Abstract 2534: Single-cell dissection of heterogeneity and dynamics of tumor ecosystem upon immune-checkpoint blockade in hepatocellular carcinoma

Abstract Background: Therapy-induced alteration of tumor ecosystem poses a major barrier for cancer regression. As the majority of patients with hepatocellular carcinoma (HCC) develops resistance to immune checkpoint blockade (ICB) therapy, we aimed to characterize heterogeneity and dynamics of multicellular tumor ecosystem following anti-PD-1 treatment using scRNA-seq. Methods: In a phase II clinical trial of pembrolizumab on advanced hepatitis B virus-related HCC patients (NCT03419481), we performed a comprehensive single-cell transcriptomic profiling of biopsy samples from 26 patients before and during treatment. The composition and distribution of immune aggregates in responsive tumors was characterized by scRNA-seq and spatial transcriptome analyses. The anti-tumor efficacy of combined therapy with anti-PD-1 and inhibition of immunosuppressive macrophage were further determined in RIL-175-derived PD-1-resistant mouse model. Results: After initial data processing and quality control, we obtained 353, 329 high-quality cells that included tumor cells, myeloid cells, lymphoid cells, endothelial cells and fibroblasts. Our analysis revealed that baseline tumor-reactive CD8+T cells exhibited a strong association with improved prognosis. Notably, treatment-induced expansion and differentiation of tumor-reactive CD8+T cells with memory-like phenotype highly correlated with durable ICB response. Moreover, we observed a co-occurrence pattern of key anti-tumor immune subsets in the tumor microenvironment (TME), which associated with ligand-receptor pair interactions. Spatial transcriptome analysis further validated the presence of immune aggregates in responsive tumors. Importantly, we derived an responsive signature from immune aggregates, which showed a strong association with improved prognosis in pan-cancer ICB datasets. Paired comparisons of pre- and on-treatment macrophage percentages in non-responders revealed a specific expansion of an immunosuppressive macrophage subset. Inhibition of this macrophage subset could overcome ICB resistance by enhancing T cell function in our PD-1 resistant model. We further observed that a tumorous epithelial-mesenchymal transition (EMT) program might contribute to therapeutic resistance via TME remodeling. Conclusions: Our study suggests that tumor-reactive CD8+T cells and immune aggregates may be associated with durable ICB response, while ICB-induced upregulation of immunosuppressive macrophages and EMThi tumor cells contributed to therapeutic evasion. These findings provide valuable insights for identification of predictive biomarkers and therapeutic strategies for ICB therapy of HCC. This project is supported by the General Research Fund (14119023, 14120621 and 14115820), the Li Ka Shing Foundation, and the CUHK Strategic Seed Funding for Collaborative Research Scheme. Citation Format: Alfred S. L. Cheng, Zhewen Xiong, Haoran Wu, Jianquan Cao, Zhixian Liang, Jingying Zhou, Joseph Jao-Yiu Sung, Kevin Yip, Stephen Chan. Single-cell dissection of heterogeneity and dynamics of tumor ecosystem upon immune-checkpoint blockade in hepatocellular carcinoma abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 2534.