Around 300,000 people are diagnosed with oral squamous cell carcinoma each year, and if the cancer spreads to lymph nodes in the neck, five-year survival drops by more than half. The problem is that cancer stem cells—a sub-population of tumour cells that can switch between a migratory and a proliferative state—drive this metastasis. Their behaviour is heavily influenced by the tumour microenvironment, the complex mix of cells and signals surrounding a tumour. But studying these interactions in living humans is extremely difficult, so researchers rely on mouse models that do not fully replicate the human tumour environment. This project replaces those mouse models with lab-grown human tumour models that mimic the microenvironment’s complexity. The team will analyse patient samples to validate the models, then use them to identify the molecular crosstalk that triggers metastasis. If successful, the work will provide an open-access resource for the broader cancer research community, offering a human-relevant alternative to animal testing. The deeper understanding of how the tumour environment controls cancer stem cell plasticity could also reveal new therapeutic targets for preventing metastasis—not just in oral cancer, but in solid tumours more broadly.
View original technical description
Oral squamous cell carcinoma (OSCC) is one of the top ten cancers worldwide, with around 300,000 cases diagnosed annually. The most reliable prognostic factor for OSCC patients is metastatic spread to the lymph nodes of the neck, which is associated with an over 50% reduction in five-year survival. Cancer stem cells (CSCs) are a tumour cell sub-population that have tumour initiating capacity and the plasticity required for the phenotype switching which drives tumour metastasis. CSCs exhibit hybrid epithelial-mesenchymal characteristics and can switch between migratory mesenchymal and proliferative epithelial phenotypes. This triggers local invasion and intravasation, leading to tumour spread, followed by establishment of a secondary tumour at a metastatic site. Understanding this metastatic cascade is critical for developing better therapeutic options and driving improved prognosis for OSCC patients. Interactions with the tumour microenvironment (TME) strongly influence this phenotypic plasticity that drives tumour spread, and present new therapeutic avenues where TME modulation may prevent tumour metastasis. Identification of essential molecular cross-talk between hybrid CSCs and the surrounding TME is now needed and, due to the complex multi-factorial nature of the in vivo human TME, this needs to be approached within an experimental system that reproduces the key aspects of this complex TME. Current approaches use xenograft and syngeneic mouse tumour models, and we propose to replace these with in vitro tumour models possessing human TME complexity. This builds on our established expertise, and presents distinct advantages over mouse models, including a human TME and the ability to analyse dynamic interactions in real-time. We will develop these models in combination with in-depth analysis of human patient samples and data, which we will use to validate our new models and support their dissemination and uptake. Functional characterisation of mechanisms of tumour-TME crosstalk controlling cancer metastasis will provide new targets whose therapeutic modulation may prevent tumour metastasis, and further advance the value of these new models to the research community. The outcomes of this study will be curated and disseminated as an open community resource, which researchers will be able to interrogate for their own studies. Alongside dissemination of our new in vitro models, this will provide a critical resource to the metastasis research community, both in OSCC and solid tumours more broadly, and will provide an alternative to in vivo models, which are currently the most common method for investigating interactions with the TME.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know