Comparative effectiveness of azathioprine and mycophenolate mofetil for myasthenia gravis (PROMISE-MG): a prospective cohort study.

Background: Myasthenia gravis is an autoimmune disorder of the neuromuscular junction. Treatment typically includes symptomatic oral cholinesterase inhibitors, immunosuppression, and immunomodulation. In addition to corticosteroids, azathioprine and mycophenolate mofetil are the most frequently used immunosuppressants in North America. We aimed to evaluate the comparative effectiveness of these two drugs, and to assess the effect of the dose and duration of treatment.

Methods: We did a prospective cohort study at 19 academic centres in Canada and the USA. We included patients (aged ≥18 years) with autoimmune myasthenia gravis, who were never treated with immunosuppressants. Treating clinicians determined the choice of medication, dose, follow-up intervals, and drug monitoring. Outcome measures and adverse events were recorded at each visit. We assessed two co-primary outcomes. The first was the patient-reported Myasthenia Gravis-Quality of Life 15-revised (MGQOL-15r) score, measured as the mean change from treatment initiation to the follow-up visit with the lowest score. A clinically meaningful reduction (CMR) in MGQOL-15r was defined as a 5-point decrease. The second was a composite clinical outcome of disease improvement (Myasthenia Gravis Foundation of America Post-Intervention Status Minimal Manifestations or better) and low adverse event burden (defined as grade ≤1 Common Terminology Criteria for Adverse Events). We also compared these outcomes in patients receiving an adequate dose and duration of azathioprine (≥2 mg/kg per day for at least 12 months) or mycophenolate mofetil (≥2 g per day for at least 8 months) and a lower dose or shorter duration of these agents. We used propensity score weighting with generalised linear regression models. This study is registered with ClinicalTrials.gov (NCT03490539).

Findings: Between May 1, 2018, and Aug 31, 2020, 167 patients were enrolled; 85 did not receive azathioprine or mycophenolate mofetil and were excluded. Four were excluded from outcome analyses because they had scores of 0 on an outcome measure at treatment initiation. Of the 78 patients included in analyses, 47 received mycophenolate mofetil (median follow-up 25 months [IQR 13·5-31·5]) and 31 received azathioprine (median follow-up 20 months [IQR 13-30]). The mean change in MG-QOL15r was -10·4 (95% CI -18·9 to -1·3) with mycophenolate mofetil and -6·8 (-17·2 to 3·6) with azathioprine (mean difference -3·3, 95% CI -7·7 to 1·2; p=0·15). 38 (81%) of 47 patients receiving mycophenolate mofetil and 18 (57%) of 31 receiving azathioprine had a CMR in MG-QOL15r (risk difference 24·0%; 95% CI -0·2 to 48·0; p=0·052). The clinical composite outcome was achieved in 22 (47·7%) of 47 patients who received mycophenolate mofetil and nine (28·1%) of 31 who received azathioprine (risk difference 19·6%, 95% CI -4·9 to 44·2; p=0·12). Descriptive analysis did not find a difference in the proportion of patients reaching a CMR in MG-QOL15r between the adequate dose and duration group and the lower dose or shorter duration group. Adverse events occurred in 11 (32%) of 34 patients who received azathioprine and nine (19%) of 48 who received mycophenolate mofetil. The most frequent adverse events were hepatotoxicity with azathioprine (five [15%] of 34) and gastrointestinal disturbances (seven [15%] of 48) with mycophenolate mofetil. There were no study-related deaths.

Interpretation: More than half of patients treated with azathioprine and mycophenolate mofetil felt their quality of life improved; no difference in clinical outcomes was noted between the two drugs. Adverse events associated with azathioprine were potentially more serious than those with mycophenolate mofetil, although mycophenolate mofetil is teratogenic. Lower than recommended doses of azathioprine might be effective, with reduced dose-dependent adverse events. More comparative effectiveness studies are required to inform treatment choices in myasthenia gravis.

Funding: Patient-Centered Outcomes Research Institute, Myasthenia Gravis Foundation of America.