
==== Front
Discov Oncol
Discov Oncol
Discover Oncology
2730-6011
Springer US New York

39231877
1287
10.1007/s12672-024-01287-4
Analysis
Natural killer (NK) cells-related gene signature reveals the immune environment heterogeneity in hepatocellular carcinoma based on single cell analysis
Ye Zhirong 1
Li Wenjun 2
Ouyang Hao 3
Ruan Zikang 4
Liu Xun 5
Lin Xiaoxia 380404642@qq.com

4
Chen Xuanting chenxuanting012@163.com

1
1 grid.410560.6 0000 0004 1760 3078 Department of Hepatobiliary and Pancreatic Surgery, The Second Affiliated Hospital of Guangdong Medical University, No. 12, Minyou Road, Xiashan District, Zhanjiang, 524000 Guangdong China
2 https://ror.org/00hagsh42 grid.464460.4 Department of Anesthesia, Zhongshan Hospital of Traditional Chinese Medicine Affiliated to Guangzhou University of Traditional Chinese Medicine, Zhongshan, 528400 China
3 Department of Clinical Laboratory, Dongguan Binhaiwan Central Hospital, Dongguan, 523903 Guangdong China
4 https://ror.org/05ptrtc51 grid.478001.a Department of Hepatobiliary Surgery, The People’s Hospital of Gaozhou, No. 89, Xiguan Road, Gaozhou, Maoming, 525200 Guangdong China
5 Department of Clinical Laboratory, The People’s Hospital of Xingning, Meizhou, 514500 Guangdong China
4 9 2024
4 9 2024
12 2024
15 40624 6 2024
28 8 2024
© The Author(s) 2024
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
The early diagnosis of liver cancer is crucial for the treatment and depends on the coordinated use of several test procedures. Early diagnosis is crucial for precision therapy in the treatment of the hepatocellular carcinoma (HCC). Therefore, in this study, the NK cell-related gene prediction model was used to provide the basis for precision therapy at the gene level and a novel basis for the treatment of patients with liver cancer. Natural killer (NK) cells have innate abilities to recognize and destroy tumor cells and thus play a crucial function as the “innate counterpart” of cytotoxic T cells. The natural killer (NK) cells is well recognized as a prospective approach for tumor immunotherapy in treating patients with HCC. In this research, we used publicly available databases to collect bioinformatics data of scRNA-seq and RNA-seq from HCC patients. To determine the NK cell-related genes (NKRGs)-based risk profile for HCC, we isolated T and natural killer (NK) cells and subjected them to analysis. Uniform Manifold Approximation and Projection plots were created to show the degree of expression of each marker gene and the distribution of distinct clusters. The connection between the immunotherapy response and the NKRGs-based signature was further analyzed, and the NKRGs-based signature was established. Eventually, a nomogram was developed using the model and clinical features to precisely predict the likelihood of survival. The prognosis of HCC can be accurately predicted using the NKRGs-based prognostic signature, and thorough characterization of the NKRGs signature of HCC may help to interpret the response of HCC to immunotherapy and propose a novel tumor treatment perspective.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12672-024-01287-4.

Keywords

Hepatocellular carcinoma
Precision therapy
Natural killer cells
Prognostic signature
Machine learning
Immunotherapy
issue-copyright-statement© Springer Science+Business Media, LLC 2024
==== Body
pmcIntroduction

About 70 to 80% of the liver cancers are histologically defined as hepatocellular carcinoma (HCC), a malignant disease with high morbidity and mortality rates [1]. However, up to date treatment regimens, the prognosis of patients affected by HCC remains poor [2]. Others include chronic viral hepatitis, alcoholism, metabolic disorders such as the metabolic syndrome, and obesity which has been rising in incidence rate and linked with HCC mortality [3]. These discrepancies in the prognosis of patients in the same stage of cancer imply the role of cancer heterogeneity [4]. With the development of omics technologies, the molecular mechanism of HCC has been well described in the recent years [5]. Despite the fact that several data-driven signatures have been established for the prediction of the clinical outcomes of HCC, their usage is still restricted. Therefore, there is a pressing need for the development of novel multi-gene signatures to better assess the prognosis of HCC.

Immunotherapy leverages the immune system, and has been recently heralded as an innovative approach in the battle against cancer. Unlike cytotoxic T cells that depend on the presence of human leukocyte antigen-I (HLA-I) for targeting, Natural Killer (NK) cells exhibit the ability to recognize and eliminate tumor cells independently of HLA-I, affirming their crucial role in tumor surveillance and elimination [6]. Representing a unique fraction of peripheral blood leukocytes, NK cells account for approximately 5–15% and are predominantly characterized by high expression levels of CD56, a neural cell adhesion molecule isoform, while lacking CD3 and T cell receptors. NK cells are capable of secreting cytokines and inducing cell death through various effector mechanisms, thus orchestrating immune responses against both virus-infected and cancerous cells. The distinct marker CD56 is pivotal in defining the biological attributes of NK cells and has been essential in understanding their complex roles in diseases such as cancer and inflammatory conditions. As a specialized lymphocyte group, NK cells can hinder cancer progression by swiftly detecting and destroying various abnormal cells without the need for prior sensitization or recognition of specific cancer antigens [7]. Despite the diverse strategies that tumors use to evade endogenous NK cell-mediated immunity, ex-vivo NK cell stimulation, expansion, and genetic manipulation techniques offer promising approaches to enhance their antitumor potential. These methods equip NK cells with robust cytotoxic capabilities [8]. Future NK cell therapeutics are expected to feature enhanced activation signals and proliferative capacities, reduced inhibitory controls, and improved tumor-homing abilities [9]. Research on NK cells is currently very active in the field of cancer treatment. An in-depth understanding of NK cell biology is likely to enhance our current knowledge of tumor therapy and potentially facilitate the discovery of new immunotherapeutic strategies.

Individualized therapy plans are crucial for patients with HCC. Unlike traditional RNA sequencing, single-cell RNA sequencing (scRNA-seq) offers the unique ability to detect genomic signals from individual cells. This technology allows for high-throughput genomic data acquisition from each cell, reducing the averaging of gene expression across a heterogeneous mix of cells typical in conventional RNA-seq. As a result, scRNA-seq facilitates precise characterization of individual cells, enabling the identification of distinct cellular subsets and variant expression profiles within single cells [10–12]. This precision yields more accurate results as gene expression is not obscured by the pooled cell population. The scRNA-seq method is particularly effective for analyzing the heterogeneity of tumor cells by focusing on the specific characteristics of each cellular subpopulation [13]. Studies combining bulk RNA sequencing with scRNA sequencing are becoming increasingly prevalent. These studies aim to analyze tumor heterogeneity and develop predictive models for assessing patient prognosis and treatment efficacy [14, 15]. However, only a few studies have successfully constructed robust models validated through combined bulk and single-cell experiments.

Despite extensive research on NK cells within the context of HCC, comprehensive data regarding the attributes of NK cells and NK cell-related genes (NKRGs) remain limited, this scarcity extends to their roles in prognosticating and enhancing the outcomes of immunotherapy treatments in HCC [16, 17]. This investigation utilized scRNA-seq and RNA-seq data from publicly available HCC datasets. T cells and NK cells identified within clusters 0, 3, 4, 5, 7, 19, 23, and 26 underwent analysis to establish an NKRGs-based risk profile for HCC. Thereafter, using Uniform Manifold Approximation and Projection (UMAP), we visualized the patterns of marker gene expression across the clusters and their relation. Moreover, the immunophenotypes and the response to immunotherapies according to the newly generated NKRGs signature were assessed, supporting the NKRGs signature. A nomogram that included independent clinical predictors and results from the model was used to create the nomogram that could be used to predict the survival probabilities of patients with HCC. By developing this approach it becomes possible to determine the HCC patients with high mortality risk and, therefore, the healthcare providers will be able to provide individualised and specific treatment management for the specific patient based on his/her clinical and molecular parameters. These studies therefore not only contribute greatly to the identification of the biological pathways that are involved in HCC but also provide a basis for the development of new treatments for the disease. They help in the creation of better, individualised treatment regimens for the management of this illness in patients worldwide suffering from this disease.

