Although all cancers are molecularly distinct, many share common driver mutations. Pan-cancer analysis, utilizes next-generation sequencing (NGS), pan-cancer model systems, and pan-cancer projects such as The Cancer Genome Atlas (TCGA), to assess frequently mutated genes and other genomic abnormalities that are common among many cancer types, regardless of the tumor origin, providing new directions for tumor biology research. However, there is currently no study that has objectively analyzed the results of pan-cancer studies on cancer biology. For this study, 999 articles on pan-cancer published from 2006 to 2020 were obtained from the Scopus database, and bibliometric methods were used to analyze citations, international cooperation, co-authorship and keyword co-occurrence clusters. Furthermore, we also focused on and summarized the application of pan-cancer in breast cancer. Our result shows that the pan-cancer studies were first published in 2006 and entered a period of rapid development after 2013. So far, 86 countries have carried out international cooperation in sharing research. Researchers form the United States and Canada have published the most articles and have made the most extensive contribution to this field, respectively. Through author keyword analysis of the 999 articles, TCGA, biomarkers, NGS, immunotherapy, DNA methylation, prognosis, and several other keywords appear frequently, and these terms are hot spots in pan-cancer studies. There are four subtypes of breast cancer (luminalA, luminalB, HER2, and basal-like) according to pan-cancer analysis of breast cancer. Meanwhile, it was found that breast cancer has genetic similarity to pan-gynecological cancers, such as ovarian cancer, which indicates related etiology and possibly similar treatments. Collectively, with the emergence of new detection methods, new cancer databases, and the involvement of more researchers, pan-cancer analyses will play a greater role in cancer biology research.