The necroptosis-related signature and tumor microenvironment immune characteristics associated with clinical prognosis and drug sensitivity analysis in stomach adenocarcinoma

TL;DR

This study identifies a new model using necroptosis-related long noncoding RNAs to predict stomach cancer prognosis and guide immunotherapy.

Contribution

A novel risk score model using necroptosis-related lncRNAs for STAD prognosis and drug sensitivity is developed and validated.

Findings

21 lncRNAs were identified as part of a risk score model predicting survival in stomach adenocarcinoma patients.

High-risk patients showed lower survival rates and higher immune infiltration compared to low-risk patients.

29 drugs with varying sensitivity across tumor clusters were identified, suggesting potential for personalized treatment.

Abstract

Purpose: Necroptosis plays an important role in the tumorigenesis, development, metastasis, and drug resistance of malignant tumors. This study explored the new model for assessing stomach adenocarcinoma (STAD) prognosis and immunotherapy by combining long noncoding RNAs associated with necroptosis.

Methods: Patient clinical data and STAD gene expression profiles were curated from The Cancer Genome Atlas (TCGA). Immune-related genes were sourced from a specialized molecular database. Perl software and R software were used for data processing and analysis. Necroptosis-related lncRNAs in STAD were pinpointed via R’s correlation algorithms. These lncRNAs, in conjunction with clinical data, informed the construction of a prognostic lncRNA-associated risk score model using univariate and multivariate Cox regression analyses. The model’s prognostic capacity was evaluated by Kaplan-Meier survival curves and validated as an independent prognostic variable. Further, a nomogram incorporating this model with clinical parameters was developed, offering refined individual survival predictions. Subsequent analyses of immune infiltration and chemosensitivity within necroptosis-related lncRNA clusters utilized an arsenal of bioinformatic tools, culminating in RT-PCR validation of lncRNA expression.

Results: Through rigorous Cox regression, 21 lncRNAs were implicated in the risk score model. Stratification by median risk scores delineated patients into high- and low-risk cohorts, with the latter demonstrating superior prognostic outcomes. The risk model was corroborated as an independent prognostic indicator for STAD. The integrative nomogram displayed high concordance between predicted and observed survival rates, as evidenced by calibration curves. Differential immune infiltration in risk-defined groups was illuminated by the single sample GSEA (ssGSEA), indicating pronounced immune presence in higher-risk patients. Tumor microenvironment (TME) analysis showed that cluster-C3 had the highest score in the analysis of the three TMEs. Through the differential analysis of immune checkpoints, it was found that almost all immune checkpoint-related genes were expressed differently in various tumor clusters. Among them, CD44 expression was the highest. By comparing all drug sensitivities, we screened out 29 drugs with differences in drug sensitivity across different clusters. Risk score gene expression identification results showed that these lncRNAs were abnormally expressed in gastric cancer cell lines.

Conclusions: This investigation provides a robust methodological advance in prognosticating and personalizing immunotherapy for STAD, leveraging quantitatively derived tumor cluster risk scores. It posits the use of necroptosis-related lncRNAs as pivotal molecular beacons for guiding therapeutic strategies and enhancing clinical outcomes in STAD.

Figures