Abstract
Background: Based on sequencing studies the molecular landscape of AML has been unraveled. Novel prognostic scores combining molecular mutations and karyotype have been proposed (Grimwade et al. Blood, 2016, Döhner et al. NEJM 2015). However, these proposed classification systems differ in several aspects and yet no consensus has been established which genetic information is required for prognostication in AML today.
Aims: 1) Test the prognostic value of a panel of molecular markers in addition to a cytogenetic score in a large cohort of AML patients. 2) Determine the proportion of patients with a suitable molecular marker for disease monitoring (MRD) applying molecular genetics.
Patients and Methods: 867 de novo AML cases younger than 60 years were investigated (median age: 48 years, median follow up of 41 months). All patients were evaluated for karyotype, KMT2A-PTD, FLT3-ITD and in addition for mutation status of ASXL1, CEBPA, DNMT3A, NPM1, RUNX1 and TP53 according to the proposal by Grimwade et al. Blood 2016.
Results: First, AML were classified according to the refined MRC cytogenetic classification with AML with t(15;17)/PML-RARA regarded as a separate group (n=89 (10%), 90% overall survival (OS) at 5 years). 89 cases (10%) were assigned to the favourable risk group (t(8;21)/RUNX1-RUNX1T1: n=42; inv(16)/t(16;16)/CBFB-MYH11: n=47), 570 (68%) to the intermediate risk group and 119 (14%) to the adverse risk group. OS at 5 years was 66%, 53% and 28%, respectively, and differed significantly between all four subgroups (for all comparisons p<0.001). Next, the following subgroups were separated: CEBPA double mutated (dm) cases (n=44 (5%); OS at 5 years: 83%), NPM1mut/FLT3-ITD- AML (n=181 (21%); OS at 5 years: 62%), and NPM1mut/FLT3-ITD+ AML (n=137 (16%); OS at 5 years: 47%; for all comparisons between these 3 groups p<0.001). Thus, prognosis of CEBPAdm cases was comparable to PML-RARA+ AML. OS in NPM1mut/FLT3-ITD- AML is comparable to CBF-leukemias. In NPM1mut AML no prognostic impact of DNMT3Amut was found. In all these 3 groups defined on molecular genetics no prognostic impact of additional karyotype information on OS was observed. Next, in the remaining cases of the cytogenetic intermediate risk group (n=209) the prognostic impact of mutations in ASXL1, DNMT3A, RUNX1, TP53, KMT2A-PTD and FLT3-ITD was evaluated. In multivariate Cox regression analysis mutations in TP53 (relative risk (RR): 3.5; p=0.04), ASXL1 (RR: 2.2, p=0.004), and FLT3-ITD (RR: 1.8; p=0.04) were independently associated with shorter OS. OS at 5 years was 25% in cases carrying at least one of these mutations compared to 54% in cases with none of these mutations (p=0.001). Within the adverse cytogenetic risk group cases with either a complex karyotype (n=27) or a KMT2A (n=25) or MECOM rearrangement (n=14) had the worst outcome compared to the remaining cases (OS at 5 yrs: 19% vs 54%, p=0.02). In the remaining cases the presence of at least one mutation in either ASXL1, TP53 or FLT3-ITD was associated with worse outcome (OS at 5 yrs: 33% vs 74%, p=0.04). Thus, AML with complex karyotype, KMT2A or MECOM rearrangements had the worst prognosis, while cases with adverse cytogenetics and at least one mutation in either ASXL1, TP53 or FLT3-ITD have a slightly better outcome which is comparable to AML with intermediate risk cytogenetics harbouring one of these mutations. OS in AML with adverse cytogenetics without mutations in ASXL1, TP53 or FLT3-ITD is not worse than in AML with intermediate cytogenetics also lacking these mutations.
A fusion gene or a molecular mutation as target for MRD monitoring was present in 791 patients (91%) when all genes analysed were taken into account. If only markers showing prognostic relevance, i.e. fusion genes, mutations in NPM1, CEBPA, FLT3-ITD, ASXL1 and TP53 were considered a MRD marker was still available in 726 cases (84%).
Conclusions: 1) In AML a prognostication system is feasible based on the identification of t(15;17)/PML-RARA, t(8;21)/RUNX1-RUNX1T1, inv(16)/t(16;16)/CBFB-MYH11, 11q23/KMT2A rearrangements, 3q26/MECOM rearrangements, complex karyotype, and mutation status of NPM1, CEBPA, ASXL1, TP53, and FLT3-ITD (figure 1). 2) The analysis of these parameters allows to identify an MRD marker in 84% of patients. 3) The analysis of additional genes may be required in a comprehensive AML work-up as soon as novel targeted treatment strategies will become available.
Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Fasan:MLL Munich Leukemia Laboratory: Employment. Perglerová:MLL2 s.r.o.: Employment. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.
