A New Multi-Factor Risk Score System Depending on Gene Mutation and Overexpression for Predicting the Outcome in AML1-ETO-Positive Acute Myeloid Leukemia

Acute myeloid leukemia with AML1-ETO-positive (AE-AML) is a heterogeneous myeloid tumor with a big scale of long-term survival. Lots of factors have been reported to have an effect on survival; however, there is not a multi-factor risk score system to balance the significance of different prognostic factors and guide the risk-stratification and precise treatment on patients. We enrolled 64 patients with newly diagnosed as AE-AML, detested their gene mutation with the next-generation sequencing (NGS), monitored the expression of amyloid precursor protein (APP) expression and minimal residual disease (MRD) using quantitative real-time polymerase chain reaction (qPCR). The results showed that 47/64 (73.4%) patients presented gene mutations at diagnosis and 6/11 underwent genetic alteration at relapse, in which c-KIT mutation showed up the most frequently. Univariate factors analysis showed that the clinical features of onset, including EML, bone marrow cellularity rate ≥90%, APP overexpression, additional gene mutations, especially mutations in MET, ASXL1, DNMT3A, MLH1 and c-KIT (exon 8, 17), badly affected on the RFS and OS, while MDAC-based regimens for consolidation after CR and obtaining major molecular remission (MMR) after two courses of consolidation were the protective factors for recurrence-free survival (RFS) and overall survival (OS). Multivariate analysis showed ASXL1 mutation (RFS: HR=4.16, 95%CI= 1.02-17.04, P=0.048), MET mutation ( RFS: HR=6.01, 95%CI= 1.19-30.36, P=0.030; OS: HR=7.16, 95%CI= 1.36-37.70, P=0.020) and APP overexpression (RFS: HR=6.10, 95%CI= 1.05-35.42, P=0.044) were the adverse factors for survival; Obtaining MMR after two courses of consolidation was the protective factor for RFS (HR=0.03, 95%CI= 0.00-0.28, P=0.002) and OS (HR=0.05, 95%CI= 0.01-0.51, P=0.011). We defined each risk factor with score one, and divided all patients into score 0-1, 2 and 3-4 groups, and found that the RFS and OS rate significantly decreased with the risk factors increasing (the 3-year RFS: 82.0%±9.0% vs. 26.0%±6.0% vs. 0, P<0.001; the 3-year OS: 96.0%±4.0% vs. 31.0%±7.0% vs. 0, P<0.001). Both of the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) value of the new multi-factor risk score system were the smallest, compared with those of the four single factors and the NCCN risk system, suggesting that this new system might have great priority in risk stratification. In conclusions, most patients with AE-AML underwent gene mutations at diagnosis and experienced genetic alteration at relapse. The new multi-factor score risk system combined with ASXL1 and MET mutation, APP overexpression and MMR at the second-course consolidation might have great priority in risk stratification.