Comparing Two Medicines for Adults with Myasthenia Gravis - The PROMISE-MG Study [Internet]

Excerpt

Background: Myasthenia gravis (MG) is an autoimmune neuromuscular disease caused by autoantibodies directed at the neuromuscular junction. The estimated prevalence is 14 to 40/100 000. There are no comparative effectiveness studies of immunosuppressive treatments for MG.

Objectives: The Prospective Multicenter Observational Cohort Study of Comparative Effectiveness of Disease-Modifying Treatments for Myasthenia Gravis was designed to compare patient-reported outcomes (PROs) and clinician-reported outcomes (CROs) in patients with MG who receive oral azathioprine (AZA) vs oral mycophenolate mofetil (MMF) and to compare PROs and CROs in patients with MG receiving an adequate dosage and duration of AZA or MMF vs patients receiving AZA or MMF at lesser dose/duration.

Methods: A new-user design was used. Patients with MG were recruited when first seen at the study site and followed for up to 36 months. There were no study-imposed interventions; patients were treated with standard of care determined by the treating clinician. The primary outcome was the change from baseline in the Myasthenia Gravis Quality of Life 15 (revised) (MG-QOL15r), a patient-reported (PRO) MG-specific quality-of-life outcome measure. The co-primary outcome measure was a composite clinician-reported outcome (CRO) with two components: clinical improvement defined as achieving the Myasthenia Gravis Foundation of America Post-Intervention Status minimal manifestation or better and adverse effects of treatments being no more than grade 1 Common Terminology Criteria for Adverse Events side effects. The co-primary outcome was the proportion of patients in each treatment group achieving both components of this composite measure. Secondary outcomes were the changes from baseline in the MG Composite (MGC), MG-Activities of Daily Living Scale (MG-ADL), MG-Manual Muscle Testing (MG-MMT), and hospitalizations for MG. Clinically meaningful reductions in the measures were MG-QOL15r, ≥5 points; MG-ADL, ≥2 points; MGC, ≥3 points; and MG-MMT, ≥3 points. The interventions were oral AZA vs oral MMF and treatment with an adequate dosage and duration of AZA and MMF vs lesser dosage and duration. The adequate dosage and duration for AZA was 2 mg/kg/day for 12 months; for MMF the adequate dosage and duration was 2 g/day for 8 months. Descriptive statistics were used for baseline characteristics after multiple imputation of missing data. Wilcoxon rank-sum test was used for continuous variables, and chi-square test or Fisher exact test was used for categorical variables. Propensity score overlap weighting was used to evaluate the comparative effectiveness of the two agents. A weighted generalized linear regression model with appropriate link functions was used. Robust variance estimators were used to derive 95% CIs. Significance was set at P ≤ .05. Secondary outcomes were not adjusted for multiple comparisons.

Results: A total of 167 patients were enrolled; 82 received either MMF or AZA; four were excluded since their outcome measures suggested good disease control at onset of treatment with AZA/MMF, leaving 78 patients in the primary analysis set (47 patients, 60% received MMF and 31, 40% received AZA. A clinically meaningful change in MG-QOL15r was noted in 80.8% of those receiving MMF and 57.1% of those receiving AZA (risk difference [RD], 23.7%; 95% CI, 11% to 35%; P = 0.0003). There was a 3.2-point mean difference in the MG-QOL15r score reduction from start of treatment favoring MMF (95% CI, −7.7 to 1.18; P = 0.15), representing a 35% greater change favoring MMF (95% CI −82.5%-11%; P = 0.13). Approximately 48% of MMF treated patients vs. 28% of AZA treated patients achieved PIS-MM or better with grade 1 or less CTCAE medication side effects (mean difference 19.6% favoring MMF, 95% CI, 5% to 44%, p=0.12). The secondary outcomes were not significantly different between MMF and AZA. Ten of 78 (12.8%) patients required hospitalization for MG, 5 each in the AZA and MMF treated group (AZA vs. MMF 16% vs 10.6%). For MMF, the median time to achieve clinically meaningful improvement ranged from 9.5 months for MG-QOL15r to 3.68 months for MG-MMT, with 5.88 months for MG-ADL and 5.72 months for MGC. For AZA, median times were: MG-QOL15r - 10.25 months, MG-MMT - 3.45 months, MG-ADL - 5.72 months and MG-MMT - 3.45 months. The doses used in patients who did and did not achieve meaningful reduction in MGC were largely similar for both drugs. The dose range of MMF was similar to usually recommended doses, 1-3 gm per day in both groups. However, the doses of AZA were below the usually recommended doses of 1-3 mg/kg in both groups. AEs to AZA were noted in 11/34 (32%) patients. Hepatotoxicity, manifested by elevated serum levels of liver enzymes, was the most common AZA AE, seen in five patients, followed by influenza-like hypersensitivity reaction in three. MMF AEs were reported in 9/48 (19%) patients. The most common MMF AE was gastrointestinal disturbance (7/9). The median duration to maximal effect on MG-QOL15r was 9.5 months for MMF and for AZA, 10.25 months. Thirty-nine of 78 (50%) received adequate dosage and duration, 32 (41%) MMF, and 7 (8%) AZA. No substantial differences were noted between adequate dose and duration vs. lesser dose/duration.

Conclusions: From 57% to 89% of patients treated with either MMF or AZA achieved meaningful reduction in one or more of MG-QOL15r, MGC, MG-ADL or MG-MMT. The proportion of patients on MMF who obtained a clinically meaningful reduction in MG-QOL15r was significantly greater than those on AZA, but the co-primary and secondary outcomes were not significantly different between drugs, despite a larger effect size with MMF in many measures. Statistical imprecision, likely due to the small numbers, precludes the identification of significant differences between the drugs. AEs were potentially more serious in patients who received AZA.

Limitations: The generalizability of these results may be limited because this study was performed in tertiary academic centers. The number of patients on AZA and MMF was lower than planned because the total enrollment was low and the number of patients with ocular MG higher than anticipated; ocular MG is not usually treated with AZA or MMF.. The COVID-19 pandemic affected follow-up, with missed visits and missing data for CROs, which could not be performed during remote visits.