; P < .001) after adjusting for center and thymectomy. The rapid-tapering regimen allowed sparing of a mean of 1898 mg (95% CI, −3121 to −461 mg) of prednisone over 1 year (ie, 5.3 mg/d per patient, P = .03). The number of serious adverse events did not differ significantly between the slow- vs rapid-tapering group (13 [22%] vs 21 [36%], P = .15).
Conclusions and Relevance
In patients with moderate to severe generalized MG who require high-dose prednisone with azathioprine therapy, rapid tapering of prednisone appears to be feasible, well tolerated, and associated with a good outcome.
Trial Registration
ClinicalTrials.gov Identifier: NCT00987116
This parallel-group, randomized, single-blind clinical trial compares rapid- with slow-tapering dosing regimens of prednisone in patients with generalized myasthenia gravis who have achieved minimal manifestation status.
Introduction
Myasthenia gravis (MG) is an autoimmune disorder of the neuromuscular junction characterized by fluctuating weakness.1 It is often associated with autoantibodies directed toward nicotinic acetylcholine receptors and, less commonly, toward the muscle-specific tyrosine kinase protein or the lipoprotein-related protein 4.1 Currently, moderate (class III of the Myasthenia Gravis Foundation of America [MGFA] classification), severe (class IV), or mechanically ventilated (class V)2 generalized autoimmune MG that is not controlled by cholinesterase inhibitors (ie, pyridostigmine) are treated with corticosteroids and an immunosuppressant,3,4,5,6,7 usually azathioprine,7 and less frequently mycophenolate mofetil.6 There are various means of administering prednisone5; however, to our knowledge, no specific dosing protocol has been validated.
The prednisone dosage in treatment of moderate to severe MG is commonly gradually increased to 0.75 mg/kg on alternate days and reduced progressively when minimal manifestation status (MMS) is reached7,8 (eTable 1 in Supplement 1). This recommended regimen leads to high and prolonged corticosteroid treatment, as the mean daily prednisone dose exceeds 30 mg/d at 15 months and 20 mg/d at 36 months.7,8 Although effective, long-term use of corticosteroids is often associated with significant complications.9 Reducing or even discontinuing prednisone treatment without destabilizing MG is therefore a therapeutic goal in generalized MG. Thymectomy and use of immunosuppressive agents help reduce the cumulative dose of prednisone, but the prednisone dose usually remains relatively high for several years.7,8,10,11 We wondered whether different regimens could help wean patients with generalized MG from corticosteroid therapy without compromising efficacy. Therefore, we carried out a multicenter, single-blind, randomized clinical trial to determine whether higher initial corticosteroid doses followed by rapid tapering could increase the proportion of patients achieving MMS without using prednisone therapy at 12 months and without relapse at 15 months in comparison with the standard slow-tapering regimen.7,8
Methods
Study Design
The MYACOR trial was a 2-arm, parallel-group, rater-blinded, randomized clinical trial conducted in 7 French university hospitals. The overall study duration for an individual participant was 15 months. All investigators were specialized in the management of MG. The trial was conducted and reported according to the protocol, which is available in Supplement 2. Ethics approval was granted by the French ethics board Comité de Protection des Personnes Ile de France XI. The trial was conducted in accordance with the Helsinki Declaration.12 Written informed consent was obtained from all participants prior to enrollment; participants did not receive financial compensation. Data were deidentified. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline for randomized clinical trials.
Participants
The original inclusion criteria were (1) age 18 to 80 years; (2) a positive serologic test for acetylcholine receptors or muscle-specific tyrosine kinase protein antibodies, or, in the case of negative serologic tests, a positive edrophonium test and abnormal repetitive nerve stimulation; (3) an MGFA clinical classification of III, IV, or V2 (eTable 2 in Supplement 1); and (4) agreement to attend monthly assessments over 15 months. Exclusion criteria included (1) ongoing treatment with corticosteroids or azathioprine, (2) having taken prednisone in the past 2 months and azathioprine in the year before inclusion, (3) any contraindication to the use of corticosteroid or azathioprine therapy, (4) other associated disease that required treatment with corticosteroids or azathioprine, (5) invasive thymoma, (6) weight more than 100 kg, and (7) pregnancy. Participants could be taking appropriate anticholinesterase therapy and could undergo thymectomy.
In April 2010, 10 months after enrollment began, the exclusion criterion regarding delay of prednisone interruption before inclusion was reduced from 2 months to 1 month to facilitate enrollment. In June 2012, the study duration was extended for a further 12 months to attain the planned number of participants.
Randomization and Blinding
Participants were randomly assigned with a 1:1 ratio to the rapid- or slow-tapering arms of the prednisone protocol through a web-based centralized system, with randomization stratified by site. From April 2010, randomization was also stratified on preexisting thymectomy. Prescribing of prednisone and azathioprine began immediately after randomization (month 0).
