A meta-metastasis analysis identifies pan-cancer markers and therapeutic targets

By October 29, 2017July 17th, 2020News


Targeting metastasis has the potential to reduce the lethality of cancer, instead making it a manageable disease. While metastasis has been studied extensively in individual forms of cancer, less attention has been paid to aspects of metastatic progression conserved across cancers. Here, we identify mechanisms consistently associated with metastasis in primary tumors from 4106 patients across 12 different cancers via comparison of primary-site tumors with no lymph node involvement to those with staging in other tissues or lymph nodes.



Leveraging RNA-seq data collected by The Cancer Genome Atlas (TCGA), we identify genes differentially expressed between primary tumors from patients with primary-site, node-negative disease (M0, N0) compared to primary tumors from patients with node-positive disease (N1, N2, or N3) and/or distant metastases (M1). Differentially expressed genes were first determined within each cancer type, to reduce tissue- or disease-specific effects. Subsequently via meta-analysis, we combine these results to identify commonality across 12 different types of cancer.



Twenty-six genes passed a threshold of adj. P < 0.05 in both the meta-analysis and individually in at least three types of cancer, while another 45 passed a threshold of adj. P < 0.05 in both the meta-analysis and individually in at least two types of cancer. These 71 genes differing most consistently between primary tumors that have metastasized and those that have not include both genes previously known to be associated with metastasis, and genes for which this is the first report of an association with metastasis. The ranked group of 71 genes represents a prioritized list of genes that warrant further consideration as targets for inhibiting metastasis.



Metastasis is a universal problem in cancer. In the pan-cancer analysis described here, we have identified signatures that distinguish primary tumors that ultimately metastasize from those that do not. A pan-cancer analysis has identified 71 genes most consistently associated with metastasis across different types of cancer. The fact that these genes are conserved across cancer types means that they may be more associated with fundamental underlying mechanisms of metastasis and therefore represent more robust intervention points, and also suggests that therapies targeting these mechanisms will be more broadly applicable to a wider set of patients.