The decision to treat a disease through surgery, drugs or other means is not to be taken lightly. A doctor will need to weigh up the patient’s wishes and general health against the potential benefits and side-effects of the of the treatment. There may be some cases where certain treatments are more appropriate than others, but identifying the best course of action is not straightforward as there are often multiple factors to consider.
The primary treatment decision for localised prostate cancer is whether to treat the disease through surgery or radiotherapy, both of which have associated risks and side-effects, or put the patient onto an “active surveillance” programme where the disease is closely monitored. Currently that decision is made based on tumour staging, histopathological evaluation of tissue biopsies (the Gleason grade) and blood biomarkers (the PSA level). These methods are effective at identifying highly aggressive tumours, yet they may not be as accurate for tumours that pose a lesser risk, which is where the majority of prostate cancer diagnoses occur. As a result, both doctors and patients may opt for more aggressive treatment, which may well not be necessary in many cases.
Researchers in the Translational Data Science group
have used advanced data science methods to discover patterns of genetic alterations that provide clues into how a prostate tumour is evolving and can now more accurately determine how advanced the tumour is. We are now working with CRUK to develop AI-driven algorithms to quantify the degree of evolutionary progression and present it in a way that in a way that can be understood by doctors and patients. Preliminary results show that our method provides a significant improvement in risk evaluation when used alongside the current methods. We are ultimately aiming to introduce these findings into clinical practice, enhancing the decision-making process to ensure that each patient receives the most appropriate treatment based on a more precise understanding of their individual tumour.