Abstract A010: Tumor-reactive T cell clonal dynamics across treatment time points drive resistance and progression to frontline chemoimmunotherapy in advanced gastric cancer

Abstract Adding aPD1 to 5-FU/platinum in advanced gastroesophageal adenocarcinomas (GEA) has yielded modest and heterogeneous results. Understanding cooperativity between these two treatment modalities will inform novel combination treatments for GEA. Towards this end, we conducted a trial (n = 47) sequentially adding pembrolizumab to 5-FU/platinum in previously untreated advanced GEA. Using serial biopsy of the primary tumor at baseline, after one cycle of 5-FU/platinum, and after the addition of pembrolizumab we profiled 66,816 tumor-resident T cells with matched single-cell RNA-seq and single-cell TCR-seq. Stratifying patients into slow- (> 6 months) and fast-progressing (< 6 months) groups based on progression-free survival allowed us to quantify clonal and transcriptional differences in T cell subpopulations across three time points. We annotated CD4 and CD8 T cell phenotypes according to canonical marker genes and found evidence of 17 different subpopulations along various stages of memory and exhaustion trajectories. We found differences in the abundance of T cell subtypes between slow- and fast-progressors at various treatment time points. T cells with exhausted phenotypes were more abundant in slow progressors in the post-chemo and post-immunotherapy time points. Additionally, we found changes in phenotypic abundances between clones shared between time points. For example, clonotypes that were predominantly of the progenitor exhausted phenotype pre-immunotherapy had a notable proportion of shared post-ICB clones exhibiting an activated phenotype. Previous studies have highlighted exhausted T cells as having important roles in maintaining tumor immunity as these tend to be subject to chronic antigen stimulation due to their specificity to tumor antigen. For this reason, we identified tumor-reactive T cell clonotypes among our data using literature-defined proxies such as CXCL13 expression, exhaustion status and degree of post-therapeutic clonal expansion, in addition to transcription signature-based predictions using machine learning tools such as predicTCR and TRTpred. We found that tumor-reactive T cell clonotypes tended to be present in slow progressors across time points. CXCL13 had variable patterns of upregulation and downregulation depending on whether a particular tumor-reactive clonotype was expanding or contracting at a specific treatment time point. Moreover, we quantified whether these tumor reactive clonotypes shared clones with non-exhausted T cell subsets prior to treatment or whether they were newly present in the tumor microenvironment (TME) post-treatment to determine the relative contributions of phenotypic transition and peripheral replenishment to tumor reactive T cells in the gastric cancer tumor microenvironment. Our data provide an invaluable resource to understand T cell behavior in the context of chemoimmunotherapy and propose strategies to drive anti-tumor immunity to expand the portion of patients benefiting from immunotherapy approaches. Citation Format: Samuel J. Wright, Minae An, Byung Hoon Min, You Jeong Heo, Milan Parikh, Lynn Bi, Hyuk Lee, Taejun Kim, Song-Yi Lee, Jeonghyeon Moon, Matthew R. Strickland, Woong Yang Park, Won Ki Kang, Kyoung-Mee Kim, Seung Tae Kim, Ryan J. Park, Arnav Mehta, Samuel Klempner, Jeeyun Lee. Tumor-reactive T cell clonal dynamics across treatment time points drive resistance and progression to frontline chemoimmunotherapy in advanced gastric cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Tumor Immunology and Immunotherapy; 2024 Oct 18-21; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2024;12(10 Suppl):Abstract nr A010.