Correction: Lymphocyte exhaustion in hepatocellular carcinoma: a dynamic evolution across disease stages
Year: 2026
Type: erratum
Abstract: The advent of immune checkpoint inhibition therapies has revolutionized cancer treatment over the past decade. These therapies utilize monoclonal antibodies targeting inhibitory checkpoints expressed by immune cells or their ligands expressed by tumor cells. By disrupting these receptor/ligand interactions, the functional inhibition of immune cells can be reversed, restoring effective anti-tumor activity (1). Since the identification of Cytotoxic T-Lymphocyte Antigen 4 (CTLA-4) and Programmed Cell Death Protein 1 (PD-1) as therapeutic targets, the number of recognized checkpoints has grown, and significant efforts are underway to discover new molecules that regulate immune system dynamics (2). Several combinations of immune checkpoint inhibitors (ICIs) and/or anti-angiogenics are in the current list of agents approved by the FDA for the treatment of a wide range of malignancies (3). In hepatocellular carcinoma (HCC) the combinations of atezolizumab/ bevacizumab, tremelimumab/durvalumab, nivolumab/ipilimumab, and camrelizumab/rivoceranib are the first-line treatments for unresectable HCC that have been shown to improve survival in phase III trials (4)). However, not all HCC patients are suitable candidates for these combinations and tyrosine kinases scheme is still indicated in more than 25% of patients (5,6). Therefore, ongoing research into the HCC microenvironment is crucial to enhance understanding of its biology and progression.Immune checkpoints are not confined to a single cell type and can be expressed in various cell populations depending on physiological conditions or anatomical compartments (7,8). This is the case for PD-1 and DNAX accessory molecule (DNAM-1), which are expressed in NK and T cell populations respectively (9,10).Human circulating NK cells (cNKs) constitute 5-15% of lymphocytes under homeostatic conditions and are classified into CD56 bright /CD16 neg (CD56 bright ) and CD56 dim /CD16 pos (CD56 dim ) NK cells (11). While CD56 dim NK cells are abundant in peripheral blood, CD56 bright NK cells are the predominant population in the liver. The CD56 dim NK cell population predominantly mediates the killing of target cells by secreting perforin and granzymes (12), whereas the CD56 b r i g h t NK cell population exhibits immunoregulatory and cytokine-producing capacity (13). Recent evidence points to CD56 bright and CD56 dim NK cells originating from two different progenitor lineages (14).The liver hosts a unique lymphocyte composition, with NK cells comprising up to 50% of intrahepatic lymphocytes, many of which exhibit tissue-resident characteristics and cannot recirculate (15). Human liver-resident NK (lrNK) cells are distinguished by transcription factors T-bet and Eomes, and surface markers like CXCR6 and CD69 (16,17). Moreover, lrNK possess more lytic granules, expresses TRAIL and FasL for apoptosis, and produces higher TNF and GM-CSF levels than cNK cells, compensating for reduced pro-inflammatory cytokine production to maintain liver tolerance (18)(19)(20)(21)(22)(23). In liver tumors, lrNK cytotoxicity is impaired, as shown by NKG2D downregulation in HCC (24,25).Lymphocyte Activation Gene-3 (LAG-3) and T cell Immunoglobulin and Mucin domain-containing-3 (TIM-3) are immune checkpoint markers that suppress effector T cell activity and enhance regulatory T cells (Tregs) (26,27). LAG-3 inhibits T cells via CD3 crosslinking, while TIM-3 acts as a negative regulator of inflammatory T cells and is expressed on Tregs. Beyond T cells, immune checkpoints also regulate NK cells, balancing activation and inhibition. For instance, NKG2A inhibits NK cells by recognizing HLA, while NKG2D activates them by targeting stress-induced proteins on malignant cells (28). CD16, present in NK dim cells, is a key NK cell activating receptor mediating antibodydependent cytotoxicity (29). Additionally, DNAM-1 enhances NK cytotoxicity, while T-cell immunoglobulin and ITIM domain (TIGIT) and CD96 inhibit it by competing for shared ligands like CD155 and CD112 (30,31).Our study aims to analyze the tumor microenvironment in resected HCC samples via immunohistochemistry and flow cytometry, correlating findings with tumor stage and progression. By profiling lymphocytes in tumoral and non-tumoral liver tissues from patients with varying HCC stages (early, advanced and advanced beyond progression), we sought to uncover immune interactions and identify potential biomarkers of therapeutic response.This study included 24 early HCC (eHCC) patients who underwent tumor resection, along with one additional patient treated with tumor biopsy and