SEARCH

SEARCH BY CITATION

This issue of Arthritis & Rheumatism contains a report by the European Vasculitis Study Group (EUVAS) on the results of the second multicenter trial of antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis (1). This randomized open-label trial compared methotrexate (MTX) with cyclophosphamide (CYC) for induction of remission in “early systemic” ANCA-associated vasculitis. More than 30 years after the first report of successful remission induction with MTX in Wegener's granulomatosis (WG) (2), the current trial validates this treatment approach by showing that MTX is not inferior to CYC for remission induction in this subset of vasculitis patients. This important report justifies a review of what we know and don't know about the use of MTX in ANCA-associated vasculitis.

MTX for which patient?

  1. Top of page
  2. MTX for which patient?
  3. MTX for remission induction in WG
  4. MTX for remission maintenance
  5. REFERENCES

The interpretation of the literature on the topic is complicated by the lack of consistent disease definitions and subset designations. First, it should be clarified that, with the exception of 6 patients with microscopic polyangiitis (MPA) enrolled in the trial by the EUVAS investigators (1) and a recent report on a small number of patients (n = 11) with Churg-Strauss syndrome (3), all published data on MTX use in ANCA-associated vasculitis relate solely to WG. Second, the reader needs to be aware of the fact that different authors use different terms to describe very similar patient populations. On the other hand, the same term used by different authors may not necessarily designate exactly the same patient population. To facilitate the interpretation of the studies reviewed here, the disease subgroup designations are included in Table 1. Furthermore, to reflect specific author attribution, disease subgroup designations are placed in quotation marks throughout this report, and details on specific patient populations are provided as necessary for the appropriate interpretation of the reported findings. For a review of the ongoing lively discourse on the topic of disease subgroup definitions and of data-driven approaches to achieve better international consensus in this area, the reader is referred to other recent publications (4–6).

Table 1. Studies of methotrexate for remission induction
No. of patientsDisease subgroupRenal involvement (%); serum creatinine, mean mg/dl% complete remission (% responders)Concomitant medicationRef.
  • *

    Three of the 49 patients were diagnosed as having microscopic polyangiitis.

42Non–life threateningYes (50); 1.4 (range 0.8–2.5)71Prednisone 1 mg/kg/day, gradual taper to alternate-day dosing, then to 09, 12
17Generalized, not immediately life threateningYes (11); not specified35 (59)Prednisolone, median dosage 10 mg/day (range 5–50)10
19Non–life threateningYes (47); 0.97 (range ≤1.2)74 (89)Prednisone, 0.5–1 mg/kg/day, tapered to 20 mg/day by 2 months, to 0 by 6 months11
52Limited, i.e., non–life or organ threateningYes (29); 0.97 (range SD 0.2)88Prednisone, 0.5–1 mg/kg/day, tapered to 20 mg/day by 2 months, to 0 by 6 months13, 17
49*Early systemic, i.e., non–life or organ threateningYes (27); 0.96 (range 0.5–1.43)90Prednisolone, 1 mg/kg/day, tapered to 15 mg/day by 3 months, to 0 by 12 months1

The trial conducted by the EUVAS investigators was originally known in the vasculitis community as the “non-renal vasculitis alternative treatment with methotrexate (NORAM) trial” (7). One hundred patients with newly diagnosed WG or MPA were enrolled. Patients qualified for participation in the trial if they had “early systemic disease,” which the investigators defined as having 1 or more organ systems affected, constitutional symptoms, and either an elevated erythrocyte sedimentation rate or elevated C-reactive protein level. Patients with organ involvement or life-threatening disease manifestations were excluded. Despite the original acronym of the trial, renal involvement was only an exclusion criterion if deemed severe enough to be organ threatening, i.e., the serum creatinine level was >1.7 mg/dl, red cell casts were present, or proteinuria was in excess of 1.0 gm/day. The clinical spectrum of the trial participants seems similar to that in the subjects of previous reports on the use of MTX for remission induction in WG (Table 1) (2, 8–13). In those reports, the patient cohorts were labeled as having “non–life or organ threatening” or “limited” disease, with various nuances of the definitions.

