A Phase I/II Trial of the Combination of Azacitidine and Gemtuzumab Ozogamicin for Treatment of Relapsed Acute Myeloid Leukemia

In view of the single-agent activity of azacitidine and gemtuzumab ozogamicin (GO) in patients with relapsed acute myeloid leukemia, we studied a combination strategy in a phase I/II trial to potentially further improve outcomes. GO/azacitidine therapy was relatively well tolerated and resulted in response rates similar to those with GO used as monotherapy at higher doses. Introduction: Treatment with hypomethylating agent therapy might enhance anti-CD33 monoclonal antibodymediated cytotoxicity against acute myeloid leukemia (AML) blasts through epigenetic effects on Syk and SHP-1 expression. Patients and Methods: In the present phase I/II study, we treated patients with relapsed or refractory AML with azacitidine, followed by 2 doses of gemtuzumab ozogamicin (GO) at 6 mg/m(2), the Food and Drug Administration-approved dose and schedule at study initiation. We sought to determine the maximum tolerated dose and clinical activity of this combination therapy. Secondarily, we aimed to determine whether baseline Syk and SHP-1 expression can be used as predictive biomarkers of treatment response. Results: The established maximum tolerated dose was azacitidine 75 mg/m(2) daily for 6 consecutive days, followed by GO 6 mg/m(2) on days 7 and 21. Of the 50 evaluable patients, 12 (24%) obtained complete remission (CR) or CR with incomplete peripheral blood recovery (CRp). No dose-limiting toxicities were observed in phase I, and no patient developed hepatic sinusoidal obstructive syndrome. Although no significant correlation was found between Syk and SHP-1 expression and the clinical response to combination therapy, in vitro studies repeatedly demonstrated that azacitidine-treated AML cells had an increased response to GO treatment. Conclusion: Our study found that the combination of GO with azacitidine is relatively well tolerated, with response rates similar to those with GO monotherapy at higher doses. Differences in the GO drug schedule, dose level, and frequency might explain the discrepant response rates between our study and others, suggesting that the optimal GO dose remains unclear, especially when combined with hypomethylating agent therapy.