Background While cancer immunotherapy has mainly focused on exploiting CD8 T cells given their role in the direct elimination of tumor cells, increasing evidence highlights the crucial roles played by CD4 T cells in anti-tumor immunity. However, their very low frequency, the lack of robust algorithms to predict peptide binding to MHC class II molecules and the high polymorphism of MHC class II molecules render the study and use of circulating tumor antigen-specific CD4 T cells challenging. In this regard, the HLA-DRB3*02:02 gene encoding an HLA allele that is expressed by half of the Caucasian population, offers a way to identify CD4 T cell-defined tumor antigens with broad cancer patient coverage.

Methods Here, we aim to identify, isolate and functionally characterize ‘quasi-universal’ human tumor antigen-specific HLA-DRB3*02:02-restricted CD4 T cells in cancer patients. Using an algorithm we recently developed in house,¹ tumor-associated antigenic peptides binding to this allele are identified. We have generated a large collection of HLA-DRB3*02:02-restricted CD4 T cell clones of different tumor-antigen specificities. We will perform in vitro co-cultures of CD4 T cell clones with tumor cells to measure cytokine secretion, their tumor cell killing and their phenotypic profile (PD-1, TIM3, TIGIT, 4-1BB, CD40L, LAG3, VISTA, OX40). We will sequence and clone the TCR of the most promising candidates for adoptive cell transfer therapy. Lastly, we will directly evaluate the presence of these cells ex-vivo and longitudinally monitor them in patients.

Results N/A

Conclusions Together, these results should contribute valuable targets for coordinated CD4 and CD8 T cell-based immunotherapy of cancer.

Reference

1. Racle, J., et al., Robust prediction of HLA class II epitopes by deep motif deconvolution of immunopeptidomes. Nat Biotechnol 2019. 37(11): p. 1283–1286.