The Use of Cyclosporine in Association with Chemotherapy As Induction Treatment in Patients with Acute Myeloid Leukemia (AML) and High Rhodamine Efflux at Diagnosis Results in Higher Complete Hematological Remission Rates, but Does Not Prolong Overall Survival

The overexpression of P-glycoprotein (Pgp) in leukemic blasts of Acute Myeloid Leukemia (AML) is associated with lower complete remission (CR) rates and shorter disease-free (DFS) and overall survival (OS). Cyclosporine A (CSA) is a competitive inhibitor of Pgp capable of reduce the P-gp mediated drug efflux. List et al. reported that CSA addition to treatment with cytarabine and daunorubicine reduced the resistance to induction treatment (IT) and prolonged DFS and OS. The effect was greater in AML patients with higher Pgp expression (Blood 2001; vol 98, n 12: 3212-20). Based on this study, we tested the effect of the association of CSA to IT in cases with high Pgp expression, which was established using the Rhodamine effux test as previously described (Pétriz, J. and García-López, J. Leukemia 1997; vol 11: 1124-30). The cut off value of the mean fluorescence intensity adopted to define high P-gp expression was ≥ 1.10. A total of 21 high P-gp AML patients were randomized in two groups: 1. Daunorubicin (DNR) 60mg/m² in continuous infusion -CI- 3x+Cytarabine 100 mg/m² CI 7x + CSA 16mg/m² CI 3x; (CSA group) and 2. DNR 60mg/m² CI 3x + Cytarabine 100 mg/m² CI 7x (Conventional IT group). Patients who did not achieve Complete Hematological Remission (CHR) in the first cycle of chemotherapy received a second course. The 2 groups did not differ regarding age (mean ± S.D.: 44.5 ± 11.9 vs. 48.5 ± 15.2 years in CSA and conventional IT group, respectively), leukocyte counts at diagnosis (44,200 ± 56,100 vs. 21,540 ± 20,200/µl), frequency of de Novo AML (81.8 vs. 80%) and of Core Binding Factor Leukemia cases (18.2 vs. 10%). The median follow up among survivors was of 6.2 years. The mean serum levels of CSA at 24 hours after infusion was of 1,157 ng/mL (range: 4 - 1,600ng/mL), which was similar to that reported by List et al. The CHR rate after the first cycle of induction was higher in the group using CSA (63.6% vs. 30%; p = 0.09) but the DFS (2.05 years vs. not reached; p= 0.27) and OS (4.08 vs. 21.12 months; p = 0.19) were higher in conventional IT group. These results may reflect the effect of the second course of induction in patients who failed to achieve CHR (CHR after second IT: 0% vs. 33.3%, p=0.34, in CSA and conventional IT group, respectively) and higher relapse rate in CSA group (42.8% vs. 33.3%, p=0.78). Furthermore, there was a lower frequency of induction deaths (18.2% vs. 9%; p = 0.59) and of hyperbilirubinemia (bilirubin > 1,5x the upper limit of normality) in the Conventional IT group (6 vs. 0 patients; p = 0.02). There were no cases of acute cardiac toxicity. The study was interrupted due to the cost of CSA and the lack of significant improvement in the outcomes. In conclusion, despite the higher rate of CR after one cycle of chemotherapy, there was not improvement in long term survival of AML patients with high Pgp expression treated with CSA in combination with anthracyclines and cytarabine.