A Novel Matrisomal-Related LncRNA Signature Associated With Survival Outcome and Immune Evasion in Patients With Gastric Cancer

Comprehensive analysis to identify pseudogenes/lncRNAs-hsa-miR-200b-3p-COL5A2 network as a prognostic biomarker in gastric cancer

OBJECTIVE

Gastric cancer is one of the most common and deadly types of cancer. The molecular mechanism of gastric cancer progression remains unclear.

MATERIALS AND METHODS

Four hub genes were identified through GEO and TCGA database screening and analysis. Prognostic analysis revealed that COL5A2 was the most likely to affect the prognosis of gastric cancer among the four hub genes. The relationships between COL5A2 and clinical variables and immune cell infiltration were analyzed. Then, COL5A2 was analyzed for single-gene differences and related functional enrichment. Using the starBase database for prediction and analysis, miRNAs and pseudogenes/lncRNAs that might combine with COL5A2 were identified; thus, the ceRNA network was constructed. Finally, the network was verified by Cox analysis and qPCR, and a nomogram was constructed.

RESULTS

First, we found that COL5A2, COL12A1, BGN and THBS2 were highly expressed in gastric cancer. COL5A2 had statistical significance in overall survival (OS), disease-specific survival (DSS), and progression-free interval (PFI) analysis. Immune infiltration analysis suggested that COL5A2 might influence the changes in the tumor immune microenvironment. The StarBase database was used to predict that 3 pseudogenes and 7 lncRNAs might inhibit the hsa-miR-200b-3p-COL5A2 axis in gastric cancer. The pseudogenes/lncRNA-hsa-miR-200b-3p-COL5A2 ceRNA network was identified and verified using Cox regression analysis and PCR. Finally, we constructed a nomogram.

CONCLUSIONS

We elucidated the regulatory role of the pseudogenes/lncRNA-hsa-miR-200b-3p-COL5A2 network in gastric cancer progression and constructed a nomogram. These studies may provide effective treatments and potential prognostic biomarkers for gastric cancer.

Comprehensive transcriptomic analysis of immune-related eRNAs associated with prognosis and immune microenvironment in melanoma

Background

Recent evidence suggests that enhancer RNAs (eRNAs) play key roles in cancers. Identification of immune-related eRNAs (ireRNAs) in melanoma can provide novel insights into the mechanisms underlying its genesis and progression, along with potential therapeutic targets.

Aim

To establish an ireRNA-related prognostic signature for melanoma and identify potential drug candidates.

Methods

The ireRNAs associated with the overall survival (OS-ireRNAs) of melanoma patients were screened using data from The Cancer Genome Atlas (TCGA) via WGCNA and univariate Cox analysis. A prognostic signature based on these OS-ireRNAs was then constructed by performing the least absolute shrinkage and selection operator (LASSO) Cox regression analysis. The immune landscape associated with the prognostic model was evaluated by the ESTIMATE algorithm and CIBERSORT method. Finally, the potential drug candidates for melanoma were screened through the cMap database.

Results

A total of 24 OS-ireRNAs were obtained, of which 7 ireRNAs were used to construct a prognostic signature. The ireRNAs-related signature performed well in predicting the overall survival (OS) of melanoma patients. The risk score of the established signature was further verified as an independent risk factor, and was associated with the unique tumor microenvironment in melanoma. We also identified several potential anti-cancer drugs for melanoma, of which corticosterone ranked first.

Conclusions

The ireRNA-related signature is an effective prognostic predictor and provides reliable information to better understand the mechanism of ireRNAs in the progression of melanoma.