The tapering of the prednisone dose depended on the MGFA postintervention status2 in both trial groups. Since the MMS and improved status were key categories of the MGFA postintervention classification for the conduct of the prednisone tapering, their definitions were specified using objective criteria (eTable 3 in Supplement 1) to standardize the clinical assessment. Postintervention status according to MGFA classification was assessed monthly by an investigator (including T.S., S.D., B.E., A.N.P., D.F., and B.C.) (ie, rater) blinded to group assignment. The MGFA postintervention status was then transmitted to a second distinct investigator (ie, prescriber), who prescribed prednisone according to the trial group (slow or rapid tapering) and MGFA postintervention status. A summary table of prednisone dose calculation was provided to the prescriber. The rater and prescriber were 2 distinct evaluators.
Prednisone treatment was given orally and began in the hospital. On discharge from the initial hospitalization and at the end of each monthly consultation, the necessary amount of prednisone for the following month was provided by the hospital pharmacy. The patient was asked to record the daily tablets taken in an adherence logbook and to return any tablets not taken.
Procedures
In the slow-tapering group, prednisone was given on alternate days, as described elsewhere.8 The initial dose was 10 mg and was increased by increments of 10 mg every 2 days up to 1.5 mg/kg of body weight on alternate days without exceeding 100 mg. This dose was maintained until MMS was reached and then reduced by 10 mg every 2 weeks until a dosage of 40 mg was reached, with subsequent slowing of the taper to 5 mg monthly. In the case of severe prednisone adverse effects reported by the blinded rater, the prednisone dose was decreased in the same or a faster way, according to the prescriber’s decision. If MMS was not maintained, the alternate-day prednisone dose was increased by 10 mg every 2 weeks until MMS was restored, and the tapering resumed as described above 4 weeks later (eTable 6 in Supplement 1).
In the rapid-tapering group, oral prednisone was immediately started at 0.75 mg/kg/d (eTable 6 in Supplement 1). The prednisone tapering protocol depended on the MGFA postintervention status.2 First, if the patient reached MMS at 1 month, the dose of prednisone was reduced by 0.1 mg/kg every 10 days up to 0.45 mg/kg/d, then 0.05 mg/kg every 10 days up to 0.25 mg/kg/d, then in decrements of 1 mg by adjusting the duration of the decrements according to the participant’s weight with the aim of achieving complete cessation of corticosteroid therapy within the planned time frame (ie, 126-140 days for this third stage of tapering) as reported in eTable 7 in Supplement 1. Second, if the state of MMS was not reached at 1 month but the participant had improved according to the MGFA postintervention status, the prednisone dose was decreased by 0.1 mg/kg every 20 days until the dose of 0.45 mg/kg/d was achieved, then 0.05 mg/kg every 20 days up to 0.25 mg/kg/d, and then by decrements of 1 mg as described in the preceding paragraph. If the participant reached MMS during this tapering process, the tapering of prednisone was similar to the sequence described in the first stage. Third, if MMS was not reached and the participant had not improved, the initial dose was maintained for the first 3 months; beyond that time, a decrease in the prednisone dose was undertaken as follows: decrease of 0.1 mg/kg every 20 days up to 0.45 mg/kg/d, then by 0.05 mg/kg every 20 days up to 0.25 mg/kg/d, after which the prednisone dose was not reduced further. However, if the participant improved, the tapering of prednisone followed the sequence described in the second category. In the event of major adverse effects noted by the blinded rater, the prednisone dose was reduced in the same manner as described in the first category.
In the best-case scenario, prednisone would be discontinued on day 326 in the slow-tapering regimen and before day 200 in the rapid-tapering regimen in a 60-kg patient. This theoretical difference indicates that the rapid-tapering regimen was designed for achieving a clinically significant reduction in corticotherapy duration, in comparison with the recommended slow-tapering regimen.7,8
In the event of an MG exacerbation (eTable 1 in Supplement 1), the participant was hospitalized and the dose of prednisone was routinely doubled. In the event of a more moderate aggravation, the prednisone dose was increased to the previous dose recommended in that participant’s prednisone tapering regimen (eTable 6 in Supplement 1).
For both groups, azathioprine was started at 50 mg/d for 1 week, then increased by 50 mg/d weekly to a maximum dose of 3 mg/kg/d, without exceeding 200 mg/d. Azathioprine therapy was interrupted in the event of severe adverse effects and recommended to be replaced by mycophenolate mofetil. We recommended that the oral dose of pyridostigmine did not exceed 300 mg/d. Plasmapheresis or intravenous immunoglobulin (IVIG) therapy was permitted for MG exacerbation but not for maintaining MMS.
