An international, phase III, randomized trial of mycophenolate mofetil in myasthenia gravis

Myasthenia gravis (MG) is an antibody-mediated, T cell-dependent autoimmune disorder that targets the acetylcholine receptor (AChR) at the neuromuscular junction in 85–90% of patients. The use of corticosteroids and immunosuppressants such as azathioprine has transformed MG management.¹ However, many patients do not tolerate or respond sufficiently to these drugs.² In addition, corticosteroids—the main current therapy for generalized MG³—have formidable side effects. These problems have prompted researchers to look at newer immunosuppressants,⁴ including mycophenolate mofetil (MMF).

The first report of MMF use in MG was its rapid induction of pharmacologic remission in a patient refractory to prednisone plus azathioprine or cyclosporine.⁵ The efficacy and safety of MMF in patients with MG have since been evaluated in case reports,^6–11 open-label prospective^12,13 and retrospective cohort^14–17 studies, and one double-blind randomized pilot study.^18,19 However, no international, large, randomized controlled trial of MMF—or of any other immunosuppressive drug—has been performed in MG.^20,21

Here we report the results of a phase III trial designed to study the efficacy and safety of MMF as a steroid-sparing adjunct in patients with mild-to-moderate, generalized, AChR antibody-positive MG. We also discuss methodologic and practical issues arising from one of the largest ever therapeutic trials in MG and implications for future studies.

METHODS

This prospective, randomized, double-blind, placebo-controlled, parallel-group, international, multicenter, 36-week trial assessed the efficacy and safety of adjunct MMF (CellCept, Roche Pharma AG, Grenzach-Wyhlen, Germany) in improving disease control while reducing corticosteroid dose in patients with MG. The study ran from August 2004 to October 2006. Initially, patients were recruited at 25 neurology centers in the United States, Canada, United Kingdom, Germany, and The Netherlands. In 2005, a further 18 centers in Ukraine, Israel, Spain, India, Italy, France, Russia, Serbia and Montenegro, and Mexico were added because of difficulties in recruiting. All patients gave written informed consent to participate in the study, which was conducted in accordance with the Declaration of Helsinki and the principles of Good Clinical Practice.

RESULTS

Patients.

Of the 176 patients enrolled, 88 (50%) were randomized to the MMF group and 88 (50%) to placebo. A total of 144 patients (81.8%) completed the study: 73 (83.0%) in the MMF and 71 (80.7%) in the placebo group (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Patient flow and reasons for withdrawal throughout the study

*All patients who were randomized received the allocated intervention and were included in the intention-to-treat analysis of efficacy. IVIg = IV immunoglobulin; MMF = mycophenolate mofetil; PE = plasma exchange.

There were no substantial differences in patient demographics or disease characteristics between the two treatment groups at baseline (table 1). Exposure to study drug after randomization, analyzed to assess compliance, was similar in the MMF group (treatment duration 229.7 ± 71.9 days; mean daily dose 1.97 ± 0.1 g/day) and the placebo group (236.0 ± 48.2 days; 1.97 ± 0.2 g/day).

View this table:

• View inline
• View popup
• Download powerpoint

Table 1 Baseline patient demographics and disease characteristics in the mycophenolate mofetil (MMF) and placebo treatment groups (intention-to-treat population)

Responses to treatment.

Both MMF and placebo groups responded to treatment. There were, however, no significant differences in any of the primary or secondary measures of response between the two groups. The primary endpoint of treatment response was achieved by 39 patients (44.3%) in the MMF group compared with 34 patients (38.6%) in the placebo group (p = 0.541; ITT population). On regression analyses, this outcome was unaffected by sex, age at disease onset, duration of MG, previous thymectomy, or center effects. The reasons why patients failed to achieve treatment response are given in table 2. Results across the two treatment groups for the secondary endpoints of QMGS (figure 2), ADL (figure e-1), and investigator and patient global assessments (figure e-2) are summarized in table 3. All were similar and consistent with those obtained for the primary endpoint.

View this table:

• View inline
• View popup
• Download powerpoint

Table 2 Reasons for failure to achieve a treatment response in the mycophenolate mofetil (MMF) and placebo groups (intention-to-treat population)

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Mean total Quantitative Myasthenia Gravis Score (QMGS) in the mycophenolate mofetil (MMF) and placebo groups

The endpoint includes the week 36 and early termination visits. Early termination analyses included the observed cases (no imputation) and missing data imputed using the last observation carried forward method for the week 36 visit.

