Original Article
Complicated prognostic values of CCL28 in breast cancer by subtype
Abstract
Background: The expression of CCL28 and its relationship with clinical outcomes remain unclear in the setting of heterogeneous breast cancer. The purpose of the current study was to identify the expression characteristics of chemokine CCL28 in breast cancer, with a focus on its prognostic relevance to different subtypes.
Methods: First, we investigated the expression of CCL28 in 150 breast cancer patients immunohistochemically and assessed the impact of CCL28 on relapse-free survival (RFS) in the whole cohort and different clinical subtypes [defined by hormone receptor (HR), and HER-2 status] by univariate and multivariate analysis. Furthermore, the other two cohorts comprised of 863 patients from the Cancer Genome Atlas (TCGA) database and 1,764 patients from the Kaplan-Meier plotter database, respectively, were chosen to validate the prognostic values of CCL28 in breast cancer.
Results: Those with positive CCL28 expression had improved RFS in luminal-like (HR positive, any HER-2 status) subtype (P=0.052) but had impaired RFS in triple-negative cases (P=0.019), after adjustment with tumor size and lymph node status. Consistently, multivariate analysis in the TCGA cohort revealed improved disease-free survival (DFS) among patients with high expression of CCL28 in luminal-like subtype (P=0.043) and decreased DFS in patients expressing high CCL28 in triple-negative cases (P=0.010). The subsequent analysis of the Kaplan-Meier plotter cohort also demonstrated that CCL28 was a favorable prognostic factor for luminal-like cases [luminal A (P<0.001) and luminal B (P=0.031)], but a poor prognostic indicator for the patients with triple-negative phenotype (P<0.001).
Conclusions: CCL28 was a favorable prognostic factor for luminal-like cases and detrimental for triple-negative subtype, indicating that the same chemokine may play different or even opposite roles in the recurrence and metastasis of different molecular subtypes of breast cancer.
Methods: First, we investigated the expression of CCL28 in 150 breast cancer patients immunohistochemically and assessed the impact of CCL28 on relapse-free survival (RFS) in the whole cohort and different clinical subtypes [defined by hormone receptor (HR), and HER-2 status] by univariate and multivariate analysis. Furthermore, the other two cohorts comprised of 863 patients from the Cancer Genome Atlas (TCGA) database and 1,764 patients from the Kaplan-Meier plotter database, respectively, were chosen to validate the prognostic values of CCL28 in breast cancer.
Results: Those with positive CCL28 expression had improved RFS in luminal-like (HR positive, any HER-2 status) subtype (P=0.052) but had impaired RFS in triple-negative cases (P=0.019), after adjustment with tumor size and lymph node status. Consistently, multivariate analysis in the TCGA cohort revealed improved disease-free survival (DFS) among patients with high expression of CCL28 in luminal-like subtype (P=0.043) and decreased DFS in patients expressing high CCL28 in triple-negative cases (P=0.010). The subsequent analysis of the Kaplan-Meier plotter cohort also demonstrated that CCL28 was a favorable prognostic factor for luminal-like cases [luminal A (P<0.001) and luminal B (P=0.031)], but a poor prognostic indicator for the patients with triple-negative phenotype (P<0.001).
Conclusions: CCL28 was a favorable prognostic factor for luminal-like cases and detrimental for triple-negative subtype, indicating that the same chemokine may play different or even opposite roles in the recurrence and metastasis of different molecular subtypes of breast cancer.
