OBJECTIVE: To assess the efficacy of systemic corticosteroids (CS) alone as first-line treatment in patients with Churg-Strauss syndrome (CSS) without poor-prognosis factors, as defined by the Five-Factors Score (FFS), and to compare the efficacy and safety of oral azathioprine (AZA) versus intravenous pulse cyclophosphamide (CYC) as adjuvant immunosuppressive therapy for treatment failure or relapse. METHODS: This multicenter, prospective, randomized, open-label therapeutic trial included 72 patients with newly diagnosed CSS (FFS of 0) treated with CS alone. At treatment failure or relapse, patients were randomized to receive 6 months of oral AZA or 6 pulses of CYC. Analyses were performed according to an intent-to-treat strategy. RESULTS: The mean +/- SD followup was 56.2 +/- 31.7 months. Among the 72 patients studied, 93% achieved remission with CS therapy alone, and 35% relapsed, mainly during the first year of treatment. Among the 19 patients randomized to additional immunosuppression because of treatment failure or relapse, 5 of 10 receiving AZA and 7 of 9 receiving pulse CYC achieved remission, but the difference was not statistically significant. Survival rates in all patients at 1 and 5 years were 100% and 97%, respectively. At the end of followup, 79% of the patients whose disease was in remission required low-dose CS therapy, mainly to control respiratory disease. CS-related adverse events were observed in 31% of the 72 patients. CONCLUSION: In CSS patients with an FFS of 0, survival was excellent, confirming the predictive value of the FFS in this disease. First-line therapy with CS achieved remission in most patients, but relapses were common, and one-third of them required additional immunosuppressive therapy. AZA or pulse CYC was fairly effective in treating CS-resistant disease or major relapses. Over the long term, most patients continued to take oral CS, which might explain the high rate of CS-related adverse events.