Project Leader: Dr. David Thomas
Peter Mac Laboratory Head: Prof. D Bowtell.
When curable, sarcomas are often treated with intensive radiotherapy before surgical resection. Radiotherapy has toxic side-effects. The aim is to utilize genomic and gene expression profiling for prediction of patient’s response to radiotherapy. In the first stage of experiments the tissue will be collected (via biopsy) and then subjected to irradiation in the laboratory conditions (ex vivo). The irradiated samples will be hybridized to cDNA arrays against corresponding non-irradiated tissue, and the gene expressions will be observed. Preliminary studies using cell lines will establish the optimal time-course, radiation dose, and sensitivity and specificity of the use of arrays to assay a transcriptional response to irradiation. In parallel, the CGH profile of the whole genome as well as the expression profile within the primary tumour will be collected. Then these datasets will be used, independently and once integrated, for a) creation of predictive models by application of supervised learning algorithms and b) identification of critical pathogenetic genes from correlations between genomic and gene expression profiles.
The primary goal is the discovery of a compact set of markers which can be used for development of test suitable for practical applications. The second goal is to understand the molecular basis for radiation responsiveness in tumours, which might have some additional application for this ubiquitous form of cancer treatment.