In a previous issue of Arthritis Care & Research, Slot et al (1) reported their experience with a retrospective cohort of 128 patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV). Patients evaluated between 1990 and 1996 (cohort 1, n = 84) were treated with daily cyclophosphamide (CP) and glucocorticoids. In contrast, those evaluated after 1996 (cohort 2, n = 44) received an alternative remission maintenance approach, switching from CP to azathioprine (AZA) after remission had been achieved (2). The investigators analyzed the 2 patient subsets according to treatment regimen and ANCA status. Cohort 2 included 33 patients who had been proteinase 3-ANCA (PR3-ANCA) and classic-ANCA (C-ANCA) positive, of whom 13 remained ANCA positive at the time of the treatment switch. This persistently ANCA-positive subset of patients had a higher rate of relapse compared with the 20 patients who had become ANCA negative by the time of the AZA switch (P = 0.04). What are the implications of this observation for current clinical practice and research?

Additional details of the 2 regimens employed and the timing of disease flares are relevant to the proper interpretation of this study. The patients in cohort 1 were treated with CP until remission and then underwent slow tapers of the medication such that the mean duration of CP use was 18 months. In contrast, the patients in cohort 2 received CP only for a mean period of ∼6 months (range ∼3–17 months), followed by conversion to AZA. That medication was then tapered by 25 mg every 3 months, resulting in a mean total duration of AZA treatment of approximately 21 months. Overall, 41% of the patients in the combined cohorts relapsed. The relapse rates at 2 and 4 years followup were not different between the 2 groups. Most of the observed relapses occurred after the cytotoxic agents had been tapered.

This study has several important strengths:

  • 1
    The sizeable number of patients.
  • 2
    The lengthy median followup period (5.3 years; range 1–11.5 years).
  • 3
    Consistency of the CP protocol in cohort 1 with the regimen widely considered the standard of care up until the late 1990s; namely, daily CP (3, 4). This facilitates comparisons with other cohorts in the literature.

The paper also has some weaknesses:

  • 1
    Comparison of 2 treatment regimens applied during different time periods.
  • 2
    Inclusion of several forms of AAV that are often recognized as being clinically distinct (e.g., Wegener's granulomatosis, microscopic polyangiitis, Churg-Strauss Syndrome, or renal-limited vasculitis). Because of its generally striking, highly characteristic eosinophilia, the Churg-Strauss syndrome marks a significant departure from the other diseases included in the study. The authors do not provide details about the number of patients with each individual clinical diagnosis within the 2 cohorts. It is quite possible that responses to treatment and durations of remission may vary across AAV subsets.
  • 3
    Omission of details related to the severity of disease relapses and toxicities of the therapeutic regimens. These data are very difficult to collect accurately in a retrospective study.
  • 4
    Variability in the timing of the switch to AZA (3–17 months after CP therapy). Particularly pertinent to the authors' major conclusion—the impact of ANCA status at the time of treatment switch on the risk of subsequent disease flare—is whether the duration of CP treatment varied among the 33 PR3-ANCA–positive patients in cohort 2.

One potential interpretation of the results of this study is that ANCA status should guide the duration of CP treatment; i.e., that PR3-ANCA–positive patients should be continued on CP as long as they remain ANCA positive, regardless of their clinical status. The authors rightly caution, however, that the known risks of chronic CP therapy make lengthy treatment unappealing. There are at least 2 major problems associated with prolonged CP. First, previous studies have demonstrated that some patients who are persistently ANCA positive achieve and maintain clinical remissions for prolonged periods (5, 6). Therefore, to continue CP solely because of ANCA positivity would be to expose some individuals to unwarranted toxicity. Slot et al (1) note that based on their data, for every 3 continuously PR3-ANCA–positive patients treated with prolonged CP, 2 would be treated unnecessarily. Second, some adverse events associated with CP, e.g., bladder cancer, leukemia, lymphoma, myelodysplasia, and infertility, are irreversible (4). These toxicities are not only associated with their own risks of morbidity and mortality, they also preclude further CP use and therefore deprive the patient of the most effective agent currently known for the treatment of life-threatening AAV.

This study raises other questions about future potential treatment strategies. Most relapses in the 2 cohorts occurred after the tapering and discontinuation of CP or AZA. Would more prolonged treatment with AZA have led to more sustained remissions (and ultimately less morbidity and mortality), regardless of ANCA status at the time of starting the maintenance agent? The answer to this question and the optimal length of remission maintenance therapy in AAV remain unknown. However, given the high risk of relapse in certain forms of AAV (e.g., Wegener's granulomatosis), we question if it might be reasonable to continue AZA or other lower-risk remission maintenance agents (7) indefinitely. If not reasonable for all patients with AAV, such a strategy might be advisable for certain high-risk patients (e.g., those with recurrent renal flares and/or significant renal damage from preexisting disease). The risk/benefit considerations of prolonged remission maintenance therapy are not known.

Additional information from this study, albeit difficult to acquire retrospectively, would have addressed the issues described above. Slot et al (1) do not provide information about the severity of relapses that followed tapering and discontinuation of therapy. Were the relapses in the AZA group mild and easily treated, or were they sufficiently severe as to justify the risks of extended CP therapy? The question of prolonged maintenance therapy cannot be answered by this study or others currently available, but it remains an important issue for future investigations.

The observations of Slot and colleagues (1) provide further testimony to the absence of cure for at least a subset(s) of patients with AAV. There remains a need to identify safe and effective therapies that modulate or irreversibly alter critical pathophysiologic pathways in AAV in a specific fashion. Until such therapies become available, we are challenged to balance the potential risks of active AAV with the potent and often toxic effects of agents we provide for our patients. Definitive answers on the role of ANCA status in guiding treatment decisions await the conduct of carefully designed prospective studies.


  1. Top of page