Identification of methylation sites and signature genes with prognostic value for luminal breast cancer

Abstract

Background: Robust and precise molecular prognostic predictors for luminal breast cancer are required. This study aimed to identify key methylation sites in luminal breast cancer, as well as precise molecular tools for predicting prognosis.

Methods: We compared methylation levels of normal and luminal breast cancer samples from The Cancer Genome Atlas dataset. The relationships among differentially methylated sites, corresponding mRNA expression levels and prognosis were further analysed. Differentially expressed genes in normal and cancerous samples were analysed, followed by the identification of prognostic signature genes. Samples were divided into low- and high-risk groups based on the signature genes. Prognoses of low- and high-risk groups were compared. The Gene Expression Omnibus dataset were used to validate signature genes for prognosis prediction. Prognosis of low- and high-risk groups in Luminal A and Luminal B samples from the TCGA and the Metabric cohort dataset were analyzed. We also analysed the correlation between clinical features of low- and high- risk groups as well as their differences in gene expression.

Results: Fourteen methylation sites were considered to be related to luminal breast cancer prognosis because their methylation levels, mRNA expression and prognoses were closely related to each other. The methylation level of SOSTDC1 was used to divide samples into hypo- and hyper-methylation groups. We also identified an mRNA signature, comprising eight transcripts, ESCO2, PACSIN1, CDCA2, PIGR, PTN, RGMA, KLK4 and CENPA, which was used to divide samples into low- and high-risk groups. The low-risk group showed significantly better prognosis than the high-risk group. A correlation analysis revealed that the risk score was an independent prognostic factor. Low- and high- risk groups significantly correlated with the survival ratio in Luminal A samples, but not in Luminal B samples on the basis of the TCGA and the Metabric cohort dataset. Further functional annotation demonstrated that the differentially expressed genes were mainly involved in cell cycle and cancer progression.

Conclusions: We identified several key methylation sites and an mRNA signature for predicting luminal breast cancer prognosis. The signature exhibited effective and precise prediction of prognosis and may serve as a prognostic and diagnostic marker for luminal breast cancer.

Keywords: Luminal breast cancer; Methylation; Prognosis; SOSTDC1; mRNA.

Conflict of interest statement

Ethics approval and consent to participate

The data of breast cancer patients we used in this study are all from the publicly accessible TCGA database. Separate ethics committee approval is not required.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures

Fig. 1

A flow chart showing the analysis process of this study

Fig. 2

Comparison of prognosis, risk scores and expression patterns of signature genes. a and b Kaplan–Meier survival curves of the low- and high-risk groups between the TCGA and GEO samples. Survival curves of low- and high-risk groups are indicated as black and red lines, respectively. P-value indicates significance for the log-rank test. c and d Distribution of risk scores, overall survival time and expression profiles of signature genes in the TCGA and GEO samples. Expression profiles are shown as heatmaps

Fig. 3

Difference in expression of signature genes. a Comparison of expression between the low- and high-risk groups. b Comparison of expression among the normal, hypomethylated and hypermethylated groups. ‘***’ indicates P-value < 0.001

Fig. 4

Prognosis of low- and high-risk groups in Luminal A and Luminal B samples from the TCGA and the Metabric cohort dataset. a Kaplan–Meier survival curves of low- and high-risk groups divided by eight signature genes in Luminal A and Luminal B samples from the TCGA database, respectively. The black line indicates the low-risk group, and the red line indicates the high-risk group. b Kaplan–Meier survival curves of low- and high-risk groups divided by eight signature genes in Luminal A and Luminal B samples from the Metabric cohort, respectively. The black line indicates the low-risk group, and the red line indicates the high-risk group

Fig. 5

Prognosis of the low- and high-risk groups in PR-negative and PR-positive samples. a Kaplan–Meier survival curves of the low- and high-risk groups in PR-negative samples. The black line indicates the low-risk group, and the red line indicates the high-risk group. P-value indicates the significance of the difference between the two groups. b Kaplan–Meier survival curves of the low- and high-risk groups in PR-positive samples. The blue line indicates the low-risk group, and the violet line indicates the high-risk group. P-value indicates the significance of the difference between the two groups. c Combination of the Kaplan–Meier survival curves from (a) and (b)

Fig. 6

Functional annotation of genes differentially expressed between the low- and high-risk groups. a Hierarchical clustering analysis of the expression levels of the top 20 positively and negatively related genes. b GO analysis of negatively (upper) and positively (lower) related genes. c KEGG pathway analysis of significantly correlated genes