Materials and methods

Preparation for RNA-seq data

The scRNA-seq data for HCC were obtained from dataset GSE149614 with the samples being taken from primary HCC and adjacent non-tumoural liver tissues. The raw RNA-Seq data, and the clinical information about patients with HCC was downloaded from TCGA database and an extra dataset linked to HCC from GEO database with accession number GSE76427. For the purpose of a more detailed analysis, we chose samples which met our inclusion criteria and had a minimum of 30 days of survival. This approach helped us to gain a deeper insight into the molecular mechanisms of HCC and to analyse potential therapeutic targets, as well as to improve the identification of putative prognostic biomarkers [19]. For the current study, we obtained the scRNA-seq data from the previous published dataset GSE149614 wherein we collected ten tumor and eight non-tumor liver tissue samples. In addition, the bulk RNA-seq data of TCGA-HCC and GSE76427 with 115 and 342 HCC patients were used in this paper. The analysis of the data and the application of statistical models was performed using R program (version 4. 2. 0), which is a versatile tool for the analysis of datasets, control of data quality and identification of differentially expressed genes. Color Fig. 1 illustrates the study procedure and methodology in a schematic diagram for the purpose of improving the comprehension of the examined methods and enhancing the possibility of replicating the study [20].Fig. 1 Flow chart

Single-cell RNA-seq data processing

Version 4.1.1 of the ‘Seurat’ R package was employed to analyze scRNA-seq data. Tailored for scRNA-seq data handling, this software facilitates the integration of diverse scRNA-seq datasets and delineates distinct cellular clusters through transcriptomic profiling. Utilization of ‘Seurat’ facilitated a comprehensive analysis of complex cellular dynamics in the progression and immune modulation of HCC. This analysis enabled the identification of novel gene signatures and biological pathways, which may be therapeutically targeted to enhance clinical outcomes for HCC patients [21, 22]. Data quality in this scRNA-seq study was evaluated using a methodology aligned with established best practice guidelines in the field, as detailed in previous research. This quality control measures adopted involved removing cells that did not meet the minimum criteria for read depth, expressed genes per cell, percentages of reads assigned to mitochondrial or ribosomal RNA genes, and proportion of total transcripts attributed to the hemoglobin gene [18]. Then, procedures such as normalization, elimination of batch effects, reduction of dimensions, clustering, and annotation of cells were methodically executed [23–25]. Following this, T/NK cells were isolated and re-sorted using the previously described methods. The function “FindMarkers” was then utilized to identify genes with high variability in NK cells, with thresholds set for log-fold change (≥ 0.25), minimum percentage (≥ 0.1), and p-value (< 0.05). Genes that exhibited significant variations were designated as NKRGs and were selected for in-depth analysis.

Development and validation of the prognostic model

To ensure unbiased and comprehensive sampling of the study population in the TCGA-HCC dataset, samples were divided using a randomization method that allocated them at a ratio of 7:3 between the training set and the internal validation set, respectively. This strategy minimized selection bias and supported the robustness of the prognostic models derived from the study. The HCC samples from dataset GSE76427 were employed as an external validation set. The research commenced with a univariate Cox regression analysis to identify NKRGs that correlated significantly with patient survival outcomes (p < 0.05). Thereafter, the least absolute shrinkage and selection operator (LASSO) method was utilized to sift through potential NKRGs that predict patient survival with precision. A multivariate Cox regression analysis was then conducted to develop a prognostic model, which incorporated these factors to compute a risk score for each patient. This risk score formula (∑i=1kβiSi) was generated based on the regression coefficients of the chosen NKRGs.

To evaluate the effectiveness of the prognostic model, the study conducted Kaplan–Meier analysis and chi-squared tests to assess its capacity for distinguishing survival outcomes across various groups. A time-dependent receiver operating characteristic (tdROC) analysis further quantified the model predictive accuracy for patient survival, comparing these results with traditional clinical indicators. The model generalizability and utility in clinical settings were also tested, particularly in distinguishing survival outcomes among patients with HCC categorized by different clinical characteristics and risk scores. The role of the prognostic model as an independent predictor of patient outcomes was investigated through univariate and multivariate Cox regression analyses. Furthermore, the consistency index (C-index) was utilized to gauge the model predictive strength and efficacy relative to clinical features alone. A nomogram incorporating the model and clinical features was subsequently developed and validated, providing an easy-to-use tool for predicting survival rates at one, three, and five years for HCC patients, thereby supporting clinicians in patient management decisions.

Enrichment analysis and gene mutation analysis

To elucidate the molecular mechanisms and pathways linked to various risk score groups, a differential gene expression analysis was conducted, targeting genes with significant changes in expression (|logFC|> = 1.5) coupled with a stringent false discovery rate threshold (p < 0.05). Subsequent investigations employed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) to identify pathways and biological processes enriched among these differentially expressed genes (DEGs) [26, 27]. Next, the utilization of the “Maftools” R package facilitated the exploration and quantification of mutational profiles across the different risk score classifications. Utilizing this software facilitated the discovery of new insights into genetic alterations across various HCC subgroups, which may contribute to the molecular phenotype observed. This advancement aids in elucidating the underlying pathophysiology of HCC [28, 29].

Assessment of tumor immune microenvironment

Tumor mutation burden (TMB) scores were evaluated across different groups to predict immunotherapy outcomes [30, 31]. Calculating TMB involves several key steps. First, DNA is extracted from tumor tissue or blood samples and subjected to whole-genome sequencing (WGS) or targeted genomic sequencing (WTS) to identify all possible gene mutations. Next, bioinformatics tools and algorithms are used to analyze the sequencing data, identifying and annotating mutations, including point mutations and small insertions/deletions (indels). Then, the number of mutations per million base pairs (Mb) in the sample is calculated, typically by dividing the number of identified mutations by the total length of the analyzed genomic regions (in millions of base pairs). After that, data is standardized and calibrated to account for technical biases and sample-to-sample variations, ensuring the consistency and accuracy of the mutation burden calculation. Finally, the calculated TMB values are correlated with clinical data to assess their impact on tumor immunotherapy response and prognosis. Survival disparities among various TMB categories were analyzed [32]. We conducted comprehensive analyses using multiple algorithms to explore the correlations between risk score categories and immune cell infiltration patterns in HCC [33–39]. Through single-sample gene set enrichment analysis (ssGSEA), variations in immune response among different HCC variants were assessed. This method also facilitated the evaluation of immune-related gene sets and pathways, providing a detailed view of the molecular biology of the immune system in each tumor [18]. Researchers investigated the expression levels of immune checkpoint inhibitors (ICI).

Identification of anti-tumor medicines

In the clinical management of HCC, a comprehensive evaluation was performed on the most commonly administered chemotherapeutic agents. Utilizing the “pRRophetic” R package, the study calculated the median inhibitory concentrations (IC50) for these drugs, widely incorporated in HCC therapy [29, 40].

Statistical analysis

R software (version 4.1.1) was used to gather the required figures and conduct the study. To compare the two groups’ differences, the t-test was utilized. To determine correlation coefficients, Spearman's technique was employed. Statistical significance was set at p < 0.05.

Result

Operation for scRNA-seq data

The scRNA-seq dataset facilitated clustering, dimensional reduction, batch effect correction, normalization, quality control, and cell annotation (Supplementary Figures 1A and B). Marker genes such as NKG7, CD3E, and CD3D identified T cells and NK cells as being significantly overexpressed in clusters 0, 3, 4, 5, 7, 19, 23, and 26. Consequently, T cells and NK cells from these clusters were reanalyzed using the same procedures. UMAP plots illustrated the differential expression of each marker gene (Fig. 2A) and depicted the spatial distribution of the clusters (Fig. 2B). Additionally, cellular composition analysis categorized the cells into CD4+ T cells,NK cells, and CD8+ T cells (Fig. 2C). Using the Bioconductor package “FindMarkers,” highly variable genes in NK cells were identified with criteria set at logFC ≥ 0.25, min.pct ≥ 0.1, and an adjusted p-value < 0.05. This analysis defined a set of 256 DEGs as NKRGs, which were designated for further analysis in our study (Supplementary Table S1). These results reveal the important role of T cells and NK cells in the immune microenvironment, particularly highlighting that the genes significantly expressed in specific clusters may be closely related to tumor immune responses.Fig. 2 A Expression levels for individual marker genes, which are critical in defining various cell populations. B Visualization depicting cluster distribution, delineated by distinct gene expression markers. C Annotations of cell types such as CD4+ T cells, CD8+ T cells, and NK cells, providing insights into the distinct functional states within each subgroup. The precise characterization of these cell populations and their functional annotations enhance our understanding of the cell composition within the biological system under study, contributing to the comprehensiveness of our findings

Development and validation of the prognostic model

The primary dataset, comprising 242 HCC samples, was employed to develop a prognostic model. Validation of this model utilized an internal set with 100 HCC samples and an external set consisting of 115 HCC samples, encompassing the entire dataset. Initially, a univariate Cox regression analysis was conducted to discern potential prognostic biomarkers within NK cells, selecting 66 NKRGs with significant prognostic value indicated by p-values below 0.05. Univariate Cox regression, a pivotal statistical tool in survival analysis, facilitates the recognition of key biomarkers, essential for precision medicine and the development of targeted therapies (Fig. 3A). Subsequent analyses using LASSO identified 25 promising NKRGs (Fig. 3B), and a refined prognostic model incorporating 12 of these genes was established through multivariate Cox regression analysis (Fig. 3C).Fig. 3 A Prognostic NK-cell related genes (NKRGs; n = 66) using univariate Cox analysis. B Candidate NKRGs (n = 25) using LASSO analysis. C A model containing 12 NKRGs using multivariate Cox analysis

