Adoptive transfer of autologous patient-derived T cells genetically modified to express tumor-specific T cell receptors (TCRs) can mediate potent therapeutic effects. This is particularly true if antigens with exclusive expression in the tumor are targeted with TCRs of optimal high avidity. However, current efforts in this field often rely on targeting of so-called onco-fetal proteins overexpressed in certain tumors. In this case specificity is limited, bearing the risk of on-target toxicity in other organs and the difficulty to identify suitable high avidity TCRs due to tolerance mechanisms towards ‘self’.

Within this project, we aim at employing a new class of neo-epitopes derived from frameshift mutations in tumor suppressor genes for the treatment of pancreatic cancer. Such epitopes have high tumor specificity, can be shared between patients with the same epitope-presenting HLA allele, and the presence of high avidity TCRs should not be prevented by central tolerance. As proof of concept, we are addressing the ring finger protein RNF43 as therapeutic target, which has recently been shown by us to act as a tumor suppressor by negatively regulating the Wnt signaling pathway. RNF43 is inactivated in many gastrointestinal tumors, including pancreatic cancer, often by frameshift mutations.

We propose to generate recombinant T cells specific for neo-epitopes within the major shifted reading frames of RNF43 found in pancreatic cancer patients, and to validate their in vitro as well as in vivo potency. We have already identified clusters of HLA-A201 and HLA-B0702-restricted epitopes derived from RNF43 frame shifts by in silico prediction as well as proteasome-processing assays. For these promising epitope candidates, we have started to search for epitope-specific T cells from peripheral blood of healthy donors. With combinatorial MHC multimer-staining even extremely rare epitope-specific T cell populations can be single cell sorted and their complete TCR sequences extracted by single cell PCR (TCR-SCAN). These TCRs will be further characterized for specificity, TCR avidity (using the so-called MHC Streptamer Koff-rate assay) and efficacy by transgenically expressing them in host T cells. Especially high avidity TCRs will be subsequently tested for their reactivity towards pancreatic tumor cells displaying associated frameshift mutations in the RNF43 gene. The most efficient TCRs will be further explored for in vivo efficacy in NSG mice bearing transplanted tumors as well as in newly generated mouse models based on tissue-specific expression of mutated RNF43 in pancreatic cancers of HLA-humanized mice.

These studies will provide first proof-of-concept for targeting frameshift mutation-derived epitopes with engineered T cells for the treatment of pancreatic cancer. It is likely that efficacy of this approach can further benefit from combination with strategies overcoming tumor-induced immunosuppression. Such future studies will take advantage of the broad expertise within this CRC on most relevant milieu factors.