%0 Generic
%A Criado Moronati, Elvira Magdalena
%C Heidelberg
%D 2024
%F heidok:35393
%K Tumor-infiltrating lymphocytes; Adoptive cell therapy
%R 10.11588/heidok.00035393
%T Characterization of tumor-infiltrating CD8+ T cells towards clinical application
%U https://archiv.ub.uni-heidelberg.de/volltextserver/35393/
%X Tumor-infiltrating lymphocytes (TILs) have been successfully used in adoptive cell therapies (ACT) for various cancer types. However, only a small fraction of TILs mediates antitumor responses. Therefore, the enrichment of tumor-reactive TILs prior to the ex vivo expansion during cell manufacturing for ACT may improve their clinical efficacy. An enrichment based on an a priori tumor-reactive T cell phenotype represents an appealing approach as it is independent of the target antigen. Therefore, my PhD project aimed to identify a cell surface marker that can enrich tumor-reactive CD8+ TILs for clinical use.    For this purpose, I comprehensively characterized CD8+ TILs using different technologies. First, I screened the expression of activation/exhaustion surface markers as a proxy for tumor reactivity on TILs from freshly dissociated tumors by flow cytometry. I focused on CD39 as this population contained higher frequencies of activated/exhausted TILs than their negative counterpart. I developed a presorting workflow to sort CD39+ and CD39- TIL subsets, which were expanded to test their tumor reactivity in vitro. The co-culture of TILs and autologous tumor cells revealed the presence of a small population of cytokine-secreting, tumor-reactive CD8+ TILs in the CD39+ subset in one out of three tumor samples tested, whereas CD39- TILs showed no reactivity in any sample. These results suggest that CD39 can select for tumor-reactive T cells, but it needs to be further confirmed in a larger cohort of patients. The combined analysis of TCR repertoire and gene expression by single-cell RNA sequencing showed that dominant clonotypes in the expanded CD39+ T cell subset had in the tumor an activated, dysfunctional state and exhibited tumor-reactive T cell transcriptome signatures. Last, I studied the phenotype of TILs in close proximity to tumor cells in tissues as a potential indicator of tumor recognition. To this end, I used our newly developed multiplexed MACSima imaging cyclic staining (MICS) technology to examine the spatial distribution of not only TIL subsets but also other cell types in the tumor microenvironment. CD39 was highly expressed on non-T cell types such as endothelium and fibroblasts which overlapped with T cells, challenging the identification of bona fide CD39+ T cells. Therefore, I used other activation/exhaustion markers that correlated with CD39 expression, including CD137, PD1, and CD103, to label potential tumor-reactive CD8+ TILs. These TIL subsets tended to be in close proximity to tumor cells compared to stromal areas or tertiary lymphoid structures, supporting that T cells expressing those markers may be interacting with or recognizing tumor cells.    In summary, the data presented in my PhD thesis highlights the potential of CD39 as a marker of tumor-reactive CD8+ TILs. Future studies are needed to validate these findings in a larger cohort of patients and tumor types, enabling the development of improved manufacturing processes of tumor-reactive TILs for ACT.