A newly introduced prognostic model was utilized to compute survival probabilities for patient cohorts categorized as low-risk and high-risk. Comparative analyses encompassing the training set, internal validation, full cohort, and external validation demonstrated significantly enhanced survival probabilities for the low-risk cohort as compared to the high-risk group (Fig. 4).The model demonstrated a robust predictive ability, with AUC values exceeding 0.7, outperforming all integrated clinical parameters (Fig. 5). Remarkably, classified in the low-risk score category demonstrated improved survival outcomes (Fig. 6A). The risk score was established as an independent predictor of prognosis following analyses using both univariate and multivariate Cox regression models (Fig. 6B). Then compared with traditional clinical attributes, the model demonstrated superior predictive efficacy, which was confirmed by a higher C-index (Fig. 7A). A significant correspondence was observed between the survival rates forecasted and those actually recorded (Fig. 7B). A clinical nomogram was developed to integrate these features with the prognostic model, thus enhancing the accuracy of survival predictions (Fig. 7C). The results affirm the model's exceptional stability and its broad applicability across diverse clinical scenarios. These results indicate that the newly established prognostic model demonstrates exceptional accuracy and stability in predicting the survival probabilities of HCC patients, particularly with a high consistency between the predicted and actual survival rates in the low-risk patient group. Compared to traditional clinical parameters, this model shows superior prognostic predictive ability, providing strong support for personalized treatment.Fig. 4 The patients assigned with a low-risk score showed significantly higher probabilities of survival in several sets, including the training set, internal validation set, complete set, and external validation set

Fig. 5 The receiver operating characteristic curve area for the model was > 0.7 in the training, internal validation, and entire sets

Fig. 6 A Individuals classified within the low-risk group demonstrated superior survival probabilities across all clinical subcategories. B Both univariate and multivariate Cox regression analyses confirmed the prognostic independence of the risk score

Fig. 7 A Comparative analysis using the C-index revealed that our model predicts outcomes more accurately than does the assessment based on clinical characteristics alone. B Our predicted and actual survival rates agreed strongly in the correlation plot, indicating the potential of the model as a reliable predictor of outcomes in this population. C A nomogram integrating both our model and key clinical features improved the accuracy of survival probability predictions

Enrichment and gene mutation analyses

298 DEGs were delineated across varying risk score categories (Supplementary Table S2). The investigation into these DEGs highlighted marked enrichment across numerous functional categories pertinent to oncology. Notably, enhancements were observed in areas such as cell adhesion, immune response, cytokine interaction, and key oncogenic pathways, including those involving mitogen-activated protein kinase (MAPK) and the protein kinase B (AKT)/phosphoinositide 3-kinase (PI3K) signaling axis (Fig. 8A). These findings are further elaborated in Supplementary Tables S3 and S4. Mutational profiling exposed the ten genes most subject to mutations, specifically TTN, TP53, LRP1B, ARID1A, MUC16, SYNE1, CSMD3, FLG, FAT4, and PCLO, with TTN, responsible for encoding titin, surfacing as the most frequently altered, linked to potential disruptions in sarcomere architecture and functionality (Fig. 8B). Comparative mutational analysis between the two risk score groups unveiled pronounced immune heterogeneity. The higher-risk cohort exhibited substantially greater mutational loads compared to their lower-risk counterparts. Interestingly, specific genes in the high-risk category demonstrated escalated mutation frequencies, suggesting their roles in exacerbating disease progression and elevating risks of detrimental outcomes.Fig. 8 A An analysis of pathways within the DEGs indicated marked enrichment in key biological processes, including cell adhesion, cytokine interactions, immune responses, and the mechanisms of tumorigenesis. B In our study, ten genes, including TTN, MUC16, TP53, ARID1A, LRP1B, CSMD3, SYNE1, FAT4, FLG, and PCLO, were found to be the most often mutated, as identified through a comprehensive genetic analysis. C The high-risk scores group had more pronounced TMB scores, indicating that the group would respond more favorably to immunotherapy. D and E Survival rates differed significantly across groups classified by both TMB and risk scores, suggesting that the integration of these metrics may offer improved prognostic precision

Evaluation of the immunological environment

The investigation demonstrated that the high-risk score cohort manifested substantially elevated TMB levels, suggesting enhanced responsiveness to immunotherapy (Fig. 8C). Divergent survival outcomes were observed across groups categorized by varying TMB and risk scores, supporting the prognostic utility of integrating these metrics (Fig. 8D and E). An observed correlation linked higher risk scores with increased numbers of regulatory T cells, CD4+ T cells, neutrophils, M0 and M1 macrophages, and B cells, while inverse correlations were identified for endothelial cells, M2 macrophages, and CD8+ T cells (Fig. 9A). The analysis also revealed significant variations in immune functions such as cytolytic activity, pro-inflammatory cytokine production, MHC class I expression, and type I and II interferon responses along the continuum of risk scores. Individuals in the higher-risk categories exhibited pronounced cytolytic activities, suggesting a vigorous antitumor immune response (Fig. 9B). Substantial variations were detected in the gene expression profiles linked to immune checkpoints across the spectrum of risk scores. Immune checkpoints are critical for modulating immune responses by either stimulating or inhibiting T cell activity, thereby maintaining immune tolerance and averting autoimmune diseases. The findings showed that differences in gene expression related to immune checkpoints varied significantly across risk categories, affecting the heterogeneity of the tumor immune environment (Fig. 9C). Our findings indicated that the high-risk patient group exhibits a strong antitumor immune response, particularly with high TMB levels associated with enhanced responsiveness to immunotherapy. The differences in immune checkpoint gene expression further highlight the heterogeneity of the tumor immune environment, reinforcing the value of integrating risk scores with immune characteristics for prognostic assessment.Fig. 9 A Correlation analyses revealed positive associations between the risk score and several immune components, including regulatory T cells, CD4+ T cells, neutrophils, M0 and M1 macrophages, and B cells, negative correlations were observed with endothelial cells, M2 macrophages, and CD8+ T cells. B Significant variations in immunological activities, with notable cytolytic activity, inflammation promotion, responses of type I and II interferon (IFN), as well as MHC class I responses, were observed between different risk score groups in our study. C Marked disparities in the expression of genes related to immune checkpoints were evident among the groups differentiated by risk scores. These variations were statistically underscored with significance markers:*p < 0.05, **p < 0.01, and ***p < 0.001

Selection of anti-tumor medicines

Explorations in immunotherapy have been augmented by the search for novel, tailored pharmaceuticals and chemotherapeutic compounds. In addition, a comprehensive evaluation of diverse antineoplastic substances and precision agents has been conducted, enabling the formulation of individualized treatment schemes for distinct patient cohorts (Figs. 10 and 11). These results indicated that by screening antitumor drugs and precision agents, personalized treatment plans can be developed for different patient groups, thereby enhancing the effectiveness of immunotherapy.Fig. 10 Identification of chemotherapeutics commonly used in clinical practice

Fig. 11 Identification of novel candidate compounds

Discussion

HCC is a lethal disease characterized by high incidence rates and unfavorable outcomes [2]. In the realm of tumor immunotherapy, NK cells are being explored as potential candidates due to their effectiveness. Research indicates a correlation between the compromised function of these immune cells and their altered metabolic states within liver tumors, contributing to enhanced cancer aggressiveness and metastatic capabilities [17]. The strategy of utilizing both T cells and NK cells equipped with chimeric antigen receptors (CAR-Ts) targeting tumor-specific antigens has proven beneficial in various cancers, including HCC. Innovations in engineered T cells expressing CARs that target the CD147 antigen, also referred to as Basigin, have demonstrated significant anti-tumor efficacy by attacking multiple HCC cell lines in vitro.

