The prognostic marker KRT81 is involved in suppressing CD8 + T cells and predicts immunotherapy response for triple-negative breast cancer

Abstract

Triple-negative breast Cancer (TNBC) is an aggressive subtype lacking estrogen, progesterone, and HER2 receptors. Known for limited targeted therapies, it poses challenges and requires personalized treatment strategies. Differential analysis revealed a significant decrease in keratin 81 (KRT81) expression in non-TNBC samples and an increase in TNBC samples, lower KRT81 expression correlated with better TNBC patient outcomes. It emerged as an independent predictive factor for TNBC, with associations found between its expression and clinically relevant features. We further developed a nomogram for survival probability assessment based on Cox regression results, demonstrating its accuracy through calibration curves. Gene annotation analysis indicated that KRT81 is involved in immune-related pathways and tumor cell adhesion. KRT81 is associated with immune cell infiltration of Follicular helper T cells (Tfh) and CD8 + T cells, suggesting its potential impact on the immunological microenvironment. The study delved into KRT81's predictive value for immunotherapy responses, high expression of KRT81 was associated with greater potential for immune evasion. Single-cell RNA sequencing analysis pinpointed KRT81 expression within a specific malignant subtype which was a risk factor for TNBC. Furthermore, KRT81 promoted TNBC cell proliferation, migration, invasion, and adhesion was confirmed by gene knockout or overexpression assay. Co-culture experiments further indicated KRT81's potential role in inhibiting CD8 + T cells, and correlation analysis implied KRT81 was highly correlated with immune checkpoint CD276, providing insights into its involvement in the immune microenvironment via CD276. In conclusion, this comprehensive study positions KRT81 as a promising prognostic marker for predicting tumor progression and immunotherapy responses in TNBC.

Keywords: CD8 + T cells; Triple-negative breast cancer; immune microenvironment; immunotherapy response; keratin 81; prognostic gene.

Graphical abstract

Figure 1.

Identification of KRT81 as a valuable prognostic marker for TNBC. (a) Screening of optimal factors of the KRT family for prognosis prediction through LASSO Cox regression analysis. (b) Analyzing the HR value of the KRT-related prognostic genes by univariate Cox regression. (c) Comparison of the KRT81’s expression between the normal, non-TNBC, and TNBC samples. (d, e) the survival and ROC curves of KRT81 for predicting OS of TNBC in the TCGA cohort. (f, g) the survival and ROC curves of KRT81 for predicting OS of TNBC in the METABRIC cohort.

Figure 2.

KRT81 prediction ability verification and construction of a nomogram. (a, b) Verification of the independent predictive ability of KRT81 via univariate and multivariate Cox regression analysis. (c, d) Construction and validation of a nomogram included clinical features and expression of KRT81 for predicting OS of TNBC patients.

Figure 3.

Functional annotation by GSEA algorithm. (a, b) the down- and up-regulated pathways involved in KRT81 were conducted by the GSEA method, respectively.

Figure 4.

Correlation analysis of KRT81 and immune infiltration. (a) Differences in stromal, immune, and ESTIMATE scores between the high- and low-KRT81 groups were examined by the ESTIMATE method. (b) Differences in 22 types of immune cell infiltration were calculated via the CIBERSORT algorithm. (c, d) the correlation of KRT81 expression and immune cell infiltration. (e) The correlation analysis between KRT81 expression and immune checkpoints. The larger the circle, the higher the gene expression. The redder the color, the more significant the difference.

Figure 5.

Validation of KRT81 as a predictor of immunotherapy response and drug sensitivity in TNBC patients. (a) The differences in TIDE, MSI, exclusion, and dysfunction scores between the high- and low-KRT81 groups. (b) The differences in IPS scores between the high- and low-KRT81 groups. (c) Comparing the expression levels of KRT81 in patients with non-response and response to immunotherapy. (d) Correlation analysis of KRT81 expression level and sensitivity of common chemotherapy drugs for TNBC.

Figure 6.

Analysis of the KRT81’s distribution in the cell subtypes by scRNA-seq GSE176078 dataset. (a, b) Visualization of cell clustering and annotation by t-SNE plots. (c) Expression of KRT81 in the different cell clusters. (d) Analysis of the relationship between the cell clusters and TNBC prognosis by univariate Cox regression.

Figure 7.

The effects of knocking down KRT81 on MDA-MB-231 cell line. (a) Verification of the KRT81’s expression in the clinical tissues by IHC, the non-TNBC tissue was obtained from the normal margin during the surgery. (b) The relative expression level of KRT81 of MDA-MB-231 treated with overexpressed or blank plasmid. (c) The proliferation curves of the MDA-MB-231 cell line. (d) Cell apoptosis rate between the sh-NC and sh-KRT81 groups. (e) The cell cycle distribution of MDA-MB-231. (f) Effects of knocking down KRT81 on cell migration (the upper), invasion (the middle), and adhesion (the lower), respectively.

Figure 8.

The effects of KRT81 overexpression on BT-20 cell line. (a) Relative expression level of KRT81 of BT-20 cell line treated with overexpressed plasmid or vector. (b) The proliferation curves of BT-20 cell line. (c) Cell apoptosis rate between the vector and oe-KRT81 groups. (d) The cell cycle distribution of BT-20 cell line. (e) Effects of overexpressed KRT81 on cell migration (the upper), invasion (the middle), and adhesion (the lower), respectively.

Figure 9.

KRT81 is involved in suppressing CD8 + T cells via CD276 (a) Effects of gene alteration of KRT81 expression on proliferation (Ki-67) and cytokineS secretion (IFN-γ and TNF-α) of CD8 + T cells in co-culture system. (b) Effect of KRT81 knockdown on CD276 expression in MDA-MB-231 cell line (the left), Effects of overexpression of KRT81 on CD276 expression in BT-20 cell line (the right).