A Post Hoc Landmark Analysis of Survival Probabilities in a Multinational Phase III Trial of Decitabine in Older Patients with Newly Diagnosed Acute Myeloid Leukemia

In a recent large, international, multicenter, randomized phase III trial (NCT00260832), the hypomethylating agent decitabine elicited a significantly greater response rate (complete remission [CR] rate plus CR without platelet recovery [CRp]) of 17.8%, with an odds ratio of 2.5 (95% CI: 1.4, 4.8; P=.001), at the time of the primary analysis (clinical cut-off [CCO] date October 28, 2009, at which time 396 deaths had occurred) compared with a response rate of 7.8% for patient's treatment choice (TC), with physician's advice, of low-dose cytarabine or supportive care (SC) in older patients with newly diagnosed acute myeloid leukemia (AML). A nonsignificant trend toward an overall survival (OS) benefit was observed with decitabine (median 7.7 months) versus TC (5.0 months). A subsequent analysis on a more mature data set (CCO date October 29, 2010; 446 deaths) confirmed this trend, finding that the median OS data were the same (Kantarjian et al. J Clin Oncol 2012;30:2760). This post hoc analysis aimed to further delineate the OS trend seen with decitabine by conducting a landmark analysis to determine OS for each treatment group in the trial at fixed time points based on the mature data set.

In this multicenter, randomized, open-label phase III study, patients aged ≥65 years with de novo or secondary AML received decitabine 20 mg/m² intravenously once daily for 5 days every 4 weeks or TC (SC or cytarabine 20 mg/m² subcutaneously once daily for 10 days every 4 weeks). The primary efficacy population comprised all randomized patients. In the landmark analysis, OS at 3, 6, 12, 18, and 24 months postrandomization was estimated for each treatment group using the Kaplan-Meier method. Hazard ratios (HRs) were calculated using a Cox regression model; P values for the difference between the decitabine and TC arms were calculated using a 2-sided log-rank test.

In total, 485 patients were randomized to receive decitabine (n=242) or TC (n=243: 215 cytarabine; 28 SC). This was a high-risk AML population: at baseline, median patient age was 73.0 years (range, 64.0–91.0; 70.9% were ≥70 years), 59.4% were men, 35.3% had secondary AML, 36.0% had poor-risk AML, 25.8% had ECOG PS of 2 or higher, and median baseline blasts in bone marrow were 46.0%. In the post hoc landmark analysis using the mature data set (2010 CCO), a survival advantage was seen with decitabine at each time point (Table), with significant differences in OS at 6, 18, and 24 months (P=.009, .027, and .019, respectively).

At fixed time points over 2 years, treatment with decitabine showed a trend toward improved OS compared with TC.

CI, confidence interval. <>*Survival rates based on Kaplan-Meier product limit estimates.

Three patients with no prerandomization choice available and later randomized to the decitabine group were excluded from the analysis.

Note: Analysis based on the mature data set (2010 CCO).

Off Label Use: Decitabine is a nucleoside metabolic inhibitor indicated for treatment of patients with myelodysplastic syndromes (MDS) including previously treated and untreated, de novo and secondary MDS of all French-American-British subtypes (refractory anemia, refractory anemia with ringed sideroblasts, refractory anemia with excess blasts, refractory anemia with excess blasts in transformation, and chronic myelomonocytic leukemia) and intermediate-1, intermediate-2, and high-risk International Prognostic Scoring System groups. Delaunay:Novartis, Genzyme: Consultancy. Jones:Eisai Inc: Employment. Barrak:Eisai Inc.: Employment. Kantarjian:Eisai: Research Funding.