Conclusion
Our findings suggest that there are shared physiopathologic processes and molecular profiles between RA and LUAD. RALUADS represents an excellent prognosis predictor and immune-related biomarker, which can be applied to select potential effective drugs and for LUAD patients with RA.
Methods
We obtained gene expression profiles and clinical information of patients with RA and LUAD from GEO and TCGA datasets. We performed differential analysis and Weighted Gene Co-expression Network Analysis (WGCNA) to discover the shared genes between RA and LUAD. Then, COX regression and LASSO analysis were employed to figure out genes significantly associated with survival. qRT-PCR and Western blot were utilized to validate the expression level of candidate genes. For clinical application, we constructed a nomogram, and also explored the value of RALUADS in characterizing immune infiltration features by CIBERSORT and xCell. Finally, responses to different drug therapy were predicted according to different RALUADS.
Results
Our analysis identified two gene sets from differentially expressed genes and WGCNA gene modules of RA and LUAD. Filtered by survival analysis, three most significant shared genes were selected, CCN6, CDCA4 and ERLIN1, which were all upregulated in tumors and associated with poor prognosis. The three genes constituted RA and LUAD score (RALUADS). Our results demonstrated that RALUADS was higher in tumor patients and predicted poor prognosis in LUAD patients. Clinical nomogram combining RALUADS and other clinicopathological parameters had superior performance in survival prediction (AUC = 0.722). We further explored tumor immune microenvironment (TME) affected by RALUADS and observed RALUADS was closely related to the sensitivity of multiple immune blockades, chemotherapy and targeted drugs.