Article Text

Deoxyspergualin in relapsing and refractory Wegener’s granulomatosis
  1. O Flossmann1,
  2. B Baslund2,
  3. A Bruchfeld3,
  4. J W Cohen Tervaert4,
  5. C Hall7,
  6. P Heinzel5,
  7. B Hellmich6,
  8. R A Luqmani7,
  9. K Nemoto5,
  10. V Tesar8,
  11. D R W Jayne1
  1. 1
    Vasculitis Unit, Addenbrooke’s Hospital, Cambridge, UK
  2. 2
    Rheumatology Clinic, Rigshospitalet, Copenhagen, Denmark
  3. 3
    Department of Nephrology, Karolinska University Hospital at Huddinge, Stockholm, Sweden
  4. 4
    Department of Clinical and Experimental Immunology, University Hospital Maastricht, Maastricht, The Netherlands
  5. 5
    Euro Nippon Kayaku GmbH, Frankfurt/Main, Germany
  6. 6
    Department of Rheumatology, University Hospital Schleswig-Holstein, Lübeck, Germany
  7. 7
    Rheumatic Diseases Unit, Western General Hospital, Edinburgh, UK
  8. 8
    Department of Nephrology, General Faculty Hospital, Prague, Czech Republic
  1. Dr O Flossmann, Box 118, Vasculitis Unit, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK; oflossmann{at}


Objectives: Conventional therapy of Wegener’s granulomatosis with cyclophosphamide and corticosteroids is limited by incomplete remissions and a high relapse rate. The efficacy and safety of an alternative immunosuppressive drug, deoxyspergualin, was evaluated in patients with relapsing or refractory disease.

Methods: A prospective, international, multicentre, single-limb, open-label study. Entry required active Wegener’s granulomatosis with a Birmingham vasculitis activity score (BVAS) ⩾4 and previous therapy with cyclophosphamide or methotrexate. Immunosuppressive drugs were withdrawn at entry and prednisolone doses adjusted according to clinical status. Deoxyspergualin, 0.5 mg/kg per day, was self-administered by subcutaneous injection in six cycles of 21 days with a 7-day washout between cycles. Cycles were stopped early for white blood count less than 4000 cells/mm3. The primary endpoint was complete remission (BVAS 0 for at least 2 months) or partial remission (BVAS <50% of entry score). After the sixth cycle azathioprine was commenced and follow-up continued for 6 months.

Results: 42/44 patients (95%) achieved at least partial remission and 20/44 (45%) achieved complete remission. BVAS fell from 12 (4–25), median (range) at baseline to 2 (0–14) at the end of the study (p<0.001). Prednisolone doses were reduced from 20 to 8 mg/day (p<0.001). Relapses occurred in 18 (43%) patients after a median of 170 (44–316) days after achieving remission. Severe or life-threatening (⩾ grade 3) treatment-related adverse events occurred in 24 (53%) patients mostly due to leucopaenias.

Conclusions: Deoxyspergualin achieved a high rate of disease remission and permitted prednisolone reduction in refractory or relapsing Wegener’s granulomatosis. Adverse events were common but rarely led to treatment discontinuation.

Statistics from

Wegener’s granulomatosis is a primary systemic vasculitis of unknown cause, characterised by granulomatous and necrotising inflammation affecting most frequently the upper and lower respiratory tract and kidneys, but it may affect any organ system. Circulating antineutrophil cytoplasm antibodies (ANCA) with specificity for proteinase-3 (PR3-ANCA) are detectable in the sera of most patients at the time of diagnosis. Without treatment, Wegener’s granulomatosis has a mortality of 80% at one year.1

Combination therapy with cyclophosphamide and high-dose corticosteroids is successful in inducing remission in over 90%,2 3 but subsequent relapse occurs in approximately 70% of patients by 5 years despite remission maintenance therapy with alternative immunosuppressive agents, including methotrexate and azathioprine.4 5 Treatment with cyclophosphamide carries a high risk of substantial cumulative drug toxicity, in particular long-term malignancy (especially bladder cancer), infertility and increased susceptibility to infection.69 Alternative treatments are required for patients who fail to achieve a full initial remission, relapse on maintenance therapy, do not tolerate conventional therapy, or have a high previous exposure to cyclophosphamide.

