Development and therapeutic manipulation of the head and neck cancer tumor environment to improve clinical outcomes

Abstract

The clinical response to cancer therapies involves the complex interplay between the systemic, tumoral, and stromal immune response as well as the direct impact of treatments on cancer cells. Each individual's immunological and cancer histories are different, and their carcinogen exposures may differ. This means that even though two patients with oral tumors may carry an identical mutation in TP53, they are likely to have different pre-existing immune responses to their tumors. These differences may arise due to their distinct accessory mutations, genetic backgrounds, and may relate to clinical factors including previous chemotherapy exposure and concurrent medical comorbidities. In isolation, their cancer cells may respond similarly to cancer therapy, but due to their baseline variability in pre-existing immune responses, patients can have different responses to identical therapies. In this review we discuss how the immune environment of tumors develops, the critical immune cell populations in advanced cancers, and how immune interventions can manipulate the immune environment of patients with pre-malignancies or advanced cancers to improve therapeutic outcomes.

Keywords: CD4; CD8; HNSCC (head and neck squamous cell carcinoma); OSCC (oral squamous cell carcinoma); TIL (tumor infiltrating lymphocytes); immunotherapy; pre-malignancies.

Figure 1

Immune intervention during tumorigenesis and prior to conventional therapy. HPV+ tumorigenesis provides a model to show potential interventions prior to conventional therapy for HNSCC that can impact outcome. 1. Protective HPV vaccination can prevent the initial transformation that can result from HPV infection. 2. Early immunotherapy of dysplasia can enhance HPV-specific responses or direct new immune responses to genes involved in progression to prevent further progression. 3. Malignant tumors can have a poor pre-existing immune response that is associated with poor prognosis following conventional therapy. 4. Immunotherapy can convert the tumor environment by directing immune responses to the tumor, and this has the potential to convert patients to an improved prognosis. 5. Even where early immunotherapy fails to prevent tumorigenesis, it may result in a tumor with a strong-pre-existing immune response that would be predicted to have a better prognosis following conventional therapy.

Figure 2

Immune regulation during infection and related immunotherapy effects on pre-existing anti-tumor immune responses. (A) (i) The response to an infection, in this example HPV infection of basal epithelial cells (1), can result in a series of events that result in immunity. Innate sensing of infection (2) can lead to locoregional inflammation (2), as well as antiviral responses in the infected cell. Inflammation combined with antigen release can permit DC migration to lymph nodes (3), so while non-specific cells continue to recirculate through the tissue (4), new immune responses can be initiated in the draining lymph node (5). (ii) Following expansion of antiviral T cells in the draining lymph node, these cells can be preferentially recruited to the site of infection via the impact of inflammation on the vasculature (1). Chemokines released by infected cells can recruit T cells through the stroma to the infection site (2). Cognate recognition of infected cells (3) can lead to their death (4). Dying cells are actively cleared, which can initiate repair and shut down inflammation in the infection site (5). (B) (i) The immune response to cancer, either in early dysplasia or in a malignant tumor, can result in infiltration of tissue resident memory T cells (1), as well as stromal infiltration of T cells, though local immune suppression (2) results in tumor growth dominating over tumor destruction. Ongoing cancer cell death, due to hypoxia or limited growth factor availability can lead to antigen release (3), though stromal suppression can limit the quality of antigen presentation. T cells can continue to recirculate through tumors (4), though they may be restricted to the stroma and poorly interact with cancer cells, and stromal cells can support further invasion of the cancer into surrounding tissues (5). (ii) Immunotherapy with checkpoint inhibitors (1) may derepress local T cells or recirculating T cells to permit destruction of cancer cells. Inflammatory cytokines released by activated T cells can repolarize macrophages in the tumor to limit invasion or immune suppression (2). Alternatively, immunotherapies can be selected to deplete or block the inhibitory effect of suppressive macrophages or Treg (3). Where pre-existing immunity is limited, vaccine or adoptive transfer approaches may provide tumor-specific T cells (4), which can be targeted to the tumor environment with locoregional immunotherapies that generate inflammation in the tumor environment (5).

Figure 3

CD4 T cell help, adjuvant signals, and immunotherapy to enhance CD8 T cell responses to tumors. (A) The classic three cell model of CD8 T cell activation relies on antigens being presented to both CD4 and CD8 T cells by dendritic cells. Innate adjuvants present in the infection improve antigen processing and presentation, as well as upregulate costimulatory molecules. Highly antigenic targets increase the likelihood of high affinity antigens being available for both CD8 targeting and CD4 T cell help. (B) The balance of positive and negative stimuli dictate whether T cells can contribute to anti-tumor responses. Broadly, highly antigenic tumors with endogenous adjuvant support can improve the likelihood of T cell responses, which are countered by a range of negative regulatory features of the tumor environment. (C) High affinity antigen-specific T cell responses have the potential to stimulate T cells independently, but the likelihood of passing the critical activation thresholds is increased by CD4 help and adjuvant signals to cross-presenting DC and directly presenting cancer cells. A range of suppressive factors in the tumor can decrease the ability of T cells to pass the critical activation threshold to control tumors, and current immunotherapies can overcome these limitations to overcome suppression and bring new T cells into the response that were formerly unable to cross the activation threshold to participate in tumor control.

Figure 4

Expanding tumor-specific T cell immunity via adoptive transfer. (A) Biopsies and surgical specimen can be a source of tumor infiltrating lymphocytes (TIL), which can be expanded ex vivo and tumor-specific cells isolated either by response, or by initial phenotype. Alternatively, tumor-specific T cells can be engineered if both antigen and MHC match known TCR specificities, or using chimeric antigen receptors (CAR). (B) Adoptive transfer can occur prior to conventional therapy to alter the tumor immune environment prior to further treatment, or (C) can follow conventional therapy to target residual disease. At present, with surgical samples the dominant source of TIL, adoptive transfer is generally delivered following completion of conventional multimodality treatment or in the metastatic setting.