In the present cohort, we found that the NS protein expression status was positively correlated with both ER and HER2 status and was a powerful prognostic factor. Patients with NS-positive breast tumors had a significantly shorter DFS time than those with NS-negative tumors (P = 0.020, Figure
2), and multivariate analysis for DFS showed that NS positivity had an independent impact as a prognostic indicator among breast cancer patients (P = 0.036, Table
2). To our knowledge, this is the first report to show the clinical implication of NS protein expression in invasive breast cancers.
Although several studies have shown the important roles of NS in the pathogenesis of various cancer types
[8–13] as well as the maintenance of cancer stem cells
[14, 15], no direct evidence is yet available to support that NS is a marker of cancer stem cells. Currently, molecules such as CD44, CD133, ALDH1, and CXCR4 have been found to be potential markers of cancer stem cells
[20–25]. Furthermore, the expression of these stem cell markers has been shown to be a poor prognostic indicator in several human cancer types
[24, 26–30]. Based on these observations, our results show that high NS expression is a powerful indicator of poor outcome, consistent with the idea that NS may be a breast cancer stem cell marker.
The limitations of the present study included the retrospective analyses and the heterogeneity of adjuvant treatments. Therefore, one should pay careful attention when interpreting these results. Further studies using a uniformly treated patient cohort are required to clarify the role of NS in breast cancer stem cells.
We found that the patient group with tumors coexpressing NS and p53 had shorter DFS times than the patient group with tumors negative for either NS or p53. GTP binding modulates the movement of NS from the nucleoli to the nucleoplasm; NS then binds p53 at its N-terminal basic domain, which results in the suppression of p53 function
[6, 7]. Because prolongation of the half-life of most of the mutated p53 protein induces its nuclear accumulation, it is generally believed that the p53 pathway does not fully function in tumors with high p53 nuclear immunoreactivity
[31–33]. This evidence leads to the assumption that p53 function would be profoundly suppressed in tumors coexpressing NS and p53. Our results show the validity of this concept and that functional crosstalk between NS and p53 may also occur in vivo.
Currently, we cannot explain the correlation between NS expression and p53 expression. Although several studies have shown that NS modulates the expression of wild-type p53
[34, 35], the role of NS in breast cancers with mutant p53 has not yet been evaluated. Further research is needed to elucidate the correlation.
We found that the NS expression status was positively correlated with both ER and HER2 status and also found a significant prognostic implication of NS expression for patients with luminal-type tumors and those with HER2-type tumors, except for those with triple-negative tumors. NS was first identified as a gene upregulated in MCF-7 cells upon 17β-estradiol treatment
; therefore, our inclusion of subgroup analysis among patients with luminal-type tumors was reasonable. To our knowledge, this is the first report to demonstrate the possible association between NS and HER2. Zhang G et al. showed that NS is required for the expression of EGF and EGFR in an esophageal squamous carcinoma cell line
. Presumably, NS is required for the expression of HER2 in a manner similar to that for EGFR. We found no survival impact of the NS expression status among patients with triple-negative tumors, who show higher rates of mutated p53 than patients with luminal-type or HER2-type tumors
. NS can function in the presence of wide-type p53
; therefore, the expression status of NS may have survival impact only for the luminal-type and HER2-type tumors.