Inclusion Criteria:

• - Patients with a diagnosis of MPA or GPA independently of ANCA status, - Patient aged of 18 years or older, - Patients with newly-diagnosed disease or relapsing disease at the time of screening, with an inactive disease defined as a BVAS = 0, - Patients receiving maintenance infusion of rituximab 500 mg at 6 and 12 months after the start of vasculitis induction.

• - Patients receiving 5-10 mg/day of prednisone at screening, - Patient able to give written informed consent prior to participation in the study.

• - At Inclusion visit day, patient must be between 5 and 10 mg/day prednisone and at randomization visit day (D1), patient must be at 5 mg/day prednisone.

Exclusion Criteria:

• - Patients with EGPA, or other vasculitides, defined by the ACR criteria and/or the Chapel Hill Consensus Conference, - Patients with vasculitis with active disease defined as a BVAS >0, - Patients with acute infections or chronic active infections (including HIV, HBV or HCV), - Patients with active cancer or recent cancer (<5 years), except basocellular carcinoma and prostatic cancer of low activity controlled by hormonal treatment, - Pregnant women and lactation.

Patients with childbearing potential should have reliable contraception for the all duration of the study,

• - Patients with other uncontrolled diseases, including drug or alcohol abuse, severe psychiatric diseases, that could interfere with participation in the trial according to the protocol, - Patients included in other investigational therapeutic study within the previous 3 months, - Patients suspected not to be observant to the proposed treatments, - Patients who have white blood cell count ≤4,000/mm3, - Patients who have platelet count ≤100,000/mm3, - Patients who have ALT or AST level greater than 3 times the upper limit of normal that cannot be attributed to underlying MPA-GPA disease, - Patients unable to give written informed consent prior to participation in the study.

- Patients with contraindication to use rituximab,

Immunosuppressive therapy of granulomatosis with polyangiitis (GPA, Wegener's) and microscopic polyangiitis (MPA) has transformed the outcome from death to a strong likelihood of disease control and temporary remission. However, most patients have recurrent relapses that lead to damage and require repeated treatment associated with long-term morbidity and death. Rituximab has been shown to be as effective as cyclophosphamide to induce remission and maintenance of remission in severe GPA and MPA patients, with an acceptable safety profile. Although rituximab is becoming the standard of care for maintenance therapy in these patients, relapse still occurs and the optimal duration of prednisone therapy remains debated. On the one hand, most US studies use early withdrawal (6-12 months) because of feared side effects. On the other hand, most European trials propose late withdrawal (>18 months) given a lower observed relapse rate on long-term low dose glucocorticoids treatment. In a systematic review and meta-analysis, glucocorticoids regimen was the most significant variable explaining the variability between the proportions of ANCA-associated vasculitis patients with relapses. Nevertheless, it was an indirect estimation of treatment effect because of the absence of dedicated randomized trial. This meta-analysis concluded that combined longer-term (i.e. >12 months) use of low dose prednisone or nonzero glucocorticoids target is associated with a 20% reduction of relapse compared to early withdrawal (i.e. ≤12 months). The relapse rate in patients with early glucocorticoids (10-12 months) withdrawal was provided in two studies and was of 37 and 34%, respectively. By contrast, the relapse rate in patients with late prednisone withdrawal (18-24 months) and receiving rituximab as maintenance treatment was 14% at 24 months in the MAINRITSAN trial. Of note, the decision to withdraw glucocorticoids after 18 months was left to physician's discretion in this study and two thirds of the nonsevere relapses occurred when patients were off prednisone. The trial detailed here is the first prospective trial evaluating the length of glucocorticoid administration as remission adjunctive treatment for patients with GPA or MPA.

Arms

Experimental: Prednisone 5mg/day extended of 12 additional months

Prednisone 5mg/day will be administered from Day 1 to Month 12

Placebo Comparator: Placebo 5mg/day extended of 12 additional months

Placebo 5mg/day will be administered from Day 1 to Month 12

Interventions

Drug: - Prednisone 5mg/day extended of 12 additional months

Prednisone 5mg/day orally during 12 Month + 1 mg/week tapering until 0mg.

Drug: - Placebo 5mg/day extended of 12 additionnal months.

1mg/week orally tapering Prednisone until Month 1 + Placebo orally 5mg/day until Month 13