Outcomes
The primary outcome was the proportion of participants having reached MMS without prednisone at 12 months and having not relapsed or taken prednisone between months 12 and 15. Prespecified secondary outcomes were (1) the cumulative dose of prescribed prednisone over 12 months, (2) the proportion of participants with MMS at 12 months, (3) the delay in reaching MMS, (4) the rate and time of MG worsening and exacerbations within the first 15 months, (5) the type and rate of prednisone-related complications, and (6) the frequency of IVIG and plasmapheresis over the 15-month study period.
Follow-up assessments were scheduled monthly until month 12 and then at month 15. Assessments included MG status, prednisone dose, and azathioprine dose. Prednisone- and azathioprine-associated complications were blindly assessed monthly by a general examination and standard biological tests.
Myasthenia gravis status was assessed by the blinded rater, using the MGFA classification,2 the Myasthenic Muscle score,13 and the MG Activities of Daily Living scale14 (eTables 2, 4, and 5 in Supplement 1). Hospitalization, treatment with IVIG or plasmapheresis for an MG exacerbation, and hospitalization for thymectomy were documented.
Statistical Analysis
The trial was powered to detect an absolute difference of 25 percentage points in the proportion of patients achieving the primary outcome between arms: an increase from 15% with slow tapering to 40% with rapid tapering, with power 80% and a 2-sided 5% significance level. Findings were considered significant at P < .05. Accordingly, the study was planned to recruit 114 participants.
Binary outcomes were analyzed by computing the Mantel-Haenszel risk difference and risk ratio stratified by group of centers (1 large center vs the others) and thymectomy at inclusion. For continuous outcomes, adjusted mean differences with 95% CIs were estimated using linear mixed models with random effects for the stratum (combination of group of centers and thymectomy), except when the data distribution was skewed, as ascertained by a Shapiro-Wilk test. In that case, a linear mixed model on ranked observations was used and the 95% CI for the adjusted mean difference was obtained by bootstrapping. For time-to-event outcomes, death was considered as a competing risk. The hazard ratio was estimated in a proportional cause-specific hazards model with a stratified baseline hazard. Proportions of participants with severe adverse events were compared by Fisher exact tests.
The primary analysis followed the intention-to-treat principle. Accordingly, missing components for the primary outcome were handled by multiple imputation; details are given in the Statistical Analysis Plan of the protocol (Supplement 2). Several sensitivity analyses for the primary outcome were performed. No missing data were specifically imputed for secondary outcomes but were considered when they were imputed to calculate the primary outcome. Since no hierarchical testing approach or correction for multiple testing was used, analyses of secondary efficacy outcomes should be considered as exploratory.
Two subgroup analyses for the primary outcome were not planned in the protocol but decided before data analysis. Those subgroups were participants with or without thymectomy at inclusion and participants with MG duration of less than 2 years or 2 years or more at inclusion. Analyses were conducted using R software, version 3.6.1 (R Foundation for Statistical Computing).
Results
Between June 1, 2009, and July 31, 2013, 2291 patients with generalized MG were evaluated; of these, 2086 patients did not fulfill the inclusion criteria, 87 declined to participate, and 1 patient registered after trial closure. A total of 117 patients were randomized, 58 to the slow-tapering and 59 to the rapid-tapering regimen. A total of 113 participants completed the study; 3 participants in the slow-tapering group and 1 participant in the rapid-tapering group dropped out prematurely (Figure 1).
Figure 1. Study Flow Diagram.
The population included 62 men (53%) and 55 women (47%); median age was 65 years (interquartile range [IQR], 35-69 years). Treatment groups were well balanced at baseline for age, MG characteristics, and thymectomy (Table 1). The primary outcome analysis showed that the proportion of participants who reached MMS free of corticosteroid treatment without relapsing at 15 months was significantly higher in the intervention group (23 [39%]) compared with the control group (5 [9%]) (risk ratio, 3.61; 95% CI, 1.64-7.97; P < .001). Subgroup analyses showed no interaction with thymectomy or MG duration (Table 2 and eTable 8, eTable 9, and the eFigure in Supplement 1).
