Multi-Omics-Based Analysis of the Effect of Longevity Genes on the Immune Relevance of Colorectal Cancer

Abstract

Background: Colorectal cancer (CRC) ranks as the third most prevalent cancer globally, with its incidence and recurrence rates steadily rising. To explore the relationship between CRC and longevity-associated genes (LAGs), and to offer new therapeutic avenues for CRC treatment, we developed a prognostic model based on these genes to predict the outcomes for CRC patients. Additionally, we conducted an immune correlation analysis. Methods: We conducted a comprehensive analysis of the effects of 81 LAGs in CRC by integrating multiple omics datasets. This analysis led to the identification of two distinct molecular subtypes and revealed that alterations in LAGs across various layers were linked to clinicopathological features, prognosis, and cell infiltration characteristics within the tumor microenvironment (TME). The training and validation cohorts for the models were derived from the TCGA-COAD, TCGA-READ, and GSE35279 datasets. Subsequently, we developed a risk score model, and the Kaplan-Meier method was employed to estimate overall survival (OS). Ultimately, we established a prognostic model based on five LAGs: BEDN3, EXOC3L2, CDKN2A, IL-13, and CAPN9. Furthermore, we assessed the correlations between the risk score and factors such as immune cell infiltration, microsatellite instability, and the stem cell index. Results: Our comprehensive bioinformatics analysis revealed a strong association between longevity genes and CRC. The risk score derived from the five newly identified LAGs was determined to be an independent prognostic factor for CRC. Patients categorized by this risk score demonstrated significant differences in immune status and microsatellite instability. Conclusions: Our comprehensive multi-omic analysis of LAGs highlighted their potential roles in the tumor immune microenvironment, clinicopathological features, and prognosis, offering new insights for the treatment of CRC.

Keywords: bioinformatics; colorectal cancer; immunization; longevity; multi-omics.

Figure 1

Analysis of cell longevity-related genes in TCGA. (A) Longevity-related genes differences between tumor and normal tissues in the TCGA database, *, p < 0.05; **, p < 0.01; ***, p < 0.001. (B) locations of COAD and READ alterations in the longevity-related genes on 23 chromosomes. (C) loss and gain of copy number of longevity-related genes (D) waterfall diagram of longevity-related genes mutations. TCGA, The Cancer Genome Atlas; COAD, Colon adenocarcinoma; READ, Rectum adenocarcinoma.

Figure 2

Primary typing of longevity-related genes. (A) Prognostic network of longevity DEGs; (B) a total of 698 patients with CRC were divided into six clusters according to the consensus clustering matrix (k = 6); (C) Kaplan–Meier overall survival curves of six clusters; (D) Take the intersecting genes for clusters A, C, E, and F; (E) bar chart of Gene Ontology (GO) pathways; (F) network diagram of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways.

Figure 3

Second typing of cell longevity-related genes (gene clusters). (A) Forest plot showing prognosis-related genes. (B) A total of 698 patients with CRC were divided into two clusters according to the consensus clustering matrix (k = 2); (C) Kaplan–Meier overall survival curves for two clusters; (D) Box-and-line plot showing differential expression of longevity-related genes in A and B component phenotypes, *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 4

Clinical analysis of the prognostic models. (A) Sanggi diagram of the prognostic model; (B,C) Risk score of the longevity-related genes clusters and gene clusters; (D) longevity-related DEGs in the risk groups, *, p < 0.05; **, p < 0.01; ***, p < 0.001; (E) nomograph focusing on factors, ***, p < 0.001; (F) predicted 1-, 3-, and 5-year survival calibration curve.

Figure 5

Validation of prognostic model. (A) Survival curve of the training, test and all group; (B) receiver operating characteristic (ROC) curve of the training, test and all group; (C) risk heatmap of the survival-related genes in the training, test and all group; (D) risk curve of the survival-related genes in the training, test and all group; (E) survival state diagram of the survival-related genes in the training, test and all group.

Figure 6

Analysis of the immune microenvironment of the prognostic models. (A) Correlation graph between the risk score and each immune cells; *, p < 0.05; **, p < 0.01; ***, p < 0.001. (B–H) Correlation graph between the risk score and each immune cells; (I) differential analysis of tumor immune microenvironment. (J,K) Relationship between risk scores and microsatellite instability; (L) Scatterplot Demonstrates Association of Risk Score with Tumor Stem Cell Index.