Deoxyspergualin is an antiproliferative drug derived from Bacillus laterosporus, with effects on lymphocyte and macrophage function and neutrophil production. It is licensed in Japan for the treatment of acute and accelerated rejection in renal transplant recipients.10 A preliminary trial has suggested that deoxyspergualin is effective in the treatment of refractory ANCA-associated vasculitis, with a response rate of 70% and a good safety profile.11 This phase II trial was conducted to assess further the efficacy and safety of deoxyspergualin in patients with refractory or relapsing Wegener’s granulomatosis.


Patients were recruited from seven university hospitals in six European countries. The study was conducted according to the standards set out in the Declaration of Helsinki and the guidelines of good clinical practice. Ethics approval was granted by the institutional and national ethics committees. Informed written consent from each patient was obtained before enrolment.

Study design

This multicentre, single-limb, open-label, prospective trial aimed to determine the response rate and safety of deoxyspergualin in patients with refractory or relapsing Wegener’s granulomatosis. The total study duration was 12 months.


Patients were eligible for the study if they were aged between 18 and 80 years and had a documented diagnosis of Wegener’s granulomatosis according to the American College of Rheumatology criteria12 and the Chapel Hill Consensus Conference definition.13 Their disease had to be active, as measured with the Birmingham vasculitis activity score (BVAS) in which clinical manifestations caused by active vasculitis are scored on a list of predefined organ-specific items.14 A minimum BVAS of 4 was required, which requires at least one predefined major, ie, organ threatening, or at least three minor disease manifestations despite receiving at least 3 months previous treatment with cyclophosphamide and/or at least 6 months treatment with methotrexate. Patients had to be able to self-administer the study drug. Exclusion criteria included pregnancy, breastfeeding or not using appropriate contraception if of childbearing age, active infection, HIV, hepatitis B surface antigen or hepatitis C antibody positivity, or an underlying medical condition that would have posed an unacceptable risk of immunosuppression. In addition, previous treatment with deoxyspergualin or a requirement for additional immunosuppression other than corticosteroids, liver dysfunction, white blood count (WBC) less than 4000 cells/mm3, haemoglobin less than 8 g/dl and platelet count less than 100 000 cells/mm3 precluded entry.

Drug regimen

Immunosuppressive agents were withdrawn at study entry. Deoxyspergualin, 0.5 mg/kg per day, was self-administered by subcutaneous injection for up to 21 days each cycle, with a minimum drug-free period of 7 days between cycles for a total of six cycles. The full blood count was checked twice each week and the cycle terminated early if the total WBC fell below 3000 cells/mm3 or 4000 cells/mm3 at the discretion of the investigator. Treatment with deoxyspergualin was only restarted once the WBC had recovered to above 4000 cells/mm3. Deoxyspergualin could be withdrawn and the treatment period terminated after three cycles if the patient had achieved complete remission.

At inclusion the daily prednisolone dose had to be within the range of 7.5–60 mg/day. The prednisolone dose was tapered according to clinical response at the discretion of the investigator. The maximum allowed prednisolone dose was 25 mg/day after one month from entry.

After completion of the treatment period with deoxyspergualin, azathioprine was started at a dose of 2 mg/kg per day in addition to the continuation of prednisolone. For patients intolerant of or relapsing on azathioprine and with a good response to deoxyspergualin during the trial, deoxyspergualin could be re-introduced. Other treatments could be used during the follow-up period at the discretion of the investigator if clinically required.

The use of trimethoprim/sulphamethoxazole at a maximal dose of 480 mg/day for Pneumocystis jiroveci prophylaxis was permitted but not mandatory.


Evaluations were performed at screening, study entry within 4 weeks of screening, twice during each treatment cycle, 2 weeks following the last cycle and at 3 and 6 months after the end of the treatment period. They included BVAS, full blood count, electrolytes, urea, creatinine, liver function tests, C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR). Disease damage measured by the vasculitis damage index (VDI),15 quantification of proteinuria and ANCA were assessed at entry and every 3 months. Investigators had to demonstrate proficiency in the use of BVAS and VDI before commencing the study. The medical outcomes study 36-item short-form general health survey (SF-36) was completed at entry and after the treatment period.16

Outcome measures

The primary efficacy outcome measure was remission of vasculitis. Complete remission was defined as a BVAS of 0 sustained for at least 2 months. Partial remission was defined as a reduction in BVAS of 50% or more, sustained for at least 2 months, when compared with the BVAS at entry.

