背景：急性骨髓性白血病（AML）具高度基因異質性，儘管治療進展，長期存活率仍不理想，亟需新型預後標誌物以改善治療策略。方法：整合三個GEO資料集（GSE121169、GSE149237、GSE63270），篩選AML與健康對照間的差異表現基因（DEGs），並進行功能富集分析（GO、KEGG、Cancer Hallmarks）、蛋白互作網絡分析（STRING、Cytoscape）及樞紐基因排序（Degree、MNC）。利用Kaplan–Meier Plotter與UCSC Xena評估預後影響，並在TCGA AML隊列中分析TRIP13與FLT3、PML/RARA及NRAS突變的關聯。結果：共鑑定118個DEGs，富集於有絲分裂調控與基因組不穩定途徑。在十個樞紐基因中，僅高表現TRIP13與較差總生存顯著相關。TRIP13在NRAS突變AML中表現升高，並呈現預後依賴突變背景的差異性：於NRAS野生型預測較差存活，於NRAS突變型則與較佳預後相關。結論：TRIP13為具背景依賴性的AML預後標誌物，可優化現有風險分層，並具潛力成為治療靶點，需進一步機制驗證。

Background: Acute myeloid leukemia (AML) is a genetically heterogeneous hematologic malignancy characterized by uncontrolled proliferation of immature myeloid cells. Despite advancements in targeted therapies, long-term survival rates remain poor. Identifying novel biomarkers is essential for improving prognosis and guiding treatment. Methods: We performed an integrative analysis of three publicly available Gene Expression Omnibus (GEO) datasets (GSE121169, GSE149237, and GSE63270) to identify differentially expressed genes (DEGs) between AML and healthy control samples. Functional enrichment analyses, including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, Cancer Hall-marks, protein-protein interaction (PPI) network analysis (STRING and Cytoscape), and hub gene identification (Degree and Maximum Neighborhood Component (MNC) algorithms) were performed. Prognostic impact was assessed using Kaplan–Meier Plotter and UCSC Xena. TRIP13 expression was further analyzed in relation to FLT3, PML/RARA, and NRAS mutation status in The Cancer Genome Atlas (TCGA) cohorts. Results: We identified 118 overlapping DEGs enriched in mitotic regulation and genome instability pathways. Among ten consensus hub genes, only high TRIP13 expression was significantly associated with worse overall survival. TRIP13 levels were elevated in NRAS-mutant AML and showed context-dependent prognostic effects: predicting poor survival in NRAS wild-type patients but better outcomes in NRAS-mutant cases. Conclusion: TRIP13 is a context-dependent prog-nostic biomarker in AML, potentially improving risk stratification beyond current mutation-based models. These findings highlight TRIP13 as a candidate for targeted therapy, warranting further mechanistic validation. Further functional studies are warranted to clarify its mechanistic role in leukemogenesis and treatment response.