In elderly patients with secondary leukemia, poor therapeutic response and low overall survival have been attributed mainly to age and to the primary resistance of leukemic cells to chemotherapy. Modulation of resistance has been attempted in different studies, but the results have been contradictory. We conducted an open, randomized multicenter clinical trial involving patients more than 60 years old with secondary leukemia preceded by a myelodysplastic syndrome. The induction chemotherapy regimen included idarubicin, cytarabine, and etoposide (group A); randomization involved simultaneous administration of cyclosporin-A per os (group B). Fifty-five patients were evaluated, 26 in group A and 29 in group B. Overall complete remission was achieved in 40% of the patients, 27% vs 52% in groups A and B, respectively (p=0.01). Leukemia-free survival was more favorable in patients who received cyclosporin-A, 12 vs 7 months for groups B and A, respectively (p=0.03). In a follow up period of 30 months, 7 out of 55 patients (13%) were alive, 4 of whom were in complete remission. Five out of the 7 alive patients were randomized in group B and had received cyclosporin-A. Treatment failure was higher in group A [19 of 26 patients (73%)] than in group B with CsA [14 of 29 patients (48%)] (p<0.0001). Treatment-related toxicity/mortality was 13%. Modulation of drug resistance by CsA in elderly people suffering from secondary acute leukemia may improve the outcome of chemotherapy without increasing drug toxicity and treatment-related mortality.