Secondary outcome measures were duration of clinical response, ie, time from response to relapse, laboratory markers (CRP, ESR, urine analysis, ANCA), damage as measured by the VDI and patient function as measured by the SF-36 score. A major relapse was defined as the new appearance of a major BVAS item, indicating threatened vital organ function, a minor relapse represented the new appearance of at least three other BVAS items.

Adverse events were recorded following clinical evaluation, laboratory tests and patient diaries. The severity of events was graded according to predefined criteria as mild, moderate, severe or life threatening.17 Any event irrespective of severity was classified as a serious adverse event if it required or prolonged hospitalisation, was fatal or life threatening, caused permanent disability, led to the delivery of a child with a congenital abnormality or was a malignancy.

Statistical analysis

It was planned to enrol 42 patients to provide 34 evaluable patients completing at least three treatment cycles. The sample size was determined in order to achieve a 17% precision for a two-sided 95% CI with an expected response rate of 50%.

Data were analysed using the SPSS version 12 statistical package. All results are expressed as median and range. Comparisons of variables were made with the Wilcoxon signed rank test. A p value of 0.05 or less was considered statistically significant. Time to remission and relapse was calculated according to Kaplan–Meier.



Forty-five patients were enrolled from December 2003 until January 2005. Demographic data are summarised in table 1. Individual demographic, treatment and outcome data are shown in supplementary table 1 available online only. The diagnosis of Wegener’s granulomatosis was supported by histology in 40/45 (89%) and 40/45 (89%) were ANCA positive at study entry. The five patients in whom histological confirmation was not available were all ANCA positive at entry.

Table 1 Demographic and clinical data

The treatment period according to protocol was completed by 38/45 (85%) patients. Six-month follow-up data after the end of the treatment period or earlier withdrawal of the study drug were available for 42/45 (93%) patients.


There were six withdrawals from the trial protocol. Two patients were withdrawn as a result of relapses in the third and sixth cycles. One patient did not respond and was withdrawn after the fifth cycle. The treatment was stopped in one after partial remission due to adverse events. One patient withdrew consent during the first cycle. One patient died after completing the third cycle.

One patient was excluded from the efficacy analysis as the BVAS at entry was 3 after being 7 at screening. This patient was included in the safety report.

Primary endpoint

Partial remission was achieved in 42/44 patients (95%) after a median of 22 days (range 22–134). Twenty (45%) attained complete remission after a median of 78 days (range 50–186).


Relapses occurred in 18 (43%) of 42 patients who had achieved at least partial remission after a median of 170 days (range 44–316) after remission. Two relapses occurred during treatment with deoxyspergualin. The other relapses occurred in the 6-month follow-up period while patients were receiving maintenance therapy with azathioprine9 or mycophenolate mofetil.7 One patient had a major relapse with biopsy-confirmed renal vasculitis during the fourth cycle and was subsequently withdrawn. The other relapses were minor (fig 1).

Figure 1

Kaplan–Meier analysis of (A) time to partial remission, (B) time to complete remission and (C) time to relapse. The ↔ indicates the duration of treatment with deoxyspergualin (DSG).

Renal involvement

Sixteen patients had an active urinary sediment at entry. Haematuria resolved in 10 patients by the end of the treatment phase (p = 0.002).

There was no significant change in the creatinine value of these patients between enrolment 130 μmol/l (58–299) and the end of the treatment period 132 μmol/l (72–279) including the patients who had persistent haematuria.

Drug dosing

The median cycle length was 20 days (4–28) and the median drug-free period between cycles was 11 days (6–25). The main reason for a shortened cycle was neutropaenia. There was no difference in the length of cycles and drug pauses between the first and the sixth cycles.

The prednisolone dose one month before entry and at entry was 11.25 mg/day (5–60) (median, range) and 20 mg/day (5–100). Following six treatment cycles with deoxyspergualin, the median prednisolone dose had fallen to 8 mg/day (2.5–80; p<0.001). The reduction was sustained with a prednisolone dose of 8.75 mg/day (2.5–40) after 6-months follow-up (fig 2).

Figure 2

Average daily prednisolone doses.

Adverse events

Forty-eight severe or life-threatening (⩾ grade 3) treatment-related adverse events occurred in 24 (53%) patients (table 2). The main side effects were irritation at the injection site, reversible bone marrow suppression, oropharyngeal pain and infections. The majority of infections involved the respiratory and urinary tracts or were caused by mucocutaneous candidiasis, which responded to antibiotic and/or antifungal therapy.

