Supplementary file 2_Molecular signatures bidirectionally link myocardial infarction and lung cancer.xlsx

Myocardial Infarction (MI) and lung cancers are major contributors to mortality worldwide. While seemingly diverse, the two share common risk factors, such as smoking and hypertension. There is a pressing need to identify bidirectional molecular signatures that link MI and lung cancer, in order to improve clinical outcomes for patients. In this study, we identified common differentially expressed genes between MI and lung cancer. Specifically, we identified 1,496 upregulated and 1,482 downregulated genes in the MI datasets. By focusing on the 1,000 most upregulated and downregulated genes in Lung Adenocarcinoma (LUAD) and Lung Squamous Cell Carcinoma (LUSC), we identified 35 genes that are common across MI, LUAD, and LUSC. Functional enrichment analysis revealed shared biological processes, such as “inflammatory response” and “cell differentiation.” The Cox proportional hazards model demonstrated a significant association between the shared genes and overall survival in lung cancer patients, as well as with smoking history in these patients. In addition, a machine learning model based on the expression of the shared genes distinguished MI patients from controls, achieving an AUROC of 0.72 and an AUPRC of 0.86. Finally, based on drug repurposing analysis, we proposed FDA-approved drugs potentially targeting the upregulated genes as novel therapeutic options for the co-occurring conditions of MI and lung cancer. Overall, our findings highlight the similarities in molecular makeup between lung cancer and MI.

心肌梗死（Myocardial Infarction, MI）与肺癌是全球范围内致死的主要病因。尽管二者看似病理特征迥异，但共享吸烟、高血压等共同危险因素。当前亟需识别连接心肌梗死与肺癌的双向分子特征，以改善患者的临床结局。本研究中，我们鉴定出了心肌梗死与肺癌之间共有的差异表达基因。具体而言，我们在心肌梗死数据集中共鉴定出1496个上调基因与1482个下调基因。通过聚焦肺腺癌（Lung Adenocarcinoma, LUAD）与肺鳞状细胞癌（Lung Squamous Cell Carcinoma, LUSC）中排名前1000的上调与下调基因，我们筛选出了35个同时在心肌梗死、肺腺癌与肺鳞状细胞癌中共享的基因。功能富集分析显示，二者共享多个生物学过程，如"炎症应答"与"细胞分化"。Cox比例风险模型（Cox proportional hazards model）结果显示，这些共享基因与肺癌患者的总生存期以及患者的吸烟史均存在显著关联。此外，基于共享基因表达谱构建的机器学习模型可有效区分心肌梗死患者与健康对照，其受试者工作特征曲线下面积（Area Under the Receiver Operating Characteristic Curve, AUROC）达0.72，精确率-召回率曲线下面积（Area Under the Precision-Recall Curve, AUPRC）达0.86。最后，通过药物重定位分析，我们提出了若干经美国食品药品监督管理局（Food and Drug Administration, FDA）批准的、可靶向上调基因的药物，作为心肌梗死与肺癌共病的潜在新型治疗方案。综上，本研究结果揭示了肺癌与心肌梗死在分子构成层面的相似性。