Table 1. Demographic and Clinical Characteristics of Participants at Baseline.
| Characteristic | No. (%) | |
|---|---|---|
| Slow tapering (n = 58) | Rapid tapering (n = 59) | |
| Age, median (range), y | 58 (22-81) | 56 (18-80) |
| Sex | ||
| Female | 35 (60) | 20 (34) |
| Male | 23 (40) | 39 (66) |
| BMI, mean (SD) | 23.1 (4.4) | 24.6 (5.2) |
| Autoantibody profile | ||
| Anti-AChR | 45 (78) | 45 (79) |
| Anti-MuSK | 7 (12) | 8 (14) |
| Early MG onseta | 27 (47) | 29 (49) |
| Mean disease duration, median (IQR), y | 0.9 (0.5-1.8) | 1.0 (0.7-3.1) |
| Thymectomy | 22 (38) | 18 (31) |
| Years before inclusion, median (IQR) | 0.4 (0.1-8.8) | 0.2 (0.1-5.5) |
| Thymoma | 11 (58) | 8 (14) |
| MGFA classb | ||
| III | 28 (48) | 27 (46) |
| IV | 20 (34) | 15 (25) |
| V | 10 (17) | 17 (29) |
| MG-ADL score, median (IQR) | 7 (5-13) | 8 (5-14) |
| Myasthenic Muscle score, median (IQR) | 63 (47-80) | 58 (45-77) |
| Treatment at enrollment | 19 (33) | 18 (31) |
| Pyridostigmine, median (IQR), mg | 220 (60-300) | 180 (40-300) |
| Comorbidities | ||
| Hypertension | 11 (19) | 16 (27) |
| Diabetes | 2 (3) | 8 (14) |
| Osteoporosis | 4 (7) | 2 (3) |
| Psychological disorder | 1 (2) | 1 (2) |
Abbreviations: AChR, acetylcholine receptor; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); IQR, interquartile range; MG, myasthenia gravis; MGFA, Myasthenia Gravis Foundation of America classification; MG-ADL, Myasthenia Gravis Activities of Daily Living; MuSK, muscle-specific tyrosine kinase.
Myasthenia gravis onset before age 40 years.
Class III of the MGFA indicates moderate generalized weakness; class IV, severe generalized weakness; and class V, respiratory failure.
Table 2. Treatment Effect for the Primary and Secondary Outcomes.
| Characteristic | No. | Slow tapering (n = 58), No. (%) | No. | Rapid tapering (n = 59), No. (%) | Adjusted effect, RD (95% CI) | P value |
|---|---|---|---|---|---|---|
| Primary outcomea | 58 | 5 (9) | 59 | 23 (39) | 3.61 (1.64 to 7.97) | <.001 |
| Prespecified secondary outcomes | ||||||
| From inclusion to month 15 | ||||||
| Exacerbation | 58 | 6 (11) | 57 | 6 (11) | 0.1 (−11.5 to 11.4) | .99 |
| Time to exacerbation, median (IQR), d | 58 | 28 (20 to 74) | 59 | 204 (72-259) | HR, 0.70 (0.45 to 1.07) | .10 |
| Worsening, No. (%)b | 58 | 26 (50) | 57 | 30 (54) | 3.7 (−16.3 to 23.8) | .72 |
| Time to worsening, median (IQR), db | 58 | 223 (35 to 304) | 59 | 159 (105 to 218) | HR, 1.39 (0.80 to 2.42) | .24 |
| Treatment with IVIG or plasmapheresis, No. (%) | 58 | 6 (11) | 59 | 6 (10) | −0.1 (−11.6 to 11.4) | .99 |
| Time to IVIG or plasmapheresis, median (IQR), d | 58 | 42 (20 to 74) | 59 | 204 (72-259) | HR, 1.06 (0.33 to 3.40) | .92 |
| Prednisone-related complications, No. (%) | 56 | 51 (91) | 58 | 54 (93) | 2.4 (−8.6 to 13.5) | .66 |
| Azathioprine-related complications, No. (%) | 56 | 41 (73) | 57 | 38 (67) | −7.2 (−23.5 to 9.1) | .38 |
| At 12 mo | ||||||
| Minimal manifestation status, No. (%) | 58 | 35 (60) | 59 | 33 (56) | −3.1 (−20.6 to 14.5) | .73 |
| Time to achieve minimal manifestation status, median (IQR), d | 58 | 132 (89 to 227) | 59 | 155 (68-246) | HR, 0.70 (0.45 to 1.07) | .10 |
| Cumulated dose of prednisone, mean (SD), mg | 52 | 10 562 (3573) | 57 | 8724 (3955) | MD, −1898 (−3121 to −461) | .003 |
| Nonprespecified secondary outcomes | ||||||
| At 12 mo | ||||||
| MG-ADL score, mean (SD)c | 52 | 1.1 (2.6) | 50 | 1.5 (2.2) | NA | .05e |
| Myasthenic Muscle score, median (IQR)d | 51 | 93.2 (11.7) | 50 | 91.8 (11.0) | .22e | |
| Patients treated with prednisone, imputed, No. (%) | 58 | 48 (83) | 59 | 19 (32) | <.001 | |
| Prednisone dose, imputed, median (IQR), mg | 58 | 8 (2 to 15) | 59 | 0 (0-3) | <.001f | |
| Patients treated with azathioprine, No. (%) | 48 | 34 (71) | 47 | 39 (83) | .14 | |
| Azathioprine dose, median (IQR), mg | 48 | 150 (0 to 150) | 47 | 150 (100-150) | .43f | |
| Patients treated with pyridostigmine, No. (%) | 48 | 37 (77) | 47 | 35 (74) | .74 | |
| Pyridostigmine dose, median (IQR), mg | 48 | 120 (22 to 240) | 47 | 120 (5-240) | .74f | |
| At 15 mo | ||||||
| Minimal manifestation status, No. (%) | 55 | 37 (67) | 57 | 29 (51) | NA | .05 |
| MG-ADL score, mean (SD)c | 44 | 0.8 (1.5) | 46 | 1.7 (2.2) | .004c | |
| Myasthenic Muscle score, median (IQR)d | 44 | 89.8 (20.3) | 46 | 91.5 (11.7) | .91e | |
| Patients treated with prednisone, No. (%) | 53 | 28 (53) | 56 | 19 (34) | .07 | |
| Prednisone dose, median (IQR), mg | 53 | 2 (0 to 10) | 56 | 0 (0-4) | .06f | |
Abbreviations: HR, hazard ratio; IQR, interquartile range; IVIG, intravenous immunoglobulin G; MD, mean difference; MG-ADL, Myasthenia Gravis Activities of Daily Living; NA, not applicable; RD, risk difference.
