Distinctive role of SIRT1 expression on tumor invasion and metastasis in breast cancer by molecular subtype.

Research paper by Yul Ri YR Chung, Hyojin H Kim, Soo Young SY Park, In Ae IA Park, Ja June JJ Jang, Ji-Young JY Choe, Yoon Yang YY Jung, Seock-Ah SA Im, Hyeong-Gon HG Moon, Kyung-Hun KH Lee, Koung Jin KJ Suh, Tae-Yong TY Kim, Dong-Young DY Noh, Wonshik W Han, Han Suk HS Ryu

Indexed on: 26 May '15Published on: 26 May '15Published in: Human Pathology


The aim of this study was to evaluate silent mating type information regulation 2 homolog 1 (SIRT1) expression levels by subtype and evaluate its predictive power of axillary lymph node metastasis (LNM) and its association with clinical outcome. A total of 427 patients diagnosed with invasive ductal carcinoma were chosen, immunohistochemical staining for SIRT1 expression was performed on tissue microarrays, and in vitro experiments with each intrinsic subtype of human breast cancer cell line were carried out. Increased expression of SIRT1 in hormone receptor-positive breast cancer and HER2 breast cancer subtype significantly correlated with lower risks of LNM. On the contrary, in triple-negative breast cancer, increased SIRT1 expression was more frequently observed in LNM-positive subgroup than LNM-negative subgroup. Combination of statistically significant, independent parameters including SIRT1 revealed predictive performance for LNM with area under the curve of 0.602, 0.587, and 0.726 for hormone receptor-positive breast cancer, HER2 breast cancer, and triple-negative breast cancer subtype, respectively. Inhibition of SIRT1 expression with small interfering RNA suppressed tumor invasion in MDA-MB-231, specifically. This is the first study to examine SIRT1 expression in breast cancer by subtype, and we have observed the potentially different role of SIRT1 gene having tumor-suppressive or tumor-promoting influence depending on the subtype; thus, different associations between SIRT1 expression and prognosis by subtype should be considered in its target therapy.

More like this: