Antitumor CD8+T cells in hepatocellular carcinoma: Present but exhausted


  • Ka-Kit Li,

    1. NIHR Biomedical Research Unit and Centre for Liver Research, University of Birmingham, Birmingham, UK
    Search for more papers by this author
  • David H. Adams

    Corresponding author
    1. NIHR Biomedical Research Unit and Centre for Liver Research, University of Birmingham, Birmingham, UK
    • Address reprint requests to: David H. Adams, FRCP, F.Med.Sci., Professor of Hepatology, National Institute for Health Research Biomedical Research Unit, University of Birmingham, Birmingham, UK. E-mail:

    Search for more papers by this author

  • See Article on Page 1415

  • Potential conflict of interest: Nothing to report.




cytotoxic T-lymphocytes antigen-4


dendritic cell


hepatocellular carcinoma


interferon gamma


myeloid-derived suppressive cells


major histocompatility complex


programmed death receptor-1


tumor associated antigens


tumor-associated macrophages


regulatory T cells.

Hepatocellular carcinoma (HCC) is the second leading cause of global cancer deaths and the incidence is rising.[1] Advances in medical treatment have resulted in improved survival for many common malignancies,[2] but not HCC. The lack of curative treatments and consequent dismal survival rates make elucidating HCC biology and translating this into effective treatments a priority. There is good evidence that HCC stimulates an immune response, making immunotherapy in which the host's immune system is stimulated to attack the cancer a potential option. However, despite efficacy in animal models, most clinical trials of tumor immunotherapy have been disappointing. The article from Robert Thimme's group3 in this issue highlights mechanisms that inhibit effective antitumor immunity in HCC and suggests immunotherapeutic options against a formidable foe.

The host immune system has the potential to recognize a wide range of tumor-associated antigens (TAA) including alpha-fetoprotein (AFP).[4] Moreover, there is emerging evidence that the presence of TAA-specific T cells is associated with an improved response to dendritic cell (DC) vaccination and better prognosis.[5, 6] However, despite mounting a detectable immune response, tumor progression frequently occurs, and understanding why the immune response fails to control the tumor is a crucial step if immunotherapy is to become clinically effective. Many mechanisms have been implicated in cancer immune evasion.[7] These include down-regulation of major histocompatility complex (MHC) molecules and TAA on tumor cells, inhibition of effector immune cells functions, the recruitment and induction of immunosuppressive immune cells, and the disruption of effective tumor antigen presentation as a consequence of reduced costimulatory molecule expression and interference of antigen processing by DC (Fig. 1).

Figure 1.

The diagram provides a simplified overview of how tumor cells are able to evade the host's immune response. The immune system has the potential to eliminate tumor cells by way of the activation of effector cells such as CD8+T cells with the help of dendritic cells (DCs) and helper CD4+T cells. Multiple mechanisms have been identified that tumors use to evade the immune system, these include: 1) The down-regulation of MHC molecule and TAA on tumor cells. 2) The secretion of chemokines by tumor and tumor-associated stromal cells to recruit suppressive cells such as regulatory T cells (Tregs), immature DC (iDC), tumor associated macrophage (TAM), and myeloid derived suppressor cell (MDSC). 3) Reduced costimulatory molecule expression and interference of antigen processing by DC. 4) Inhibition of helper CD4+T cells. 5) Inhibition of effector cells by TAM, MDSC, and Tregs through a variety of mechanisms, including production of arginase, reactive oxygen species (ROS), and the immunosuppressive cytokines interleukin-10 (IL-10) and transforming growth factor beta (TGF-β). 6) Activation of inhibitory receptors by its ligands such as PD-L1 on tumor cells, resulting in impaired production of inflammatory cytokines interleukin-2 (IL-2), tumor necrosis factor (TNF), interferon-γ (IFN-γ), and cytotoxic mediators perforin and granyzme B by effector CD8+T cells. 7) The induction of Tregs and MDSC by iDC though secretion of immunosuppressive cytokines and production of indoleamine 2,3-dioxygenase (IDO).

In this month's issue of Hepatology, Flecken et al.[3] report new findings on the antitumor response in HCC with important therapeutic implications. The authors describe a subset of CD8+T cells specific for the TAAs AFP, glypican-3 (GPC-3), melanoma-associated gene-A1 (MAGE-A1), and New York-esophageal squamous cell carcinoma-1 (NY-ESO-1) in the blood, liver, and tumors of patients with HCC, and quantified the magnitude and breadth of the antitumor CD8+T-cell response. Previous studies have reported an association between the strength of the antitumor T-cell response and patient survival and the present work illustrates that a broad immune response early in disease is associated with a better prognosis. This has implications for the development of therapeutic vaccines. Many vaccine trials have used a specific antigen or antigenic epitope to prime T-cell responses. However, the identification of multiple immunodominant epitopes across different tumor antigens suggests that vaccines should incorporate several antigens if they are to be effective. In the context of DC-based vaccination, this study supports the use of tumor lysate to pulse DCs based on the notion that multiple tumor lysate proteins favor the development of a broader antitumor immune response.

