The present invention relates to a method for predicting the prognosis of a breast cancer patient. More specifically, to provide information needed to predict the prognosis of a breast cancer patient, the method for predicting the prognosis of breast cancer including the following steps of the present invention comprises: (a) obtaining a biological sample from a breast cancer patient; (b) measuring the mRNA expression level of matrix metallopeptidase 11 (MMP11) and the mRNA expression level of cluster of differentiation 2 (CD2) from patient information or the sample of step (a); (c) normalizing the gene mRNA expression levels selected and measured in step (b); and (d) predicting the prognosis of breast cancer by combining the gene expression levels normalized in step (c), wherein overexpression of the MMP11 indicates a bad prognosis, and overexpression of CD2 indicates a good prognosis.The method of the present invention has an effect of being capable of more accurately predicting the future prognosis of metastasis, recurrence, or metastatic recurrence in breast cancer patients, and in particular, has a very excellent ability to predict the prognosis of HER2-type breast cancer, the prognosis of which is very poor, and thus can be usefully used to provide clues for the direction of future treatment of breast cancer.