This retrospective study included a cohort of 645 elderly patients (aged 60 years and above) who were diagnosed with AML, spanning from July 2016 to December 2024 at the Fujian Medical University Union Hospital. Patients with available genetic mutation profiles were included in this study, whereas those lacking documented molecular analysis data were systematically excluded. A retrospective analysis of patient medical records enabled systematic compilation of multidimensional clinical datasets encompassing, including hematologic parameters and biochemical profiles, genetic aberrations, therapeutic protocols, objective treatment responses and long-term survival. This study was conducted in accordance with the principles of the Declaration of Helsinki. This retrospective cohort study utilizing anonymized clinical data was classified as non-interventional research under Declaration of Helsinki. In accordance with ICH-GCP guidelines regarding secondary data use, the requirement for written informed consent was formally waived by the Institutional Review Board of Fujian Medical University Union Hospital (approval no. 2024KY278), as the research posed no additional risks and preserved participant confidentiality.

The diagnosis of AML was established in accordance with the criteria provided in the 2016 revision of the World Health Organization (WHO) Classification of Tumours of Hematopoietic and Lymphoid Tissues, specifically pertaining to myeloid neoplasms and acute leukemia (21). Secondary AML (s-AML) was defined as AML that arose from pre-existing myeloproliferative disorders, such as myelodysplastic syndromes (MDS), myeloproliferative neoplasms (MPN), and chronic myelomonocytic leukemia (CMML). Patients with the blast phase of chronic myeloid leukemia (CML) were not included in this study. Therapy-related AML (t-AML) was identified as AML that developed following prior chemotherapy or radiation therapy for another malignancy. The chromosomal karyotype analysis was performed using R-banding technique in our center, and a complex karyotype was defined as the presence of three or more chromosomal aberrations. Gene mutations analysis were mainly conducted through targeted next-generation sequencing (NGS) using a custom-designed myeloid-focused gene panel (Kangsheng Global Medical Technology, Beijing) covering clinically actionable genes. Sequencing was performed on the Illumina NovaSeq X Plus platform, and the bioinformatic processing utilized an automated pipeline for hematologic neoplasm genomic reporting. The specific gene regions analyzed for mutation detection are summarized in Supplementary Table 1.

In this study, the “IC” regimen denotes the standard 3 + 7 protocol. Following the achievement of complete remission (CR) or complete remission with incomplete recovery (CRi) after IC treatment, intermediate or high-dose cytarabine was used as a consolidation therapy. Hypomethylating agent (HMA)-based therapies were administered either as monotherapy or in combination with low-dose chemotherapy regimens. For patients who achieved CR with HMA-based treatments, the same regimen was continued for consolidation. The azacitidine-venetoclax regimen utilized the standard dosage and administration. For patients who achieved CR with this combination, the concurrent administration of AZA and VEN was sustained until disease progression or intolerable adverse effects arose. A detailed summary of the initial treatment protocols for these patients, including the specific therapeutic approaches utilized, is provided in Supplementary Table 2.

The effectiveness of therapeutic interventions was assessed in accordance with the 2022 edition (5th) of the European LeukemiaNet (ELN) response criteria for AML (8). Complete blood count measurements were obtained at the commencement of each treatment cycle, and bone marrow aspirations were conducted at intervals of one or two cycles to evaluate the therapeutic response. The overall response rate (ORR) was determined by combining the rates of CR, CR with incomplete hematologic recovery (CRi), and partial remission (PR). To identify minimal residual disease (MRD) in AML, an 8-color flow cytometric protocol was utilized at our center, employing the FACSCanto flow cytometer from BD company, United States. The protocol was designed with a lower limit of detection set at 10⁻⁴. Relapse was defined as the return of leukemic cells in the bone marrow or peripheral blood after achieving CR/CRi, indicated by blast cell percentages greater than 5%, or the appearance of extramedullary lesions. The duration of remission (DOR) was calculated as the time span from the achievement of CR/CRi or PR to the occurrence of relapse or disease progression. Overall survival (OS) was defined as the time period from the date of diagnosis to the date of death or the last date of follow-up.