In vivo assessments using xenograft and patient-derived xenograft models in rodents have confirmed the effectiveness of therapies based on CAR-T cells targeting CD147 and CAR-NK cells, leading to the complete elimination of HCC tumors [41–44]. It has been established that enhanced cytokine signaling intensifies the efficacy of cetuximab-driven NK cell antibody-dependent cellular cytotoxicity [45, 46]. Research increasingly recognizes the crucial function of CD47 in health and oncology, noting that elevated CD47 expression aligns with less favorable cancer outcomes and positions its pathways as potential therapeutic targets. Blockading CD47 notably strengthens antitumor activities primarily through the stimulation of the CD103+ dendritic cell (DC)-NK cell pathway [47]. Oxygen-deprived environments have shown that CD47 inhibition fosters the engulfment of tumor DNA by CD103+ DCs [48]. The engagement of CD103+ DCs is pivotal for NK cell mobilization, which is integral to the immune assault on tumors. This mechanism triggers the secretion of CXCL9 and interleukin (IL)-12 from these activated DCs, leading to increased interferon production, augmented granzyme B output, and elevated TNF production along with enhanced NKG2D activation in NK cells. Concurrently, a marked reduction in the inhibitory receptor NKG2A alleviates suppression on NK cells, thus heightening their cytotoxic action against cancer cells [49, 50]. Elevated NKG2A levels correlate with NK cell fatigue and predict negative outcomes in liver cancer patients [51]. Further research highlights that CD160 levels are inversely related to NK-cell function and are linked to unfavorable outcomes in HCC cases [52]. It has been discovered that microRNA-146a (miR-146a) significantly influences NK cell performance by downregulating signal transducer and activator of transcription 1 (STAT1), vital for controlling immune response genes [53]. Identification of the CD3 epsilon subunit (CD3E) by Yinghui Hou and collaborators has been proposed as a potential diagnostic and therapeutic marker in HCC [54]. These developments enhance the understanding of HCC pathogenesis and introduce novel therapeutic targets. Nevertheless, the specific extent of metabolic dysfunction experienced by NK cells in HCC has yet to be defined.

In recent investigations, a subgroup characterized by elevated risk scores exhibited a significantly higher prevalence of genetic mutations. Studies have pinpointed particular genes with mutations that lead to compromised immune responses and diminished survival rates among patients. Research conducted by Wang and colleagues has shown that the TTN gene, often mutated in liver cancer, plays a vital role in promoting immune escape and tumor progression, as these mutations are linked with reduced immune cell penetration and poorer clinical outcomes [55]. Research by Kim et al. has revealed that alterations in several genes, including TP53, OBSCN, TTN, MUC5B, and CSMD1, are notably linked with increased mortality rates in HCC among Asian populations [56]. Research led by Bing Liu and associates has demonstrated a significant association between mutations in the MUC16 gene and increased TMB levels in HCC, suggesting these mutations could reliably forecast the development of the disease. Extensive statistical and bioinformatics studies indicate that these genetic alterations not only correlate with higher TMB but also may function as an independent prognostic marker for assessing HCC risk. These mutations are implicated in the modulation of several oncogenic pathways, including MAPK, Wnt, and PI3K/Akt, which facilitate tumor growth, metastasis, and chemoresistance [57]. Gene Set Enrichment Analysis (GSEA) has revealed a notable effect of mutations in MUC16 on the metabolic activities within HCC cells [58, 59]. Research conducted by Li and colleagues indicates that the most commonly altered genes in these cells include CTNNB1, TP53, TTN, MUC16, and ALB [60]. There is a well-established link between gene mutations and poor survival outcomes, with recent data indicating that patients with a wild-type genotype have higher overall survival rates. Patients within the highest decile of TMB have been shown to have a poorer prognosis compared to those in the lower 90%, highlighting the prognostic importance of TMB in clinical settings [61, 62]. These findings underscore the necessity for further research into the regulatory mechanisms of these risk-associated genes in the context of HCC therapy heterogeneity.

NK cells offer significant promise in cancer therapeutics. The inhibitory influence of the tumor microenvironment (TME) remains the primary barrier to their efficacy [6, 63]. In recent years, therapies targeting the TME have diversified the available cancer treatments. However, due to incomplete knowledge of the cellular composition and immune landscape of HCC, these strategies have yet to be fully integrated into clinical practice. The cohort with higher risk scores exhibited significantly increased ssGSEA scores compared to their low-risk counterparts, underscoring the association between heightened risk scores and ssGSEA. This indicates a likelihood of improved immunotherapy outcomes and more accurate prognostic evaluations for patients in the high-risk category. Further investigation into the immune landscape showed an inverse correlation between the risk score and the presence of endothelial cells, M2 macrophages, and CD8+ T cells. Conversely, B cells, neutrophils, M0 macrophages, M1 macrophages, regulatory T cells, and CD4+ T cells displayed positive correlations with the risk score. High-risk tumors may exhibit a more complex immune environment, characterized by enhanced inflammation and immune cell diversity. These immune landscape features suggest that high-risk tumors might employ different mechanisms to evade immune surveillance. Investigating the roles of these immune cells and their relationship with tumor prognosis can provide valuable insights for developing more effective treatment strategies. Activation of the Inducible T cell COStimulator ligand (ICOSL) in ICOS-positive group 2 innate lymphoid cells (ICOS + ILC2a) enhances inflammation through increased synthesis of IL-13 [64]. Blocking ICOS and heat shock protein 70 (HSP70), downstream effectors of ICOS, in ILC2s, significantly reduces tumor growth rates, potentially due to decreased tumor-promoting factors associated with ILC2s. Moreover, suppressing ICOS and HSP70 induces shifts in the TME, favoring anti-tumor immune responses and reducing immunosuppressive mechanisms. These observations indicate that targeting ICOS and HSP70 in ILC2s may enhance anti-tumor immunity and modify the tumor-promoting environment [65]. Garnelo et al. identified a functional link between T and B cells within tumors, correlating with increased local immune activation and improved outcomes in HCC [64]. Evidence indicates that a precursor to carcinogenesis involves atypical macrophage infiltration within the immune system. Furthermore, the assessment of a risk score alongside M0 macrophage profiles correlates strongly with RAC1 activity, providing a reliable metric for forecasting the prognosis across various cancer stages [66].

Through the use of scRNA-seq, investigations were conducted by various scientists on how APOC1 inhibition enhances anti-PD1 immunotherapy efficacy in HCC and promotes macrophage transformation through the ferroptosis pathway [67]. In the realm of immunological responses including cytolytic and pro-inflammatory activities, marked disparities were observed among various risk score categories [68, 69]. Marked variations were detected in gene expression profiles linked to immune checkpoints within distinct risk score classifications. A variety of tumor cells and cells associated with tumors secrete factors including transforming growth factor-beta, interleukin-6, prostaglandin E2, interleukin-10, and indoleamine 2,3-dioxygenase. These factors are known to inhibit NK cell activity in the TME, acting either directly or indirectly [9, 70]. By utilizing these cytokines and other molecules, tumors are capable of hindering the activation of receptors that stimulate NK cells [71]. Inhibitory signals from receptors on the surfaces of several solid tumors can activate NK cells [72–74]. The primary consequence of these interactions is an alteration in the equilibrium between activation and inhibition signals in NK cells, which is crucial for their activation. The influence of the TME on NK cell metabolism must also be considered, as it plays a pivotal role in their functionality as effectors.

HCC emerges within an environment characterized by immune suppression, rendering it immunogenic [4]. Both stimulatory and inhibitory checkpoint molecules, discovered in recent research, play crucial roles within the immune system. Notably, inhibitory checkpoints curtail immune responses against tumors in various solid malignancies [16, 75]. The optimal therapeutic approach for localized liver malignancy involves an integrated strategy utilizing ICIs, and treatments combining chemoembolization, radiofrequency ablation, or radioembolization with ICIs are subjects of ongoing research [75–78]. Research from Phase I to Phase III has evaluated the efficacy of monoclonal antibodies directed at immune checkpoint proteins, including those that inhibit CTLA-4 and PD-1. The findings indicate a significant improvement in survival rates and patient responses, leading to FDA approval of several therapies utilizing these antibodies for various cancers, such as melanoma, renal cell carcinoma, and non-small-cell lung cancer. These approvals have catalyzed further investigations into innovative and potent ICIs for oncology applications [75, 79]. The Phase III IMbrave clinical study confirmed that the combination of atezolizumab and bevacizumab is superior as a frontline treatment for advanced HCC compared to the previously favored sorafenib, significantly boosting both overall and progression-free survival rates for those treated. The benefits of this combination therapy were consistent across different patient subgroups, and its safety profile was both manageable and generally well-tolerated. Thus, the integration of ICIs with anti-angiogenic agents represents an innovative and effective strategy for managing advanced HCC, potentially shaping future therapeutic developments [80–82].

While this study offers valuable insights, it is crucial to consider the limitations that may restrict the broader applicability of these findings. The development of NKRGs and the associated risk signature derived from retrospective datasets in publicly accessible databases underscores the need for validation through prospective cohort studies. Furthermore, this research primarily investigates the predictive capabilities of the NKRGs-based risk signature without examining its impact on HCC progression, which necessitates additional investigative efforts. Finally, a key limitation of this study is that while single-cell RNA sequencing provides insights into the expression of NK cell-related genes, it does not directly measure the abundance of NK cells in the HCC tumor microenvironment.

