A Randomized, Placebo-Controlled, Double Blind Trial of the MDR Modulator, Zosuquidar, during Conventional Induction and Post-Remission Therapy for Pts > 60 Years of Age with Newly Diagnosed Acute Myeloid Leukemia (AML) or High-Risk Myelodysplastic Syndrome (HR-MDS): ECOG 3999.

A major limitation of therapeutic modulation of MDR1/Pglycoprotein mediated drug efflux is the pharmacokinetic interaction that increases the toxicity of chemotherapy. Zosuquidar (Z) is a highly selective and potent third generation P-gp modulator which demonstrated only modest reduction of anthracycline clearance in phase I trials. Thus, we conducted a randomized placebo-controlled double-blind trial to compare the outcome of pts > 60 years of age, with newly diagnosed AML or HR-MDS, who received cytarabine 100 mg/m2/d as a continuous infusion (d1 - 7), daunorubicin 45 mg/m2/d as a 10 – 15 minute injection (d1 - 3) and either Z (550 mg/m2; arm A) or placebo (arm B) infused starting 1 hour prior and continuing 5 hours following daunorubicin. Pts who achieved a complete remission received a single cycle of intermediate-dose cytarabine (1.5 gm/m2 BID [once daily for pts > 70] for 6 days) followed by a second cycle identical to induction. A total of 442 patients were eligible for analysis among 449 enrolled; 219 received Z and 223 placebo. P-gp expression on pre-treatment marrow myeloblasts was determined by the ratio of rhodamine accumulation in the presence or absence of Z in vitro and MRK-16 binding. The arms were comparable in terms of median age (69.3 years with 45% of patients > 70 yrs), frequency of secondary AML (37%) and frequency of poor-risk cytogenetics (40%), and distribution of P-gp status. However, there was an imbalance in the distribution of ECOG performance status: 59 pts (27%) in arm A were ECOG PS 2–3 compared to 29 (13%)in arm B (Fisher’s exact test, p = 0.0005). The primary outcome of interest was overall survival (O.S.). As of 6/06 315 deaths have occurred (159 on arm A and 156 on arm B). The median survivals are 7.7 and 9.4 months, respectively (p=0.45). The frequency of adverse events were similar between the two arms with the exception of generally mild to mild reversible neurotoxicity (ataxia, hallucinations, confusion) which occurred in 27% of pts in arm A compared to 10% in arm B. Analyses using the Cox PH model were performed to explore relationships between treatment arm, baseline patient characteristics, and outcome. The median O.S. of pts with high P-gp status in arm A and B were 4.6 m and 8.3 m respectively (p=0.2). The discrepancy in outcome may be explained by the imbalance in ECOG PS: 43% of pts assigned to placebo were a PS of 0 compared to 23% in arm A. Both cytogenetics risk group (p<0.01) and high Pgp ratio (p=0.002) were significantly associated with OS in both arms. There was a significant association between cytogenetic risk group and P-gp level (p=0.02). In Cox modeling relating cytogenetics and P-gp level to OS, only the cytogenetic risk remained significant. The lack of benefit of Z in this trial may be due to an insufficient duration of inhibition of P-gp in leukemic cells, since Z has a short half-life. Alternatively, the association between MDR ratio and poor-risk cytogenetics suggests that other mechanisms of resistance to daunorubicin limit the clinical benefit of P-gp inhibition in AML.