Preview

PDF, English Download (6MB) | Terms of use

Abstract

Tumor-derived material in blood samples is informative concerning molecular alterations in respective tumor tissues. Thus, biomarkers detected by liquid biopsy can guide clinicians in designing personalized therapies and moreover may serve as a proxy for treatment response and success. However, robust biomarkers for the detection of aberrant oncogene activity in tumor cells and the identification of druggable target structures are difficult to define. To illustrate the applicability of RNA signatures as cancer-spanning biomarkers, I established an in vitro screening approach integrating RNA-seq data from siRNA screens, publicly available ChIP-seq data as well as patient expression data from different tumor entities. As exemplified for the Hippo pathway, a 4-gene long non-coding RNA (lncRNA) signature consisting of CYTOR, MIR4435-2HG, SNHG1, and SNHG17 was defined that is transcriptionally controlled by the YAP/TAZ/TEAD complex. This 4-lncRNA signature represents a robust predictor of YAP activity in several tumor types such as liver or lung cancer and its overexpression is statistically associated with poor clinical outcome. In vitro experiments showed that lncRNA signature constituents themselves contribute to the tumor-promoting properties of the Hippo/YAP/TAZ pathway. Furthermore, murine orthologues of these lncRNAs were overexpressed in YAPS127A transgenic mouse livers and lncRNA signature levels were elevated in a subgroup of human cancer tissues and serum samples. Importantly, nuclear YAP accumulation in human liver cancer tissues is significantly associated with YAP-dependent lncRNA abundance in the serum of these patients. Moreover, the signature defines responsiveness of tumor cells to YAP-directed-pharmacological inhibition. These results let me draw the following conclusions: First, lncRNA-based approaches broaden previous liquid biopsy concepts by allowing the detection of potential druggable oncogene activity in the tumor. Second, lncRNAs represent robust and sensitive biomarkers to identify patients eligible for specific oncogene-directed therapies and to monitor treatment response. Third, the lncRNA signature constituents themselves support tumorigenesis in a multi-modal manner. Fourth, the YAP/TAZ/TEAD complex represents a promising target structure to inhibit YAP/TAZ activity. Lastly, YAP and TAZ may play different roles in regulating lncRNA expression depending on the cellular context. Thus, my data underline that liquid biopsy-based detection of pan-cancer lncRNA signatures can define oncogene activity in tumor cells. I therefore conclude that serum lncRNA signatures represent novel and powerful tools for diagnostics, therapy design, as well as for monitoring treatment success.