Abstract

   We downloaded the mRNA expression profiles of patients with LUAD and
   corresponding clinical data from The Cancer Genome Atlas (TCGA)
   database and used the Least Absolute Shrinkage and Selection Operator
   Cox regression model to construct a multigene signature in the TCGA
   cohort, which was validated with patient data from the GEO cohort.
   Results showed differences in the expression levels of 120
   necroptosis-related genes between normal and tumor tissues. An
   eight-gene signature (CYLD, FADD, H2AX, RBCK1, PPIA, PPID, VDAC1, and
   VDAC2) was constructed through univariate Cox regression, and patients
   were divided into two risk groups. The overall survival of patients in
   the high-risk group was significantly lower than of the patients in the
   low-risk group in the TCGA and GEO cohorts, indicating that the
   signature has a good predictive effect. The time-ROC curves revealed
   that the signature had a reliable predictive role in both the TCGA and
   GEO cohorts. Enrichment analysis showed that differential genes in the
   risk subgroups were associated with tumor immunity and antitumor drug
   sensitivity. We then constructed an mRNA–miRNA–lncRNA regulatory
   network, which identified lncRNA [34]AL590666. 2/let-7c-5p/PPIA as a
   regulatory axis for LUAD. Real-time quantitative PCR (RT-qPCR) was used
   to validate the expression of the 8-gene signature. In conclusion,
   necroptosis-related genes are important factors for predicting the
   prognosis of LUAD and potential therapeutic targets.

   Subject terms: Computational biology and bioinformatics, Gene
   regulatory networks, Genome informatics

Introduction

   Lung cancer is one of the cancer types with the highest mortality rates
   in the world and is the leading cause of cancer-related deaths in men
   and women^[35]1. Non-small cell lung cancer (NSCLC) is the most common
   type of lung cancer, accounting for approximately 85% of all lung
   cancer cases^[36]2. Lung adenocarcinoma (LUAD) is an NSCLC subtype and
   has the highest fatality rate in nonsmokers^[37]3. Given that LUAD is
   prone to metastasis and recurrence in the early stage, the prognostic
   effect of LUAD is extremely poor, and patients with LUAD have an
   average 5-year survival rate of less than 20%^[38]4. In clinical
   practice, the tumor staging system has been widely used in guiding the
   treatment and prognosis evaluation of cancer patients^[39]5. However,
   the prognosis is usually based on inherent anatomical information.
   Owing to the heterogeneity of LUAD^[40]6, the development of the
   disease is difficult to predict. Therefore, effective prognostic
   biomarkers that can aid clinicians in making accurate LUAD diagnoses,
   predicting clinical results, and providing references for personalized