Multimodality Early Detection of Head and Neck Cancer Recurrence

The optimal frequency of post-treatment imaging surveillance for head and neck cancer patients who are at risk of recurrence is not well-defined and non-imaging biomarkers for recurrence (required for a risk-stratified imaging surveillance strategy) are lacking particularly in the HPV-negative cohort of patients.

Our goal is to derive an integrated set of imaging-multianalyte based technologies that link studies of tumour cell molecular and immune microenvironment mechanisms to establishing a robust risk model for predicting cancer recurrence.

A new prospective clinical study will be conducted and will expand on the exosomal microRNA and immunological parameter model that we have built within a recent Phase 2 clinical/translational head and neck cancer study, and can delineate a high risk patient subgroup who has significantly lower recurrence free survival compared to that classified as low risk.

We will address the potential clinical utility of peripheral blood sampling of the exosome/immunome, in combination with individual patients’ genome (CtDNA) as a longitudinal surveillance tool to achieve risk-stratified imaging surveillance.

Our multidisciplinary experts will combine our tools/expertise to work on pre-defined biological hypotheses to provide the scientific rationale for an integrated non-invasive diagnostic.

As an example hypothesis, MR elastography which can detect at an early stage, intra-lymphatic invasive form of cancers; will be combined with peripheral blood sampling of specific immune cells which may facilitate cancer metastasis through lymphovascular remodelling or lymphovascular invasion by tumour cells.

We will examine changes in intratumoural and circulating immune cells using mass cytometry (cytometry by time of flight (CyTOF)) and conventional cytometry methods.

We postulate that a combination of deep immunophenotyping by multiplexed imaging of tissues (at baseline), coupled to blood monitoring during treatment when biopsy material may not be obtainable; would be a better way to assess the risk of early recurrence than either modality alone.

For genomics, we will utilise a sequencing approach that incorporates a combination of error-suppression and signal enrichment techniques for sensitive detection of cfDNA.

Where detected, we will correlate the mean variant allele frequency of e.g. p53 mutations in cfDNA and the corresponding exosomal p53 mutant protein level, with the quantification of tissue- and circulating immune cells.

The multivariate model for early cancer recurrence risk detection will be used to risk-stratify patients prior to, or shortly after primary radical treatment.

The potential benefit is for patients at high-risk of recurrence to receive treatment intensification, such as the addition of immunotherapy to chemoradiation.