MTX for remission induction in WG

  1. Top of page
  2. MTX for which patient?
  3. MTX for remission induction in WG
  4. MTX for remission maintenance
  5. REFERENCES

Capizzi and Bertino described 2 patients with WG treated with 50 mg/week of MTX administered intravenously for several weeks, followed by an oral dosage of 15–25 mg/week (2). The reported clinical details suggest that these cases would fit into the disease subgroup of “early systemic disease” used by the EUVAS investigators. These patients had multisystem disease, constitutional symptoms, but no evidence of active renal disease when MTX therapy was started. During MTX therapy, glucocorticoids could be withdrawn, and patients remained in remission for several years. This study provided the first evidence for a beneficial effect of MTX in WG.

Two decades later, investigators at the National Institutes of Health (NIH) provided the first report on successful treatment of WG (29 patients) with glucocorticoids (prednisone, 1 mg/kg/day) and oral MTX (20–25 mg/week) (8). Three years later, they described the extended followup of 42 patients treated with this regimen, including 20 from the previously reported group of 29 (9). One-third of these patients were newly diagnosed and two-thirds were being treated for a relapse of the disease. Patients with disease activity that was considered immediately life threatening, defined as acute renal failure resulting in a serum creatinine level >2.5 mg/dl, acute pulmonary hemorrhage with an arterial PO2 of <70 mm Hg, and/or carbon monoxide diffusing capacity <70% of predicted, were excluded from treatment with this regimen. Thirty patients (71%) achieved complete remission over a median of 4.2 months (range 1–17 months). Eleven of these 30 patients (36%) relapsed, with an estimated median time to relapse of 29 months. All relapses occurred after discontinuation of prednisone; 7 patients relapsed while still receiving MTX, and 4 relapsed after MTX had been stopped. A surprisingly high rate of serious adverse events occurred in this cohort. Four patients (10%) developed opportunistic infections, 2 of which were fatal. Leukopenia occurred in 3 patients, MTX pneumonitis in 3, and stomatitis in 1. These adverse events were attributed to the lack of prophylaxis for Pneumocystis jiroveci pneumonia and the lack of folic acid substitution, both of which have since become part of the standard of care for patients receiving this regimen (14, 15).

In 1998, de Groot and coworkers reported on 17 patients with “generalized, not immediately life threatening” WG, treated with MTX at a weekly intravenous dosage of 0.3 mg/kg and concomitant oral prednisolone starting at a median dosage of 10 mg/day (range 5–50) (10). The treatment success rate in these patients was substantially lower than in the cohort reported by the NIH investigators (9). Only 10 of the 17 patients (59%) were classified as responders, and only 6 (35%) achieved a complete remission. Nevertheless, the frequency and severity of adverse events observed in this cohort was negligible. The lower efficacy and toxicity were attributed to the lower glucocorticoid dosage used in these patients (10).

Stone and colleagues reported on 19 patients with “non-life threatening” WG treated with a regimen that represented a modification of the one used by the NIH investigators (11). As in the NIH regimen, the oral prednisone dosage ranged from 20–60 mg/day (median 40) at initiation of treatment. The weekly oral MTX dosage was increased more gradually, reaching a maximum of 22.5 mg/week over the course of 7 weeks. The mean maximum MTX dosage used in these patients was 18.7 mg/week. Once clinical improvement was achieved, the daily prednisone dosage was tapered. The goal was to reduce the oral prednisone dosage to 20 mg/day within 2 months, and to then further reduce the dosage to complete discontinuation. If the patient was maintained in remission on prednisone at ≤10 mg/day, the MTX was also gradually tapered by 2.5 mg/month until complete discontinuation. Folic acid supplementation (1 mg/day) was part of the regimen, and half of the patients received Pneumocystis jiroveci pneumonia prophylaxis.

