T. Wang; Liaoning Cancer Hospital & Institute, Shenyang, Liaoning, China
Purpose/Objective(s): Lung squamous cell carcinoma (LUSC) is a common subtype of non-small-cell lung cancer. Immunotherapy has achieved considerable breakthroughs in LUSC treatment. The purpose of this study was to try to construct a signature of immune response-related genes (IRRGs) in LUSC to assess the prognosis and effects of immunotherapy and to explore the prognostic significance of thrombomodulin (THBD) in LUSC immunotherapy. Materials/
Methods: We obtained data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets of patients with LUSC. A risk model was constructed based on differentially expressed prognostic IRRGs combined with clinical characteristics to forecast the prognosis and immunotherapy effectiveness. Pretreatment formalin-fixed and paraffin-embedded (FFPE) tumor tissues were collected from 55 patients with LUSC stages IIB–IIIB who underwent successful surgical resection after neoadjuvant chemoimmunotherapy (NCIO), and THBD was detected using immunohistochemistry (IHC). Results: In the GSE126044, we identified 600 IRRGs with univariate Cox regression analysis and 53 prognostic IRRGs in TCGA-LUSC. Subsequently, least absolute shrinkage and selection, random forest, and multivariate Cox regression analyses were used to construct a risk model consisting of eight prognostic IRRGs (B3GNT3, FOSB, THBD, GABRR1, IL36RN, PTGIS, ERBB4, and POPDC3). The risk score based on the eight prognostic IRRGs was found to be an independent prognostic factor for LUSC, and a nomogram was successfully constructed. In addition, two immune subtypes were identified according to the expression of the eight genes. Cluster 2 had lower overall survival (OS), higher immune checkpoint gene expression, higher immunoreactivity, and more abundant infiltrating immune cells than cluster 1. Additionally, IHC analysis of FFPE tissues from patients with LUSC revealed high THBD expression was closely associated with a good immunotherapeutic response, which was positively correlated with programmed cell death ligand 1 expression. Conclusion: The risk model based on the prognostic IRRGs can be used as a valid biomarker to predict the OS in LUSC. High THBD expression predicts a better response to NCIO. Thus, these findings can facilitate new therapeutic approaches and guide individualized treatment.