Primary outcome was attainment of minimal manifestation status without prednisone treatment at 12 months and without relapse or prednisone resumption at 15 months.
Worsening included Myasthenia Gravis Foundation of America postintervention status classes Worse and Exacerbation (eTable 6 in Supplement 1).
The MG-ADL assesses the main functions impaired by myasthenia gravis and ranges from 0 (no disability) to 24 (major disability).
The Myasthenic Muscle score assesses weakness of the main muscles impaired by myasthenia gravis and ranges from 0 (maximal weakness) to 100 (no weakness).
Linear mixed models on ranks adjusted for ranked baseline score. All analyses were adjusted for group of center and thymectomy at baseline.
Linear mixed models on ranks.
Rapid tapering allowed sparing of a mean of 1898 mg (95% CI, −3121 to −461 mg) of prednisone over 1 year per patient (ie, 5.3 mg/d; P = .03). Changes in prednisone dose according to tapering group are depicted in Figure 2. The proportion of participants treated with prednisone and the prednisone dose tended to be lower in the rapid-tapering group at 15 months (proportion: slow tapering, 28 [53%] vs rapid tapering, 19 [34%]; P=.07; dose: slow tapering, 2 mg [IQR, 0-10 mg] vs rapid tapering, 0 mg [IQR, 0-4 mg]; P = .06). If the proportion of participants with MMS at 12 months did not significantly differ between the 2 groups (slow tapering, 35 [60%] vs rapid tapering, 33 [56%]; P = .73), it was greater at 15 months in the slow-tapering group (slow tapering, 37 [67%] vs rapid tapering, 29 [51%]; P = .05). The rate of MG exacerbation (slow tapering, 6 [11%] vs rapid tapering, 6 [11%]; P = .99) or worsening (slow tapering, 26 [50%] vs rapid tapering, 30 [54%]; P = .72) between inclusion and 15 months did not differ significantly between the 2 groups, nor did the use of plasmapheresis or IVIG (slow tapering, 6 [11%] vs rapid tapering, 6 [10%]; P = .99). The doses of azathioprine (slow tapering, 150 mg [IQR, 0-150 mg] vs rapid tapering, 150 mg [IQR, 100-150 mg]; P = .43) and pyridostigmine (slow tapering, 120 mg [IQR, 22-240 mg] vs rapid tapering, 120 mg [IQR, 5-240 mg]; P = .74) also did not differ significantly at 12 months (Table 2).
Figure 2. Changes in Prednisone Dose According to Tapering Group.
Three deaths were reported; 2 occurred in the slow-tapering group (Table 3). Serious adverse events were not statistically significantly more frequent in the slow- vs rapid-tapering group (13 [22%] vs 21 [36%]; P = .15) (Table 3). The overall number of complications did not differ significantly between the 2 groups (slow tapering, 22% vs rapid-tapering, 36%; P = .15) (Table 3; eTable 10 in Supplement 1). Diabetes was more frequently reported in the rapid-tapering group (4 [7%] vs 2 [3%]; P = .68), but there was a greater prevalence of preexisting diabetes in this group (Table 1; eTable 10 in Supplement 1).
