Abstract 4587: Vascular-derived signature imprinted by tumor microenvironment-dependent transcriptional memory predicts colon cancer prognosis

Abstract Background: Antiangiogenic therapy is part of the guideline therapy of colorectal cancer (CRC). Surprisingly, the impact of the tumor microenvironment (TME) on tumor vessel endothelial cells (TECs) is largely unclear. The aim of this study was to investigate the presence of a TME-dependent transcriptional memory in TECs isolated from patients and to exploit it to retrieve signatures that characterize TECs in different intratumoral TMEs with impact on patient outcome. Methods: ECs from tumor and normal colon tissues, PBMCs and tumor cells were isolated from CRC patients with different prognostic TMEs. Cells were analyzed by qPCR, immunocytochemistry and multi-omics (transcriptomics, EPICmethylation chips, exome sequencing). Integrative bioinformatics was used to identify TME-dependent memory genes predicting prognosis and scRNASeq to validate endothelial gene expression in CRC tissues. Results: Ultrapure TECs were isolated from CRC with different prognostic TMEs (Th1 vs. non-Th1) and systematically compared by multi-omics. A transcriptional memory differentiating the respective TEC groups was identified. This in vivo imprinted transcriptional memory was preferentially regulated by epigenetic DNA methylation but not by genomic alterations and was different from an in vitro primed transcriptional memory to IFN-γ. Moreover, it was specific for TECs and not observed in CAFs. With integrative bioinformatics a TME-dependent memory signature was extracted and its expression in TECs in CRC tissues was confirmed by scRNASeq. Notably, the identified signature predicted the prognosis of CRC patients. Conclusion: We identified a tumor vessel-derived TME-dependent transcriptional memory signature that was manifested by epigenetic mechanisms and allowed tumor vessel-based prediction of CRC patients prognosis. Citation Format: Elisabeth Naschberger, Maximilian Fuchs, Nicholas Dickel, Meik Kunz, Charles G. Anchang, Richard Demmler, Bernt Popp, Arif B. Ekici, Steffen Uebe, Carol I. Geppert, Claudia Günther, Susanne Merkel, Vera S. Schellerer, Michael Stürzl. Vascular-derived signature imprinted by tumor microenvironment-dependent transcriptional memory predicts colon cancer prognosis. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4587.