Background Tumor-associated antigens (TAAs) and especially cancer testis antigens (CTAs) are classical tumor-specific targets for immunotherapies. As TAAs are shared between patients, strategies aiming to exploit these targets are scalable and potentially applicable across different types of cancer. Loss of target antigens and other mechanisms of immune escape have limited the success of CTA-directed immunotherapy. CAR T cells and other highly effective cellular therapies have renewed the interest in TAAs. Especially combined targeting of multiple antigens appears highly promising as recently shown in lymphoma. In our study, we aimed to characterize CTA-expression patterns and their impact on endogenous T-cell responses, T-cell abundance and antigen-presentation in esophago-gastric adenocarcinoma (EGA).

Materials and Methods 41 treatment-naïve EGA patients were included in our study. RNA of tumor and patient-matched healthy tissue was isolated and used for NanoString based RNA expression analysis of 26 CTAs and 25 genes associated with antigen-presentation. Based on CTA expression, 10 peptide pools were selected and co-cultured with peripheral blood mononuclear cells (PBMCs, n=21) to determine cellular anti-tumor immune responses in a FluoroSpot assay. T-cell abundance was assessed using immunohistochemistry (CD3, CD8) and digital image analyses of tumor area and invasive margin. Autologous CD40 activated B cells were used to expand antigen-specific T cells using peptide pools of CTAs.

Results NanoString analysis revealed pronounced differences regarding CTA expression, with CEP55 and MAGEA3/6 showing strong expression, while NY-ESO-1 or MAGEA1 were only weakly expressed. 68.3% (28/41) of the patients showed expression of ≥ 5/10 analyzed TAAs simultaneously. In line with the frequent expression, 75.0% of the patients showed a cellular response against at least one of the TAAs. T-cell responses were most frequently detected to Survivin and NY-ESO-1 (65.0% and 52.6% of patients, respectively), while only 20.0% responded to CEP55 or TTK peptide pools. Overall, 6/20 patients showed cellular responses against ≥5 TAAs simultaneously. We found a strong correlation of T-cell abundance and antigen-presentation. In addition, patients with a high Immune-Score showed increased TAA expression. Finally, we demonstrate feasibility of TAA-specific T-cell expansion using CD40 activated B cells as potential strategy to induce or enhance TAA immune responses in EGA.

Conclusions Our study highlights the importance of TAAs in EGA. The identified antigens are highly relevant for immunomonitoring of clinical trials and as targets for immunotherapy. Personalized immunotherapeutic strategies targeting EGA-specific or even patient specific TAAs appear highly promising in this challenging disease.

Disclosure Information M. Thelen: None. M. Garcia-Marquez: None. J. Lehmann: None. D. Keller: None. E. Preugszat: None. M. von Bergwelt-Baildon: B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Modest; Astellas, Roche, MSD. D. Speakers Bureau/Honoraria (speakers bureau, symposia, and expert witness); Modest; BMS. F. Consultant/Advisory Board; Modest; BMS. H.A. Schlößer: B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Significant; Astra Zeneca.