In contrast to the 2 other reported cohorts (9, 10), all patients in the study by Stone and colleagues were newly diagnosed, previously untreated patients (11). Similar to the NIH cohort, half of the patients had evidence of active glomerulonephritis at the beginning of therapy, but the maximal serum creatinine level was only 1.2 mg/dl (11). The clinical response rate was similar in both groups of patients; 89% experienced clinical improvement, whereas 74% achieved remission. However, only 2 patients (11%) had stable, complete remission that lasted 1 month beyond discontinuation of immunosuppressive drugs. The frequency of serious adverse events was much lower than reported from the NIH. This was probably the result of several factors in combination. The patients had a lesser degree of renal insufficiency. None of the patients had received any immunosuppressive therapy preceding the beginning of prednisone and MTX combination therapy. All patients received folic acid supplementation, and at least half of them received prophylaxis for Pneumocystis jiroveci pneumonia.

All of the described WG cohorts treated with MTX for remission induction included patients with some degree of mild renal involvement (Table 1). Twenty-one (50%) of the 42 NIH patients had renal involvement (9). Only 1 patient (2%) of that cohort developed acute renal failure, which occurred at the time of disease relapse. A long-term outcome analysis of the NIH cohort indicated that patients who had active renal involvement with a mean pretreatment serum creatinine level of 1.4 mg/dl (range 0.8–2.5) experienced stabilization or improvement of renal function over the course of the entire followup period (median 76 months, range 20–108) (12). Only 1 patient (5%) from the cohort reported by Stone and colleagues developed severe necrotizing glomerulonephritis while receiving MTX (11). This is in contrast to the 17 patients described by de Groot and colleagues, in 5 (29%) of whom the disease progressed to active glomerulonephritis during MTX therapy (10). Taken together, these observations suggest that glucocorticoids (prednisone) given at a starting dosage of 0.5–1 mg/kg/day followed by a dosage tapering over months are critical for the success of an MTX-based remission induction regimen.

Based on these reports, comprising a total of 78 patients, the combination of oral daily prednisone with oral weekly MTX has emerged as accepted standard therapy for “non–life or organ threatening” WG (9–11). This regimen was incorporated into the design of the Wegener's Granulomatosis Etanercept Trial (WGET) (16). In this double-blind trial, every participant was randomized to receive either etanercept or placebo in addition to standard therapy. Because the study medication had no beneficial effects (17), the trial results provide valuable lessons about standard therapy. Of the 52 participants with “limited (non–life or organ threatening)” disease who received the combination of oral prednisone and weekly MTX with a target dosage of 25 mg/week for remission induction, 46 (88%) achieved complete remission, which lasted for at least 6 months in 35 patients (67%). This success rate is consistent with expectations based on the earlier reports (9, 11).

The EUVAS investigators are to be commended for performing the first and only randomized controlled trial that compares MTX plus glucocorticoids versus the old “gold standard” combination of oral CYC plus glucocorticoids. In view of the significant morbidity associated with prolonged and repeated CYC use (18, 19), it is a relief to learn that an MTX-based remission induction regimen is indeed “noninferior” to a CYC-based regimen in patients with early systemic, non–life-threatening, or limited disease (1). Six months after initiation of therapy, 89.8% of patients in the MTX arm of the study had achieved remission. For practical clinical purposes, this remission rate is as good as or better than previously reported rates in US studies (9, 11, 17), and equivalent to the 93.5% achieved in the CYC arm.

However, the authors state that the time to remission was shorter with the use of CYC than with MTX (1). While this difference did not reach statistical significance when the trial population was analyzed as a whole, it was significant in those patients who were sicker at entry (with a disease extent index higher than the median of 10). CYC also seemed to reduce the time to remission significantly in patients with lung involvement. Unfortunately, the description of the trial results is not detailed enough to determine whether the conclusion that “the MTX regimen used in the present study was less effective for induction of remission in patients with extensive disease and pulmonary involvement” is indeed valid.

