Investigating anti-tumour T cell responses in nasopharyngeal carcinoma to refine vaccine-based immunotherapies

Lead Research Organisation: University of Birmingham
Department Name: Cancer Sciences


Nasopharyngeal carcinoma (NPC) is a type of cancer that occurs at the back of the nose. It is unusually common in some parts of the world, including countries like Malaysia where it constitutes a very significant health problem. NPC can be cured, but in many cases the cancer comes back and this is usually fatal. Furthermore, patients who do survive are often left with significant long-term side effects caused by the toxic treatments. Therefore, new treatments are needed to reduce deaths from NPC and improve the quality of life for survivors. All patients with NPC have a virus, known as the Epstein-Barr virus, or EBV for short, in their tumour cells. We have developed a vaccine that can recognise EBV with the aim of boosting the patients' own immune response to the tumour. Early tests of our vaccine in people show that it is safe and that it can boost immunity against the virus. In this proposal we want to understand more about immunity to EBV in patients with NPC so that we can modify how we deliver our vaccine and eventually combine it with other drugs to improve its therapeutic effects. To do this we will be using new sequencing technologies to probe how patient's T cells, the body's main weapon against virus-infected cells, respond to EBV, for example before and after treatment, enabling us to 'track' individual T cell responses over time. At the same time, we will explore the nature of potential inhibitory pathways present in the tumour so that we might eventually target these in combination with our vaccine making it more effective.

Technical Summary

We have developed a therapeutic vaccine, known as MVA-EBNA1/LMP2, that expresses a fusion protein encoding almost all known CD4 T-cell targets in EBNA1 and multiple CD8 T-cell epitopes from LMP2. This vaccine has already been shown to be safe and to exhibit promising therapeutic activity in early phase clinical trials. Knowledge of the factors in the blood and tumour tissues of NPC patients that might limit T cell responses will be required not only to maximize the therapeutic efficacy of our vaccine, but also to identify biomarkers of likely treatment response for eventual patient stratification.

To address these gaps in knowledge we will first use multiplex immunohistochemistry to provide a detailed characterisation of the immune micro-environment of NPC, including the expression of immune checkpoints. We will use our established EBV-specific T cell clones and NPC cells, including lines we have recently established that more faithfully recapitulate primary disease, to explore the activity of these checkpoints, initially focusing on those mediated by fibroblasts. Finally, we will use next generation sequencing to describe how the TCR repertoire in the blood of NPC patients varies throughout the course of disease, and how it compares to the TCR repertoire we observe in tumour tissues. Importantly, using our cultured immune assays we will be able to interrogate these data to define T cell responses to relevant EBV and cellular tumour antigens.

Planned Impact

This study will reveal new insights into the immune pathogenesis of nasopharyngeal carcinoma (NPC), a disease that is very common in South East Asia, including Malaysia. The Bidayuh natives in Sarawak have the highest incidence of NPC in the world, and NPC is the most common cancer among Sarawakian men. The cancer disproportionately affects the vulnerable groups, with the lowest social class reported to have 4 fold higher risk of the disease. NPC is associated with significant mortality and morbidity which result in a major impact on the immediate and extended families as over 90% of the patients supported households in some way. This study will specifically document the existence and function of immune checkpoints in biopsies from NPC patients and describe for the first time how T-cell receptor (TCR) specificities vary in the blood of NPC patients during the course of their disease and how they compare to the TCR repertoire we identify in the tumour; importantly, we will also be able to interrogate these data to define T cell responses to specific tumour antigens in both compartments. This knowledge will not only be important for our understanding of the pathogenesis of NPC, but will also help refine T cell therapies being developed for NPC patients, including our own therapeutic EBV vaccine, currently in early phase clinical trials.

A further impact of this project will be the transfer of scientific knowledge, generic and specific research skills, and technologies in the field of cancer immunology and EBV immunotherapy from the UK to Malaysia. We expect that this, coupled with the enhancement of our research network in Malaysia, will provide opportunities to translate our findings into better treatments for patients in Malaysia.

We also expect the outputs of this research to be relevant to the pharmaceutical industry and to biotech companies. The EBV vaccine is being developed for commercial use by Cancer Research Technology, who manage all IP. Eventually, this should lead to the development of additional new therapeutic agents, for example, drugs which will improve the efficacy of our vaccine when used in combination. New, more effective, therapies for patients with NPC will reduce the burden of disease, particularly for the vulnerable groups and will have a significant economic impact in Malaysia and in other parts of the world where NPC is common.
Description Progress Summary:

Nasopharyngeal carcinoma (NPC) tumour cells contain the Epstein-Barr virus genome and consistently express the viral proteins, EBNA1 and LMP2, with a proportion of cases expressing an additional viral protein LMP1. The tumours are richly infiltrated with diverse lymphocyte subsets including T-cells. NPC patients possess detectable T-cell responses to EBNA1 and LMP2 in their peripheral blood. The aim of this part of the work was to determine whether NPC tumours contained T-cells specific to the viral proteins expressed in the malignant cells. Knowing whether T-cells specific for EBV-encoded tumour antigens are present within NPC tumours is important for the rational development of immunotherapies - for example, agents designed to overcome local immunosuppression, such as anti-PD1 checkpoint inhibitors, are entirely dependent upon such cells being present for their mode of action.

