Mycophenolate Mofetil for Maintenance Therapy of Wegener's Granulomatosis and Microscopic Polyangiitis

Wegener's granulomatosis (WG) and microscopic polyangiitis (MPA) are systemic vasculitides with small to medium vessel involvement. Together with idiopathic pauci-immune necrotizing glomerulonephritis (NCGN), which is considered a renal limited variant of MPA, they are the most common cause of rapidly progressive glomerulonephritis. These entities are frequently associated with antineutrophil cytoplasmic autoantibodies (ANCA), and are therefore categorized together as ANCA-associated systemic vasculitides (AASV) (1). Despite some differences in clinical presentation and prognosis between them, AASV are treated following a common therapeutic concept.

Untreated generalized WG has a poor prognosis with a mortality of 80% within the first year after diagnosis (2). The disease became treatable only when alkylating agents were combined with corticosteroids (3,4). The combination of oral cyclophosphamide (CYC) in a dose of 2 mg/kg per d with oral corticosteroids (OCS) in a starting dose of 1 mg/kg per d became the standard treatment of WG from the 1970s onward (5,6).

To stabilize remission and prevent relapses, CYC is usually maintained together with tapered OCS for up to 12 mo or longer after induction of remission. Together with the amount of CYC that will be administered for treatment of relapses, this policy results in a high cumulative dosage of CYC and increases drug-related toxicity. Serious CYC-related long-term complications have been identified by retrospective analyses of AASV patients (7,8,9). The increased risk for malignancies, e.g., 30-fold for bladder carcinoma and 10-fold for lymphomas, is the most serious complication of CYC treatment and will certainly contribute to the still-reduced life expectancy of WG patients (10). This dilemma has stimulated the search for less toxic but equally effective drugs to replace CYC for treatment of AASV (11,12).

Mycophenolate mofetil (MMF) is a strong immunosuppressive drug with rather low toxicity due to its lymphocyte-selective mode of action (13). Mycophenolic acid, the active metabolite of MMF, reversibly inhibits inosin-monophosphate-dehydrogenase, a key enzyme of “de novo” purine synthesis (14,15). Lymphocytic proliferation and function relies almost exclusively on de novo purine synthesis, whereas most other cells can also use the salvage pathway. MMF has been approved for prophylaxis and treatment of renal transplant rejection (16,17,18,19,20), for which it proved to be superior over azathioprine (AZA) (16,20). Encouraged by a preliminary experience with MMF in four AASV patients (21), this pilot study with MMF for maintenance therapy of AASV in patients with generalized disease was performed.

Materials and Methods

Results

Eleven consecutive patients with newly diagnosed and untreated AASV (WG, n = 9) and MPA, n = 2) were recruited for the study (Table 1). Their median age was 58 yr (range, 19 to 70) and they had active generalized disease (Table 1) with biopsy-proven pauci-immune necrotizing glomerulonephritis and a high median BVAS1 score of 24 (range, 10 to 34).

At diagnosis, renal failure (serum creatinine > 150 μmol/L) was present in 6 of 11 patients (Table 2), and in patients 1, 5, and 11 segmental glomerular crescents were found histologically. Patients 1 through 5, 7, and 11 presented with pulmonary infiltrates unresponsive to antibiotics; in patient 3, artificial respiration had to be performed for 4 d because of respiratory failure. CRP was elevated to a median of 12 mg/dl (range, 3 to 45), and the median hemoglobin was reduced to 9.5 g/dl (range, 7.3 to 13). All patients with WG were positive for C-ANCA with anti-PR3 antibody and patients with MPA for P-ANCA with anti-MPO antibody, respectively.

Standard induction therapy was given for a mean of 14 (range, 9 to 19) wk, in which a mean cumulative CYC dosage of 11.9 g (4.3 to 18) was reached (Table 3). Modifications of the standard induction therapy included additional plasma exchange or intravenous pulses of CYC or methylprednisolone (Table 3). In one patient, the induction therapy was switched to MMF after 9 wk because of persistent severe CYC-associated leukopenia.

