Treatment without cytarabine (ara-C) is increasingly being used as initial therapy for acute myelogenous leukemia (AML) particularly in patients (pts) deemed ineligible, usually on the basis of age, for standard ara-C-containing therapy; prime examples are the hypomethylating agents decitabine (DAC) and 5-azacitidine (Aza). This practice raises the question whether it is useful for patients whose disease fails to respond to hypomethylating agents to receive ara-C as salvage. Thirty-five pts treated from 2001 to 2006 failed to respond to initial therapy with decitabine/azacitidine. Twenty-five pts subsequently received ara-C as salvage and 10 received other salvage regimens. We compared their complete remission (CR) rates (Table 1). Although more pts who received non ara-C salvage had poor prognosis karyotype, CR rates in pts with standard-risk karyotype receiving ara-C and non ara-C salvage therapy were 7/20 (35%) vs. 0/6 (0%) respectively. Although this difference was not significant (p=.15) the 95% confidence limits for the true difference, computed so as to account for the small sample size, was [−.69, .11] indicating as much evidence for a true difference of 58% as for a true difference of 0% since zero and 58% are equidistant from the midpoint of the confidence interval (−0.29). Although future improvements in results with non ara-C containing salvage therapies will impact decision, the results suggest that it is worthwhile to give pts who fail initial therapy with decitabine or azacitidine, standard dose ara-C salvage provided the pts do not have −5/−7 (CR rate 0/5).

Table 1
Induction/salvagePatientsMedian age−5 / −7CR
DAC or Aza/ara-C 25 67 5 (20%) 7 (28%) 
DAC or Aza/no ara-C 10 71 4 (40%) 
Induction/salvagePatientsMedian age−5 / −7CR
DAC or Aza/ara-C 25 67 5 (20%) 7 (28%) 
DAC or Aza/no ara-C 10 71 4 (40%) 

Author notes

Disclosure: No relevant conflicts of interest to declare.

Sign in via your Institution