Abstract
Objectives . We have assessed indications, duration and tolerability of treatment with mycophenolate mofetil (MMF) in patients with diffuse cutaneous systemic sclerosis (dcSSc), and compared clinical outcome with a control cohort treated with other immunosuppressive drugs.
Methods . The clinical records of 109 patients treated with MMF and 63 control subjects receiving other immunosuppressive drugs were reviewed. Data covering a 5-yr period from commencement of treatment or until last assessment date were collected.
Results . MMF and control groups were well-matched in terms of basic demographic and clinical parameters. Treatment with MMF was very well tolerated. Of all patients, 12% experienced adverse reactions with gastrointestinal (GI) tract disturbances and infections being most frequent. MMF was discontinued due to disease stabilization in 9%, side effects in 8% and no effect on the disease activity in 14% of the patients. There was a significantly lower frequency of clinically significant pulmonary fibrosis in the MMF-treated cohort ( P = 0.037) and significantly better 5-yr survival from disease onset and from commencement of treatment ( P = 0.027 and P = 0.012, respectively). There was no significant difference between the two groups in terms of modified Rodnan skin score and forced vital capacity (FVC) change.
Conclusions . MMF is very well tolerated and appears to be at least as effective as the other current therapies for dcSSc. Our results provide support for further evaluation of MMF in a prospective trial.
Introduction
Although the ultimate pathology of systemic sclerosis (SSc) involves vascular damage and organ-based fibrosis, there is substantial evidence of immunological dysfunction and autoimmunity in scleroderma patients [ 1 ]. SSc has the highest case-specific mortality of all the autoimmune rheumatic conditions, and recent studies support the use of immunosuppressive treatment for the severe forms of the disease.
Mycophenolate mofetil (MMF) is an antiproliferative immunosuppressive agent metabolized in the body to mycophenolic acid, which inhibits the de novo synthesis of purines and thus suppresses T- and B-lymphocyte proliferation. It was shown to provide better protection against acute renal and cardiac transplant rejection than azathioprine (AZA), with better long-term patient and graft survival [ 2–6 ], and it was also used successfully as rescue therapy in patients with steroid-resistant acute and chronic hepatic graft rejection [ 7 ]. MMF is currently licenced for use in prophylaxis of renal, cardiac and hepatic transplant rejection in combination with ciclosporin A (CsA) and corticosteroids. In recent years, MMF was shown to be effective in the treatment of lupus nephritis (LN), and is now also used for other systemic lupus erythematosus (SLE) manifestations [ 8–11 ].
The present study reviews indications, duration and tolerability of treatment with MMF in patients with SSc. A historical control cohort treated with other standard immunosuppressive therapies is used to compare frequency of major organ-based complications and survival.
Patients and methods
Study cohort
All of the subjects fulfilled the American College of Rheumatology (ACR) classification criteria for SSc. Through our pharmacy records, we identified 109 patients treated with MMF. A control group of 63 patients was selected through filtering our scleroderma database. The inclusion criteria were diffuse disease, actively treated with immunosuppressive drugs (other than MMF) and no major immunosuppressive treatment prior to first assessment. The clinical records of all subjects were retrospectively reviewed.
Clinical assessment
For baseline in both groups, we used the last assessment before the treatment was initiated.The collected data covered the period from baseline to the most recent assessment date for MMF-treated patients and a period of 5 yrs from the commencement of treatment or until the last assessment date for the control group. In addition to basic demographic and clinical features, we recorded details about treatment regimen and any adverse events. Disease onset was defined by the first non-Raynaud's manifestation. Skin involvement was quantified by modified Rodnan skin score (mRss). Pulmonary involvement was defined as presence of any degree of fibrosing alveolitis demonstrated on high-resolution computed tomography (HRCT). Clinically significant pulmonary fibrosis (PF) was defined as a documented 15% reduction in FVC from baseline or FVC < 55%. Heart involvement was defined as haemodynamically significant pericardial disease or arrhythmia, or new onset congestive heart failure. Renal involvement was defined as documented renal crisis (RC). Gastrointestinal (GI) involvement was recorded as present when there was any degree of involvement of any part of the GI tract, and muscle involvement was defined as the presence of muscle weakness and the increased levels of creatine kinase (CK).
Statistical analysis
For the statistical analysis we used MINITAB 14 statistical software. Log-rank test was used to compare the time to major event in the two groups of patients. A value of P < 0.05 was considered to be statistically significant.