When remission was achieved and medication was switched to maintenance therapy, new or deteriorated disease activity was absent in all patients (BAVS1 = 0), but grumbling disease activity of variable degree was still present, as indicated by a median BVAS2 of 5 (range, 0 to 11). The CRP had fallen to a median of 4 (range, 0 to 10). In 10 of the 11 patients, permanent organ damage attributable to AASV was present at remission as indicated by a median VDI of 2 (range, 0 to 5). Renal function could be preserved in all patients; at remission, the serum creatinine had fallen below 150 μmol/L in all but one of the six patients with renal failure (Table 2).

During maintenance therapy, no further decline of renal function occurred. In patient 1, renal function recovered only partially, and the serum creatinine stabilized between 170 and 190 μmol/L until the end of the trial. After switching from CYC to MMF 2 g/d, all but one patient remained in stable remission for 15 mo (BVAS = 0). The relapse occurred in patient 4 in the 14th month of maintenance therapy with recurrence of peripheral neuropathy, arthralgias, and constitutional symptoms (increase of BVAS1 score to 12). This patient with a high initial BVAS1 score of 34 had needed a prolonged course of induction therapy (18 wk) and had considerable, although declining, grumbling disease activity throughout maintenance treatment.

BVAS2 further decreased during maintenance therapy in all patients and became negative in 6 of 11 patients. Proteinuria gradually declined during remission treatment from a median of 0.5 g/d (range, 0.08 to 2.8) to a median of 0.2 g/d (range, 0.03 to 1.4) in all patients, and fell below 0.5 g/d in all except three patients (Table 2).

OCS could be reduced step by step as indicated in the guidelines, and a median daily dosage of 5 mg was reached by the end of the study. In three patients, OCS were discontinued several months before the end of the study (Figure 1). ANCA titers became negative in 6 of 11 patients by the end of the induction therapy, and two additional patients became negative during remission therapy. Three patients remained ANCA-positive throughout the study period. The patient with relapse was among those who became ANCA-negative early during induction therapy, but had an elevation of ANCA prior to and at clinical relapse.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Daily dosages of oral corticosteroids (OCS) during maintenance therapy. In three patients, OCS were discontinued before the study end.

A few adverse effects, probably drug-related, were observed during remission therapy, including abdominal pain (n = 3), diarrhea (n = 2), respiratory infection (n = 2), and leukopenia (<3000/μl) (n = 2). Leukopenia occurred in the second and third month of treatment with MMF and was reversible with dosage reduction to 1 g/d. The reduced dose was subsequently maintained until the end of the study. In one patient, MMF was stopped in the 14th month of treatment when a cytomegalovirus-colitis was diagnosed histologically. The patient remained in remission with low-dose OCS, and the colitis was successfully treated with ganciclovir.

Discussion

In this prospective pilot study of patients with WG and MPA, maintenance of remission for 15 mo was achieved with a combination of MMF and low-dose OCS. Only 1 of 11 recruited patients relapsed and the medication was generally well tolerated. The standard treatment of active AASV is a combination of oral CYC (2 mg/kg per d) and OCS (starting dose, 1 mg/kg per d), which leads to high remission rates and effective prevention of relapses, but carries the risk of serious acute and long-term adverse effects (7,8,9).

How these complications of AASV therapy can be avoided is currently unsolved. Strategies aim at reducing the total dosage and/or treatment time of CYC or replacing the drug from the beginning of treatment by a less toxic but still effective drug. The total dose of CYC can be significantly reduced by giving the drug as monthly intravenous pulses instead of daily oral therapy. Protocols using intravenous pulses of CYC were indeed associated with fewer and less serious adverse effects (25,26,27,28,29). However, although the remission rates of intravenous pulse and daily oral therapy are about the same, intravenous protocols were shown to be less effective (30,31) for prevention of relapses in WG patients. In contrast, when mixed populations of WG and MPA patients (26,27,28) and patients with panarteritis nodosa or Churg-Strauss syndrome (28,32) were studied, both treatment modalities were equivalent also for prevention of relapses. It remains therefore controversial whether giving less CYC by intravenous pulse protocols can be recommended for all patients with AASV.