Conclusion

A detailed investigation into NKRGs reveals their profound impact on the diversity of the TME, as well as on the clinicopathological traits and diverse prognostic mechanisms in patients with HCC. There is a pressing need for additional research to clarify how immunotherapy and prevalent anticancer agents such as Axitinib, Docetaxel, Bexarotene, and AltSchool contribute to treatment efficacy. These findings highlight the importance of NKRGs in clinical settings and suggest innovative strategies for merging immunotherapy with conventional anticancer treatments in managing HCC.

Supplementary Information

Supplementary Material 1. Table S1. 256 highly variable genes. Table S2. List of 298 Genes Exhibiting Differential Expression. Table S3. Analysis of Gene Ontology. Table S4. Examination of Genes and Genomes through the Kyoto Encyclopedia. Table S5. The siRNA sequences. Table S6. Primers and their sequences.

Supplementary Material 2. Figure S1. (A and B) Methodological approaches employed in the analysis of scRNA-seq data, comprising batch effect correction, quality control, normalization, clustering, dimensionality reduction, and cell type identification.

Acknowledgements

We gratefully thank all the members who helped with data collection and analysis.

Author contributions

Zhirong Ye and Xuanting Chen conceived the idea. Wenjun Li and Hao Ouyang designed the study. Zikang Ruan, Xun Liu Xiaoxia Lin collected the data. Zhirong Ye and Xiaoxia Lin drafted the manuscript. Xuanting Chen reviewed and corrected the manuscript. All authors contributed to the article and approved the submitted version.

Data availability

The datasets could be downloaded from the TCGA (https://portal.gdc.com) and GEO (http://www.ncbi.nlm.nih.gov/geo/) website.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Zhirong Ye, Wenjun Li and Hao Ouyang contributed equally to this article.
==== Refs
References