These subset analyses raise a number of points for consideration. Are patients labeled as having “early systemic” disease with a high disease extent index (≥10) comparable with patients labeled by others as having “limited” disease, or would they have been categorized as having “severe” disease (13)? With respect to the efficacy of MTX in patients with lung disease, we are not told how many patients in the study by the EUVAS group formed the basis for this subset analysis, or what the distribution of different types of pulmonary involvement was in the 2 treatment arms. For instance, infiltrates caused by alveolar hemorrhage are expected to resolve much more readily with treatment than are large pulmonary masses or cavities. These considerations further illustrate that the appropriate interpretation and comparability of study results are critically dependent on a clear and detailed clinical description of the study populations, particularly if the clinical spectrum is as heterogeneous as in ANCA-associated vasculitis.

MTX for remission maintenance

  1. Top of page
  2. MTX for which patient?
  3. MTX for remission induction in WG
  4. MTX for remission maintenance
  5. REFERENCES

What do we know about the efficacy of MTX for remission maintenance? The NORAM trial was designed as a remission induction and not a remission maintenance trial (1). Yet, the results raise questions about how best to maintain long-term disease remission.

Emerging recognition of the high relapse rate in WG and the significant long-term toxicity associated with continued CYC use (18) has prompted the search for more benign alternatives. Encouraged by the efficacy of MTX for remission induction (9), the logical next step was to investigate this agent as an alternative to CYC for remission maintenance. In 1999 Langford and colleagues reported on a cohort of 31 patients in whom CYC was replaced with MTX (once remission had been achieved with 3–6 months of remission induction therapy) (20). The observed remission maintenance rate after 21 months of followup was 80%. This rate appeared equivalent to the rate that had been achieved with continued use of CYC in a historical control cohort of 60 patients, and led to the wide acceptance of MTX as the preferable alternative to CYC for remission maintenance in WG.

The long-term efficacy of MTX for remission maintenance, however, seems to leave a lot to be desired. Reinhold-Keller and coworkers reported that 36.6% of a cohort of 71 patients treated with MTX experienced a relapse within a median of 19.4 months (range 1–49 months) (21). Two-thirds of these flares involved the kidneys. These reported disease and renal relapse rates are very similar to the long-term followup observations reported by Langford and colleagues from the NIH cohort of 42 patients treated with MTX. Of those patients, 52% had a relapse within a median of 15 months (range 5–54 months) after prednisone and CYC treatment was replaced with MTX, following remission induction (22). Two-thirds of the patients were still receiving MTX for remission maintenance, and had been in remission for <2 years, when the flare occurred. Furthermore, 72% of the relapses involved the kidneys (22).

The studies by Reinhold-Keller et al (21) and Langford et al (22) confirm that long-term MTX use in WG is well tolerated and safe, provided that folic acid supplementation and Pneumocystis jiroveci pneumonia prophylaxis are given, that dose adjustments are implemented based on renal function, transaminase values, and leukocyte counts, and that the drug is given with extreme caution in patients with any degree of renal insufficiency (21, 22). Since neither of the 2 studies identified specific predictors for relapse, continued careful monitoring of all patients receiving maintenance MTX treatments, with particular emphasis on screening for early signs of glomerulonephritis, seems crucial for the prevention of an adverse renal outcome.

In the current EUVAS trial, immunosuppressive therapy was discontinued completely 12 months after initiation (1). This time is earlier than suggested by current recommendations for the treatment of WG, namely that a 3–6-month remission induction phase should be followed by a longer remission maintenance phase of at least 12–15 months (23). Customarily, newly diagnosed patients receive at least 18 months of treatment with immunosuppressive drugs before discontinuation of remission maintenance therapy is considered. However, it must be acknowledged that the current approach to remission maintenance therapy is based on empiricism and not on results of clinical trials. The optimal duration remains unknown. Similarly, it is unclear whether all patients need prolonged remission maintenance therapy, or whether it is possible to identify specific subsets of patients for whom this is unnecessary.