radiofrequency ablation. It also included 22 advanced HCC (aHCC) patients who received percutaneous biopsies prior to initiating systemic treatment with sorafenib, and 15 patients with advanced HCC beyond progression (aHCCbp) who underwent percutaneous biopsies after sorafenib failure and were subsequently treated with nivolumab (anti-PD-1). eHCC patients were defined as BCLC stage 0 or single BCLC A, aHCC patients as BCLC stage B or C, and aHCCbp as BCLC stage B-C beyond progression. Early recurrence was defined as recurrence within one year of HCC resection. All patients were recruited in our center between May 2018 and November 2020. All patients provided written informed consent before enrolment. The study was approved by the Institutional Review Board (HBC/2017/ 0026 and HCB/2020/0208) and complied with the provisions of the Good Clinical Practice guidelines and the Declaration of Helsinki. Patients' characteristics are described in Supplementary Table S1. In some cases, lymphocytes were then isolated by density gradient centrifugation using Lymphoprep [Stem Cell Technologies, 07851] according to manufacturer's instructions.The whole biopsy isolation product, or 500.000 cells from liver resection isolation product, were stained according to previously published protocols (https://www.bdbiosciences.com/en-us/ resources/protocols/cell-surface). For intracellular staining, FOXP3 Transcription Factor Staining Buffer Set [eBioscience, 00-5523-00] was used according to manufacturer's instructions (https://www.thermofisher.com/es/es/home/references/protocols/ cell-and-tissue-analysis/protocols/staining-intracellular-antigensflow-cytometry.html).The following lymphocyte cells were identified based on combination of surface markers (Supplementary Table S2): T cells, T CD4 + cells, cytotoxic T CD8 + cells, Treg, B cells Natural Killer cells and NK-like CD3 + cells. NK cells were further grouped into CD56 +bright and CD56 +dim , which have been reported as granulocytic and cytotoxic, respectively; and into CXCR6 + CD69 + which are associated with liver-residency. The gating strategy used was previously described by our group (32). Briefly, initial gating was performed on single events to exclude cell doublets and aggregates. Lymphocytes were then selected based on their characteristic forward scatter (FSC) and side scatter (SSC) properties. Viable cells were identified and selected using a VivaFix 353/442, (BioRad). T cells were defined by expression of CD3 and further subdivided into CD4 + helper T cells, CD8 + cytotoxic T cells, and regulatory T cells (Tregs), the latter identified by FOXP3 expression. B cells were identified by CD19 expression, while natural killer (NK) cells were defined by the expression of CD56 and CD16. CD56 + cells were defined as lymphocytes expressing the neural cell adhesion molecule (NCAM). This population predominantly includes NK cells (CD3 - CD56 + ) but may also encompass a population of "NK-like" cells identified based on co-expression of CD3 with CD56 markers. NK and NK-like cells were further categorized into CD56bright and CD56dim subsets. Within each defined immune population, expression of immune checkpoint molecules, including PD-1, DNAM-1, LAG-3, and CD69, was evaluated to assess activation and exhaustion phenotypes. Cell populations labelled with plus or minus signs refer to those cells with (plus) or without (minus) visible expression of the specific markers. In this study, immune exhaustion in T cells was defined by the expression of multiple inhibitory receptors (PD-1, TIM-3, LAG-3, CD39), while for NK cells, exhaustion was inferred based on increased expression of inhibitory receptors (e.g., NKG2A) and reduced expression of cytotoxicityassociated markers such as CD16. Markers like CXCR6 and CD69 were used to define liver-resident NK populations. We acknowledge that NK cell exhaustion remains a debated concept and interpret these profiles as indicative of functional impairment rather than definitive exhaustion.Mean Fluorescence Intensity (MFI) was computed for all parameters in every analysis and normalized over the Sphero ™ Rainbow Calibration Particles (8 peaks), 3.0 -3.4 µm.All antibodies and reagents used are detailed in Supplementary Table S2 in the Supplementary Material. Due to sample limitations, different antibody combinations were used for biopsy-isolated products and liver resection-isolated products (Supplementary Table S3). Not all markers could be used in all samples due to cell number limitations. The total sample number analyzed for each marker is detailed in the Results and Figures sections.The formalin fixed paraffin embedded (FFPE) samples of the 24 eHCC patients were exhaustively reviewed, analysing all the