Secondary Acute Myeloid Leukemia (AML), Including Therapy-Related AML, Associates with Worse Treatment Outcomes, Than De Novo AML Even After Hematopoietic Stem Cell Transplantation, and Its Unfavorable Behavior Does Not Associated with Functional Activity of p-Glycoprotein, MDR1

Secondary acute myeloid leukmia (AML; sAML) is a distinct disease entity among AMLs. It usually includes AMLs with previous history of myelodysplastic syndrome (MDS) or myeloproliferative neoplasm (MPN), and therapy-related AMLs (tAMLs) developed after exposure to leukemogenic chemotherapy or radiation therapy. It has been known that its prognosis is very poor with median survival of less than 1 year, and that allogeniec hematopoietic stem cell transplantation (HSCT) is the only curative treatment modality with improving clinical outcomes. The current study evaluated clinical characteristics, treatment outcomes in patients with sAML, the impact of cytogenetic risk group or functional activity of p-glycoprotein, related to multidrug resistance 1 (MDR1), on the treatment outcomes in sAML compared to those in de novo AML (dAML).

A total of 441 consecutive AML patients diagnosed at the Samsung Medical Center, Seoul, Korea between Jan 2002 and Jun 2010 were included in this retrospective study. Out of them, 76 patients (17.2%) were sAML, among whom 39 patients (8.8%) had previous history of MDS or MPN and 37 patients (8.4%) were tAML. The overall survival (OS) and event free survival (EFS) were evaluated according to the cytogenetic risk group and the performance of HSCT, in addition to other clinical risk factors. Furthermore, P-glycoprotein assay for multidrug resistance 1 (MDR1) functional activity was performed. MDR1 functional activity was calculated by verapamil-inhibited rhodamine-123 efflux activity minus uninhibited rhodamine-123 efflux activity, and positivity was determined as equal to or over 5% MDR activity.

In sAML, a trend of higher frequency of unfavorable cytogenetic group was noted (23.7% in sAML vs 13.7% in dAML). The OS and EFS were significantly shorter in sAML than in dAML. Median OS duration was 12.8 months in sAML, but not reached in dAML (p<0.001), while that of EFS was 8.5 months in sAML, vs. 28.1 months in dAML (p<0.001).

Confined to intermediate cytogenetic risk group, sAML group showed shorter OS than dAML group (8.0 months vs 55.1 months; p<0.001). However, no difference of OS was noted between sAML vs dAML in favorable (not reached vs not reached; p=0.4) or unfavorable cytogenetic risk groups (16.5 months vs 11.2 months; p=0.7), respectively. With respect to EFS, when confined to intermediate cytogenetic risk group, sAML group showed shorter EFS than dAML group (5.7 months vs 19.9 months; p<0.001). No difference of EFS was noted between sAML vs dAML in unfavorable cytogenetic risk groups (12.6 months vs 7.9 months; p=0.9). Of interest, in favorable cytogenetic risk group, adverse impact of sAML over dAML on EFS was suggested (15.4 months vs not reached, p=0.1).

The patients receiving HSCT for sAML showed better survival compared to those not receiving HSCT for sAML (p<0.001). Among patients receiving HSCTs, OS of sAML patients was still shorter than that of dAML patients (28.3 months vs not reached, p=0.01). Even more, among patients without receiving HSCT, OS of sAML patients was significantly shorter than that of dAML patients (4.4 month vs 28.3month, p<0.001), thus still requiring further improvement of HSCT outcomes in sAML.

In terms of MDR1 activity, no difference of MDR1 positivity was noted between sAML (60.6% [n=20/33]) and dAML (61.9% [n=117/189]; p=0.8). Also, mean MDR1 activity was indifferent between in sAML and dAML groups (12.5±2.9% in sAML, 13.4±1.8% in dAML, p=0.8). This finding suggested that MDR1 may not be resopnsbile for drug resistance in patients with sAML, thus other mechanism needs to be investigated for unfavorable characteristics of sAML.

The current study suggested that 1) the patients with sAML showed a poorer outcome especially in intermediate cytogenetic risk group; 2) HSCT could improve treatment outcomes of sAML patients, but transplant outcome in sAML patients is still inferior to that in dAML patients receiving HSCT; 3) MDR1, one of important resistance mechanism of AML, may not be the case for resistance in sAML.

No relevant conflicts of interest to declare.