View this table:

• View inline
• View popup
• Download powerpoint

Table 3 Responses to treatment: Efficacy results for the mycophenolate mofetil (MMF) and placebo groups on secondary outcome measures (intention-to-treat population)

DISCUSSION

Patients in the MMF group showed similar improvements to those in the placebo group on all efficacy measures examined. Although at first sight some data on softer endpoints, such as prednisone dose reduction, decline in AChR antibody titers, and mean length of hospitalization, seemed to suggest potential benefits of MMF over placebo, these differences were not significant. Thus, prednisone plus MMF for 36 weeks was not superior to prednisone alone in managing the selected study population of mild-to-moderate, AChR antibody-positive, generalized MG.

One possible conclusion is that MMF is ineffective as a steroid-sparing agent in the treatment of MG, contradicting evidence from smaller, less rigorous studies and the reported experience of many clinicians.^6–19 Several factors could account for this incongruity.

Treatment response was achieved by only 44% of patients in the MMF group, which is strikingly lower than the 70–75% response rate reported in a review of the use of MMF in MG.² One reason for the low response rate may be that our definition of treatment response was so rigorous that it masked lesser but clinically useful improvements.

The initiation of the steroid taper, as early as 2 weeks in patients who had achieved MM or PR by that time, may have been too early for a clinical effect of MMF to become apparent. Also, the study period of 36 weeks may have been too short, particularly as the proportion of patients who responded increased with duration of treatment and did not appear to have reached a plateau at week 36 in either group (figure e-6). An open-label pilot study of MMF in MG reported initial improvement at 2–8 weeks,¹² although a longer average of 10.8 weeks (range 4–40 weeks) was reported from a retrospective analysis of 85 patients.¹⁷ Treatment response to other immunosuppressants used in MG can take even longer. For example, azathioprine took more than 18 months to maximally benefit patients with MG in a trial of prednisolone vs azathioprine plus prednisolone.²⁵ In patients with mild disease in whom prednisone had been recently started, improvement may have been prednisone-related, making it hard to demonstrate additional benefits from MMF. Longer-term trials would be needed to demonstrate delayed initial responses and to study the long-term effects on maintenance of disease control and prevention of relapse months after prednisone discontinuation.

Our study entry criteria favored inclusion of older patients, those with milder disease, and (given the warnings against MMF use in pregnancy²⁶) men, and thus may have selected an unrepresentative MG population less likely to respond to MMF. They also had MG for an average of 3 years (by which time MG is usually well-controlled) and had been treated with differing doses and durations of corticosteroids before entering the study. Data are unavailable to stratify our results by prednisone dose before randomization to assess potential confounding effects of different pre-randomization prednisone regimens.

Our study was designed with the expectation of detecting at least a 30% difference in treatment response rate between the MMF and placebo groups. Smaller but still clinically meaningful differences may have been missed; their detection would require an even larger sample size than the 136 patients needed to power our study.

A further reason why our study failed to confirm a steroid-sparing effect of MMF may lie in the unexpected finding that the prednisone dose could be tapered relatively rapidly to 7.5 mg/day and remain effective for the duration of the study in many placebo-group patients. This target prednisone dose was chosen as it is considered to be below the level where chronic use would severely compromise bone mineral density.²⁷ The effect of MMF may have been masked by this corticosteroid dose and, potentially, cholinesterase inhibitors. This finding suggests that the current prednisone doses used in clinical practice may be higher than needed.

MMF was well tolerated with a safety profile consistent with that found in previous studies, except for a lower than expected incidence of diarrhea.^26,28

The completion of this study confirms that a large, international, multicenter, phase III study can be successfully performed in a rare, orphan autoimmune neurologic disease. As the first such drug trial in MG to incorporate MGFA recommendations²² on clinical research standards in full, it should aid the design of future studies of MG immunotherapies.

ACKNOWLEDGMENT

The authors thank Tom Potter and Sam Coates of Caudex Medical (supported by Aspreva Pharmaceuticals) for their editorial assistance with the preparation of the article.