histological features (Table 1) from all the tumours for each case, and non-tumoral parenchyma was also reviewed. All the clinical data (Supplementary Table S1), as well as follow-up data from all the patients were collected for further correlations with histological and immunohistochemical (IHC) data. Tissues suitable for IHC examination were identified in H&Estained slides, choosing one slide of tumoral and non-tumoral parenchyma of each case, and the paraffin blocks in the archives of our pathology department were accessed. Two sections (4 mm) were prepared on positively charged slides from each paraffin block.All the antibodies were validated by conventional IHC using tonsil as a control tissue in an automated immunostainer (DAKO; Agilent Autostainer Link 48), and also in which dilution the antibody worked best in case for concentrated antibodies. The process of deparaffinization, rehydration and epitope retrieval were done with PT-Link Rinse Station (Agilent). IHC was performed against CD3 (Polyclonal Rabbit Anti-Human, Readyto-Use, Agilent), CD4 (Monoclonal Mouse anti-Human, Clone 4B12, Ready-to-Use, Agilent), CD8 (Monoclonal Mouse Anti-Human, Clon C8/144B, Ready-to-Use, DAKO), FOXP3 (1:100 Monoclonal Mouse anti-Human, clone 86D, Anacrom Diagnostics), CD68 (Monoclonal Mouse anti-Human, clone KP1, Ready-to-use, Agilent), CD163 (1:100 Monoclonal Mouse anti-Human, clone 10D6, Biocare Medical), CD38 (1:50, Mouse monoclonal anti-human, clone 38C03, Gennova), CD56 (Monoclonal Mouse anti-Human, clone 123C3, Agilent), CD69 (1:500 Monoclonal Rabbit anti-Human, clone EPR21814, Abcam), CD16 (1:100 Monoclonal Rabbit anti-Human CD16 SP175, Abcam), CXCR6 (1:50 Polyclonal Rabbit anti-Human, Abcam, overnight incubation) and PD-L1 (1:50 Monoclonal Mouse anti-Human, clone 22C3, DAKO). The chromogen used was AEC (3-amino-9-ethylcarbazole, Ready-to-Use, DAKO) and hematoxylin as a counterstain, except for PD-L1 that was performed with hematoxylin-diaminobenzidine (DAB) staining.Multiple immunostains were done using the protocol SIMPLE (sequential immunoperoxidase labelling and erasing) previously described (33) for each of the chosen slides in each case. Positive cell detection in pathology images was analyzed using QuPath software. Initially, it was necessary to validate the effectiveness of the stripping technique eliminating antibodies and preserving antigens using control tissues. Using the SIMPLE protocol, we successfully performed up to four immunohistochemical stains on the tissue Beyond tissue was different were to different cell populations in one T lymphocytes FOXP3 and CD38 and and NK cells CD16, CD69, each the slides were before the SIMPLE protocol to the antibody and the All the slides were with slide in We compartments for each and of the of in each histological whole slide were of was wide (Supplementary all positively stained cells using the QuPath The parameters for cell detection and were for each based on the of and were to We cell density number of stains for 24 patients the compartments in cell density with due to or are as and for data were as or and for not In analysis and was used for and were using and The was The between was using was defined as an than or was used to analyze the between immune cell analysis was performed using flow were expressed as and range were described as and between two for or were by was used to were by the for or or with for All were with was used for all except for the of lrNK and their that were and using an histological analysis of from the 24 recurrence within the year of resection, while of the histological characteristics analyzed are in Table that in eHCC the and were (Supplementary S1), we evaluated by IHC specific of immune cells were expressed in these In we the of the findings with early recurrence (Table These to assess the of immune cell cell density and into the immune that to tumor recurrence and as biomarkers for we on all patients some in the of immune population in tumor and can be data in Table For were present in all compartments (Table and or Figures PD-L1 expression was in the tumoral in and in non-tumoral parenchyma (Table We also identified NK cells as and were in the and non-tumoral parenchyma (Table a follow-up we which immune population were to early findings that patients with early recurrence identified as cells, present in the non-tumoral parenchyma (Table whereas these cells were in patients with the of early recurrence (Table was the higher and of cells, including NK or cells, in the tumor of these potential immune cell were to identify the best of early The an of a significant The of this as by the under the characteristic was for early further the of NK cells we to flow Moreover, to the of cells in