Table 3. Serious Adverse Events.
| Event | Slow tapering (n = 58) | Rapid tapering (n = 59) | P valuea | ||
|---|---|---|---|---|---|
| Patients, No. (%) | No. of events | Patients, No. (%) | No. of events | ||
| Any serious adverse event | 13 (22) | 23 | 21 (36) | 34 | .15 |
| Deathb | 2 (3) | 0 | 1 (2) | 0 | .62 |
| Serious adverse events | |||||
| Infection | 3 (5) | 3 | 9 (15) | 10 | .13 |
| Diabetes | 2 (3) | 2 | 4 (7) | 4 | .68 |
| Injurious fall | 2 (3) | 2 | 3 (5) | 3 | >.99 |
| Pancreatitis | 2 (3) | 2 | 3 (5) | 3 | >.99 |
| Respiratory insufficiency | 3 (5) | 3 | 1 (2) | 1 | .36 |
| Osteoporosis | 2 (3) | 2 | 1 (2) | 1 | .62 |
| Hemorrhagic ulcer | 1 (2) | 4 | 1 (2) | 1 | >.99 |
| Hepatitis | 0 | 0 | 2 (3) | 2 | .50 |
| Manic episode | 1 (2) | 1 | 1 (2) | 1 | >.99 |
| Myopathy | 1 (2) | 1 | 1 (2) | 1 | >.99 |
| Cancer | 1 (2) | 1 | 0 | 0 | .50 |
| Cardiac arrest | 0 | 0 | 1 (2) | 1 | >.99 |
| Confusion | 0 | 0 | 1 (2) | 1 | >.99 |
| Delusion | 1 (2) | 1 | 0 | 0 | .50 |
| Electrolyte abnormalities | 0 | 0 | 1 (2) | 1 | >.99 |
| Heart failure | 0 | 0 | 1 (2) | 1 | >.99 |
| Leukopenia | 1 (2) | 1 | 0 | 0 | .50 |
| Neutropenia | 0 | 0 | 1 (2) | 1 | >.99 |
| Pulmonary embolism | 0 | 0 | 1 (2) | 1 | >.99 |
| Stroke | 0 | 0 | 1 (2) | 1 | >.99 |
P values compare the number of patients with at least 1 event. Events related to the course of the disease (eg, myasthenia gravis crisis) were not analyzed as serious adverse events because they were already part of efficacy criteria.
The cause of death was malignant stroke in 1 patient and acute respiratory failure in 2 patients.
Discussion
The main goal of this study was to assess whether prednisone can be safely and rapidly discontinued in patients with generalized MG treated with azathioprine. We developed a rapid-tapering regimen and compared it with standard slow tapering, which has been used for assessing prednisone-sparing effect of azathioprine7 and thymectomy,8 and since recommended by international experts.4 We found that slow tapering allows for discontinuing prednisone earlier than previously reported (ie, 15 months). The rapid-tapering regimen enabled an even-faster prednisone discontinuation (ie, before 12 months), with a 4-fold increase in the proportion of patients having reached MMS at 12 months, not receiving corticosteroid therapy, and without relapsing at 15 months. On the other hand, we found that the 2 regimens were comparable in terms of MG status, prednisone tapering at 15 months, and corticosteroid-related adverse effects.
Prednisone tapering depends on the efficacy of azathioprine, which had similar dosing between the 2 groups throughout the study period. Our results indicate that azathioprine has greater immunosuppressive and prednisone-sparing effects than previously reported.7 It is plausible that prednisone tapering would differ from another immunosuppressive agent and therefore the distinction between the 2 regimens would not be observed. However, azathioprine remains the first-line immunosuppressant usually recommended,4 indicating that our study is relevant for a large proportion of patients with MG. Moreover, our study will help in the design of trials on the corticosteroid-sparing-effect of new immunosuppressive agents.
It is noteworthy that prednisone was discontinued earlier than expected in the slow-tapering group. The mean daily prednisone doses at 12 and 15 months were 8 and 2 mg in the slow-tapering arm but above 20 mg in a trial on thymectomy.8 This difference indicates that the slow-tapering regimen in our trial has been fairly applied. It is likely that the monthly follow-up enabled optimal adjustment of the prednisone dose in both therapeutic groups.
Our findings cannot be ascribed to a confounding effect of MG characteristics15,16 or thymectomy. Myasthenia gravis disease duration, immunologic subtypes, and severity did not statistically significantly differ between these 2 groups. The subgroup analyses did not reveal any interaction with thymectomy for the primary outcome. This finding does not contradict the efficient role of thymectomy in controlling MG and in prednisone sparing.6 Therefore, the better outcome of the intervention group seems to be related to the differences in prednisone administration compared with the control group: (1) immediate high dose vs a slow increase of the prednisone dose, (2) daily vs alternate-day dosing, (3) earlier tapering initiation (at first clinical improvement vs MMS), and (4) faster tapering. The structure of the study did not allow us to identify which of these factors was responsible.