The authors go on to demonstrate that TAA-specific CD8+T cells expanded from the blood of patients with HCC have markedly impaired IFN-γ secretion and cytolytic activity. Interestingly, the same patients had preserved effector CD8+T cells against Epstein-Barr virus (EBV) and cytomegalovirus (CMV), suggesting that this is a specific tumor effect. This is important because it shows that the TAA-specific CD8+T cells are present, they have not been deleted, and thus it is potentially possible to restore their effector function. However, to do this it is important to understand the underlying mechanisms responsible for their dysfunction. These include T-cell exhaustion as a consequence of chronic stimulation with TAA, or an effect of the tumor environment preventing full and sustained T-cell activation, both of which would explain the differential effects on tumor-specific and antiviral T cells.

CD8+T-cell exhaustion had been well described in chronic viral infections.[8, 9] In healthy subjects, acute viral infection results in activation and proliferation of antigen-specific CD8+T cells that kill infected cells and clear the virus. This is followed by the rapid contraction of viral specific effector CD8+T cells, leaving a small population of memory T cells to provide long-term immunity against the same virus. In chronic human immunodeficiency virus (HIV), hepatitis B virus (HBV), or HCV infection, persistent antigen stimulation results in loss of CD8+T-cell cytolytic function and impaired secretion of interleukin (IL)−2, tumor necrosis factor (TNF), and interferon-gamma (IFN-γ). CD8+T-cell exhaustion and dysfunction has also been reported in human cancer[10, 11] and the identification of dysfunctional TAA-specific CD8+T cells in the current study provides further evidence that this occurs in HCC. The inhibitory receptor programmed death receptor-1 (PD-1) is a marker of exhausted CD8+T cells and activation by its ligand PDL-1 in the tumor environment contributes to T-cell exhaustion. Other inhibitory receptors implicated in T-cell exhaustion include T-cell immunoglobulin and mucin-domain-containing-molecule-3 (TIM-3), lymphocyte-activation gene-3 (LAG-3), and cytotoxic T-lymphocytes antigen-4 (CTLA-4).[8] The current study reports PD-1 and TIM-3 on most, but not all, TAA-specific CD8+T cells after in vitro expansion, suggesting that multiple pathways may be involved.

The colocalization of inhibitory receptors and their ligands within the tumor environment may partially explain the differences seen in TAA responses in CD8+T cells isolated from the tumor and those isolated from blood or uninvolved liver tissue. Tumor-stromal cells can mediate immunosuppression by modulating DC activation and differentiation in favor of DCs that generate regulatory T cells (Treg) rather than effector cells. Such effects may be mediated by changes in the balance between costimulatory and coinhibitory signals; i.e., a relative lack of molecules crucial for full T-cell activation such as CD80 and CD86 plus overexpression of inhibitory receptors such as CTLA-4 and PD-1. The importance of CTLA-4 and PD-1 has translated into therapy and blocking both receptors appears to have an additive antitumor effect in patients with melanoma.[12] Thus, therapeutic vaccination in HCC may be more effective if combined with blockade of inhibitory receptors.

TAM and tumor-stromal cells generate an immunosuppressive microenvironment by secreting immunosuppressive cytokines including IL-10 and transforming growth factor beta (TGF-β) or chemokines that recruit MDSC or Treg. In the current article the authors found that although Treg depletion restored the proliferative capacity of TAA-specific CD8+T cells, it did not restore IFN-γ secretion, suggesting that Treg depletion in itself is unlikely to be enough to generate effective antitumor immunity.

The work by Flecken et al. emphasizes the complexity of antitumor immunity and suggests that to be effective, immunotherapy will need to combine several approaches targeting multiple pathways. It is not enough to deplete Treg or even to stimulate an antitumor immune response because these approaches alone are unlikely to generate effective immunity in the face of a hostile tumor microenvironment and active tumor immune escape. However, combination therapy that uses a vaccination strategy aimed at generating sustained and broad T-cell immunity combined with checkpoint blockade to maintain T-cell effector function, as well as strategies dampening regulatory cells, may be more effective. This is particularly true if immunotherapy is combined with radiofrequency ablation (RFA) or transarterial chemoembolization (TACE), which themselves drive antitumor immune responses through the release of tumor antigens and the generation of danger signals that activate the innate immune system.[13] In support of this the current study found a greater TAA-specific CD8+T-cell response in patients who had received TACE.

In summary, the current study highlights the complexity of immune regulation in the context of a hostile tumor microenvironment and offers insights into how immunotherapy may be improved in the future.

  • Ka-Kit Li, M.B.Ch.B., MRCP, and David H. Adams, FRCP, F.Med.Sci.

  • NIHR Biomedical Research Unit and Centre for Liver Research, University of Birmingham Birmingham, UK