Cluster Analysis According to Immunohistochemistry is aRobust Tool for Non-Small Cell Lung Cancer and Revealsa Distinct Immune Siqnature-defined Subgroup

摘要

In medicine, clustering is known as the grouping of objects by calculating similarities; thus, objects in one cluster are more closely related than those found in other clusters. Especially in association with high-throughput protein and gene analyses, the utilization of clustering has become increasingly popular for characterizing different subgroups of cancer. Particularly in breast cancer and in lymphomas, cluster analysis has been reported as a powerful tool for discovering different classes.1'2 Sub-classification of heterogenous cohorts makes it possible to identify distinct prognostic or predictive subgroups that are, for example, potentially linked to specific treatment outcomes. In breast cancer, gene expression analyses based on clustering have made it possible to distribute morphologically similar tumors into groups showing different prognosis as well as different treatment responses.3'4 In the future, such expression profiles will likely also be incorporated into the tumor classification systems.