Older Patients with AML Live Longer with Azacitidine

Robert Osborne

August 2014, Vol 5, No 6 - EHA 2014


Milan, Italy—Older patients with acute myeloid leukemia (AML) gained a “clinically significant,” if not statistically significant improvement in survival when treated with azacitidine (Vidaza) rather than with conventional therapy, reported Hervé Dombret, MD, hematologist, Hôpital Saint Louis in Paris, France, at the European Hematology Association meeting.

Azacitidine therapy led to a median overall survival (OS) of 10.4 months compared with 6.5 months with conventional treatment. An analysis that ended with a switch to a new therapy led to a 5-month difference in survival, which was statistically significant, said Dr Dombret.

“Although this study did not reach statistical significance for its primary end point of overall survival, azacitidine demonstrated a clinically meaningful improvement in median overall survival,” said Dr Dombret.

OS with censoring at subsequent AML therapy did demonstrate a clinically significant benefit for azacitidine, and patients had substantial improvement in 1-year survival.

Historically, older patients with AML have a poor prognosis with standard therapy and a median survival of 2 to 8 months, said Dr Dombret. Poor health status and comorbid conditions rule out treatment with intensive chemotherapy for many older patients.

Building on earlier positive results with this drug, Dr Dombret and colleagues conducted a phase 3, multicenter, randomized trial involving patients aged ?65 years with newly diagnosed AML associated with >30% blasts in bone marrow; patients had intermediate- or poor-risk cytogenetics and were not eligible for stem-cell transplantation.

All patients received best supportive care and were randomized to azacitidine, low-dose chemotherapy, intensive chemotherapy with cytarabine and an anthracycline, or to best supportive care alone.

The primary end point was OS, but investigators also planned a survival analysis with data censoring at the start of subsequent therapy for AML. The secondary end points included 1-year survival, event-free survival, and relapse-free survival. The final analysis included 488 patients (median age, 75 years) with a median blast involvement of 70%.

The difference in the primary (unstratified) analysis translated into a 16% reduction in the hazard for survival, which failed to reach statistical significance. The investigators prespecified a survival analysis stratified by Eastern Cooperative Oncology Group performance status and cytogenetic risk. That analysis yielded a 15% reduction in the hazard, which also failed to achieve statistical significance.

The censored survival analysis resulted in median OS times of 12.1 months in the azacitidine arm and 6.9 months in the conventionally managed patients. The difference represented a 25% reduction in the survival hazard and was a significant outcome.

With respect to secondary end points, the data showed an absolute 12% difference in 1-year survival in favor of the azacitidine arm (46.5% vs 34.2%, respectively). The difference retained significance in a stratified analysis.

The subgroup analysis showed a consistent benefit of azacitidine versus conventional therapy: larger benefits were seen in patients under age 75 years, and in women.

Overall, 69 patients in the azacitidine arm and 75 patients in the conventional arm received subsequent therapy. Patients in the azacitidine arm lived longer with subsequent therapy. The median OS from first subsequent treatment was approximately 8 months with azacitidine and 3 to 4 months with conventional therapy.

Secondary outcomes included an event-free survival of 6.7 months with azacitidine versus 4.8 months with conventional therapy, and relapse-free survival times of 9.3 months and 10.5 months for the groups, respectively. Complete responses were reported in 28% of patients with azacitidine versus 25% of patients with conventional treatments; the rates of stable disease were also similar.

“The safety profile was consistent with that previously observed with azacitidine.” Grade 3/4 hematologic toxicity was similar among the groups.