基于细胞焦亡基因构建肾上腺皮质癌预后模型及免疫微环境分析

To construct a prognostic model and analyze the immune microenvironment of adrenocortical carcinoma based on pyroptosis gene

目的:建立一个细胞焦亡基因相关的肾上腺皮质癌预后模型,分析其免疫微环境并实验验证。方法:运用TCGA数据库和GTEx数据库筛选焦亡相关的差异表达基因,Cox回归和Lasso分析构建ACC预后风险模型,外部队列验证预后价值及临床相关性,对差异表达基因进行GO-KEGG和GSEA富集分析,探究致癌可能的相关通路,ssGSEA算法、Pearson相关性分析研究ACC的免疫微环境,实时荧光定量及免疫组化实验验证预测结果。结果:Lasso分析成功构建了ACC风险模型,Cox多因素回归分析显示风险评分可作为评价预后的独立因素。高风险患者具有更低的OS和PFI,时间依赖性ROC曲线下面积(AUC)均大于0.5。风险模型还与T、M、N临床分期相关,对外部队列进行同样的风险分组,显示高风险患者预后更差,进一步证实风险模型的普适性。富集分析显示E2f、MYC等致癌通路被显著上调。高风险组中多数免疫细胞浸润较差,且MHC分子、趋化因子等分子分泌下降,不利于抗癌免疫微环境的形成。TP53和DNNB1蛋白提示为ACC的潜在治疗靶点。实时荧光定量和免疫组化实验验证了模型的普适性。结论:本研究利用生物信息学方法,成功构建出肾上腺皮质癌风险模型,外部验证良好,具有较准确的预后能力。免疫微环境分析得出ACC患者常处于免疫抑制状态,因此热门免疫检查点抑制剂效果不佳。TP53和DNNB1癌蛋白在高风险患者中表达显著,可作为潜在治疗靶点,以改善免疫治疗失效的现状。希望该风险模型能为肾上腺皮质癌的预后识别及靶向治疗提供新思路。

Objective:A prognostic model of adrenal cortical carcinoma associated with pyroptosis gene was established,and its immune microenvironment was analyzed and verified by experiment.Methods:TCGA database and GTEx database were used to screen the differentially expressed genes related to pyroptosis.The prognostic risk model of ACC was constructed by Cox regression analysis and Lasso analysis.The prognostic value and clinical correlation relevance were explored based on the model and verified by external cohort.The differentially expressed genes were enriched by Go-KEGG and GSEA enrichment analysis,and the related carcinogenic pathways were explored.ssGSEA algorithm and Pearson correlation analysis were used to study the immune microenvironment of ACC.Real-time fluorescence quantification and immunohistochemical experiments verified the prediction results.Results:The ACC risk model was successfully constructed by Lasso analysis.Cox multivariate regression analysis showed that the risk score could be used as an independent prognostic factor to evaluate the prognosis.High-risk patients had lower OS and PFI,and the area under curve(AUC)of the time-dependent ROC curve was greater than 0.5.The risk model was also correlated with T,M and N clinical stages.The same risk grouping for the external cohort also showed a worse prognosis for high-risk patients,further confirming the universality of the risk model.Enrichment analysis showed that E2f,MYC and other oncogenic pathways were significantly up-regulated.In the high-risk group,the infiltration of most immune cells was poor,and the secretion of MHC molecules and chemokines was decreased,which was not conducive to the formation of anti-cancer immune microenvironment.TP53 and DNNB1 proteins were suggested as potential therapeutic targets for ACC.Real-time fluorescence quantification and immunohistochemical experiments verified the universality of the model.Conclusion:In this study,the risk model of adrenocortical carcinoma was successfully constructed by using bioinformatics method,which was well verified externally and had relatively accurate prognostic ability.Immune microenvironment analysis showed that ACC patients were often immunosuppressed,resulting in poor efficacy of popular immune checkpoint inhibitors.The expression of TP53 and DNNB1 oncoproteins is significant in high-risk patients,which can be used as potential therapeutic targets to improve the failure of immunotherapy.It is hoped that this risk model can provide a new idea for the prognostic identification and targeted therapy of adrenocortical carcinoma.