Abstract
Abstract Background: Epithelial-mesenchymal transition (EMT) is a well-known multistep process of cancer cell invasion and metastasis, as well as treatment resistance. Our group has been reporting the predictive role of scores generated using Gene Set Variation Analysis (GSVA) of Hallmark pathways in breast cancer. In this study, we hypothesized that EMT score high breast cancer is aggressive and is associated with poor clinical outcome. Material and Methods: The clinicopathological data and transcriptome data of breast cancer patients from three independent large publicly available databases, The Cancer Genome Atlas (TCGA, n = 1077), The Molecular Taxonomy of Breast Cancer International Consortium (METARBRIC, n = 1904), and GSE96058 (n=3069) were utilized. Survival analyses; Overall survival (OS), Disease-specific survival (DSS) and Disease-free survival (DFS) were performed by comparing the high and low score groups. Tumor immune microenvironment was analyzed utilizing the values reported by Thorsson et al. Also, single sample Gene Set Enrichment analysis (ssGSEA) was performed between EMT high and low expression groups utilizing singe cell sequence cohorts. Results: EMT score was generated by Gene Set Variation Analysis of a Hallmark gene set and we divided each cohort into EMT score high and low groups by utilizing median as the cutoff. To our surprise, EMT score of the primary tumor was not associated with metastasis (N and M categories of cancer staging), Nottingham histological grade, nor MKI67 expression levels consistently in TCGA, METABRIC, and GSE96058. EMT score high tumors were not associated with worse DFS, DSS, OS in TCGA and METABRIC and OS in GSE96058. Analyses using xCell demonstrated that EMT score high tumors were associated with high infiltration of stromal cells such as adipocyte (p< 0.001, p< 0.001 and p< 0.001, respectively) and fibroblasts (p< 0.001, p< 0.001 and p< 0.001, respectively) in all three cohorts, TCGA, METABRIC, and GSE96058. Also, myeloid cells such as macrophages (p< 0.001, p< 0.001 and p< 0.001, respectively) and dendritic cells (p< 0.001, p< 0.001 and p< 0.001, respectively) were highly infiltrated with EMT score high tumors. Result of ssGSEA of single cell sequence cohorts revealed that cancer associated fibroblasts demonstrated highest EMT scores compared with the other cell types such as cancer cells, T-cells, B-cells, or myeloid cells. In other words, EMT score of a bulk tumor may reflect the signature from fibroblasts rather than cancer cells. Conclusion: We found that cancer associated fibroblasts rather than cancer cells are the major source of the transcriptomic signatures of EMT in the bulk tumor, which cautions us to be careful with the interpretation of the results of EMT signature from a bulk tumor. Citation Format: Yoshihisa Tokumaru, Rongrong Wu, Junko Ukai, Masanori Oshi, Yoshimi Niwa, Ryutaro Mori, Kazuaki Takabe, Manabu Futamura. Transcriptomic signature score of Epithelial-Mesenchymal Transition (EMT) of a bulk tumor may not reflect that of cancer cells [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P4-02-03.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have