Development and validation of a cuproptosis-related prognostic model for acute myeloid leukemia patients using machine learning with stacking

Our objective is to develop a prognostic model focused on cuproptosis, aimed at predicting overall survival (OS) outcomes among Acute myeloid leukemia (AML) patients. The model utilized machine learning algorithms incorporating stacking. The GSE37642 dataset was used as the training data, and the GSE12417 and TCGA-LAML cohorts were used as the validation data. Stacking was used to merge the three prediction models, subsequently using a random survival forests algorithm to refit the final model using the stacking linear predictor and clinical factors. The prediction model, featuring stacking linear predictor and clinical factors, achieved AUC values of 0.840, 0.876 and 0.892 at 1, 2 and 3 years within the GSE37642 dataset. In external validation dataset, the corresponding AUCs were 0.741, 0.754 and 0.783. The predictive performance of the model in the external dataset surpasses that of the model simply incorporates all predictors. Additionally, the final model exhibited good calibration accuracy. In conclusion, our findings indicate that the novel prediction model refines the prognostic prediction for AML patients, while the stacking strategy displays potential for model integration.

Mutation spectrum of FLT3 and significance of non-canonical FLT3 mutations in haematological malignancy

Fms-like tyrosine kinase 3 (FLT3) is frequently mutated in haematological malignancies. Although canonical FLT3 mutations including internal tandem duplications (ITDs) and tyrosine kinase domains (TKDs) have been extensively studied, little is known about the clinical significance of non-canonical FLT3 mutations. Here, we first profiled the spectrum of FLT3 mutations in 869 consecutively newly diagnosed acute myeloid leukaemia (AML), myelodysplastic syndrome and acute lymphoblastic leukaemia patients. Our results showed four types of non-canonical FLT3 mutations depending on the affected protein structure: namely non-canonical point mutations (NCPMs) (19.2%), deletion (0.7%), frameshift (0.8%) and ITD outside the juxtamembrane domain (JMD) and TKD1 regions (0.5%). Furthermore, we found that the survival of patients with high-frequency (>1%) FLT3-NCPM in AML was comparable to those with canonical TKD. In vitro studies using seven representative FLT3-deletion or frameshift mutant constructs showed that the deletion mutants of TKD1 and the FLT3-ITD mutant of TKD2 had significantly higher kinase activity than wild-type FLT3, whereas the deletion mutants of JMD had phosphorylation levels comparable with wild-type FLT3. All tested deletion mutations and ITD were sensitive to AC220 and sorafenib. Collectively, these data enrich our understanding of FLT3 non-canonical mutations in haematological malignancies. Our results may also facilitate prognostic stratification and targeted therapy of AML with FLT3 non-canonical mutations.