1. Forner A Reig M Bruix J Hepatocellular carcinoma Lancet 2018 391 10127 1301 1314 10.1016/S0140-6736(18)30010-2 29307467
Forner A, Reig M, Bruix J. Hepatocellular carcinoma. Lancet. 2018;391(10127):1301–14.29307467 10.1016/S0140-6736(18)30010-2
2. Ganesan P Kulik LM Hepatocellular carcinoma: new developments Clin Liver Dis 2023 27 1 85 102 10.1016/j.cld.2022.08.004 36400469
Ganesan P, Kulik LM. Hepatocellular carcinoma: new developments. Clin Liver Dis. 2023;27(1):85–102.36400469 10.1016/j.cld.2022.08.004
3. Hartke J Johnson M Ghabril M The diagnosis and treatment of hepatocellular carcinoma Semin Diagn Pathol 2017 34 2 153 159 10.1053/j.semdp.2016.12.011 28108047
Hartke J, Johnson M, Ghabril M. The diagnosis and treatment of hepatocellular carcinoma. Semin Diagn Pathol. 2017;34(2):153–9.28108047 10.1053/j.semdp.2016.12.011
4. Llovet JM, Kelley RK, Villanueva A, Singal AG, Pikarsky E, Roayaie S, et al. Hepatocellular carcinoma. Nat Rev Dis Primers. 2021;7(1):6.
5. Parikh ND Pillai A Recent advances in hepatocellular carcinoma treatment Clin Gastroenterol Hepatol 2021 19 10 2020 2024 10.1016/j.cgh.2021.05.045 34116048
Parikh ND, Pillai A. Recent advances in hepatocellular carcinoma treatment. Clin Gastroenterol Hepatol. 2021;19(10):2020–4.34116048 10.1016/j.cgh.2021.05.045
6. Sun C Natural killer cell dysfunction in hepatocellular carcinoma and NK cell-based immunotherapy Acta Pharmacol Sin 2015 36 10 1191 1199 10.1038/aps.2015.41 26073325
Sun C, et al. Natural killer cell dysfunction in hepatocellular carcinoma and NK cell-based immunotherapy. Acta Pharmacol Sin. 2015;36(10):1191–9.26073325 10.1038/aps.2015.41
7. Sajid M Liu L Sun C The dynamic role of natural killer (NK) cells in liver cancers: role in HCC and HBV associated HCC and its therapeutic implications Front Immunol 2022 13 887186 10.3389/fimmu.2022.887186 35669776
Sajid M, Liu L, Sun C. The dynamic role of natural killer (NK) cells in liver cancers: role in HCC and HBV associated HCC and its therapeutic implications. Front Immunol. 2022;13: 887186.35669776 10.3389/fimmu.2022.887186
8. Myers JA Miller JS Exploring the NK cell platform for cancer immunotherapy Nat Rev Clin Oncol 2021 18 2 85 100 10.1038/s41571-020-0426-7 32934330
Myers JA, Miller JS. Exploring the NK cell platform for cancer immunotherapy. Nat Rev Clin Oncol. 2021;18(2):85–100.32934330 10.1038/s41571-020-0426-7
9. Bald T The NK cell-cancer cycle: advances and new challenges in NK cell-based immunotherapies Nat Immunol 2020 21 8 835 847 10.1038/s41590-020-0728-z 32690952
Bald T, et al. The NK cell-cancer cycle: advances and new challenges in NK cell-based immunotherapies. Nat Immunol. 2020;21(8):835–47.32690952 10.1038/s41590-020-0728-z
10. Yip SH Sham PC Wang J Evaluation of tools for highly variable gene discovery from single-cell RNA-seq data Brief Bioinform 2019 20 4 1583 1589 10.1093/bib/bby011 29481632
Yip SH, Sham PC, Wang J. Evaluation of tools for highly variable gene discovery from single-cell RNA-seq data. Brief Bioinform. 2019;20(4):1583–9.29481632 10.1093/bib/bby011
11. Bridges K Miller-Jensen K Mapping and validation of scRNA-Seq-derived cell-cell communication networks in the tumor microenvironment Front Immunol 2022 13 885267 10.3389/fimmu.2022.885267 35572582
Bridges K, Miller-Jensen K. Mapping and validation of scRNA-Seq-derived cell-cell communication networks in the tumor microenvironment. Front Immunol. 2022;13: 885267.35572582 10.3389/fimmu.2022.885267
12. Chen G Ning B Shi T Single-cell RNA-Seq technologies and related computational data analysis Front Genet 2019 10 317 10.3389/fgene.2019.00317 31024627
Chen G, Ning B, Shi T. Single-cell RNA-Seq technologies and related computational data analysis. Front Genet. 2019;10:317.31024627 10.3389/fgene.2019.00317
13. Papalexi E Satija R Single-cell RNA sequencing to explore immune cell heterogeneity Nat Rev Immunol 2018 18 1 35 45 10.1038/nri.2017.76 28787399
Papalexi E, Satija R. Single-cell RNA sequencing to explore immune cell heterogeneity. Nat Rev Immunol. 2018;18(1):35–45.28787399 10.1038/nri.2017.76
14. Kuksin M Applications of single-cell and bulk RNA sequencing in onco-immunology Eur J Cancer 2021 149 193 210 10.1016/j.ejca.2021.03.005 33866228
Kuksin M, et al. Applications of single-cell and bulk RNA sequencing in onco-immunology. Eur J Cancer. 2021;149:193–210.33866228 10.1016/j.ejca.2021.03.005
15. Li X Wang CY From bulk, single-cell to spatial RNA sequencing Int J Oral Sci 2021 13 1 36 10.1038/s41368-021-00146-0 34782601
Li X, Wang CY. From bulk, single-cell to spatial RNA sequencing. Int J Oral Sci. 2021;13(1):36.34782601 10.1038/s41368-021-00146-0
16. Mizukoshi E Kaneko S Immune cell therapy for hepatocellular carcinoma J Hematol Oncol 2019 12 1 52 10.1186/s13045-019-0742-5 31142330
Mizukoshi E, Kaneko S. Immune cell therapy for hepatocellular carcinoma. J Hematol Oncol. 2019;12(1):52.31142330 10.1186/s13045-019-0742-5
17. Oura K Tumor immune microenvironment and immunosuppressive therapy in hepatocellular carcinoma: a review Int J Mol Sci 2021 22 11 5801 10.3390/ijms22115801 34071550
Oura K, et al. Tumor immune microenvironment and immunosuppressive therapy in hepatocellular carcinoma: a review. Int J Mol Sci. 2021;22(11):5801.34071550 10.3390/ijms22115801
18. Lu Y A single-cell atlas of the multicellular ecosystem of primary and metastatic hepatocellular carcinoma Nat Commun 2022 13 1 4594 10.1038/s41467-022-32283-3 35933472
Lu Y, et al. A single-cell atlas of the multicellular ecosystem of primary and metastatic hepatocellular carcinoma. Nat Commun. 2022;13(1):4594.35933472 10.1038/s41467-022-32283-3
19. Grinchuk OV Tumor-adjacent tissue co-expression profile analysis reveals pro-oncogenic ribosomal gene signature for prognosis of resectable hepatocellular carcinoma Mol Oncol 2018 12 1 89 113 10.1002/1878-0261.12153 29117471
Grinchuk OV, et al. Tumor-adjacent tissue co-expression profile analysis reveals pro-oncogenic ribosomal gene signature for prognosis of resectable hepatocellular carcinoma. Mol Oncol. 2018;12(1):89–113.29117471 10.1002/1878-0261.12153
20. Ritchie ME limma powers differential expression analyses for RNA-sequencing and microarray studies Nucleic Acids Res 2015 43 7 e47 10.1093/nar/gkv007 25605792
Ritchie ME, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7): e47.25605792 10.1093/nar/gkv007
21. Hao Y Integrated analysis of multimodal single-cell data Cell 2021 184 13 3573 3587.e29 10.1016/j.cell.2021.04.048 34062119
Hao Y, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184(13):3573-3587.e29.34062119 10.1016/j.cell.2021.04.048
22. Stuart T Comprehensive integration of single-cell data Cell 2019 177 7 1888 1902.e21 10.1016/j.cell.2019.05.031 31178118
Stuart T, et al. Comprehensive integration of single-cell data. Cell. 2019;177(7):1888-1902.e21.31178118 10.1016/j.cell.2019.05.031
23. Butler A Integrating single-cell transcriptomic data across different conditions, technologies, and species Nat Biotechnol 2018 36 5 411 420 10.1038/nbt.4096 29608179
Butler A, et al. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol. 2018;36(5):411–20.29608179 10.1038/nbt.4096
24. Hafemeister C Satija R Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression Genome Biol 2019 20 1 296 10.1186/s13059-019-1874-1 31870423
Hafemeister C, Satija R. Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression. Genome Biol. 2019;20(1):296.31870423 10.1186/s13059-019-1874-1
25. Korsunsky I Fast, sensitive and accurate integration of single-cell data with Harmony Nat Methods 2019 16 12 1289 1296 10.1038/s41592-019-0619-0 31740819
Korsunsky I, et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods. 2019;16(12):1289–96.31740819 10.1038/s41592-019-0619-0
26. Wu T clusterProfiler 4.0: a universal enrichment tool for interpreting omics data Innovation (Camb) 2021 2 3 100141 34557778
Wu T, et al. clusterProfiler 4.0: a universal enrichment tool for interpreting omics data. Innovation (Camb). 2021;2(3): 100141.34557778
27. Wang T Comprehensive analysis of nine m7G-related lncRNAs as prognosis factors in tumor immune microenvironment of hepatocellular carcinoma and experimental validation Front Genet 2022 13 929035 10.3389/fgene.2022.929035 36081998
Wang T, et al. Comprehensive analysis of nine m7G-related lncRNAs as prognosis factors in tumor immune microenvironment of hepatocellular carcinoma and experimental validation. Front Genet. 2022;13: 929035.36081998 10.3389/fgene.2022.929035
28. Mayakonda A Maftools: efficient and comprehensive analysis of somatic variants in cancer Genome Res 2018 28 11 1747 1756 10.1101/gr.239244.118 30341162
Mayakonda A, et al. Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res. 2018;28(11):1747–56.30341162 10.1101/gr.239244.118
29. Yin J Identification of molecular classification and gene signature for predicting prognosis and immunotherapy response in HNSCC using cell differentiation trajectories Sci Rep 2022 12 1 20404 10.1038/s41598-022-24533-7 36437265
Yin J, et al. Identification of molecular classification and gene signature for predicting prognosis and immunotherapy response in HNSCC using cell differentiation trajectories. Sci Rep. 2022;12(1):20404.36437265 10.1038/s41598-022-24533-7
30. Sha D Tumor mutational burden as a predictive biomarker in solid tumors Cancer Discov 2020 10 12 1808 1825 10.1158/2159-8290.CD-20-0522 33139244
Sha D, et al. Tumor mutational burden as a predictive biomarker in solid tumors. Cancer Discov. 2020;10(12):1808–25.33139244 10.1158/2159-8290.CD-20-0522