Table 2 Adverse events

Seventeen (38%) patients required a total of 23 hospitalisations. The causes were leucopaenia and/or anaemia (n  =  7), infections (n  =  7), comprising bronchopneumonia (n  =  5), of which one was caused by Pneumocystis jirovicii infection diagnosed 3 months after finishing deoxyspergualin, dental abscess (n  =  1) and Campylobacter jejuni diarrhoea (n  =  1). There were four relapses of Wegener’s granulomatosis. Two episodes of acute upper airway obstruction occurred in patients with pre-existing subglottic stenosis, both had had similar episodes before study entry. One was admitted with active Wegener’s granulomatosis during the first cycle. There was one case of food poisoning and one of diarrhoea starting in the first cycle in a patient who subsequently withdrew from the study.

Breast cancer was diagnosed in a 45-year-old woman who had been on hormone replacement therapy for 8 years.


There were two deaths; one during the treatment and one during the follow-up periods. A 25-year-old woman died of upper airway obstruction during the third cycle of deoxyspergualin. She had a pre-existing tight subglottic stenosis and had had episodes of severe upper airway obstruction before enrolment. Vasculitis had been inactive at the last assessment before her death. This case has been reported separately.18

A 20-year-old man died of active cerebral vasculitis (autopsy confirmed) resulting in intracerebral haemorrhage while receiving acenocoumarol anticoagulation for an arterial embolus of his leg. The patient had relapsed during deoxyspergualin therapy 2 months earlier and was receiving mycophenolate mofetil and high-dose steroids at the time of death.

Scoring tools and laboratory tests

Birmingham vasculitis activity score

The BVAS at screening and entry were 12 (0.72, 6–22) (median, SEM, range) and 12 (0.80, 4–25). By the end of the treatment period the BVAS had fallen to 2 (0.48, 0–14; p<0.001). At the end of follow-up the BVAS remained low at 0 (0.43, 0–9; fig 3).

Figure 3

Disease activity as measured by the Birmingham vasculitis activity score (BVAS). DSG, deoxyspergualin.

Antineutrophil cytoplasm antibodies

At study entry 35 patients had a positive cytoplasmic staining ANCA on immunofluorescence and/or a positive PR3-ANCA. Four were antimyeloperoxidase antibody (MPO-ANCA) positive and five were ANCA negative. After completion of the treatment period 34 were ANCA positive, 32 PR3-ANCA, two MPO-ANCA and nine were ANCA negative. There was no subsequent ANCA result in one patient.

Inflammatory markers

Overall there was no significant difference in CRP or ESR between entry and the end of treatment.

There was a fall in CRP from a median (range) of 17.5 mg/dl (7–88) to 9 mg/dl (1–124; p = 0.036) in those 23 patients who had an elevated CRP (>6 mg/dl) at entry.

Damage assessment

The VDI at entry was 4.5 (0–14) (median, range). At the end of the treatment and follow-up periods the VDI had increased to 5 (0–16) and 5 (0–16), respectively.

Short-form 36

The composite score for the physical health measures improved from a value of 29.6 (14.8–50.2) at study entry to 34.3 (11.6–53.7) at the end of the treatment period (p = 0.002). There was no difference in the mental component score 49.4 (21.0–62.9) versus 48.0 (23.7–62.8).


This study was performed to assess the efficacy and safety of deoxyspergualin in the treatment of refractory or relapsing Wegener’s granulomatosis. The primary outcome measure of disease response was achieved by 95% of patients and 45% reached complete remission. The improvement in disease control allowed a reduction in the prednisolone dose from a median of 20 mg/day at entry to 8 mg/day at the end of the 6-month deoxyspergualin treatment period. These results support those of a smaller trial that reported a 70% response rate in a similar patient population.11 The disease control obtained during the treatment period was sustained in 57% during the subsequent 6 months when patients received azathioprine or mycophenolate mofetil and prednisolone. These results compare favourably with the poor disease control, despite standard therapy, experienced during the 6 months before study entry and support the conclusion of a useful therapeutic effect of deoxyspergualin. The re-introduction of deoxyspergualin for subsequent relapse has been effective and longer-term exposure appears safe.19 Sixteen of 44 patients had evidence of glomerulonephritis at entry and this resolved in 10. A further open-label study of deoxyspergualin in 44 patients with crescentic glomerulonephritis, including ANCA-associated vasculitis, demonstrated improvements in proteinuria and haematuria after 4 weeks.20

