More than 12,000 men die from prostate cancer in the UK each year, and current biopsies often miss the most dangerous tumours. A new approach called radio-spatialgenomics aims to fix that by linking MRI scans directly to the genetic changes inside the prostate. The problem is that standard biopsies provide only a snapshot of a highly diverse disease. A man’s MRI might show a suspicious area, but doctors cannot tell from the scan alone whether that spot harbours aggressive, fast-spreading cancer or a slow-growing one that does not need treatment. This leads to either over-treatment—with side effects from surgery or radiation—or missed opportunities to catch the dangerous disease early. This project will merge MRI data with spatial genetic profiles from archived prostate tissue, using artificial intelligence to match specific scan features to genetic changes linked to poor outcomes. The researcher will then build a digital tool that measures those MRI patterns and test it against long-term patient records. If successful, the tool could allow doctors to pinpoint exactly which men need aggressive treatment and which can safely avoid it. That would reduce unnecessary side effects for thousands of patients and improve survival for those with truly aggressive disease. The work is a direct step toward a clinical trial that could change how prostate cancer is diagnosed and managed in the NHS.
View original technical description
Prostate cancer is the most common cancer among men in the United Kingdom, and more than 12000 die from it every year. Men with suspected prostate cancer undergo an MRI scan to identify if there are any suspicious areas which warrant a biopsy. One of the main challenges in prostate cancer is precisely identifying men with an aggressive disease that is at high risk for spreading to other parts of the body. Currently, patient-tailored approaches to prostate cancer diagnosis are limited and largely rely on patterns of genetic changes in the biopsy tissue. This offers only a snapshot of a much larger and complex disease process. Using spatial genetic profiling technology to characterise genetic changes that can be mapped back to their location within the prostate, we have gained a better understanding of the highly diverse disease process in prostate cancer. In this project, I aim to combine MRI with spatial genomic data and clinical outcomes using artificial intelligence to identify patterns of MRI changes that can be associated with more aggressive disease. This novel approach is termed radio spatialgenomics. This approach can potentially identify men who have suspicious areas on MRI that harbour more aggressive prostate cancer. These areas can be biopsied to improve diagnostic precision for clinically important prostate cancer and also offer opportunities for targeted interventions such as focal therapy and radiotherapy. Improving the accuracy of identifying aggressive prostate cancer, can better identify men suitable for more invasive treatment such as surgery. This can reduce the risk of over-treatment and side effects in others with low-risk disease. In phase 1 of this project, I aim to merge MRI with genetic data. For this, I will use prostate tissue that has been surgically removed, archived and profiled for spatial genetic changes. Following exposure to preservation chemicals, the prostate tissue undergoes structural changes. I will use artificial intelligence techniques to address these changes when comparing the prostate images before and after exposure to the chemicals. Ultimately, this will allow a prostate region within the MRI scan to be matched to underlying genetic changes. In phase 2 of this project, I will identify patterns of MRI changes in areas harbouring genetic changes associated with poor outcomes. Using computational modelling I will develop a digital tool to measure and assess MRI changes associated with aggressive disease. In phase 3, I will evaluate the performance of the digital tool developed in phase 2 in patient cohorts with MRI and long-term clinical outcomes. My studies will provide valuable information on MRI changes in prostate cancer and how they can be correlated with underlying biology. This can improve how we provide personalised or tailor-made risk assessment when we counsel men newly diagnosed with prostate cancer. This project will be an important step towards setting up a clinical trial to evaluate the developed digital tool in guiding decision-making.
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