Expression of biomarkers in the AKT pathway correlates with malignancy and recurrence in phyllodes tumours of the breast

Aims Phyllodes tumours ( PT s) of the breast are uncommon fibroepithelial neoplasms with the potential to recur and metastasise. Apart from histological grading, the expression of biological markers and its relationship with tumour behaviour have been topics of interest. The phosphatidylinositol 3‐kinase ( PI 3K)– AKT pathway regulates diverse biological functions, and is one of the most frequently deregulated pathways in cancers. Little is known of PI 3K– AKT pathway alteration in PT . We aim to investigate the alterations in different component of AKT pathway in PT s. Methods and results This study investigated the expression of four biological markers involved in this pathway ( PTEN , INPP 4B, PI 3 KCA and pAKT ) in 134 PT s by the use of immunohistochemistry. According to an immunoscore incorporating staining intensity and proportion, low epithelial INPP 4B expression ( P = 0.045) was associated with recurrence. A trend of association was found for low epithelial PTEN expression with recurrence ( P = 0.090). Interestingly, low epithelial INPP 4B expression was also associated recurred tumours ( P = 0.043). Stromal PI 3 KCA expression ( P = 0.016) and pAKT expression ( P = 0.006) were found to be correlated with increased histological grade, but an opposite trend was seen for stromal INPP 4B expression ( P = 0.018). In addition, epithelial and stromal PTEN expression, PI 3 KCA expression and pAKT expression showed strong correlations with each other ( P ≤ 0.001). Conclusions These findings indicate that alterations in AKT pathway activation may correlate with malignant transformation and recurrence in PT . Low epithelial INPP 4B/ PTEN expression is associated with shorter recurrence‐free survival. These observations suggest that the pathway may play a crucial role in the biological behaviour and progression of PT , and assessing the expression of this pathway may be of value in diagnosis, grading, prognostication, and management.