Allogeneic Stem Cell Transplantation Confers a Favorable Outcome in Patients with NPM1 Positive Acute Myeloid Leukemia: Results From a Donor Vs. No-Donor Analysis of 309 Patients Treated in the SAL AML-2003 Trial

According to retrospective analyses, the presence of a mutated Nucleophosmin-1 gene (NPM1+) in acute myeloid leukemia (AML) is associated with a favorable prognosis, particularly in the absence of an FLT3-ITD mutation (FLT3-ITD-). Therefore, AML with NPM1+/FLT3-ITD- and normal karyotype has been classified as favorable risk in current prognostic classifications. In order to assess the predictive value with regard to allogeneic stem cell transplantation (allo SCT), we compared the clinical course of 309 NPM1+ AML patients eligible for allo SCT in a donor versus no-donor analysis.

Patients diagnosed with AML, aged 18–60 years, and treated in the AML 2003 trial of the Study Alliance Leukemia (SAL) were analyzed. According to the risk-adapted treatment strategy of the trial, cytogenetically intermediate-risk (IR) and adverse-risk (AR) patients should receive an allo SCT as consolidation treatment if an HLA-identical-sibling donor (IR) or HLA-matched related or unrelated donor (AR) was available. Patients with no available donor received high-dose cytarabine-based consolidation or autologous SCT. In order to avoid selection bias in an as-treated analysis of transplanted versus non-transplanted patients, we compared relapse-free survival (RFS) and overall survival (OS) depending on the availability of a suitable donor in a donor-no-donor analysis. Survival analyses were performed using the Kaplan-Meier method including log-rank tests for significance testing. Cox regression models and Wald tests were used for multivariate analyses on the influence of potential prognostic variables on the outcomes.

Of 1182 patients enrolled in the AML 2003 trial between December 2003 and November 2009, 375 were NPM1+ (32%), and 309 patients were eligible for evaluation for the donor vs. no-donor analysis. Their median age was 49 years, 304 patients had an intermediate-risk karyotype according to MRC criteria (98%), and amongst them there were 277 patients with a normal karyotype (90%). The FLT3-ITD mutation was present in 144 patients (37%). A donor was identified for 77 patients (25%), of whom 57 actually received allo SCT as first consolidation (74%). The no-donor group consisted of 232 patients. Age, disease status, cytogenetic profile, and FLT3-ITD incidence were equally distributed between the two groups. Median follow up was 41 months (3.4 years). The 3-year RFS in the donor and no-donor groups was 72% (95%–CI 61%–82%) and 47% (95%–CI 40%–55%), respectively. (p=0.007). The OS in the donor and no-donor groups were 70% (95%–CI 59%–81%) versus 60% (95%–CI 54%–67%) after 3 years, and 70% (95%–CI 59%–81%) versus 53% (95%–CI 45%–61%) after 5 years (p=0.138). In multivariate analyses, the presence of a donor as a prerequisite for allo SCT retained its statistically significant favorable influence on RFS (HR=0.56) even after adjustment for established risk factors such as FLT3-ITD, cytogenetic risk, WBC, LDH, age, and disease status. In patients with normal karyotype and NPM1+/FLT3-ITD- (n=152), the 3-year RFS in the donor and no-donor groups was 87% (95%–CI 77%–97%) and 53% (95%–CI 42%–63%), respectively (p=0.001).

According to our results, allo SCT leads to a significantly prolonged RFS in NPM1+ AML patients with a pronounced effect even in NPM1+/FLT3-ITD- patients. The absence of a statistically significant difference in OS is most likely due to the fact that relapsed NPM1+ patients responded well to salvage treatment, particularly to allo SCT from an unrelated donor. Our data suggest that patients with NPM1+ AML who have a well-matched donor benefit from allo SCT in first remission. This hypothesis is currently being tested prospectively in a randomized controlled trial (“ETAL-1”, NCT01246752) evaluating allo SCT in all intermediate-risk AML patients with a well-matched sibling or unrelated donor identified until the achievement of first CR.

No relevant conflicts of interest to declare.