表皮生长因子受体酪氨酸激酶抑制剂（EGFR-TKIs）是肺腺癌（LUAD）患者携带活化EGFR突变的一线治疗方案。然而，EGFR-TKI耐药性的出现仍然是成功治疗的关键障碍，并且与患者预后不良相关。本研究的总体目标是应用生物信息学工具，探索EGFR-TKI耐药机制，并开发一个稳健的预测模型。利用基因芯片表达数据，在LUAD细胞基因表达平台（GEO）数据库中鉴定与吉非替尼耐药相关的基因。对功能富集分析、基因集富集分析（GSEA）和免疫浸润分析进行全面的探索以了解吉非替尼耐药机制。此外，通过Least Absolute Shrinkage and Selection Operator（LASSO）和Cox回归分析将与LUAD预后相关的基因从The Cancer Genome Atlas（TCGA）数据库与筛选出的与吉非替尼耐药相关的差异表达基因（GRRDEGs）相结合构建了GRRG_score。此外，还对不同GRRG_score组之间的肿瘤微环境（TME）特征及其与免疫浸润的相关性进行了深入分析。基于GRRG_score开发了LUAD的预后模型，并进行了验证。使用HPA数据库验证了蛋白质表达。使用CTR-DB数据库验证了基于相关基因的药物治疗预测结果。共鉴定了110个差异表达基因。对DEGs的通路富集分析显示，差异表达基因主要富集于黏液类O-糖基化生物合成、细胞因子-细胞因子受体相互作用和鞘脂代谢通路。基因集富集分析显示，与吉非替尼耐药强相关的生物过程包括细胞增殖和免疫相关通路，EPITHELIAL_MESENCHYMAL_TRANSITION、APICAL_SURFACE和APICAL_JUNCTION在耐药组中高表达；KRAS_SIGNALING_DN、HYPOXIA和HEDGEHOG_SIGNALING在耐药组中高表达。GRRG_score基于13个基因（包括HSPA2、ATP8B3、SPOCK1、EIF6、NUP62CL、BCAR3、PCSK9、NT5E、FLNC、KRT8、FSCN1、ANGPTL4和ID1）的表达水平构建。我们进一步筛选和验证了两个关键基因，即NUP62CL和KRT8，它们在预后和耐药性预测方面具有预测价值。我们的研究鉴定了几个新的GRRDEGs，并揭示了LUAD中吉非替尼耐药的潜在机制。我们的结果对开发更有效的治疗策略和LUAD患者的预后模型具有重要意义。© 2023. 作者（们）独家许可给Springer-Verlag GmbH德国分公司，Springer Nature的一部分。

Epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) are the first-line therapy for patients with lung adenocarcinoma (LUAD) harboring activating EGFR mutations. However, the emergence of drug resistance to EGFR-TKIs remains a critical obstacle for successful treatment and is associated with poor patient outcomes. The overarching objective of this study is to apply bioinformatics tools to gain insights into the mechanisms underlying resistance to EGFR-TKIs and develop a robust predictive model.The genes associated with gefitinib resistance in the LUAD cell Gene Expression Omnibus (GEO) database were identified using gene chip expression data. Functional enrichment analysis, gene set enrichment analysis (GSEA), and immune infiltration analysis were performed to comprehensively explore the mechanism of gefitinib resistance. Furthermore, a GRRG_score was constructed by integrating genes related to LUAD prognosis from The Cancer Genome Atlas (TCGA) database with the screened Gefitinib Resistant Related differentially expressed genes (GRRDEGs) using the Least Absolute Shrinkage and Selection Operator (LASSO) and Cox regression analyses. Furthermore, we conducted an in-depth analysis of the tumor microenvironment (TME) features and their association with immune infiltration between different GRRG_score groups. A prognostic model for LUAD was developed based on the GRRG_score and validated. The HPA database was used to validate protein expression. The CTR-DB database was utilized to validate the results of drug therapy prediction based on the relevant genes.A total of 110 differentially expression genes were identified. Pathway enrichment analysis of DEGs showed that the differentially expressed genes were mainly enriched in Mucin type O-glycan biosynthesis, Cytokine-cytokine receptor interaction, Sphingolipid metabolism. Gene set enrichment analysis showed that biological processes strongly correlated with gefitinib resistance were cell proliferation and immune-related pathways, EPITHELIAL_MESENCHYMAL_TRANSITION, APICAL_SURFACE, and APICAL_JUNCTION were highly expressed in the drug-resistant group; KRAS_SIGNALING_DN, HYPOXIA, and HEDGEHOG_SIGNALING were highly expressed in the drug-resistant group. The GRRG_score was constructed based on the expression levels of 13 genes, including HSPA2, ATP8B3, SPOCK1, EIF6, NUP62CL, BCAR3, PCSK9, NT5E, FLNC, KRT8, FSCN1, ANGPTL4, and ID1. We further screened and validated two key genes, namely, NUP62CL and KRT8, which exhibited predictive value for both prognosis and drug resistance.Our study identified several novel GRRDEGs and provided insight into the underlying mechanisms of gefitinib resistance in LUAD. Our results have implications for developing more effective treatment strategies and prognostic models for LUAD patients.© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.