Neoplastic stem cells (NSCs)are the source of tumor occurrence and development,and its special microenvironment is involved in tumor drug resistance,recurrence and metastasis by regulating NSCs or changing itself. Eradicating NSCs and interfering the special microenvironment may be the new therapeutic strategies for cancer.

Objective:To investigate the distribution characteristics of gene mutations and their relationship with prognosis in newly diagnosed patients with acute myeloid leukemia (AML).Methods:The clinical data of 225 newly diagnosed AML (non-acute promyelocytic leukemia) patients in the Second Affiliated Hospital (Tangdu Hospital) of Air Force Medical University from May 2016 to December 2019 were retrospectively analyzed. Thirty-four gene mutations related to AML, myelodyplastic syndrome (MDS) and myeloproliterative neoplasms (MPN) were detected by second-generation sequencing. The distribution of all gene mutations and its difference among different age groups were analyzed, and the differences in survival of patients with different gene mutations were compared. The Cox regression model was employed to analyze the survival influencing factors of patients aged <60 years old and ≥60 years old.Results:A total of 496 gene mutation sites were detected in 225 patients, with a median variant allel frequency (VAF) of 38.55% (1.00%-94.86%) and a median gene mutations of 3/case (0-8/case). The genes with high mutation frequency were ASXL1, CEBPA, NPM1, NRAS, FLT3-ITD, DNMT3A, IDH2, TET2, RUNX1, and IDH1. The gene mutation rates of TET2, SRSF2 and SF3B1 in patients aged ≥60 years old (56 cases) were higher than those in patients aged <60 years old (169 cases), and the differences were statistically significant (all P < 0.01). The proportion of patients aged ≥60 years old with 3 or more gene mutations was higher than that of patients aged <60 years old [53.6% (30/56) vs. 33.1% (56/169), χ2 = 7.44, P = 0.006]. The overall survival (OS) of patients with TP53, RUNX1 or FLT3-ITD gene mutation was worse than that of wild-type patients, the OS of patients with CEBPA double mutations was better than that of patients with CEBPA single mutation or wild-type, and the differences were statistically significant (all P < 0.05). Multivariate Cox regression analysis showed that CEBPA ( HR = 0.279, 95% CI 0.084-0.926, P = 0.037), TET2 ( HR = 2.611, 95% CI 1.115-6.111, P = 0.027) and TP53 ( HR = 3.609, 95% CI 1.159-11.234, P = 0.027) gene mutations were independent factors affecting the survival of AML patients aged <60 years old, ASXL1 ( HR = 3.523, 95% CI 1.385-8.962, P = 0.008), FLT3-ITD ( HR = 4.618, 95% CI 1.813-11.762, P = 0.001) and NRAS ( HR = 2.896, 95% CI 1.166-7.000, P = 0.022) were independent risk factors of the survival of AML patients aged ≥60 years old. Conclusions:There are differences in the distribution of gene mutations among AML patients with different age, and the elderly patients are more likely to have multiple gene mutations at the same time. In addition to the currently known CEBPA double mutations, TP53, ASXL1, RUNX1 and other gene mutations, TET2 and NRAS gene mutations may also be factors affecting the prognosis.