Allogeneic Stem Cell Transplantation in Acute Myeloid Leukemia with Normal Karyotype: A Risk Factor Analysis in 247 Patients, Based On Molecular Markers and Stage at Transplantation.

The prognosis of patients with cytogenetically normal acute myeloid leukemia (CN-AML) ranges from relatively favorable to extremely poor. Recently, based on the presence or absence of well defined mutations, molecular subgroups have been identified, which allow an estimate of a patient's prognosis at the time of diagnosis. Allogeneic stem cell transplantation (SCT) is the only curative treatment for the majority of these patients. However, only limited data is available to describe the role of alloSCT in different molecular subgroups of CN-AML, particularly in advanced stages of the disease.

We retrospectively analyzed the data on 247 patients with CN-AML, who uniformly had received the FLAMSA-RIC conditioning regimen for alloSCT in 14 European centers between 1996 and 2008.

Patients suffered from de novo AML (76%), sAML/MDS (21%), and tAML (4%). Median age was 52.1 (19-71) years. Donors were matched or mismatched family, and matched or mismatched unrelated donors in 30%, 2%, 50% and 18%, respectively. SCT was performed in untreated disease (6%), after primary induction failure (PIF, median time from diagnosis to transplantation 134 days; 23%), in first complete remission (CR1, 14%), and beyond CR1 (57%). Median follow-up of survivors was 19 months. Overall survival (OS) and leukemia-free-survival (LFS) of the entire cohort at 2 years from SCT was 51% and 47%, respectively. The disease stage at transplant was the most important variable for outcome (p=.001 for OS, <.001 for LFS): Encouraging results were achieved in patients transplanted in CR1 (2y OS and LFS: 76%), and in patients with PIF (2y OS and LFS: 69%), whereas results were inferior after transplantation in previously untreated disease (2y OS and LFS: 34%), or beyond CR1 (2y OS: 42%, LFS: 34%). Age, sex, de novo vs. secondary leukemia, donor type and CD34+ cell counts showed no influence on outcome.

Information on molecular markers was available in 183 patients (74%). As suggested by Schlenk et al. (NEJM 2008), analysis was based on two subgroups: 22 patients with isolated NPM1 mutation (NPM1mut), and 161 patients with other genotypes (FLT3 internal tandem duplication [FLT3-ITD], n=66; or wildtype FLT3/wildtype NMP1 [FLT3wt/NPM1wt], n=95). Patients with NPM1mut had a 4y OS/LFS of 75/63%. Results were not significantly different, when these patients were transplanted in PIF, CR1, or beyond CR1. Patients with other genotypes showed an OS/LFS of 51%/48% at 2y and of 40%/39% at 4y, without differences among patients with FLT-ITD and FLT3wt/NPM1wt. However, in this subgroup, outcome was highly dependent on the disease stage at SCT, with excellent results after transplantation in PIF (2y OS/LFS: 75%/74%) or in CR1 (2y OS and LFS: 76%), but inferior outcome after transplantation beyond CR1 (2y OS/LFS 38%/33%; p=.004 for OS and .001 for LFS).

Allogeneic SCT following the FLAMSA-RIC conditioning produces excellent survival rates in patients with CN-AML, particularly when performed in CR1. Encouraging results in PIF support an early transplant, regardless of molecular subgroup, when CR is not reached after double induction therapy. In patients with an NPM1 mutation, transplantation in advanced disease achieved identical results as in early stage, which supports the strategy not to transplant these patients in CR1, but to delay alloSCT until relapse has occurred. In contrast, patients with FLT3-ITD or FLT3wt/NPM1wt achieved significantly worse results when transplanted beyond first relapse, arguing in favor of transplantation in CR1 for this molecular subgroup.

Mayer:BMS: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; GSK: Consultancy; Fresenius: Consultancy; Roche: Research Funding; Pfizer: Research Funding.