Results
A comparison of the basic demographic parameters of the two patient groups are summarized in Table 1 . The only major difference between the two groups in terms of organ involvement was in the proportion of patients with renal scleroderma (23% of the MMF cohort and <5% of the controls).
Basic demographic parameters, overlap syndromes, auto antibodies and organ involvement of the MMF-treated patients and the controls
| MMF patients % (No) | Controls % (No) | |
|---|---|---|
| Patients | 100 (109) | 100 (63) |
| Female sex | 82.6 (90) | 81 (51) |
| Race | ||
| Asian | 9.2 (10) | 9.5 (6) |
| Black | 7.3 (8) | 6.3 (4) |
| Caucasian | 79.8 (87) | 79.4 (50) |
| Other | 3.7 (4) | 4.8 (3) |
| Diffuse disease | 92.7 (101) | 98.4 (62) |
| Age at onset (yrs) mean ± s.d . | 47 ± 13 | 45 ± 12 |
| Overlap | ||
| Total | 23.9 (26) | 22.2 (14) |
| >1 overlap | 3.7 (4) | 1.6 (1) |
| Polymyositis/dermatomyositis | 11.9 (13) | 14.3 (9) |
| Arthritis | 8.3 (9) | 7.9 (5) |
| Sjogren's | 1.8 (2) | 0 (0) |
| SLE | 3.7 (4) | 1.6 (1) |
| Vasculitis | 3.7 (4) | 1.6 (1) |
| Antibodies | ||
| Scl 70 | 32.1 (35) | 27 (17) |
| RNA polymerase | 23.9 (26) | 22.2 (14) |
| nRNP | 4.6 (5) | 3.2 (2) |
| U3RNP | 4.6 (5) | 6.3 (4) |
| ACA | 1.8 (2) | 1.6 (1) |
| PM/Scl | 1.8 (2) | 4.8 (3) |
| dsDNA | 4.6 (5) | 0 (0) |
| Ro | 5.5 (6) | 7.9 (5) |
| Sm | 1.8 (2) | 0 (0) |
| Non-defined Abs | 30.3 (33) | 30.2 (19) |
| Organ involvement | ||
| Skin | 100 (109) | 100 (63) |
| Raynaud's | 100 (109) | 100 (63) |
| GIT a | 78 (85) | 87.3 (55) |
| Lung | 56 (61) | 63.5 (40) |
| Joint | 19.3 (21) | 14.3 (9) |
| Muscle | 13.8 (15) | 15.9 (10) |
| Heart | 6.4 (7) | 6.3 (4) |
| Kidney | 22.9 (25) | 4.8 (3) |
| MMF patients % (No) | Controls % (No) | |
|---|---|---|
| Patients | 100 (109) | 100 (63) |
| Female sex | 82.6 (90) | 81 (51) |
| Race | ||
| Asian | 9.2 (10) | 9.5 (6) |
| Black | 7.3 (8) | 6.3 (4) |
| Caucasian | 79.8 (87) | 79.4 (50) |
| Other | 3.7 (4) | 4.8 (3) |
| Diffuse disease | 92.7 (101) | 98.4 (62) |
| Age at onset (yrs) mean ± s.d . | 47 ± 13 | 45 ± 12 |
| Overlap | ||
| Total | 23.9 (26) | 22.2 (14) |
| >1 overlap | 3.7 (4) | 1.6 (1) |
| Polymyositis/dermatomyositis | 11.9 (13) | 14.3 (9) |
| Arthritis | 8.3 (9) | 7.9 (5) |
| Sjogren's | 1.8 (2) | 0 (0) |
| SLE | 3.7 (4) | 1.6 (1) |
| Vasculitis | 3.7 (4) | 1.6 (1) |
| Antibodies | ||
| Scl 70 | 32.1 (35) | 27 (17) |
| RNA polymerase | 23.9 (26) | 22.2 (14) |
| nRNP | 4.6 (5) | 3.2 (2) |
| U3RNP | 4.6 (5) | 6.3 (4) |
| ACA | 1.8 (2) | 1.6 (1) |
| PM/Scl | 1.8 (2) | 4.8 (3) |
| dsDNA | 4.6 (5) | 0 (0) |
| Ro | 5.5 (6) | 7.9 (5) |
| Sm | 1.8 (2) | 0 (0) |
| Non-defined Abs | 30.3 (33) | 30.2 (19) |
| Organ involvement | ||
| Skin | 100 (109) | 100 (63) |
| Raynaud's | 100 (109) | 100 (63) |
| GIT a | 78 (85) | 87.3 (55) |
| Lung | 56 (61) | 63.5 (40) |
| Joint | 19.3 (21) | 14.3 (9) |
| Muscle | 13.8 (15) | 15.9 (10) |
| Heart | 6.4 (7) | 6.3 (4) |
| Kidney | 22.9 (25) | 4.8 (3) |
a GIT, gastrointestinal tract.