Although T-cell receptor (TCR) sequencing has previously been performed on NPC tumours, the online databases contain few TCR sequences from from T-cells specific for EBNA1, LMP2 and LMP1. Furthermore, only a minority of TCR sequences occur in multiple people (so called public TCR sequences). Most people possess their own 'private' repertoire of TCRs specific for a particular immune epitope. Therefore, studes that rely solely on TCR databases are likely to under-estimate the true frequency of antigen-specific T-cells in tissue samples.

We took several approaches to study intratumoural T-cells in NPC. First, we developed a TCR clone tracking procedure to identify, for any individual, the range of TCR sequences they posses capable of recognising a particular antigen (in our case EBV tumour antigens). This method involved stimulating blood peripheral blood mononuclear cells (PBMCs) separately with each viral antigen to generate polyclonal T-cell lines in vitro, restimulating the lines after 7-14 days with the appropriate antigen and then purifying antigen specific CD8 and CD4 T-cells by FACS (based on production of the cytokine TNF-a upon antigen re-stimulation). TCR sequencing of the purified T-cells then provided each individual's antigen-specific TCR repertoire. Second, we applied the aforementioned method to a range of healthy donors and patients with NPC to increase the number of EBV tumour antigen-specific T-cell sequences for our own analyses and future work. Third, we performed TCR-sequencing on 28 snap-frozen NPC tumours (and 4 non-cancerous nasopharyngeal specimens) to increase the number of NPC patients for whom tumour TCR data is available. Fourth, of these 28 patients six had a paired cryopreserved viable blood sample, thus allowing us to generate their EBV antigen-specific TCR repertoires from the blood sample and search for these sequences within their tumour. Fifth, cryopreserved cell suspensions from four NPC tumours were: i) analysed by high dimensional mass cytometry to study the range and phenotype of immune cells within the complex NPC tumour microenvironment and ii) used to generate polyclonal T-cells in order to isolate EBV-antigen-specific T-cells from the tumour itself.

Analysis of the blood and tumour TCR data demonstrates that EBNA1 and LMP2-specific T-cells are present within NPC tumours. This finding was confirmed by the generation of EBNA1- and LMP2-specific T-cells from the cryopreserved tumour biopsy. However, based on the abundance of TCR reads it is clear that the frequency of EBNA1- and LMP2-specific T-cells within the tumour is low. Mass cytometry showed a hierarchy of immune cells exist within NPC tumours (CD4 & CD8 T-cells > B cells > NK cells > macrophages). A proportion of CD8 T-cells co-expressed CD103 and CD39, markers that have previously been shown to be present on tumour specific intratumoural T-cells. These T-cells expressed high levels of PD1 but low levels of the activation marker CD38. A population of CD4+ CD25+ CD39+ T-cells (potential regulatory cells) was also detected; these cells lacked PD1. Two distinct populations of B-cells, one of which expressed the plasmablast marker CD38, were also present.
Exploitation Route Significance of findings:

Our data shows that while NPC tumours contain T-cells specific for EBNA1 and LMP2 their frequency is low. Successful immunotherapy for NPC may therefore require methods to boost the frequency of T-cells specific for these antigens, such as adoptive T-cell therapy of therapeutic vaccination. This low frequency of anti-tumour effectors may explain the poor response rates seen in clinical trials of the anti-PD1 checkpoint inhibitor in NPC.

We expect that once completed and published our data will help advance immunotherapies for patients with cancer
Sectors Healthcare

Description We have engaged with the NPC study group, Malaysia to explore patient's view of the disease and to explain how our research might benefit We are developing an instrument to measure T cell immunity in remote parts of S.E Asia based on these responses The Group have also been able to -establish a new collaboration with Professor KW Lo to explore the nature of EBV immunity in NPC patients - develop multiplex immunohistochemistry on the CODEX platform, including a 40-plex panel to explore spatial phenotypes in NPC biopsies, expanding the work started during this award -establish the methodology for TCR-seq (and BCR-seq) on FFPE tissues which can be used to further explore T cell responses in archive NPC biopsies
First Year Of Impact 2020
Sector Healthcare
Impact Types Societal