O091 Predict factors of hepatocyte growth factor family and bone morphogenetic proteins for breast cancer nodal metastasis

Introduction Lymph node metastasis is essential for the outcome of patients with breast cancer. Bone morphogenetic proteins (BMP) and hepatocyte growth factor (HGF), key to angiogenesis and metastasis of cancers, were investigated for their roles in the predication of nodal metastasis from breast cancer. Methods Expression of BMP family ligands, BMP receptors, HGF, its receptor MET and regulatory factors of HGF activation including HGF activator (HGFA), HGFA inhibitor 1 (HAI1), HAI2, Matriptase-1 and Matriptase-2 were determined in breast cancers (n=127) using real time quantitative PCR. Transcript levels of these genes were analysed in the breast cancers for their implication in lymphatic involvement using Mann-Whitney test. Predictive potential of those genes for nodal metastasis of these molecules was evaluated using both binary logistic regression and receiver operating characteristic (ROC) with SPSS (Version 27). Results With median as a cut-off value, logistic regression analyses showed that MET, Matriptase-1 and BMP15 were positively associated with nodal metastasis whereas Matriptase-2, BMP3 and HAI1 were inversely correlated with nodal involvement. After integrating these six prospective factors, ROC model returned with a significant value against nodal status (RUC=0.657, p=0.001). With the optimal ROC cut-off value to dichotomise patients, the integrated expression produced a significant prediction of nodal metastasis (p=0.006, HR=2.929). MET and HAI1 was correlated with the lymphangiogenesis marks of LYVE1, podo, prox1. Matriptase-1 was inversely correlated with podo, whilst Matriptase-2 and BMP15 showed a positive correlation with prox1. Conclusion The MET/Matriptase-1/BMP15 and the inversed Matriptase-2/BMP3/HAI1 are connected significantly with lymph node metastasis in breast cancer.