Axillary lymph node (ALN) metastasis is a common form of breast cancer metastasis. Understanding molecular mechanisms of ALN metastasis may provide new therapeutic strategy for breast cancer. To explore the functional evolution process of ALN metastasis in breast cancer, we proposed a network model to identify functional modules of ALN metastasis by clustering the pathway evolution network. First, a gene co-expression network was constructed for each ALN stage separately, and integrated gene expression, DNA methylation, and copy number variation as node attributes information. Second, integrative non-negative matrix factorization (IntNMF) algorithm was used to integrate multi-omics data to detect gene co-expression network modules. Next, the pathway evolution network for ALN metastasis was built by integrating enriched pathways within gene co-expression network modules at each stage. Finally, the topological and functional similarity-based module detection algorithm was proposed to identify functional modules from the pathway evolution network, and seven modules were obtained. Compared with other classical algorithms, the proposed algorithm achieved better similarity cohesion while satisfying the structural cohesion. This study identified functional modules of ALN metastasis by clustering the pathway evolution network, which may provide insights into exploring the functional evolution process of ALN metastasis and finding drug targets for breast cancer.