31. Hu B Analysis of immune subtypes based on immunogenomic profiling identifies prognostic signature for cutaneous melanoma Int Immunopharmacol 2020 89 Pt A 107162 10.1016/j.intimp.2020.107162 33168410
Hu B, et al. Analysis of immune subtypes based on immunogenomic profiling identifies prognostic signature for cutaneous melanoma. Int Immunopharmacol. 2020;89(Pt A): 107162.33168410 10.1016/j.intimp.2020.107162
32. Wu C Identification of ferroptosis-related lncRNA pairs for predicting the prognosis of head and neck squamous cell carcinoma J Oncol 2022 2022 7602482 10.1155/2022/1434565 35909900
Wu C, et al. Identification of ferroptosis-related lncRNA pairs for predicting the prognosis of head and neck squamous cell carcinoma. J Oncol. 2022;2022:7602482.35909900 10.1155/2022/1434565
33. Aran D Hu Z Butte AJ xCell: digitally portraying the tissue cellular heterogeneity landscape Genome Biol 2017 18 1 220 10.1186/s13059-017-1349-1 29141660
Aran D, Hu Z, Butte AJ. xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol. 2017;18(1):220.29141660 10.1186/s13059-017-1349-1
34. Finotello F Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data Genome Med 2019 11 1 34 10.1186/s13073-019-0638-6 31126321
Finotello F, et al. Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data. Genome Med. 2019;11(1):34.31126321 10.1186/s13073-019-0638-6
35. Dienstmann R Relative contribution of clinicopathological variables, genomic markers, transcriptomic subtyping and microenvironment features for outcome prediction in stage II/III colorectal cancer Ann Oncol 2019 30 10 1622 1629 10.1093/annonc/mdz287 31504112
Dienstmann R, et al. Relative contribution of clinicopathological variables, genomic markers, transcriptomic subtyping and microenvironment features for outcome prediction in stage II/III colorectal cancer. Ann Oncol. 2019;30(10):1622–9.31504112 10.1093/annonc/mdz287
36. Newman AM Robust enumeration of cell subsets from tissue expression profiles Nat Methods 2015 12 5 453 457 10.1038/nmeth.3337 25822800
Newman AM, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453–7.25822800 10.1038/nmeth.3337
37. Tamminga M Immune microenvironment composition in non-small cell lung cancer and its association with survival Clin Transl Immunol 2020 9 6 e1142 10.1002/cti2.1142
Tamminga M, et al. Immune microenvironment composition in non-small cell lung cancer and its association with survival. Clin Transl Immunol. 2020;9(6): e1142.10.1002/cti2.1142
38. Racle J Simultaneous enumeration of cancer and immune cell types from bulk tumor gene expression data Elife 2017 6 e26476 10.7554/eLife.26476 29130882
Racle J, et al. Simultaneous enumeration of cancer and immune cell types from bulk tumor gene expression data. Elife. 2017;6: e26476.29130882 10.7554/eLife.26476
39. Li T TIMER: a web server for comprehensive analysis of tumor-infiltrating immune cells Cancer Res 2017 77 21 e108 e110 10.1158/0008-5472.CAN-17-0307 29092952
Li T, et al. TIMER: a web server for comprehensive analysis of tumor-infiltrating immune cells. Cancer Res. 2017;77(21):e108–10.29092952 10.1158/0008-5472.CAN-17-0307
40. Geeleher P Cox N Huang RS pRRophetic: an R package for prediction of clinical chemotherapeutic response from tumor gene expression levels PLoS ONE 2014 9 9 e107468 10.1371/journal.pone.0107468 25229481
Geeleher P, Cox N, Huang RS. pRRophetic: an R package for prediction of clinical chemotherapeutic response from tumor gene expression levels. PLoS ONE. 2014;9(9): e107468.25229481 10.1371/journal.pone.0107468
41. Wang SJ CD147 promotes collective invasion through cathepsin B in hepatocellular carcinoma J Exp Clin Cancer Res 2020 39 1 145 10.1186/s13046-020-01647-2 32727598
Wang SJ, et al. CD147 promotes collective invasion through cathepsin B in hepatocellular carcinoma. J Exp Clin Cancer Res. 2020;39(1):145.32727598 10.1186/s13046-020-01647-2
42. Lu M Cell expression patterns of CD147 in N-diethylnitrosamine/phenobarbital-induced mouse hepatocellular carcinoma J Mol Histol 2015 46 1 79 91 10.1007/s10735-014-9602-3 25447507
Lu M, et al. Cell expression patterns of CD147 in N-diethylnitrosamine/phenobarbital-induced mouse hepatocellular carcinoma. J Mol Histol. 2015;46(1):79–91.25447507 10.1007/s10735-014-9602-3
43. Li X Enhanced glucose metabolism mediated by CD147 contributes to immunosuppression in hepatocellular carcinoma Cancer Immunol Immunother 2020 69 4 535 548 10.1007/s00262-019-02457-y 31965268
Li X, et al. Enhanced glucose metabolism mediated by CD147 contributes to immunosuppression in hepatocellular carcinoma. Cancer Immunol Immunother. 2020;69(4):535–48.31965268 10.1007/s00262-019-02457-y
44. Zhang C Synthetic biology in chimeric antigen receptor T (CAR T) cell engineering ACS Synth Biol 2022 11 1 1 15 10.1021/acssynbio.1c00256 35005887
Zhang C, et al. Synthetic biology in chimeric antigen receptor T (CAR T) cell engineering. ACS Synth Biol. 2022;11(1):1–15.35005887 10.1021/acssynbio.1c00256
45. Decaup E A tridimensional model for NK cell-mediated ADCC of follicular lymphoma Front Immunol 2019 10 1943 10.3389/fimmu.2019.01943 31475004
Decaup E, et al. A tridimensional model for NK cell-mediated ADCC of follicular lymphoma. Front Immunol. 2019;10:1943.31475004 10.3389/fimmu.2019.01943
46. Muntasell A Targeting NK-cell checkpoints for cancer immunotherapy Curr Opin Immunol 2017 45 73 81 10.1016/j.coi.2017.01.003 28236750
Muntasell A, et al. Targeting NK-cell checkpoints for cancer immunotherapy. Curr Opin Immunol. 2017;45:73–81.28236750 10.1016/j.coi.2017.01.003
47. Wang S Blocking CD47 promotes antitumour immunity through CD103(+) dendritic cell-NK cell axis in murine hepatocellular carcinoma model J Hepatol 2022 77 2 467 478 10.1016/j.jhep.2022.03.011 35367532
Wang S, et al. Blocking CD47 promotes antitumour immunity through CD103(+) dendritic cell-NK cell axis in murine hepatocellular carcinoma model. J Hepatol. 2022;77(2):467–78.35367532 10.1016/j.jhep.2022.03.011
48. Hayat S CD47: role in the immune system and application to cancer therapy Cell Oncol (Dordr) 2020 43 1 19 30 10.1007/s13402-019-00469-5 31485984
Hayat S, et al. CD47: role in the immune system and application to cancer therapy. Cell Oncol (Dordr). 2020;43(1):19–30.31485984 10.1007/s13402-019-00469-5
49. Sun B Eradication of hepatocellular carcinoma by NKG2D-based CAR-T cells Cancer Immunol Res 2019 7 11 1813 1823 10.1158/2326-6066.CIR-19-0026 31484657
Sun B, et al. Eradication of hepatocellular carcinoma by NKG2D-based CAR-T cells. Cancer Immunol Res. 2019;7(11):1813–23.31484657 10.1158/2326-6066.CIR-19-0026
50. Wang J Li CD Sun L Recent advances in molecular mechanisms of the NKG2D pathway in hepatocellular carcinoma Biomolecules 2020 10 2 301 10.3390/biom10020301 32075046
Wang J, Li CD, Sun L. Recent advances in molecular mechanisms of the NKG2D pathway in hepatocellular carcinoma. Biomolecules. 2020;10(2):301.32075046 10.3390/biom10020301
51. Sun C High NKG2A expression contributes to NK cell exhaustion and predicts a poor prognosis of patients with liver cancer Oncoimmunology 2017 6 1 e1264562 10.1080/2162402X.2016.1264562 28197391
Sun C, et al. High NKG2A expression contributes to NK cell exhaustion and predicts a poor prognosis of patients with liver cancer. Oncoimmunology. 2017;6(1): e1264562.28197391 10.1080/2162402X.2016.1264562
52. Sun H Reduced CD160 expression contributes to impaired NK-cell function and poor clinical outcomes in patients with HCC Cancer Res 2018 78 23 6581 6593 10.1158/0008-5472.CAN-18-1049 30232222
Sun H, et al. Reduced CD160 expression contributes to impaired NK-cell function and poor clinical outcomes in patients with HCC. Cancer Res. 2018;78(23):6581–93.30232222 10.1158/0008-5472.CAN-18-1049
53. Xu D miR-146a negatively regulates NK cell functions via STAT1 signaling Cell Mol Immunol 2017 14 8 712 720 10.1038/cmi.2015.113 26996068
Xu D, et al. miR-146a negatively regulates NK cell functions via STAT1 signaling. Cell Mol Immunol. 2017;14(8):712–20.26996068 10.1038/cmi.2015.113
54. Hou Y Zhang G Identification of immune-infiltrating cell-related biomarkers in hepatocellular carcinoma based on gene co-expression network analysis Diagn Pathol 2021 16 1 57 10.1186/s13000-021-01118-y 34218795
Hou Y, Zhang G. Identification of immune-infiltrating cell-related biomarkers in hepatocellular carcinoma based on gene co-expression network analysis. Diagn Pathol. 2021;16(1):57.34218795 10.1186/s13000-021-01118-y
55. Wang R Construction of liver hepatocellular carcinoma-specific lncRNA-miRNA-mRNA network based on bioinformatics analysis PLoS ONE 2021 16 4 e0249881 10.1371/journal.pone.0249881 33861762
Wang R, et al. Construction of liver hepatocellular carcinoma-specific lncRNA-miRNA-mRNA network based on bioinformatics analysis. PLoS ONE. 2021;16(4): e0249881.33861762 10.1371/journal.pone.0249881
56. Kim T Issa D Onyshchenko M Analyzing TCGA data to identify gene mutations linked to hepatocellular carcinoma in Asians Gastrointest Tumors 2022 9 2–4 43 58 10.1159/000524576 36590851
Kim T, Issa D, Onyshchenko M. Analyzing TCGA data to identify gene mutations linked to hepatocellular carcinoma in Asians. Gastrointest Tumors. 2022;9(2–4):43–58.36590851 10.1159/000524576
57. Liu B Prognostic value of MUC16 mutation and its correlation with immunity in hepatocellular carcinoma patients Evid Based Complement Alternat Med 2022 2022 3478861 36034941
Liu B, et al. Prognostic value of MUC16 mutation and its correlation with immunity in hepatocellular carcinoma patients. Evid Based Complement Alternat Med. 2022;2022:3478861.36034941