Appendix

Investigators: O. Abramsky, MD, Neurology Department, Hadassah Medical Organization, Jerusalem, Israel; S. Apostolski, MD, Institute of Neurology, University of Belgrade School of Medicine, Serbia and Montenegro; R. Barohn, MD, Department of Neurology at the Landon Center on Aging, University of Kansas Medical Center, Kansas City; A. Berthele, MD, Klinik und Poliklinik für Neurologie der Technischen Universitaet Muenchen, Klinikum Rechts der Isar, Muenchen, Germany; J.L.M. Blanco, MD, Neurology Department, Hospital General Universitario Gregorio Maranon, Madrid, Spain; E. Bogdanov, MD, Neurology and Rehabilitation Cathedra, Kazan State Medical University, Kazan, Russia; E. Ciafaloni, MD, Department of Neurology, University of Rochester Medical Center, NY; M. de Baets, MD, Department of Neurology, Academical Hospital Maastricht, the Netherlands; C. Desnuelle, MD, Federation Maladies Neuromusculaires, Hopital de l’Archet 1, Nice, France; S. Dwivedee, MD, Department of Neurology, Vidysagar Institute of Mental Health and Neurological Sciences, New Delhi, India; M. Freimer, MD, Department of Neurology, Ohio State University, Columbus; G. Garcia Ramos, MD, Instituto Nacional de Ciencias Médicas y Nutricion “Salvador Zubirán,” Mexico City, Mexico; A. Gustov, MD, Regional Hospital/State Educational Institution (postgraduate medical education) “Nizhniy Novgorod State Medical Academy,” Nizhniy Novgorod, Russia; I.K. Hart, FRCP, University Division of Neuroscience, The Walton Centre for Neurology and Neurosurgery, Liverpool, UK; D. Hilton-Jones, MD, University Department of Clinical Neurology, John Radcliffe Hospital, Oxford, UK; J. Howard Jr., MD, Department of Neurology, University of North Carolina, Chapel Hill; I. Illa, MD, Neurology Department, Hospital Ste. Creu i Saint Pau, Barcelona, Spain; M. Inghilleri, MD, Dipartimento di Scienze Neurologiche, Universita degli Studi “La Sapienza,” Rome, Italy; S. Jander, MD, Neurologische Klinik, Heinrich-Heine-Universitat, Düsseldorf, Germany; A. Kozelkin, MD, School of Neurology, Zaporozhye State Medical University, Zaporozhye, Ukraine; G. le Masson, MD, Service de Neurologie 2a2, Hopital Pellegrin/CHU Bordeaux, France; R. Malamut, MD, Neurological Associates of Delaware Valley, CCMC Ambulatory Care, Upland, PA; R. Mantegazza, MD, Department of Neurology IV, Neurological Institute “Carlo Besta” Foundation, Milan, Italy; A.K. Meena, MD, Department of Neurology, Nizam’s Institute of Medical Sciences, Hyderabad, India; M.M. Mehndiratta, MD, Neurology, GB Pant Hospital, New Delhi, India; A. Melms, MD, Neurologische Klinik, Universität Tübingen, Germany; G. Meola, MD, Dipartimento di Neurologia, Policlinico san Donato, San Donato Milanese, Milan, Italy; M. Nicolle, MD, London Health Sciences Centre, University of Western Ontario, London, Canada; S. Ravat, MD, Department of Neurology, GS Medical College and KEM Hospital, Mumbai, India; D. Richman, MD, Department of Neurology and the Center for Neuroscience, University of California, Davis; S. Rivera Nava, MD, private clinic, Guanajuato, Mexico City, Mexico; A. Sanadze, MD, Institute of General Pathology, Russian Academy of Medical Sciences, Moscow, Russia; D.S. Saperstein, MD, Department of Neurology at the Landon Center on Aging, University of Kansas Medical Center, Kansas City; B. Schalke, MD, Klinik und Poliklinik für Neurologie der Universitat Regensburg, Germany; K. Sharma, MD, Department of Neurology, University of Miami School of Medicine, FL; M. Shevnyuk, MD, 1 Bohoutivska Str., Kiev, Ukraine; Z. Siddiqi, MD, Department of Medicine, University of Alberta, Edmonton, Canada; N. Spirin, MD, Department of Neurology at the City Clinical Hospital No. 8, Yaroslavl State Medical Academy, Russia; J. Sussman, MD, Department of Neurology at the Greater Manchester Neuroscience Centre, Hope Hospital, Salford, UK; A. Syal, MD, Arizona Neurological Institute, Sun City; P. van Doorn, MD, Department of Neurology, Erasmus MC, Rotterdam, The Netherlands; J. Verschuuren, MD, Department of Neurology, Leiden University Medical Centre, the Netherlands; P.V. Voloshyn, MD, Institute of Neurology, Psychiatry and Narcology, Ams of Ukraine, Kharkiv, Ukraine; G. Wolfe, MD, Department of Neurology, The University of Texas Southwestern Medical Center at Dallas.