Abstract 6524: Single-cell landscape of cancer-associated fibroblasts in immunotherapy resistance of hepatocellular carcinoma

Abstract

Abstract Introduction: The advent of immune checkpoint blockade (ICB) therapy has revolutionized the treatment approach for hepatocellular carcinoma (HCC). However, the response rate to ICB therapy remains low (below 20%). This limited response rate can be attributed to the considerable heterogeneity observed in HCC cells and the tumor microenvironment (TME). Recent research has highlighted the pivotal role of cancer-associated fibroblasts (CAFs) in orchestrating various components within the TME of HCC. Although the significance of lipid metabolism-associated CAFs in the efficacy of ICB therapy has been demonstrated in HCC murine models [1], the functions and contributions of other CAF subtypes remain largely unknown. Therefore, identifying CAF heterogeneity could advance our comprehension of HCC TME and potentially augment ICB efficacy. Methods: We performed single-cell RNA sequencing on tumor biopsy specimens obtained from patients with advanced HBV-related HCC who had participated in a pembrolizumab (anti-PD1) phase II clinical trial (NCT03419481). Our objective was to examine the dynamic alterations in the HCC TME before and during anti-PD1 treatment (upon 2-cycle). We utilized the Seurat and Cellchat packages for bioinformatics analysis which enabled us to investigate the associations between changes in the proportions of CAF subtypes and the patients' clinical responses, as well as the potential cell-cell interactions involving CAF and other TME components. Results: We identified four CAF subtypes within the TME: cycling, inflammatory, matrix, and vascular CAFs in our anti-PD1 HCC patient cohort. Among the anti-PD1 non-responders, there was a significant increase in the proportions of cycling CAFs (cCAFs) and matrix CAFs (matCAFs) compared to the responders. Notably, matCAFs exhibited a high enrichment of Gene Ontology (GO) Terms related to the extracellular matrix (ECM), whose higher proportion correlated with poor prognosis of patients. Moreover, a distinct collagen pathway signal was found between matCAFs and two immunosuppressive cell types, namely regulatory T cells and M2 macrophages in the non-responders. Discussion and Conclusion: The presence of matCAFs within the HCC tumor may associate with the development of an immunosuppressive TME and unfavorable prognosis for patients. Targeting this specific CAF subtype holds promise as a potential strategy to overcome resistance to ICB therapy. Acknowledgments: The CUHK Strategic Seed Funding for Collaborative Research Scheme and The Li Ka Shing Foundation.