58. Felder M MUC16 (CA125): tumor biomarker to cancer therapy, a work in progress Mol Cancer 2014 13 129 10.1186/1476-4598-13-129 24886523
Felder M, et al. MUC16 (CA125): tumor biomarker to cancer therapy, a work in progress. Mol Cancer. 2014;13:129.24886523 10.1186/1476-4598-13-129
59. Aithal A MUC16 as a novel target for cancer therapy Expert Opin Ther Targets 2018 22 8 675 686 10.1080/14728222.2018.1498845 29999426
Aithal A, et al. MUC16 as a novel target for cancer therapy. Expert Opin Ther Targets. 2018;22(8):675–86.29999426 10.1080/14728222.2018.1498845
60. Li Z Mutational and transcriptional alterations and clinicopathological factors predict the prognosis of stage I hepatocellular carcinoma: prediction of stage I HCC prognosis BMC Gastroenterol 2022 22 1 427 10.1186/s12876-022-02496-3 36153509
Li Z, et al. Mutational and transcriptional alterations and clinicopathological factors predict the prognosis of stage I hepatocellular carcinoma: prediction of stage I HCC prognosis. BMC Gastroenterol. 2022;22(1):427.36153509 10.1186/s12876-022-02496-3
61. Tang B Diagnosis and prognosis models for hepatocellular carcinoma patient's management based on tumor mutation burden J Adv Res 2021 33 153 165 10.1016/j.jare.2021.01.018 34603786
Tang B, et al. Diagnosis and prognosis models for hepatocellular carcinoma patient’s management based on tumor mutation burden. J Adv Res. 2021;33:153–65.34603786 10.1016/j.jare.2021.01.018
62. Xu Q Prognostic role of ceRNA network in immune infiltration of hepatocellular carcinoma Front Genet 2021 12 739975 10.3389/fgene.2021.739975 34589117
Xu Q, et al. Prognostic role of ceRNA network in immune infiltration of hepatocellular carcinoma. Front Genet. 2021;12: 739975.34589117 10.3389/fgene.2021.739975
63. Piñeiro FJ Hepatic tumor microenvironments and effects on NK cell phenotype and function Int J Mol Sci 2019 20 17 4131 10.3390/ijms20174131 31450598
Piñeiro FJ, et al. Hepatic tumor microenvironments and effects on NK cell phenotype and function. Int J Mol Sci. 2019;20(17):4131.31450598 10.3390/ijms20174131
64. Garnelo M Interaction between tumour-infiltrating B cells and T cells controls the progression of hepatocellular carcinoma Gut 2017 66 2 342 351 10.1136/gutjnl-2015-310814 26669617
Garnelo M, et al. Interaction between tumour-infiltrating B cells and T cells controls the progression of hepatocellular carcinoma. Gut. 2017;66(2):342–51.26669617 10.1136/gutjnl-2015-310814
65. He Y Single-cell profiling of human CD127(+) innate lymphoid cells reveals diverse immune phenotypes in hepatocellular carcinoma Hepatology 2022 76 4 1013 1029 10.1002/hep.32444 35243668
He Y, et al. Single-cell profiling of human CD127(+) innate lymphoid cells reveals diverse immune phenotypes in hepatocellular carcinoma. Hepatology. 2022;76(4):1013–29.35243668 10.1002/hep.32444
66. You JA WGCNA, LASSO and SVM algorithm revealed RAC1 correlated M0 macrophage and the risk score to predict the survival of hepatocellular carcinoma patients Front Genet 2021 12 730920 10.3389/fgene.2021.730920 35493265
You JA, et al. WGCNA, LASSO and SVM algorithm revealed RAC1 correlated M0 macrophage and the risk score to predict the survival of hepatocellular carcinoma patients. Front Genet. 2021;12: 730920.35493265 10.3389/fgene.2021.730920
67. Hao X Inhibition of APOC1 promotes the transformation of M2 into M1 macrophages via the ferroptosis pathway and enhances anti-PD1 immunotherapy in hepatocellular carcinoma based on single-cell RNA sequencing Redox Biol 2022 56 102463 10.1016/j.redox.2022.102463 36108528
Hao X, et al. Inhibition of APOC1 promotes the transformation of M2 into M1 macrophages via the ferroptosis pathway and enhances anti-PD1 immunotherapy in hepatocellular carcinoma based on single-cell RNA sequencing. Redox Biol. 2022;56: 102463.36108528 10.1016/j.redox.2022.102463
68. Fang C Ferroptosis-related lncRNA signature predicts the prognosis and immune microenvironment of hepatocellular carcinoma Sci Rep 2022 12 1 6642 10.1038/s41598-022-10508-1 35459272
Fang C, et al. Ferroptosis-related lncRNA signature predicts the prognosis and immune microenvironment of hepatocellular carcinoma. Sci Rep. 2022;12(1):6642.35459272 10.1038/s41598-022-10508-1
69. Simmons DP Type I IFN drives a distinctive dendritic cell maturation phenotype that allows continued class II MHC synthesis and antigen processing J Immunol 2012 188 7 3116 3126 10.4049/jimmunol.1101313 22371391
Simmons DP, et al. Type I IFN drives a distinctive dendritic cell maturation phenotype that allows continued class II MHC synthesis and antigen processing. J Immunol. 2012;188(7):3116–26.22371391 10.4049/jimmunol.1101313
70. Terrén I NK cell metabolism and tumor microenvironment Front Immunol 2019 10 2278 10.3389/fimmu.2019.02278 31616440
Terrén I, et al. NK cell metabolism and tumor microenvironment. Front Immunol. 2019;10:2278.31616440 10.3389/fimmu.2019.02278
71. Dokouhaki P NKG2D regulates production of soluble TRAIL by ex vivo expanded human γδ T cells Eur J Immunol 2013 43 12 3175 3182 10.1002/eji.201243150 24019170
Dokouhaki P, et al. NKG2D regulates production of soluble TRAIL by ex vivo expanded human γδ T cells. Eur J Immunol. 2013;43(12):3175–82.24019170 10.1002/eji.201243150
72. Kamiya T Blocking expression of inhibitory receptor NKG2A overcomes tumor resistance to NK cells J Clin Invest 2019 129 5 2094 2106 10.1172/JCI123955 30860984
Kamiya T, et al. Blocking expression of inhibitory receptor NKG2A overcomes tumor resistance to NK cells. J Clin Invest. 2019;129(5):2094–106.30860984 10.1172/JCI123955
73. Liu X Immune checkpoint HLA-E:CD94-NKG2A mediates evasion of circulating tumor cells from NK cell surveillance Cancer Cell 2023 41 2 272 287.e9 10.1016/j.ccell.2023.01.001 36706761
Liu X, et al. Immune checkpoint HLA-E:CD94-NKG2A mediates evasion of circulating tumor cells from NK cell surveillance. Cancer Cell. 2023;41(2):272-287.e9.36706761 10.1016/j.ccell.2023.01.001
74. Shreeve N The CD94/NKG2A inhibitory receptor educates uterine Natural killer (NK) cells to optimize pregnancy outcomes in humans and mice Immunity 2021 54 6 1231 1244.e4 10.1016/j.immuni.2021.03.021 33887202
Shreeve N, et al. The CD94/NKG2A inhibitory receptor educates uterine Natural killer (NK) cells to optimize pregnancy outcomes in humans and mice. Immunity. 2021;54(6):1231-1244.e4.33887202 10.1016/j.immuni.2021.03.021
75. Donisi C Immune checkpoint inhibitors in the treatment of HCC Front Oncol 2020 10 601240 10.3389/fonc.2020.601240 33585218
Donisi C, et al. Immune checkpoint inhibitors in the treatment of HCC. Front Oncol. 2020;10: 601240.33585218 10.3389/fonc.2020.601240
76. Chen Y Advances in immune checkpoint inhibitors for advanced hepatocellular carcinoma Front Immunol 2022 13 896752 10.3389/fimmu.2022.896752 35757756
Chen Y, et al. Advances in immune checkpoint inhibitors for advanced hepatocellular carcinoma. Front Immunol. 2022;13: 896752.35757756 10.3389/fimmu.2022.896752
77. Cheng AL Challenges of combination therapy with immune checkpoint inhibitors for hepatocellular carcinoma J Hepatol 2020 72 2 307 319 10.1016/j.jhep.2019.09.025 31954494
Cheng AL, et al. Challenges of combination therapy with immune checkpoint inhibitors for hepatocellular carcinoma. J Hepatol. 2020;72(2):307–19.31954494 10.1016/j.jhep.2019.09.025
78. Wong KM King GG Harris WP The treatment landscape of advanced hepatocellular carcinoma Curr Oncol Rep 2022 24 7 917 927 10.1007/s11912-022-01247-7 35347594
Wong KM, King GG, Harris WP. The treatment landscape of advanced hepatocellular carcinoma. Curr Oncol Rep. 2022;24(7):917–27.35347594 10.1007/s11912-022-01247-7
79. Llovet JM Locoregional therapies in the era of molecular and immune treatments for hepatocellular carcinoma Nat Rev Gastroenterol Hepatol 2021 18 5 293 313 10.1038/s41575-020-00395-0 33510460
Llovet JM, et al. Locoregional therapies in the era of molecular and immune treatments for hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol. 2021;18(5):293–313.33510460 10.1038/s41575-020-00395-0
80. Yau T Efficacy and safety of nivolumab plus ipilimumab in patients with advanced hepatocellular carcinoma previously treated with sorafenib: the CheckMate 040 randomized clinical trial JAMA Oncol 2020 6 11 e204564 10.1001/jamaoncol.2020.4564 33001135
Yau T, et al. Efficacy and safety of nivolumab plus ipilimumab in patients with advanced hepatocellular carcinoma previously treated with sorafenib: the CheckMate 040 randomized clinical trial. JAMA Oncol. 2020;6(11): e204564.33001135 10.1001/jamaoncol.2020.4564
81. Wong J Ipilimumab and nivolumab/pembrolizumab in advanced hepatocellular carcinoma refractory to prior immune checkpoint inhibitors J Immunother Cancer 2021 9 2 e001945 10.1136/jitc-2020-001945 33563773
Wong J, et al. Ipilimumab and nivolumab/pembrolizumab in advanced hepatocellular carcinoma refractory to prior immune checkpoint inhibitors. J Immunother Cancer. 2021;9(2): e001945.33563773 10.1136/jitc-2020-001945
82. Kaseb AO Perioperative nivolumab monotherapy versus nivolumab plus ipilimumab in resectable hepatocellular carcinoma: a randomised, open-label, phase 2 trial Lancet Gastroenterol Hepatol 2022 7 3 208 218 10.1016/S2468-1253(21)00427-1 35065057
Kaseb AO, et al. Perioperative nivolumab monotherapy versus nivolumab plus ipilimumab in resectable hepatocellular carcinoma: a randomised, open-label, phase 2 trial. Lancet Gastroenterol Hepatol. 2022;7(3):208–18.35065057 10.1016/S2468-1253(21)00427-1
