Exploring the role of mitophagy-related genes in breast cancer: subtype classification and prognosis prediction

Abstract

Background: Breast cancer (BC) is the most common cancer among women globally and poses the leading health threat to women worldwide, with persistently high incidence rates. Mitophagy is a selective autophagy process that specifically targets mitochondria within the cell, maintaining cellular energy balance and metabolic health by identifying and degrading damaged mitochondria. Although there is an understanding of the relationship between mitophagy and cancer, the specific mechanisms remain unclear due to the complexity and diversity of mitophagy, suggesting that it could be an effective and more targeted therapeutic approach for BC. Methods: In this study, we meticulously examined the BC expression and clinical pathology data from The Cancer Genome Atlas (TCGA) to assess the expression profiles, copy number variations (CNV), and to investigate the correlation, function, and prognostic impact of 34 mitophagy-related genes (MRGs). Differentially expressed genes (DEGs) were identified based on group classification. Lasso and Cox regression were used to determine prognostic genes for constructing a nomogram. Single-cell analysis mapped the distribution of these genes in BC cells. Additionally, the association between gene-derived risk scores and factors such as immune infiltration, tumor mutational burden (TMB), cancer stem cell (CSC) index, and drug responses was studied. In vitro experiments were conducted to confirm the analyses. Results: We included 34 MRGs and subsequently generated a risk score for 7 genes, including RPLP2, PCDHGA2, PRKAA2, CLIC6, FLT3, CHI3L1, and IYD. It was found that the low-risk group had better overall survival (OS) in BC, higher immune scores, but lower tumor mutational burden (TMB) and cancer stem cell (CSC) index, as well as lower IC50 values for commonly used drugs. To enhance clinical applicability, age and staging were incorporated into the risk score, and a more comprehensive nomogram was constructed to predict OS. This nomogram was validated and showed good predictive performance, with area under the curve (AUC) values for 1-year, 3-year, and 5-year OS of 0.895, 0.765, and 0.728, respectively. Conclusion: Our findings underscore the profound impact of prognostic genes on the immune response and prognostic outcomes in BC, indicating that they can provide new avenues for personalized BC treatment and potentially improve clinical outcomes.

Keywords: breast cancer; gene signature; immune; mitophagy; prognosis.

Figure 1

Flow chart of the study design.

Figure 2

The MRGs landscape in breast cancer. (A) Comparison of MRGs between tumor and normal group; (B) Chromosomal locations of these genes; (C) The frequency of CNV gain and loss in MRGs; (D) mRNA expression of MRGs between tumor and normal tissues (t-test, **** P < 0.0001; *** P < 0.001; ** P < 0.01; * P < 0.05).

Figure 3

Identification of mitophagy Subgroup in breast cancer. (A) Correlation analysis of MRGs; (B, C) A consensus matrix heat map defining two clusters (k=2); (D) Kaplan-Meier analysis of three subtypes of OS. (E, F) PCA analysis of two mitophagy clusters. (G) The gene expression level of two mitophagy clusters (t-test, **** P < 0.0001; *** P < 0.001; ** P < 0.01; * P < 0.05).

Figure 4

Characteristics of the biological behavior in mitophagy subgroups. (A) GSVA analysis of two mitophagy clusters. (B) Immune cell infiltration of two mitophagy clusters. (C, D) GO (C) and KEGG (D) enrichment analysis of DEGs among two mitophagy clusters (t-test, *** P < 0.001; ** P < 0.01; * P < 0.05).

Figure 5

Identification of gene subtypes in BC based on DEGs. (A, B) A consensus matrix heat map defining three gene subtypes (k=3); (C) Kaplan-Meier analysis of three subtypes of OS. (D, E) The MRGs expression level of three gene subtypes (t-test, **** P < 0.0001; *** P < 0.001; ** P < 0.01; * P < 0.05).

Figure 6

Construction of mitophagy-related prognostic risk score. (A, B) Lasso regression analysis on the prognosis-related genes; (C) Multivariate Cox regression analysis; (D) The sankey diagram of the sample distribution of two mitophagy clusters, three gene subtypes and two risk score groups. (E) OS analysis of two risk groups using Kaplan-Meier in the training cohort; (F) ROC curves to predict 1, 3, and 5-year OS according to the risk score in the training cohort; (G) OS analysis of two risk groups using Kaplan-Meier in the validation set (t-test, *** P < 0.001; ** P < 0.01; * P < 0.05).

Figure 7

Development and Validation of a Prognostic Nomogram for Breast Cancer. (A) A nomogram used to predict BC OS; (B) ROC curves to predict 1-, 3-, and 5year OS according to the nomogram in the training cohort; (C) ROC curves to predict 1-, 3-, and 5year OS according to the nomogram in the verification cohort; (D) OS analysis of two nomorisk groups using Kaplan-Meier in the training cohort and validation cohort; (E) Cumulative hazard of two nomorisk groups in the training cohort.

Figure 8

TME and immune checkpoint characteristics in both risk score groups. (A) Association of risk score with immune cell infiltration; (B) Association between risk score and TME score; (C) Association between immune cell infiltration and seven genes in the risk score model; (D-G) Immunotherapy effect in the low- and high-risk groups (t-test, *** P < 0.001; ** P < 0.01; * P < 0.05).

Figure 9

TMB, CSC index and drug susceptibility analysis among two risk_score groups. (A) Correlation between risk score and TMB; (B) Correlation between risk score and CSC index; (C-I) Correlation between risk score and drug susceptibility.

Figure 10

Single cell verification of the distribution of prognostic genes in breast cancer. (A) tSNE and UMAP projections of breast cancer cells in GSE176078. Different cell types are indicated by unique colors; (B) Delineating the distribution of key genes in cell subsets.

Figure 11

RT-PCR was used to compare mRNA levels of seven prognostic mitophagy-related genes in breast cancer cells and normal mammary epithelial cells (t-test, *** P < 0.001; ** P < 0.01; * P < 0.05).