Personalised tailoring of therapy options in various cancer types has become a potential outcome of the development of –omics technologies and associated computational algorithms. Current commercial assays are costly and have limited specificity, failing to identify many breast and oesophageal cancer patients who have low to intermediate risk of recurrence or have a low probability of response to chemotherapy or radiotherapy. In addition clinical trials have failed to identify a set of biomarkers that segregate low or intermediate risk groups in these patients. This present research will address this unmet clinical need by developing a prognostic assay for application in predicting the result of the treatment of any individual breast or oesophageal cancer patient with either radiotherapy, chemotherapy, or adjuvant therapy. The project will utilize Raman spectroscopic imaging data coupled to image segmentation and data mining algorithms that will link the spectroscopic data to measures of treatment outcome in these patients. Ultimately the assay will give an estimate of the probability of tumour recurrence for an individual patient in each of the treatment types. This information will act as a contraindication for cancer treatment option in patients for whom little or no treatment response is likely to be seen, and the potential consideration of other treatment approaches by the clinician. This will have the dual clinical impact of the elimination of waste in the prescription of treatment for patients who have limited chance of response to that treatment option, and the reduction or elimination of the unnecessary toxicity burden in these patients. In addition, as the technology will be label free and will rapidly produce a prediction for an individual patient, this will improve patient throughput and reduce healthcare costs associated with misprescription of treatment.