the tumor parenchyma could we also included patients various in our This to NK cells, along with lymphocyte to these populations expression and in immune exhaustion markers between tumor and different We analyzed the expression of the different CD155 ligands DNAM-1, CD96 and which have shown in the of NK cells We have evaluated the expression of these markers in all lymphocytes isolated from eHCC All markers were in T and NK cells, with higher expression in cells CD56 or than in NK cells CD56 bright or Figures significant were for In this study, immune exhaustion in T cells was defined by the expression of multiple inhibitory receptors (PD-1, TIM-3, LAG-3, CD39), while for NK cells, we exhaustion as the of increased inhibitory receptor expression (e.g., PD-1, and markers (e.g., the of liver lymphocytes isolated from eHCC patients with aHCC and aHCCbp shown in were between tumor isolated lymphocytes and isolated lymphocytes with some shared the patient samples from all more T cells and than their Additionally, and T cells from tumor samples of eHCC patients higher PD-1 expression to their These were not in advanced (aHCC) or advanced beyond progression (aHCCbp) due to a in PD-1 expression with advanced Due to the number of lymphocytes isolated from aHCC and aHCCbp some exhaustion markers were analyzed in the eHCC more could be these we increased expression of NKG2A in NK dim and NK dim CD3 + cells, as well as increased expression of CD96 in NK dim cells Additionally, expression, a marker of T cells, was in the tumor in and T cells on liver NK (lrNK) cells, by the co-expression of we that in eHCC a higher of lrNK cells, CD56 dim and CD56 bright , than in tumor These were in aHCC patients with aHCCbp patients of CD56 dim or CD56 bright , in tumor and the we that isolated lymphocytes all HCC stages a in the number of T cells and NK dim cells along HCC progression These were not (Supplementary Additionally, we the expression of immune checkpoints such as PD-1 in different immune cell populations. Results that T and cells from advanced HCC exhibit higher PD-1 expression to the eHCC group was a and in the expression of CD16, a marker for NK and CD69, a marker of activation and tissue in NK bright cells with BCLC stage in the non-tumoral HCC These except for PD-1 expression, were not (Supplementary Figures All these data that immune exhaustion in the tumoral and to the non-tumoral tissue with to assess HCC progression liver we evaluated in the aHCC and aHCCbp In the aHCCbp group an increased of NK-like dim and NK-like bright CD3 + cells and a in NK dim and NK bright CD3 NK dim cells of the aHCCbp group higher expression of the activation marker NKG2D and of the markers than associated with liver but in reduced expression of the activation marker CD16 of these were (Supplementary Figures liver T cells also DNAM-1 expression, an activation in eHCC patients with further that T CXCR6 cells higher expression of the exhaustion markers PD-1, TIM-3, and LAG-3 in aHCC and aHCCbp These were also present in T cells (Supplementary Figures study new into the immune of the of NK and T cells in HCC progression and While in immune exhaustion markers were between and in our the potential of HCC on immune microenvironment further in of our research in the of patients into HCC for a understanding of the of immune dynamics with as well as the of lymphocytes from tumor and these stages in the group of which a of in immune checkpoints expression between and of eHCC NK dim and NK dim CD3 + cells isolated from tumor samples of eHCC patients higher expression of NKG2A and CD96 to their T CD4 + and CD8 + cells isolated from tumor higher of + cells early HCC Fluorescence isolated tumor isolated immune exhaustion and the of the HCC progression. of our study is the examination of the microenvironment an not with flow analysis in early HCC patients that the tumor and parenchyma that the tumor the immune of the non-tumoral liver from early HCC (eHCC) We more significant in the non-tumoral parenchyma than in the tumor between patient This that may the immune microenvironment the liver. findings the of immune cell populations to a higher of eHCC the detection of CD68 in the non-tumoral tissue is CD38 is a molecule in the which has been to T cell This could that in the non-tumoral liver of patients with early are signs of an present microenvironment that may tumor Moreover, we also CD56 + NK cell within tumor in patients with HCC recurrence (Table NK in tumor and immune and NK cell activity in HCC While NK cell exhaustion remains a debated and in liver markers with activation and tissue our data evidence of NK cell HCC by the of inhibitory receptors and of cytotoxic a of effector This study has limitations. the