CD25 Expression Is Associated with Inferior Clinical Outcomes in Elderly Patients with Acute Myeloid Leukemia

Background: Elderly patients with acute myeloid leukemia (AML) show unfavorable biologic characteristics and inferior clinical outcomes compared to younger patients. Because some elderly patients can benefit from intensive chemotherapy, the identification of prognostic markers is important for therapeutic decision-making. CD25 (interleukin 2 receptor alpha) has been reported to be highly expressed in leukemia stem cells and correlate with adverse outcomes in younger (less than 60 years) patients with AML. However, the significance of CD25 expression in elderly patients with AML has not yet been investigated.

Methods: We retrospectively reviewed 154 newly diagnosed AML (de novo, secondary) patients aged 60 years or over (median: 69, range: 60-85 years), admitted between 2005 and 2013 to Jichi Medical University Hospital. Blast cells from bone marrow at diagnosis were identified by flow cytometry (FCM) using CD45/side scatter properties. CD25 expression in the blast region was analyzed by FCM combined with single- and two-color staining.

Results: The two-color FCM analysis showed that CD25 was expressed in CD34, CD117, and CD13-positive blast cells. The median CD25 expression of the blast cells was 0.7% (range: 0-75.3%). On the basis of CD25 expression in bone marrow blasts from healthy volunteers (n=5; median, 3.2%; range, 1.5-5.6%), we defined CD25 expression >10% as positivity. In this setting, we identified CD25-positive blasts in 21 patients (14%). CD25+AML was characterized by the MDS/MPN history (30%) and, a low or zero frequency of favorable cytogenetics (0%), FAB M2 subtype (5%), and FAB M3 subtype (0%) compared with CD25-AML (n=133). Among the 118 patients who underwent induction chemotherapy, including 75 patients who received a combination of anthracyclin and cytarabine (Ara-C), 69 (58%) achieved complete remission (CR). CD25+AML was associated with an inferior CR rate (20 vs. 64%, respectively, P=0.003), inferior event-free survival (EFS) (1-year EFS, 0 vs. 33%, respectively, P< 0.001), and inferior overall survival (OS) (2-year OS, 13 vs. 38%, respectively, P=0.003) compared with CD25-AML. On multivariate analysis using the following factors: cytogenetics, age (over 75 years), MDS/MPN history, and intensity of induction chemotherapy, CD25-positivity was a significant risk factor for the CR rate (HR: 8.0, 95% CI: 1.7-36.6; P=0.008) and EFS (HR: 2.3, 95% CI: 1.1-4.7; P=0.021). For CD25+AML, the CR rate was higher with induction chemotherapy using a regimen of low-dose Ara-C combined with granulocyte colony-stimulating factor (50%) than that using a regimen of daunorubicine and Ara-C (0%). However, there was no significant difference in OS between patients with CD25+AML treated with induction chemotherapy and those who treated with best supportive care (2-year OS, 13 vs. 11%, respectively, P=0.29). There were no patients with CD25+AML received stem cell transplantation due to disease progression.

Conclusion: This study demonstrated that CD25 is an independent prognostic factor for elderly AML patients. The prognosis of CD25+AML patients is extremely poor; therefore, it is necessary to develop alternative approaches, such as a regime including anti-CD25 antibody, therapy targeting leukemia stem cells, and nonmyeloablative stem cell transplantation.