Based on the observed high relapse rate, the EUVAS investigators conclude that continuation of immunosuppression beyond 12 months is supported. This conclusion would indeed be appropriate if all or most flares occurred after discontinuation of immunosuppression. However, the results over the followup (shown in Figure 4 of their report [1]) indicate that 45% of patients in the MTX arm and 30% in the CYC arm experienced a relapse before maintenance therapy was discontinued. Interestingly, the rates of decline of remission maintenance (Kaplan-Meier curves) in the 2 treatment arms seem to parallel each other, suggesting that other variables may be responsible for the early separation of treatment groups. The trial data clearly confirm that continued MTX therapy does not keep all patients in remission. However, the data also suggest that not even CYC can guarantee a stable long-term remission. Clearly, the important unanswered question in the field remains: how can we keep patients in long-term remission without exposing them to substantial risks of drug toxicity?

Are there alternatives to MTX for remission maintenance? The most widely accepted contender is azathioprine (AZA). AZA is in fact the only agent that has been directly compared with continued use of CYC for this purpose, in a prospective randomized trial performed by the EUVAS group (24). Over the 18 months of observation, 15.5% of patients in the AZA arm experienced a flare, compared with 13.7% (P = 0.67) in the CYC arm (24). Disease-specific subset analysis revealed that the relapse rate in WG patients was significantly higher (18%) than that in MPA patients (8%) (P = 0.03). For WG, the relapse rates observed with AZA treatment seem to be similar to those achieved with MTX (20). Compared with MTX, AZA can be used as a safe treatment in patients with renal insufficiency. Unlike MTX, AZA can also be given in combination with trimethoprim/sulfamethoxazole (T/S) (twice-daily dosages of 160 mg AZA and 800 mg T/S). Following initial observations of a beneficial effect of T/S in WG, a prospective, randomized, placebo-controlled study indicated that this drug reduces the relapse rate of WG significantly (25–27). It may be more than coincidence that T/S, like MTX, interferes with folic acid metabolism. Whether AZA alone or possibly in combination with T/S is superior to MTX for remission maintenance remains unknown.

Mycophenolate mofetil (MMF) is also gaining popularity as a potential remission maintenance agent for ANCA-associated vasculitis. This is in spite of the high cost of the drug and a scarcity of published data. The initial report of 10 patients observed for 15 months suggested a low relapse rate (9%) (28). However, this contrasts with the experience from the NIH (29); 6 (43%) of 14 WG patients receiving MMF for remission maintenance following the standard remission induction protocol experienced a relapse, with a median time from remission of 10 months (range 1–25 months). These data make the use of MMF less appealing for remission maintenance in ANCA-associated vasculitis, except for patients who have contraindications for MTX and AZA. Other conventional immunosuppressive agents in need of further investigation as potential remission maintenance agents for ANCA-associated vasculitis include leflunomide and cyclosporine (30, 31).

In conclusion, MTX, given at a dosage of 20–25 mg per week in combination with glucocorticoids, has emerged as the standard remission induction regimen for WG in patients whose disease is classified as “limited,” “early systemic,” or “non–life or organ threatening.” MTX has now even been documented to be noninferior to CYC in that capacity (1). Whether continued, long-term MTX use is the best option for remission maintenance remains less clear. A well-designed, multicenter, long-term remission maintenance trial needs to be conducted to determine which patients need long-term continuation of immunosuppressive treatment, and which regimen has the best risk-to-benefit ratio. The current climate of collaboration is extremely favorable for the conduct of such a trial. The challenge is to secure funding for a trial that does not involve a novel biologic agent and requires that a large number of patients remain enrolled for several years.

REFERENCES

  1. Top of page
  2. MTX for which patient?
  3. MTX for remission induction in WG
  4. MTX for remission maintenance
  5. REFERENCES