Limitations
Limitations of our trial design include its single-blind nature and short follow-up duration. It was deemed nonfeasible to develop a placebo for prednisone because this would have required participants to take too many pills to mask the alternate-day intake and the reduction in prednisone dosage.17 We opted for a 15-month follow-up considering that a longer follow-up period would be too demanding for patients and increase the risk of loss to follow-up. Furthermore, this follow-up duration has recently been used for assessing the prednisone-sparing effect of methotrexate and rituximab.18,19
One may argue that both tested regimens led to a comparable status and prednisone dose at 15 months, minimizing the benefit of rapid tapering. We think that the reduction of the cumulative dose over a year (equivalent to 5 mg/d) is a clinically relevant reduction, since the risk of complications is proportional to the daily or cumulative doses of prednisone.20 Moreover, Figure 2 shows that 5 mg represents more than 25% reduction of the daily dose beyond the sixth month. This reduction is particularly important when the daily dosage decreases to below 20 mg, which is a fundamental step in prednisone tapering. The 60% rate of MMS and very low MG-ADL values at 12 and 15 months indicate that the clinical outcome was satisfactory in the 2 groups but was obtained with much less prednisone in the rapid-tapering group. Our results warrant testing of a more rapid-tapering regimen in a future trial. In the meantime, our trial provides useful information on how prednisone tapering could be managed in patients with generalized MG treated with azathioprine.
Conclusions
The findings of this randomized, rater-blinded trial support the use of rapid tapering of prednisone in patients with generalized MG requiring combined corticosteroid and azathioprine therapy. Researching the best prednisone-tapering scheme is not only a major issue for patients with MG but also for other autoimmune or inflammatory diseases, because validated prednisone-tapering regimens are scarce.
eTable 1. Myasthenia Gravis Foundation of America (MGFA)—Post-Intervention Status
eTable 2. Myasthenia Gravis Foundation of America
eTable 3. Prespecified Criteria for the Minimal Manifestation and Improved Status of the MGFA Post-Interventions Classification
eTable 4. Myasthenia Gravis of Daily Living Scale (MG-ADL)
eTable 5. Myasthenic Muscle Score (MMS)
eTable 6. Tapering Regimens
eTable 7. Protocol of Decrease in the Experimental Group
eTable 8. Proportion of Participants Having Met the Primary Outcome
eTable 9. Treatment Effect for the Primary Outcome
eTable 10. Occurrence of Any Complication
eFigure. Subgroup Analyses of the Primary Outcome
Trial Protocol and Statistical Analysis Plan
Nonauthor Collaborators. The MYACOR Investigators Collaborators.
Data Sharing Statement
References
- 1.Berrih-Aknin S, Le Panse R. Myasthenia gravis: a comprehensive review of immune dysregulation and etiological mechanisms. J Autoimmun. 2014;52:90-100. doi: 10.1016/j.jaut.2013.12.011 [DOI] [PubMed] [Google Scholar]
- 2.Jaretzki A III, Barohn RJ, Ernstoff RM, et al. ; Task Force of the Medical Scientific Advisory Board of the Myasthenia Gravis Foundation of America . Myasthenia gravis: recommendations for clinical research standards. Neurology. 2000;55(1):16-23. doi: 10.1212/WNL.55.1.16 [DOI] [PubMed] [Google Scholar]
- 3.Gilhus NE, Verschuuren JJ. Myasthenia gravis: subgroup classification and therapeutic strategies. Lancet Neurol. 2015;14(10):1023-1036. doi: 10.1016/S1474-4422(15)00145-3 [DOI] [PubMed] [Google Scholar]
- 4.Sanders DB, Wolfe GI, Benatar M, et al. International consensus guidance for management of myasthenia gravis: executive summary. Neurology. 2016;87(4):419-425. doi: 10.1212/WNL.0000000000002790 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Farmakidis C, Pasnoor M, Dimachkie MM, Barohn RJ. Treatment of myasthenia gravis. Neurol Clin. 2018;36(2):311-337. doi: 10.1016/j.ncl.2018.01.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sanders DB, Hart IK, Mantegazza R, et al. An international, phase III, randomized trial of mycophenolate mofetil in myasthenia gravis. Neurology. 2008;71(6):400-406. doi: 10.1212/01.wnl.0000312374.95186.cc [DOI] [PubMed] [Google Scholar]
