Single-cell analyses reveal key NK cell subsets associated with response to PD-L1 blockade in hepatocellular carcinoma (#278)
Introduction: Natural Killer (NK) cells constitute major subsets of intrahepatic lymphocytes, yet their role in hepatocellular carcinoma (HCC) progression and response to immune checkpoint inhibitors (ICIs) remains unclear. Moreover, the phenotype and function of NK cells infiltrating HCC in patients undergoing ICIs treatment have not been investigated. Using single-cell profiling, we characterized the intra-tumoral NK cell landscape to identify features associated with treatment response or resistance.
Methods: We utilized single-cell RNA sequencing data from three external datasets (HCC n=33, HD n=12, PMBC n=14) to derive gene expression signatures defining distinct NK cell subsets. We then characterized the immune landscape of pre-treatment biopsies from HCC patients treated with anti-PD-L1 (n=10), correlating NK cell states with single-cell profiles and responsive states. The mouse NK cell studies were conducted on a model of diethylnitrosamine- induced HCC on a long-term high-fat diet regimen.
Results: Human NK cells exhibited distinct subset heterogeneity, with CD16-high NK cells predominating in HCC patient’s peripheral blood while a more balanced distribution of CD16-high and CD56-high NK cells was observed within the liver. When Infiltrating tumours, NK cells undergo phenotypic shifts marked by an increased frequency of an ILC1-like subset (ITGA1⁺ KLRC1^high). Similar alterations were recapitulated in mouse HCC, where ILC1-like NK cells expanded — potentially via IL-15+TGF-β–mediated conversion from conventional NK cells, as supported by in vitro findings. Remarkably, analysis of pre-treatment samples from anti–PD-L1–treated HCC patients revealed that the patient responding to treatment exhibited a higher proportion of CD16-high NK cells and a lower ILC1 signature score.
Conclusion: We identified a distinct NK cell landscape, with a CD16-high NK cell subtype associated with response and an ILC1 signature linked to non-responsiveness to anti-PDL1 treatment. Our findings highlight the potential predictive value of NK cell profiling in HCC to inform strategies for optimizing treatment outcome.