Administration of methotrexate (MTX) instead of CYC from the beginning of treatment led to promising results in patients with limited WG (33,34). However, in patients with renal failure, such as in this pilot study, MTX cannot be given. AZA, when given for active AASV, was not effective (6,8). In limited WG, the antibiotic trimethoprim-sulfamethoxazol may induce remission and it may also be used as an additional drug for prevention of relapses during maintenance treatment (35,36,37).

The most widely used strategy to reduce the toxic burden of CYC is to limit the administration to the period of active disease (induction therapy) and then switch to a less toxic drug for maintenance therapy. Discontinuation of CYC at the time of remission without introduction of another immunosuppressant proved not to be safe, since it resulted in many early relapses (5). Instead of CYC, MTX was given for maintenance treatment of limited WG with good success (38), and AZA is currently investigated in a randomized controlled trial comparing it with CYC (12).

The new immunosuppressant MMF is currently studied in patients with different autoimmune diseases. In this pilot study, a low relapse rate of less than 10% within 15 mo of treatment with MMF and low-dose OCS was achieved in patients with WG and MPA. This relapse rate, which is lower than in previous treatment studies of AASV, is encouraging, but should be interpreted with caution for several reasons.

A relapse rate between 25 and 45% was published for the longer observation time of 3 to 5 yr after diagnosis of AASV (6,8) and can therefore not directly be compared with the results of this pilot study with a much shorter observation time. The median time to relapse was found to be as long as 27 mo (range, 2 to 168 mo) in a Swedish series of WG and MPA patients (8). Whether relapses occurring later than 15 mo after remission will be prevented by treatment with MMF and OCS is unknown.

Other factors that hamper the direct comparibility of outcomes between this study and published retrospective and prospective studies are differences in mixture of patient's diagnosis, severity of disease, and length and protocol of induction therapy. Finally, no common definition of relapse was applied. Maintenance immunosuppression means prolonged exposure to the side effects of the drug used, and a favorable side effect profile is therefore even more warranted than for drugs used for induction treatment. The low incidence of short-term adverse effects of MMF in this study is therefore as important as the demonstration of efficacy. A cytomegalovirus-colitis in one patient was the most serious complication. Since OCS could be reduced to less than 7.5 mg/d and even discontinued in some patients, this regimen may also help to avoid OCS-related side effects in AASV patients.

CYC-induced myelosuppression is not a rare problem in old patients with AASV. In this respect, the observation is certainly of interest that MMF could replace CYC already during induction treatment in one old patient with severe CYC-induced leukopenia. Although MMF itself may cause leukopenia, this patient could safely be treated with MMF without the need to increase the OCS dosage.

Besides short-term tolerability of MMF, the long-term safety of the drug will be crucial for the future of MMF-containing regimens in AASV. Thus far, long-term safety data from other studies look promising. The overall frequency of malignancies in the transplant studies in which MMF was combined with other immunosuppressive drugs was as low as 1%. In patients treated for rheumatoid arthritis, no malignancies occurred (39). Important data on long-term safety came from the experience in psoriasis patients with mycophenolic acid, the active compound of MMF, in the 1970s. The safety data of 85 patients who were treated with mycophenolic acid for up to 13 yr in comparatively high dosage (maximum dose of 7.4 g/d) showed no increased malignancy rate (40,41).

In conclusion, this pilot study shows that MMF is not only well tolerated but also effective in WG and MPA patients to maintain remission. Replacement of CYC in maintenance therapy of these diseases would be of great benefit, and therefore randomized trials comparing the efficacy of MMF with that of CYC should be started.