The statistical evaluation of differences between patient subgroups was carried out using the chi-square test for categorical variables, the t-test for continuous data that were normally distributed, or nonparametric tests for data that did not conform to the assumptions of parametric tests. The Kaplan–Meier method was used for univariate analysis to assess the prognostic impact of methylation-related gene mutations compared to wild-type and other clinical and laboratory indicators. Log-rank tests were conducted to compare differences between groups. All statistical computations and graphical illustrations were performed with GraphPad Prism 9.5 (GraphPad Software, San Diego, CA, United States). Statistical significance was established at a p-value of less than 0.05.

The study cohort comprised 645 individuals, with a male-to-female ratio of approximately 1.70:1, including 406 males (62.9%) and 239 females (37.1%). The age of the patients ranged from 60 to 98 years, with a median age of 67 years. Among these patients, 511 (79.2%) had de novo AML, 97 (15.0%) had s-AML, and 37 (5.7%) had t-AML. Myelodysplastic syndromes (MDS) were the most frequent precursors of s-AML, representing 73.2% of cases.

Cytogenetic analysis results were available for a subset of patients with 40.7% displaying abnormal chromosome karyotypes. Specific abnormalities included deletions of chromosome 5 or 5q- in 6.8%, chromosome 7- in 3.3%, chromosome 17- in 1.9%, chromosome +8 in 10.3%, and complex karyotypes in 17.5% of cases. As depicted in Figure 1, methylation-related gene mutations, including DNMT3A, TET2, IDH1, and IDH2, were observed in 24.0%, 22.5%, 9.1%, and 13.8% of patients, respectively. The incidence of IDH1 gene mutations was more occurred in De novo AML compared to s-AML/t-AML, and there was no significant difference in the occurrence rates of DNMT3A, TET2, and IDH2 gene mutations between these two AML subtypes. FLT3 gene mutations were found in 21.2% of patients, and other commonly detected mutations included those in NPM1, CEBPA, TP53, ASXL1, and RUNX1, with positive rates of 21.4%, 12.9%, 14.1%, 17.1%, and 13.8%, respectively. Based on genetic risk classification, the adverse group was the most prevalent, accounting for 55.3%, while the favorable and intermediate groups represented 18.0% and 26.7% of the patients, respectively. A comprehensive summary of the patients’ characteristics is provided in Table 1.

Of the 645 patients in the study cohort, 585 (90.7%) received further anti-leukemia treatment following diagnosis. The treatments included 240 patients (41.0%) undergoing IC, 43 patients (7.4%) receiving HMAs as monotherapy, 91 patients (15.6%) treated with a combination of HMAs and low-dose chemotherapy, 183 patients (31.3%) administered AZA in combination with VEN, and 28 patients (4.8%) managed with other therapeutic regimens. The baseline characteristics of patients receiving various treatment regimens were summarized in Table 2. From the table, it could be observed that patients in the IC group were relatively younger compared to those in other groups. Additionally, the proportions of s-AML/t-AML and adverse cytogenetics in the IC group were significant lower than those in the other treatment groups.

