Abstract
BACKGROUND: Despite the success of immune checkpoint inhibitor (ICI)-based combination therapies in hepatocellular carcinoma (HCC), its effectiveness remains confined to a subset of patients. The development of reliable, predictive markers is important for accurate patient stratification and further mechanistic understanding of therapy response. METHODS: We comprehensively analyzed paired single-cell RNA transcriptome and T-cell repertoire profiles from 14 HCC ascites samples, collected from 7 patients before and after treatment with the combination of sintilimab (anti-PD-1) and bevacizumab (anti-VEGF). RESULTS: We identify a widespread convergence on interferon (IFN) signaling across various immune cell lineages in treatment-responsive patients with HCC, indicating a common transcriptional state transition in the immune microenvironment linked to immunotherapy response in HCC. Strong IFN signaling marks CD8(+) T cells with larger clonal expansion and enhanced cytotoxicity, macrophages toward M1-like polarization and strong T-cell recruitment ability, dendritic cells with increased antigen presentation capacity, as well as highly cytotoxic natural killer cells and activated B cells. By translating our finding to cohorts of patients with HCC, we demonstrate the specificity of IFN-signaling in the prognosis of patients with HCC and its ability to predict immunotherapy response. CONCLUSIONS: This study provides a unique single-cell resource with clonal and longitudinal resolution during ICI therapy and reveals IFN signaling as a biomarker of immunotherapy response in HCC, suggesting a beneficial effect by combining IFN inducers with ICIs for patients with HCC.