sample in the advanced HCC (aHCC) and (aHCCbp) was due to the of biopsies in This reduced the for and while we not all we used a of without for multiple or the of the of our IHC study was the to the activation or inhibition of cells due to in for CD16, CD69, and CXCR6 markers. While flow to of these markers within tissue sections have provided additional we immune exhaustion and markers in we not assess functional molecules such as B or perforin to the cytotoxic potential of NK and T cells. these limitations, the are and a of immune exhaustion HCC stages and tissue findings on NK cell dynamics to a understanding of immune NK cells in tumor and various stages of DNAM-1, and as ligands for the CD155 in NK cell While their on T cells is well have indicated that DNAM-1 expression on T cells can a population, whereas its expression on T cells is associated with increased cytotoxicity In our analysis in eHCC we that DNAM-1, and are expressed on T and NK cells, with higher levels on T cells to NK cells. Moreover, the in NK cells, reduced CD16 expression and increased exhaustion markers such as the of tumor on NK cell and T cells increased PD-1 and expression, of immune in These findings that immune from the tumor to the and in advanced clinical the of immune in T cells have a immune checkpoint expression T cells from eHCC patients have DNAM-1 expression in the tumor In the PD-1, TIM-3, and LAG-3 expression is higher in liver T cells all HCC as to liver T cells, with higher of immune exhaustion markers in lymphocytes from advanced early HCC advanced HCC advanced HCC patients beyond isolated tumor isolated Fluorescence tumor and into non-tumoral liver a to In to immune checkpoint inhibitors may HCC may be more to while advanced stages may combination to immune as in the was that NK dim cells within the tumor of aHCCbp patients increased NKG2D expression, as to aHCC this was by a in cytotoxic an immune While NKG2D is as an activation marker in HCC and its expression has been to be reduced in liver NK (lrNK) cells from HCC some that in advanced by such as the expression of NKG2D can cancer progression rather than In this our data the of NKG2D in advanced progression. It be from a of to further assess NKG2D NK cell their cytotoxic In NKG2D such as increased in can as receptors and NKG2D expression in advanced HCC an between with HCC has been described for patients with that of are of by immune we not functional NK cell cytotoxicity and cytokine production be to the exhaustion to functional impairment in the of research in this be to the of NKG2D and its ligands in advanced In this such as and whole with or for to the and functional of immune exhaustion and interactions in findings that lrNK cells, by CXCR6 and CD69 expression, are more abundant in eHCC but with with The of lrNK cells in advanced stages in immune which could tumor our profiling of NK cell dynamics in HCC is with the of effector and of of NK cells within the tumor In our study of T cell dynamics that immune exhaustion and is in liver-resident T cells, to with an exhaustion with in the tumor and also NK-like or NK cells, in to conventional NK cells. In the of immune exhaustion markers such as CD96 or NKG2A on NK-like cells is more than on NK cells. within the the of NK dim cells with progression with an in NK-like cells. the number of NK dim cells also with that immune exhaustion is an and that the to the tumor from the of immune cells in the tumor tissues that immune is in tumor tissues from the early stages of our study includes lymphocytes from the patients with HCC different stages and not from We are that this is a of our data as it the immune of these and the of inhibitors these the we is in functional and in in the of the immune cell analysis of the tissues from the patient significant in the of the lymphocyte population and in markers of immune In aHCC was a in the number of T cells and NK dim cells, of which are crucial for cytotoxic our data the of immune in HCC and its potential for The between immune checkpoint expression and HCC progression the that targeting multiple checkpoints are in the identification of CD56 + cell density and immune exhaustion markers as potential biomarkers a for are to validate these findings and the therapeutic potential of restoring NK or NK-like cell with immune
Source: Frontiers in Immunology
Institutions: Hospital Clínic de Barcelona, Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, Consorci Institut D'Investigacions Biomediques August Pi I Sunyer, Universitat de Barcelona, Consejo Superior de Investigaciones Científicas, Institut d'Investigacions Biomèdiques de Barcelona
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7128016778
Cites:
Cited by:
Related to: