Interim report for the BAOMS/FSRF-Saving Faces Joint Fellowship
Summary of Research
Oral cancer is the commonest type of head and neck cancer and remains a debilitating and devastating disease. Its incidence is growing in the UK and advances in the management of this disease have made little impact on the overall survival for the condition. The most adverse factor for survival in oral cancer is extracapsular spread (ECS). In this instance the disease spreads from the mouth to lymph nodes in the neck and subsequently spills out from these lymph nodes.
Biopsy tissue was collected from 110 patients with oral cancer at University Hospital Aintree in Liverpool, one of the largest head and neck cancer referral centres in the UK. This tissue was used to determine the important molecular events in the development of ECS. Clinical data from patient follow-up after surgery confirmed the dire consequences of ECS. The findings were comparable with a larger previous study in which less than one quarter of the patients with ECS survived after 5 years.
Our research group has previously demonstrated a genetic expression pattern or signature for RNA, a molecule used to translate the genetic material stored in DNA. This expression signature demonstrates how much a particular or significant sets of genes are turned off or on and could be used to predict the likelihood of individuals having ECS. One of the main aims of the first year of my research was to validate this expression profile. The second aim of the project was to establish cell lines for head and neck cancer to identify differences in cell behavior between patients with ECS and patients without ECS. These cells were isolated because we expected to see differences in the behavior of the cells. Cell lines were created in the laboratory by growing cells from tumour tissue. 3 cell lines from patients with ECS and 2 cell lines from patients without ECS were established. We will examine their behavior to see if there is any difference and try to explain why there is such a difference. These will then be tested with selected interventions such as manipulating targets identified from the gene expression signature. This may give us an insight into the cause of ECS and the factors that are important in the spread of cancers. By identifying differences in these characteristics we hope to be able to identify which patients are at risk of developing ECS. We also hope to unravel the important molecular processes that cause ECS and discover potential targets for future drug therapies.
Oral cancer is the fifth commonest cancer worldwide with over 600,000 cases per year and it has increased in incidence in the UK by 24% over the last decade. It is also the most common type of head and neck cancer. Despite all the advances in treatment of cancer, very little progress has been made in surviving this condition, particularly for patients who present with metastatic disease of the lymph nodes in the neck. This research project focused on the most aggressive form of this disease. This occurs in a subgroup of patients in whom the cancer had spilled out from these metastatic lymph nodes, this is called extracapsular spread (ECS). ECS is only determined after definitive surgical treatment with missed opportunities to influence its management. Diagnosis relies on histopathology with X-ray scans and clinical examination of limited value.
Our laboratory had already carried out RNA expression analysis of tumour tissue obtained from 57 patients. A sample of expressed RNA was obtained from the tumour tissue and was placed on small dots on a slide that contained probes representing the entire expressed DNA in human cells. Genes expressed were able to bind to the probes stuck on the glass. This binding to the probe causes fluorescence that was detected. When comparing the expression to normal tissue an expression pattern was identified with an 8-gene signature for ECS. These findings create the possibility of predicting the presence of extracapsular spread from biopsy of the cancer from the mouth.
Aims of Study
The aims of the study were two-fold. Firstly, to validate the gene signature on a larger series of 93 samples of tumour tissue and secondly, to develop cancer cell lines from patients with ECS. These cells will subsequently be used to identify differences in cell behavior between cancer cells derived from patients with ECS compared to patients without ECS. Confirmed gene expression targets will be manipulated in these cell lines and physical characteristics measured. With the isolation of different cell types from individual tumour samples we also hope to identify the contribution that these different cell types make to the overall properties of cancer. The cell lines will be characterized genetically over time and with antibody staining techniques.
Clinical data including histopathology, clinical outcomes and magnetic resonance imaging (MRI) scan results were obtained from case note review. Frozen tumour tissue from 53 patients was prepared for molecular analysis. Fresh tumour tissue was also collected for primary cell culture in which tissue was placed in growth media in the laboratory and cells that grew out were available for subsequent analysis. The cell lines obtained included different cell type as well as cells from both normal and cancer tissue. They were characterized in terms of their genetic and physical attributes. Gains or losses of critical gene regions or genotyping and antibody tests for key proteins known as immunofluorescence were utilized. Very sensitive and quantitative techniques for detecting gene products called Taqman® assays were also developed to validate the gene expression signature from the initial discovery series. A larger series of 93 tumour samples were identified as having adequate quality for this analysis.
Findings to date
Clinical outcome data for this group of patients has confirmed previous findings that ECS is one of the most significant adverse predictors of survival. The MRI scan data demonstrated how poorly the current scanning techniques were in diagnosing ECS, with only 6.8% of patients accurately diagnosed. 26 samples of tissue were collected for primary cell culture. From these tissue samples 2 cell lines were established from ECS negative patients, 4 cell lines from ECS positive patients and 3 cell lines from normal tissue. Different cell types (keratinocytes and fibroblasts) were also obtained. Genetic analysis revealed some instability in the composition of the DNA in one of the cell lines over time. Corresponding normal and tumour tissue with blood samples have also been obtained for each of the successful cell lines with isolation of DNA for future comparisons.
Gene specific tests (Taqman® assays) have been completed for validation of the original 8-gene expression signature discovered in the initial exploratory samples and await analysis. Physical properties of cell line material such as cell migration will be determined with subsequent analysis of the targets identified from the expression signature. If the physical attributes fit with the expression targets in these cells the genes will then be manipulated with an expectation that this would make a difference to the cells behaviour. The cell line material will also be used to identify the contributions that different cell types make to the overall gene expression for ECS positive and ECS negative patients.
Conclusion and Clinical Implications
Findings from clinical data for the group of patients in this study demonstrated the importance of ECS as a predictor of poor survival. If the gene expression signature is validated in a larger group of patients it could provide diagnostic tools for predicting the presence of ECS as well as identifying potential targets for therapy.
Summary of findings and research fellowship experience
The key achievements to date have been the collection of clinical data confirming that ECS is the most aggressive form of oral cancer. The findings have also demonstrated that current methods of detecting this condition are inadequate. Tissue has been collected to expand the series with preparation of genetic material will be used to validate the 8-gene signature found in previous work. Tissue collected for creation of cell lines from patients with extracapsular spread will add a valuable resource to understanding the biology underlying this disease.