To assess the efficacy of systemic corticosteroids alone as first-line treatment of Churg-Strauss syndrome without poor-prognosis factors as defined by the five-factor score (FFS=0), and to compare the efficacy and safety of azathioprine vs pulse cyclophosphamide as adjunctive immunosuppressive therapy to treat failure or relapse

All patients initially treated with systemic corticosteroids alone: optional IV methylprednisolone pulse (15 mg/kg) at treatment start followed by oral prednisone (1 mg/kg/day) according to a tapering schedule. Treating physicians allowed to treat minor relapses with corticosteroids without referring the patient for randomization, as long as the prednisone dose did not exceed 0.5 mg/kg for 1 month.

Patients in whom prednisone doses could not be tapered below 20 mg, those who failed to enter remission and those who relapsed were randomized to receive either 6 months of oral azathioprine (2mg/kg/day) or 6 cyclophosphamide pulses (0.6g/m2 D1, D15, D30 then every month)

Inclusion Criteria:

• Men and women with newly diagnosed Churg-Strauss syndrome;
• absence of poor prognostic factors as defined by the five-factor score (serum creatinine > 140 μmol/l or 1.58 mg/dl, proteinuria > 1 g/day, severe gastrointestinal tract involvement, specific cardiomyopathy and/or central nervous system involvement;
• written informed consent.

Exclusion Criteria:

• age < 15 years, previously treated Churg-Strauss syndrome;
• history of cancer;
• pregnant or breast-feeding women;
• psychiatric disorders that might compromise compliance with therapy;
• contraindication to study drug;
• other ongoing therapeutic trial;
• concomitant viral hepatitis B or C or human immunodeficiency virus (HIV) infection