Dissecting The Transcriptional Profiles Of Metastatic And Primary Disease Across Cancer Types

Cancer metastasis is the principal cause of death in individuals with cancer; nevertheless, the molecular basis of metastases is poorly understood. Genetic and epigenetic changes in malignant cells and their interaction with the tumor microenvironment are presumably key events in the establishment of metastases. In this study, we pursue to provide insights into potential importance of various fundamental cancer-related pathways in the metastatic process. We have performed a pathway - based analysis on mRNA expression data (Agilent 8X60K array) from colorectal liver metastases (CLM; N = 38 patients, n = 44 metastases), breast lymph-node metastases (BCM; N = n = 43) and melanoma lymph-node metastases (MLM; N = n= 44). For each of the 186 pathways in the KEGG pathway database, each sample is represented as points in a ‘n’ dimensional space according to their gene expression values, where ‘n’ is the number of genes in the pathway. We first calculated centroids for each of the three metastases types (CLM, BCM and MLM) and then calculated the Euclidean distance between centroids to get a measure of how similar or different each cancer type is for that particular pathway. The findings were compared to the corresponding primary cancer types and normal tissue available from TCGA to uncover biological processes specific to metastases. The pathways were functionally classified into genetic information processing, cellular processes, environmental information processing and metabolic pathways as per KEGG database. Across metastases from different cancer types, pathways involved in genetic information processing (e.g. translation, transcription, protein folding and degradation, DNA replication and repair) and metabolic pathways were more similar than the pathways involved in cellular processes (e.g. cell-cycle, apoptosis, adherence junction), and signal transduction pathways (e.g. Notch, MTOR, JAK STAT) responsible for environmental information processing. The genetic information processing pathways were more similar across cancer types also in the primary setting than the pathways involved in cellular and environmental information processing, whereas the metabolic pathways were more similar across cancer types in the metastatic setting compared to the primary setting. Preliminary findings suggest that the basic molecular architecture of the primary tumors is maintained, except for the metabolic pathways, which become more similar in the metastatic setting; and hence may be a common denominator across cancer types during metastatic process. Citation Format: Laxmi Silwal-Pandit, Vigdis Nygaard, Hege Russnes, Vegar Johansen Dagenborg, Veronica Skarpeteig, Olga Ostrup, Silje Nord, Vivi Ann Flornes, Anne Hansen Ree, Kjersti Flatmark, Anne-Lise Borresen-Dale, Ole Christian Lingjaerde, Gunhild Mari Maelandsmo. Dissecting the transcriptional profiles of metastatic and primary disease across cancer types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2842. doi:10.1158/1538-7445.AM2017-2842