AFFILIATIONS
1 Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
2 Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
3 The first hospital of Jilin University, Changchun, China
4 Nanfang Hospital, Southern Medical University, Guangzhou, China
5 Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
6 Institute of Science Innovation for Blood Ecology and Intelligent Cells, Xinqiao Hospital of Army Medical University, Chongqing, China
7 Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
8 Qilu Hospital of Shandong University, Jinan, China
9 Department of Hematology and Shenzhen Bone Marrow Transplantation Public Service Platform, Shenzhen Second People’s Hospital, The First Affiliated Hospital of Shenzhen University, Shenzhen University School of Medicine, Shenzhen, China
10 The First Affiliated Hospital of Guangxi Medical University, Nanning, China
11 Yichang Central People's Hospital, The First Clinical Medical College of The Three Gorges University, Yichang, China
12 Jingzhou Hospital, Affiliated to Yangtze University, Jingzhou, China
13 The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China
14 The Second Hospital of Hebei Medical University, Shijiazhuang, China
15 Xi'an International Medical Center Hospital, Xian, China
16 The affiliated hospital of southwest medical university, Luzhou, China
17 The First People's Hospital of Yunnan Province, The Affiliated Hospital of Kunming University of Science and Technology, Kunming, China
18 Department of Epidemiology and Biostatistics, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
19 Musketeers Foundation Institute of Data Science, The University of Hong Kong, Hong Kong, China
20 Musketeers Foundation Institute of Data Science, The University of Hong Kong, Hong Kong, China
21 Pediatric Hematology and Oncology, Haley Center for Children’s Cancer and Blood Disorders, Orlando Health, Arnold Palmer Children's Hospital, Orlando, United States
22 Department of Hematology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
ABSTRACT
Background
Acute myeloid leukemia (AML) with t(9;11) (p21.3;q23.3)/KMT2A::MLLT3 is currently classified as intermediate risk despite marked clinical heterogeneity. We evaluated clinical, immunophenotypic, molecular, and treatment-related factors associated with outcomes in the largest multicenter cohort of patients with this AML subtype reported to date.
Methods
We retrospectively analyzed 183 patients with AML harboring t(9;11)/KMT2A::MLLT3 treated across multiple institutions. Overall survival (OS) and event-free survival (EFS) were estimated using Kaplan-Meier methods. Prognostic factors were evaluated using Cox proportional hazards regression. A machine learning model integrating clinical and biologic variables was developed for individualized risk prediction.
Results
The median age was 42 years, and 83.1% had de novo AML. Immunophenotyping demonstrated universal CD33 expression, frequent monocytic antigen positivity, and CD56 expression in 29.0% of cases. Co-occurring mutations were identified in 54.3% of patients, most commonly involving the RAS signaling pathway (32.2%). After a median follow-up of 1.7 years, the 2-year OS was 46.0%, with an estimated 5-year OS of 41.2%. Among patients achieving complete remission after induction I, allogeneic hematopoietic stem cell transplant (allo-HSCT) was associated with significantly improved 2-year OS (80.2% vs. 31.0%) and relapse-free survival (76.4% vs. 36.3%) compared with no transplant. Positive flow minimal residual disease (MRD) after induction II induction (53.7% vs. 33.9%) and molecular MRD after induction I (40.5% vs. 18.6%) were each associated with a higher risk of 2-year relapse. Patients treated with venetoclax plus hypomethylating agent regimen had the poorest outcomes, whereas intensive chemotherapy alone and venetoclax combined with intensive chemotherapy achieved comparable survival. On multivariable analysis, age >40 years (HR 1.04), TP53 mutation (HR 9.81), and venetoclax-based low-intensity therapy (HR 2.13) were independently associated with inferior OS, whereas CD56 expression (HR 0.36) and allo-HSCT (HR 0.21) were independently predicted improved survival. A machine learning prognostic model demonstrated strong discrimination and was deployed as an online risk calculator for individualized survival prediction.
Conclusions
AML with t(9;11)/KMT2A::MLLT3 exhibits substantial clinical and molecular heterogeneity that is not adequately captured by current intermediate-risk classification. Age, TP53 mutation, CD56 expression, treatment intensity, MRD status, and allo-HSCT independently influence outcomes. Integrating these variables into a machine learning model enables more refined risk stratification and may facilitate individualized therapeutic decision-making.