Identification of Ferroptosis-Associated Long Noncoding RNA Prognostic Model and Tumor Immune Microenvironment in Thyroid Cancer

Abstract

Background. Thyroid cancer (TC) is a rapidly increasing incidence of endocrine malignancies, occupying 3% of new cancer incidence, of which 10% has a heterogeneous prognosis. Ferroptosis is a form of cell death distinct from apoptosis, which involves antitumor drug-related research. Long noncoding RNAs (lncRNAs) could affect cancer prognosis by regulating the ferroptosis; thus, ferroptosis-associated lncRNAs are emerging as prospective biomarkers for cancer therapy and prognosis. However, the prognostic factors of ferroptosis-associated lncRNAs in this solid tumor and their mechanisms remain unknown. Methods. The TC lncRNA data were extracted from RNA sequencing files of The Cancer Genome Atlas (TCGA). Then, we performed a two-cluster analysis and grouped 502 patients with TC in a 7 : 3 ratio. Both the least absolute shrinkage and selection operator (LASSO) regression and Cox regression analysis were conducted to create and validate the ferroptosis-associated lncRNA prognostic model (Ferr-LPM). Based on the median Ferr-LPM-based risk score (LPM_score) of the training cohort, we categorized patients into high and low LPM_score groups, which were then subjected to prognostic correlation and difference analysis. We also created a nomogram and assessed its predictive ability. Furthermore, immune-related mechanisms were investigated by analyzing the tumor immune microenvironment (TIME) and applying algorithms such as CIBERSROT. Results. We built a highly accurate nomogram to promote the clinical applicability of Ferr-LPM. The area under the receiver operating characteristic curve (AUC-ROC) reached above 0.9. Survival analysis suggested that when the Ferr-LPM score was higher, the overall survival (OS) of patients within this group was shorter. Meanwhile, we found a strong association between Ferr-LPM and TIME. Interestingly, the LPM_score was inversely proportional to the tumor purity but positively related to immune checkpoint blockade (ICB) response. Conclusion. We constructed a novel ferroptosis-associated lncRNA nomogram that could highly predict the prognosis of TC patients. Ferroptosis-associated lncRNAs might possess potential functions in regulating TIME, and lncRNAs provide TC patients with new prognostic biomarkers and therapeutic targets.

The role of lactate metabolism-related LncRNAs in the prognosis, mutation, and tumor microenvironment of papillary thyroid cancer

“…The risk model for the six-LRLs signature accurately predicted the OS of PTC patients, demonstrating its potential as an independent prognostic model. Meanwhile, the AUC value of the risk model reached 0.979, which was significantly higher than the prognostic model constructed in previous studies ( 44 – 46 ), indicating that our six-LRLs signature is a better prediction model. Based on the risk model, differences in enrichment pathways, immune function, immune escape and mutations were compared between the high- and low-risk groups.…”

Integrated bioinformatics analysis of noncoding RNAs with tumor immune microenvironment in gastric cancer

“…LncRNAs have been implicated in facilitating the evasion of immune surveillance by tumor cells through multiple mechanisms, including the promotion of an immunosuppressive microenvironment 26 – 28 . Multiple studies used comprehensive analysis to determine the functions of various lncRNAs in tumor immune microenvironment (TIME) in various cancers, such as colon cancer 29 , clear cell renal cell carcinoma 30 , breast cancer 31 , 32 , lung cancer 33 , 34 , thyroid cancer 35 , adrenocortical adenocarcinoma 36 , bladder cancer 37 , and head and neck squamous cell carcinoma 38 . LncRNA NKILA has demonstrated the ability to initiate apoptosis in tumor-specific T cells, thereby impeding their capacity to infiltrate the tumor, leading to tumor immune evasion 39 .…”