AIMS: SOX2 is a key regulatory gene in embryonic stem cells. Although it has been implicated in cancer progression, its role in breast carcinoma is poorly understood. MATERIALS AND METHODS: Fifty-seven ductal carcinomas in situ (DCIS), 552 invasive breast carcinomas and 107 corresponding metastatic lymph nodes were evaluated immunohistochemically for the expression of SOX2. Its correlation with clinicopathological features, other biomarker profiles and patients' outcomes were analysed. RESULTS: SOX2 was detected in 19.0% (105 of 552) of invasive breast carcinomas and 12.3% (seven of 57) of DCIS. Expression correlated with larger tumour size (P = 0.005) and higher grade (P = 0.002). It was associated negatively with ER (P = 0.015) and PR (P = 0.046) expression, but positively with Ki67 index (P = 0.013). Interestingly, it was also associated with neuroendocrine marker expression (synpatophysin and chromogranin/synaptophysin, P = 0.048 and 0.028, respectively). Expression appeared to be independent from that of common stem cell markers, namely CD44, CD24 and aldehyde dehydrogenase 1 (ALDH1). Furthermore, a higher rate of expression was observed in metastatic lymph nodes than in the corresponding primary tumours (P = 0.034). High SOX2 expression was correlated with poor disease-free survival (log-rank=9.489, P = 0.012) and was an independent prognostic factor (HR=2.918, P = 0.015) in patients with high nodal stages. CONCLUSIONS: In summary, SOX2 expression was related to adverse breast carcinoma profile and poor outcome in selected patient groups.
AIMS: SOX2 is a key regulatory gene in embryonic stem cells. Although it has been implicated in cancer progression, its role in breast carcinoma is poorly understood. MATERIALS AND METHODS: Fifty-seven ductal carcinomas in situ (DCIS), 552 invasive breast carcinomas and 107 corresponding metastatic lymph nodes were evaluated immunohistochemically for the expression of SOX2. Its correlation with clinicopathological features, other biomarker profiles and patients' outcomes were analysed. RESULTS:SOX2 was detected in 19.0% (105 of 552) of invasive breast carcinomas and 12.3% (seven of 57) of DCIS. Expression correlated with larger tumour size (P = 0.005) and higher grade (P = 0.002). It was associated negatively with ER (P = 0.015) and PR (P = 0.046) expression, but positively with Ki67 index (P = 0.013). Interestingly, it was also associated with neuroendocrine marker expression (synpatophysin and chromogranin/synaptophysin, P = 0.048 and 0.028, respectively). Expression appeared to be independent from that of common stem cell markers, namely CD44, CD24 and aldehyde dehydrogenase 1 (ALDH1). Furthermore, a higher rate of expression was observed in metastatic lymph nodes than in the corresponding primary tumours (P = 0.034). High SOX2 expression was correlated with poor disease-free survival (log-rank=9.489, P = 0.012) and was an independent prognostic factor (HR=2.918, P = 0.015) in patients with high nodal stages. CONCLUSIONS: In summary, SOX2 expression was related to adverse breast carcinoma profile and poor outcome in selected patient groups.