Of the MMF cohort, 68% had received other immunosuppressive drugs before MMF with anti-thymocyte globulin (ATG) (32.1%), AZA (18.3%), intravenous cyclophosphamide (ivCYC) and methotrexate (MTX) (14.7% each) being the most commonly used. The immunosuppressive agents used in the control group were AZA (39.7%), ATG (31.7%), d -penicillamine (30.2%), ivCYC (27%), oral CYC (20.6%), α-interferon (12.7%) and MTX (11.1%). The most frequent indication for treatment in both groups was severe skin involvement usually in the context of active disease (81.7% of MMF cohort and 68.3% of the controls) and progressive lung fibrosis (27.5% and 47.6%, respectively). In 80% of the patients, the treatment with MMF was started within 36 months of the disease onset. Most of the remaining 20% had already received other immunosuppressive treatments and either had not responded to it or had developed side effects.
Of the patients treated with MMF, 79% received the drug for a period of 1 yr or more, 59 for 12–36 months. Sixty six per cent were still taking the drug at the time of data collection. The most commonly used treatment regimen (73.4% of the patients) was 2 g daily. Of all the patients, 24% received 1.5 g daily or less, and the majority of these had SSc renal involvement. MMF was discontinued in 34% of the patients. Nine per cent had stable disease and did not require active treatment, and 2.8% developed new problems requiring specific drug therapy. In 13.8% of the patients, MMF had no effect on the disease activity (patients had progression of their skin or lung disease despite being on full dose MMF). Twelve per cent had adverse reactions, and two-thirds of these had to stop taking MMF. GI tract disturbances and infections were most frequent (4.6 and 3.7%, respectively).There was no correlation between the severity of the side effects and the dose of MMF.
The control group had higher prevalence of PF (14.3%) at baseline compared with MMF-treated patients (7.3%), nevertheless over a 5-yr period 12% of the MMF cohort developed PF compared with 19% of the controls ( P = 0.037). Figure 1 compares the survival of the two groups of patients from disease onset and from commencement of treatment. It is important to account for the fact that after the sixth year the number of patients at risk in the MMF group falls to half the initial. The 5-yr survival from disease onset was 95.4% among the MMF cohort and 85.7% among the controls ( P = 0.027), while the 5-yr survival from the commencement of treatment among the MMF-treated patients was 91.7% compared with 77.8% in the control cohort ( P = 0.012).
Five-yr survival from disease onset and the commencement of treatment in patients treated with MMF and control patients treated with other immunosuppressive agents.
Five-yr survival from disease onset and the commencement of treatment in patients treated with MMF and control patients treated with other immunosuppressive agents.
A comparison between the mean mRss of the MMF and the control cohorts revealed a very similar degree of change from baseline (median mRss was 26 in both groups) to year 5 (median mRss of the MMF cohort was 11, while it was 15 for the controls), and the overall trend was similar to the expected as part of the natural course of the disease. Both the groups had slight improvement in the lung function from baseline to year 5 with very similar rate of increase in the median FVC, which at baseline was 83% predicted for both groups, while at year 5 it was 94% predicted in the MMF-treated patients and 92% in the control group.
Discussion
In this study, we report the largest single-centre cohort of SSc cases so far treated with MMF. We have compared them with an actively treated control group that did not receive the drug. Overall, mortality and clinically significant pulmonary involvement in patients treated with MMF were lower than in the control cohort. MMF was generally well-tolerated and the side effects experienced were not life threatening.
