山东医学高等专科学校学报

目的 探究肺腺样囊性癌分子机制，寻找诊断生物标志物与潜在治疗靶点。方法 基于临床病理标本进行免疫组化，结合疾病基因数据库构建靶点池，整合公共GEO数据集，用稳健秩聚合(RRA)方法筛选核心差异表达基因，通过功能富集、免疫浸润分析解析生物学特征与肿瘤微环境；结合蛋白质-蛋白质相互作用(PPI)网络与机器学习算法(LASSO及SVM-RFE)交叉验证锁定关键基因；最后通过逆向网络药理学与分子对接模拟预测潜在靶向活性成分。结果 筛选出340个PACC相关基因，显著富集于PI3K-Akt信号通路等。免疫浸润分析显示，肿瘤微环境呈免疫抑制状态，M2型巨噬细胞增多。经PPI网络与机器学习交叉验证，拓扑异构酶Ⅱα(TOP2A)被鉴定为关键诊断标志基因。逆向网络药理学预测显示，隐丹参酮和飞燕草素等天然活性成分可能靶向TOP2A,分子对接模拟证实其结合亲和力高。结论 本研究明确TOP2A可作为潜在诊断标志物和治疗靶点，为中药干预提供新思路。

Objective To vestigate the molecular mechanisms of pulmonary adenoid cystic carcinoma(PACC) to identify diagnostic biomarkers and potential therapeutic targets. Methods Immunohistochemistry was performed on clinical pathological specimens. A target pool was constructed by integrating disease gene databases, and public GEO datasets were utilized. The robust rank aggregation(RRA) method was employed to screen core differentially expressed genes. Biological characteristics and tumor microenvironment were analyzed through functional enrichment and immune infiltration analysis. Key genes were identified by cross-validation using protein-protein interaction(PPI) networks and machine learning algorithms(LASSO and SVM-RFE). Finally, reverse network pharmacology and molecular docking simulations were used to predict potential targeted active components. Results 340 ACC-related genes were screened, significantly enriched in pathways such as the PI3K-Akt signaling pathway. Immune infiltration analysis revealed an immunosuppressive tumor microenvironment with increased M2 macrophages. After cross-validation with PPI networks and machine learning, TOP2A was identified as a key diagnostic marker gene. Reverse network pharmacology predicted that natural active compounds such as cryptotanshinone and cyanidin may target TOP2A, and molecular docking simulations confirmed their high binding affinity.Conclusion This study clarifies the value of TOP2A as a potential diagnostic biomarker and therapeutic target, and provides new insights for traditional Chinese medicine intervention.