At least 70% of patients with Wegener’s granulomatosis pursue a relapsing course and 10% fail to achieve initial remission. Refractory disease has recently been defined in an international consensus statement to indicate either a lack of response to induction therapy, persistent lower level disease after induction therapy, or an inability to reduce steroids without disease relapse.21 This study’s population met these criteria and the median BVAS of 12 indicated multiple items of disease activity in different systems. The median VDI of 4.5 at entry reflected the irreversible damage, severity and chronicity of the patients’ previous disease course. The pattern of active organ manifestations at entry (table 1) was similar to previous studies of refractory vasculitis.2224

The number of patients who experienced severe or life-threatening adverse events (53%) was similar to the number of reported events in a recent trial.25 The majority of these events were related to laboratory values without overt clinical manifestations. Two patients in our study died, one each during the treatment and the follow-up periods. Neither of these deaths was directly attributable to deoxyspergualin. The mortality rate of 4.4% was comparable to the rate (4%) reported in recent trials of ANCA-associated vasculitis.3 25 26 The most common adverse events were infections, but only one severe infection was associated with leucopaenia. Cytopaenias occurring with deoxyspergualin were reversible within less than a week on drug withdrawal in comparison with the more prolonged leucopaenia associated with cyclophosphamide. Injection site reactions were common but did not lead to the discontinuation of deoxyspergualin. Severe gastrointestinal disturbance in three patients appeared to be related to the study drug as were the perioral manifestations of pain and stomatitis.

The mechanisms of the immunosuppressive activity of deoxyspergualin are not clearly understood. Deoxyspergualin suppresses IL-2-dependent proliferation of T lymphocytes in vitro27 and interferes with the differentiation of naive B lymphocytes into antibody-producing cells28 and antibody production through impairment of κ-light chain production by binding with a heat shock protein 70 family protein.29 30 Dysregulated antigen presentation appears to be important in Wegener’s granulomatosis, and deoxyspergualin hinders the ability of monocytes and macrophages to act as antigen-presenting cells through inhibition of the production of oxygen free radicals and downregulation of MHC class II expression.31 32 Although treatment with deoxyspergualin leads to a marked reduction in neutrophil numbers, neutrophil function is not affected.33 The induction of neutropaenia may, however, be important as neutrophil depletion has been shown to be protective against the development of glomerulonephritis in a mouse model of ANCA-associated vasculitis.34 This has to be balanced with the increased risk of serious infections caused by neutropaenia. ANCA-induced neutrophil activation is mediated by intracellular kinases including Akt-kinase and the antiproliferative effects of deoxyspergualin are mediated by the inhibition of protein synthesis and induction of apoptosis through the downregulation of Akt-kinase and p70S6-kinase.35 Deoxyspergualin had similar efficacy to cyclophosphamide and was superior to mycophenolate mofetil in attenuating nephritis in a murine model of ANCA-associated vasculitis.36

We have demonstrated a high rate of response to deoxyspergualin in patients with refractory or relapsing Wegener’s granulomatosis. The rate of relapses during the follow-up period was comparatively high, which might reflect the high-risk patients studied. However, the disease control achieved during the treatment period was largely maintained during the follow-up period and was reflected in the reduced requirement for prednisolone. Although adverse effects were frequent, they did not impact in a major way on the tolerability or efficacy of the test drug. Patients adhered to the cyclical, self-administered, subcutaneous injection regimen, confirming its feasibility in a routine clinical setting. There are few options for patients with Wegener’s granulomatosis with active vasculitis despite combination immunosuppressive and prednisolone therapy; deoxyspergualin is a new alternative treatment for this indication.


The authors would like to thank Dr Pieter van Paassen, (Maastricht, The Netherlands), Dr Romana Rysava, (Prague, Czech Republic), Ms Petra Frey (Lübeck, Germany) and Dr Leena Martola (Stockholm, Sweden) who helped in the care of the patients.


Supplementary materials


  • Additional supplemental table 1 is published online only at

  • Funding: This study was sponsored by Euro Nippon Kayaku GmbH, Frankfurt/M, Germany.

  • Competing interests: None.

  • Ethics approval: Ethics approval was granted by the institutional and national ethics committees.

  • Patient consent: Obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.