MMF was developed as an alternative to AZA with more selective mode of action and, therefore, fewer side effects [ 12 ]. It is currently licenced for use in prophylaxis of renal, cardiac and hepatic transplant rejection. Although outside its licenced indications, MMF has been used for the treatment of LN with good results. Chan et al . [ 8 , 9 ] demonstrated that in patients with class IV LN combination of MMF and prednisolone is as effective as treatment with CYC and prednisolone followed by AZA and prednisolone but has much less adverse reactions. Contreras et al. [ 10 ] used monthly pulses of CYC for induction of remission in 59 patients with class III, IV and V LN followed by MMF, AZA or quarterly ivCYC as maintenance therapy. MMF was much better tolerated, and the event- and relapse-free survival among the MMF-treated patients was significantly better than the CYC-treated group with better (although not statistically significant) patient survival. Ginzler et al. [ 11 ] in a 24-week randomized, open-label trial demonstrated that MMF was superior to ivCYC in inducing remission in 140 patients with LN. The patients receiving MMF experienced fewer infections but had greater incidence of diarrhoea.
Currently there are only two studies assessing the use of MMF in SSc. Stratton et al . [ 13 ] treated 13 patients with recent-onset dcSSc with MMF as a maintenance therapy following induction with ATG. MMF was well-tolerated with GI side effects being the most common (three patients). Although two patients withdrew from the study because of ineffective treatment and one died before MMF treatment was started, the rest of the patients completed the study demonstrating improvement in skin score but no significant change in mean FVC. More recently Liossis et al. [ 14 ] in an open-label trial treated five patients with dcSSc and a recent-onset alveolitis with MMF and low-dose prednisolone for 4–6 months. There was significant improvement in DLCO and improvement in FVC, which almost reached statistical significance [ 14 ].
Although immunomodulatory therapies are often employed in dcSSc, there have been few positive controlled trials to support their use. Most of the studies evaluating CYC therapy in SSc have been uncontrolled. The best evidence for efficacy comes from two recently completed placebo controlled trials for scleroderma-associated pulmonary fibrosis. One of these, the scleroderma lung study (SLS), evaluated 162 patients over 12 months who received either placebo or oral CYC [ 15 ]. There was statistically significant benefit in favour of active treatment for FVC and skin score. The results of the fibrosing alveolitis in scleroderma trial (FAST) [ 16 ] also suggested superiority of ivCYC followed by oral AZA compared with placebo.
The 5-yr survival from disease onset of our MMF cohort was 95.4%, while the survival in the control group was 85.7%. In a prospective study of 249 Swedish patients Hesselstrand et al. [ 17 ] found the 5-yr survival, estimated from the time of enrolment in the study to be 86% with mean disease duration at referral 4.6 yrs. Mayes et al . [ 18 ] also report 5-yr survival from disease diagnosis among 706 cases of SSc in the United States to be 77.9%. Overall the survival in our control group seems to be comparable with that reported by the different authors while the survival in the MMF group is significantly higher.
A major limitation in our study is the retrospective design, which inevitably resulted in some missing data and a few of the patients did not have FVC and mRss measurements for every year of follow-up. All of the patients included in our study are from a single centre, and the decision of which immunosuppressive agent should be used for the treatment has been made by the same group of clinicians. In addition, the control patients often had an overlap syndrome or organ involvement requiring specific immunosuppressive treatment, which led to the PF over-representation in the group. Since renal crisis often occurs within the first year in SSc, before any major treatment is started, and because MMF is the immunosuppressant of choice in patients with renal impairment, there is an overrepresentation in the MMF group that is likely to reflect patient selection rather than drug effect. In addition, the patients included had different duration of follow-up. Only 58% of the MMF patients compared with 87% of the controls had a 3-yr follow-up and even fewer (26% of the MMF and 49% of the controls) had 5-yr follow-up.
In conclusion, our experience with MMF suggests that it is very well-tolerated, and may have comparable efficacy to other immunosuppressive agents currently used for the treatment of scleroderma. This could be more formally explored in prospective studies that compare similar patient cohorts treated with MMF or other immunosuppressive drugs, and ultimately by double-blind randomized controlled trials. Based on the encouraging results from this retrospective analysis further evaluation is warranted.
Acknowledgements
This study was supported by an unrestricted educational grant from Aspreva pharmaceuticals. Clinical research activities within the department are funded by Arthritis Research Campaign, The Raynaud's and Scleroderma Association and The Scleroderma Society.
This work was supported in part by an unrestricted educational grant from Aspreva Pharmaceuticals.
Comments