- 7.Palace J, Newsom-Davis J, Lecky B; Myasthenia Gravis Study Group . A randomized double-blind trial of prednisolone alone or with azathioprine in myasthenia gravis. Neurology. 1998;50(6):1778-1783. doi: 10.1212/WNL.50.6.1778 [DOI] [PubMed] [Google Scholar]
- 8.Wolfe GI, Kaminski HJ, Aban IB, et al. ; MGTX Study Group . Randomized trial of thymectomy in myasthenia gravis. N Engl J Med. 2016;375(6):511-522. doi: 10.1056/NEJMoa1602489 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pascuzzi RM, Coslett HB, Johns TR. Long-term corticosteroid treatment of myasthenia gravis: report of 116 patients. Ann Neurol. 1984;15(3):291-298. doi: 10.1002/ana.410150316 [DOI] [PubMed] [Google Scholar]
- 10.Howard JF Jr, Utsugisawa K, Benatar M, et al. ; REGAIN Study Group . Safety and efficacy of eculizumab in anti-acetylcholine receptor antibody-positive refractory generalised myasthenia gravis (REGAIN): a phase 3, randomised, double-blind, placebo-controlled, multicentre study. Lancet Neurol. 2017;16(12):976-986. doi: 10.1016/S1474-4422(17)30369-1 [DOI] [PubMed] [Google Scholar]
- 11.Howard JF Jr, Bril V, Burns TM, et al. ; Efgartigimod MG Study Group . Randomized phase 2 study of FcRn antagonist efgartigimod in generalized myasthenia gravis. Neurology. 2019;92(23):e2661-e2673. doi: 10.1212/WNL.0000000000007600 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Krleza-Jerić K, Lemmens T. 7th revision of the Declaration of Helsinki: good news for the transparency of clinical trials. Croat Med J. 2009;50(2):105-110. doi: 10.3325/cmj.2009.50.105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sharshar T, Chevret S, Mazighi M, et al. Validity and reliability of two muscle strength scores commonly used as endpoints in assessing treatment of myasthenia gravis. J Neurol. 2000;247(4):286-290. doi: 10.1007/s004150050585 [DOI] [PubMed] [Google Scholar]
- 14.Wolfe GI, Herbelin L, Nations SP, Foster B, Bryan WW, Barohn RJ. Myasthenia gravis activities of daily living profile. Neurology. 1999;52(7):1487-1489. doi: 10.1212/WNL.52.7.1487 [DOI] [PubMed] [Google Scholar]
- 15.Mao ZF, Mo XA, Qin C, Lai YR, Olde Hartman TC. Course and prognosis of myasthenia gravis: a systematic review. Eur J Neurol. 2010;17(7):913-921. doi: 10.1111/j.1468-1331.2010.03017.x [DOI] [PubMed] [Google Scholar]
- 16.Evoli A, Bianchi MR, Riso R, et al. Response to therapy in myasthenia gravis with anti-MuSK antibodies. Ann N Y Acad Sci. 2008;1132:76-83. doi: 10.1196/annals.1405.012 [DOI] [PubMed] [Google Scholar]
- 17.Collinson N, Tuckwell K, Habeck F, Chapman M, Klearman M, Stone JH. Development and implementation of a double-blind corticosteroid-tapering regimen for a clinical trial. Int J Rheumatol. 2015;2015:589841. doi: 10.1155/2015/589841 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Pasnoor M, He J, Herbelin L, et al. ; Methotrexate in MG Investigators of the Muscle Study Group . A randomized controlled trial of methotrexate for patients with generalized myasthenia gravis. Neurology. 2016;87(1):57-64. doi: 10.1212/WNL.0000000000002795 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Beat MG. Phase II trial of rituximab in myasthenia gravis. ClinicalTrials.gov identifier: NCT02110706. Updated March 6, 2020. Accessed December 28, 2020. https://www.clinicaltrials.gov/ct2/show/NCT02110706
- 20.van Staa TP, Leufkens HG, Abenhaim L, Zhang B, Cooper C. Oral corticosteroids and fracture risk: relationship to daily and cumulative doses. Rheumatology (Oxford). 2000;39(12):1383-1389. doi: 10.1093/rheumatology/39.12.1383 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Myasthenia Gravis Foundation of America (MGFA)—Post-Intervention Status
eTable 2. Myasthenia Gravis Foundation of America
eTable 3. Prespecified Criteria for the Minimal Manifestation and Improved Status of the MGFA Post-Interventions Classification
eTable 4. Myasthenia Gravis of Daily Living Scale (MG-ADL)
eTable 5. Myasthenic Muscle Score (MMS)
eTable 6. Tapering Regimens
eTable 7. Protocol of Decrease in the Experimental Group
eTable 8. Proportion of Participants Having Met the Primary Outcome
eTable 9. Treatment Effect for the Primary Outcome
eTable 10. Occurrence of Any Complication
eFigure. Subgroup Analyses of the Primary Outcome
Trial Protocol and Statistical Analysis Plan
Nonauthor Collaborators. The MYACOR Investigators Collaborators.
Data Sharing Statement