Among the 585 patients who received anti-leukemia therapy, 54 (9.2%) were lost to follow-up, and 66 (11.3%) died during the induction phase. As illustrated in Figure 2, the median DOR was 16.0 months, with 1-year, 3-year, and 5-year DOR rates of 55.4, 22.8, and 15.6%, respectively. The median OS was 21.2 months, with corresponding 1-year, 3-year, and 5-year OS rates of 58.0%, 40.6%, and 30.4%. Across the entire cohort, the induction treatment CR rate was 51.2%, the PR rate was 13.3%, the ORR rate was 64.5%, and the MRD-negative rate was 30.1%. Obviously, patients treated with IC showed the most favorable treatment response and long-term outcomes, with a CR rate of 58.7% and median DOR and OS of 19.8 months and 35.4 months, respectively, and a 5-year OS of 38.2%. The AZA and VEN combination therapy also resulted in a satisfactory response, with a CR rate of 56.3%, which was comparable to the IC group (p = 0.7331). HMA monotherapy had the least favorable outcomes, with a CR rate of 17.6% and median DOR and OS of only 5.2 months and 10.8 months, respectively. In contrast, the combination of HMAs with low-dose chemotherapy was more effective, with a CR rate of 39.7% and median DOR and OS of 18.0 and 16.0 months, respectively. The highest proportions of MRD-negative responses were observed in patients treated with IC and AZA + VEN, at 37.5% and 32.0%, respectively, while all other treatment protocols resulted in MRD-negative rates of less than 20%. The comprehensive treatment response and prognosis for patients undergoing various therapeutic protocols were summarized in Table 3.

To delve deeper into the impact of demethylating gene mutations on the prognostic outcomes of elderly AML patients, as well as their influence on the therapeutic efficacy of the IC and AZA plus VEN regimen, we performed a univariate prognostic analysis. As they were showed at Figure 3, our findings suggested that mutations in DNMT3A and TET2 were inclined to correlate with an adverse prognosis. Specifically, the median OS for patients with DNMT3A mutations, as opposed to those without, were 12.8 months and 22.3 months, respectively, yielding a p-value of 0.1922. Similarly, individuals harboring TET2 mutations experienced a median survival of 13.7 months, in contrast to 21.2 months for their mutation-negative counterparts, with a p-value of 0.2081. Moreover, the poorest prognostic outcomes were observed in patients with concurrent DNMT3A and TET2 mutations, exhibiting a median survival of merely 6.2 months and a 2-year overall survival rate of 23.9%. In stark contrast, patients with IDH1 and IDH2 mutations exhibited a trend toward improved long-term prognoses. Those with IDH1 mutations had a median survival of 25.2 months, while the median survival for IDH2 mutation carriers had not yet been reached, compared to 18.0 months for wild-type patients, with a p-value of 0.1709.

As presented at Figure 4, patients harboring the DNMT3A mutation exhibited an induction remission rate that was nearly identical to that of their counterparts without the mutation, with respective complete remission (CR) rates of 53.6% and 50.6%, yielding a p-value of 0.5868. The therapeutic efficacy of the IC regimen and the combination of azacitidine with venetoclax were found to be equivalent, with CR rates of 58.0% and 59.0%, respectively, (p > 0.9999). Patients with the TET2 mutation experienced a CR rate of 42.2%, which was lower than the 53.9% observed in non-mutated patients (p = 0.0594). Additionally, the induction CR rate in the IC group was 56.5%, surpassing the 37.0% rate in the azacitidine plus venetoclax group, with a p-value of 0.1468. Notably, patients who carried both DNMT3A and TET2 mutations had an induction CR rate of merely 35.5%, which below the rates for those with isolated mutations of either DNMT3A or TET2 (p = 0.0974). The effectiveness of the IC regimen was similar with the azacitidine plus venetoclax regimen (CR rate: 41.2% vs. 37.5%, p > 0.9999). In contrast, patients with IDH1 or IDH2 mutations demonstrated a markedly higher induction CR rate compared to wild-type patients, with CR rates reaching 71.1% for IDH1 mutation carriers and 68.8% for IDH2 mutation carriers. For IDH1 mutation carriers treated with the IC regimen, the CR rate was as high as 82.4%, accompanied by a MRD-negative rate of 70.6%. Among IDH2 mutation carriers, the induction CR rate reached 68.8%, with CR rates of 68.0 and 76.9% for the IC and azacitidine plus venetoclax regimen regimens, respectively (p = 0.5414).