Development and validation of a TRP-related gene signature for overall survival prediction in lung adenocarcinoma

Abstract

The transient receptor potential (TRP) channel is a type of channel protein widely distributed in peripheral and central nervous systems. Genes encoding TRP can be regulated by natural aromatic substances and serve as a therapeutic target for many diseases. However, the role of TRP-related genes in lung adenocarcinoma (LUAD) remains unclear. In this study, we used data from TCGA to screen and identify 17 TRP-related genes that are differentially expressed between LUAD and normal lung tissues. Based on these differentially expressed genes (DEGs), we classified all patients with LUAD into two subtypes. Significant differences in prognosis, clinical features, and immune cell infiltration characteristics were observed between the two subtypes. Subsequently, a prognostic signature with 12 genes was established by applying the least absolute shrinkage and selection operator (LASSO) Cox regression method, and all patients with LUAD were classified into low- and high-risk groups. Patients with LUAD in the low-risk group had a significantly longer survival time than those in the high-risk group (p < 0.001), which was confirmed by LUAD data from the GSE72094 and GSE68571 validation datasets. Combined with clinical characteristics, the risk score was found to be an independent predictor of overall survival (OS) in patients with LUAD. Additionally, patients with high TRP scores exhibited poorer clinical characteristics and immune status while showing a sensitive response to chemotherapeutic agents. In conclusion, the TRP score is a promising biomarker for determining the prognosis, molecular subtype, tumor microenvironment, and guiding personalized treatment in patients with LUAD.

Surrogate Biomarker Prediction from Whole-Slide Images for Evaluating Overall Survival in Lung Adenocarcinoma

“…Gene expression signatures have been widely used in research to evaluate OS in LUAD [ 15 , 16 , 17 , 18 , 19 , 20 , 28 ]. For example, Zhou et al established a three-gene signature that achieved a C-index of 0.638 in TCGA-LUAD [ 28 ].…”

“…Extensive research has been dedicated to identifying prognostic gene expression signatures from transcriptomic data in LUAD. Various methods such as Cox regression, random survival forests, and deep neural networks have been implemented by researchers with the aim of stratifying patients into low- and high-risk groups [ 15 , 16 , 17 , 18 , 19 , 20 ]. Researchers have also explored the use of multimodal artificial intelligence (AI) models aimed at evaluating survival, integrating clinical, genomic, and histopathological information [ 21 , 22 ].…”

Identification of the Hub Genes Involved in Stem Cell Treatment for Intervertebral Disc Degeneration: A Conjoint Analysis of Single-Cell and Machine Learning

“…The least absolute shrinkage and selection operator (LASSO) model is a regressive analytical arithmetic method that employs regularization to improve the accuracy of the predictive process. The “glmnet” package in R was employed to carry out the LASSO analysis to identify the genes linked to the ability to discriminate between treatment and normal samples [ 20 ]. Support vector machine (SVM) is a kind of monitored machine learning technology that is utilized widely for classification and regression analysis.…”