Deciphering the immunosuppressive landscape of high-grade serous ovarian cancer
Date
2022-01-19
Authors
Smazynski, Julian
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
High-grade serous ovarian cancer (HGSC) remains the most common and lethal subtype of ovarian cancer with a 5-year survival rate of ~30%, highlighting an urgent need for new treatments. Cancer immunotherapy has emerged as an efficacious strategy aimed at harnessing the exquisite capabilities of our immune system to destroy malignant cells. However, the development of more effective immunotherapies is hampered by our limited understanding of the phenotype of bona fide tumor-reactive T cells versus irrelevant bystanders. Further, T cells that exhibit tumor specificity appear to encompass a tissue resident memory (TRM) phenotype but combat a harsh immunosuppressive tumor microenvironment, often leading to an exhausted phenotypic state and evasion of immune-mediated destruction. These insights have led to rapid clinical implementation of so-called “checkpoint blockade” therapies that re-invigorate T cell-mediated tumor destruction by blocking surface inhibitory receptors or ligands. Thus, by identifying the phenotype of prognostically favourable TRM T cells and the immunosuppressive networks they face, my thesis work tackles a critical challenge in designing the next generation of therapeutic interventions for this disease.
To address this challenge, I hypothesized that (1) the TRM phenotype could be modulated for improving adoptive T cell therapy; (2) TRM TIL characterized by the co-expression of CD103, PD-1, and CD39 in HGSC provide improved prognostic benefit indicative of enriched tumor reactivity; (3) the TIGIT/CD155 signalling axis plays a crucial role in shaping the immunosuppressive landscape impeding TRM T cells in HGSC. Firstly, I developed methods for modulating the TRM phenotype on expanded human and murine T cells for adoptive cell therapy and assessed the therapeutic impact of these phenotypes. Secondly, we applied high-dimensional flow cytometry, single-cell sequencing, and multiplexed immunofluorescence to primary human HGSC specimens to explore the single-cell phenotypic profiles and prognostic significance of tumor-infiltrating T cells co-expressing three putative markers of tumor reactivity: CD39, CD103, and PD-1. These ‘triple-positive’ T cells exhibited a highly activated/exhausted phenotype and superior prognostic value relative to all other T-cell subsets, suggesting these markers enrich for tumor-reactive clones. Furthermore, these triple-positive cells exhibited heightened expression of the inhibitory checkpoint TIGIT, which plays a prominent role in tumor-mediated immune suppression. Finally, to explore the therapeutic implications of this finding, we investigated the relationship between the TIGIT signaling axis on TIL and prognosis in HGSC. Once again utilizing high-dimensional flow cytometry, multi-color histological imaging, and gene expression profiling we found T cells from HGSC frequently express TIGIT ex vivo and post-clinical expansion. Further, CD155, the dominant ligand for TIGIT, was largely expressed on malignant epithelium in HGSC and showed a negative association with immune infiltration. Thus, TRM T cells represents a compelling immunotherapeutic immune subset in HGSC and one that could be bolstered by immune checkpoint inhibition of the TIGIT/CD155 axis.