Using AI to improve the way we diagnose and treat prostate cancer (VANGUARD PATH study)

Grant information

Reference: MA-TIA24-001
Researchers and Institutions:
Professor Clare Verrill, University of Oxford
Professor Richard Bryant, University of Oxford
Dr Srinivasa Rao, University of Oxford
Professor Jonathan Aning, North Bristol NHS Trust
Award: £1,872,042

What you need to know

• This study will explore how artificial intelligence (AI) can improve prostate cancer diagnosis by helping doctors better understand how aggressive a man's cancer is and how it might behave over time.
• Using a new AI tool, the researchers aim to give patients and doctors clearer, more personalised information to guide treatment choices, potentially avoiding unnecessary side effects or under-treatment.
• The project will test this AI in NHS hospitals to see how well it works in real life, and how patients and doctors feel about using it, paving the way for smarter, more tailored prostate cancer care.

Why are we funding this research?

Prostate cancer diagnosis can be challenging because current methods rely heavily on human judgement and imperfect scoring systems. This has a knock-on effect on treatment decisions, and can lead to either over-treatment, causing unnecessary side effects, or under-treatment, which allows the cancer to progress.

This research aims to solve this problem by using artificial intelligence (AI) to improve the accuracy of prostate cancer diagnosis and treatment planning. The AI can analyse biopsy samples in ways that humans can't, providing more detailed and precise information about the cancer. For example, AI may be able to predict how aggressive the cancer will be based on tiny changes to the cancer cells’ shape.

This helps doctors make better decisions about whether to treat the cancer (and, if so, how) or to monitor it instead, personalising each man's treatment to his specific situation.

The project will test this AI in three NHS sites, ensuring that the technology works effectively in actual clinical settings.

What will the researchers do?

Firstly, the team will make sure that the AI tool (called the ArteraAI Prostate Biopsy Assay) can be rolled out at three NHS sites in Oxford, Bristol, and Glasgow. They’ll check that the tool is compatible with computer systems, can report its results promptly and doesn’t disrupt any existing processes in place. The team will also make sure that its results are consistent across the three sites.

Next, the researchers will test the AI tool on prostate biopsy samples from men who have already been diagnosed and treated, and have at least five years of follow-up data. This will enable the team to compare the AI’s predictions to what actually happened (for example, which cancers are likely to spread rapidly without urgent treatment).

Once they show that the tool can make accurate predictions about which cancers need treating or monitoring, the team will test it out on biopsies from men as they are diagnosed – focusing on particularly difficult cases where it is hard for doctors to decide which treatment is best.

The AI will analyse biopsy samples in ways that humans can't, extracting detailed information about the cancer cells’ appearance and structure, and come up with a risk score that can guide doctors to the best treatment.

Importantly, this AI analysis will be done alongside the usual report prepared by a human and won’t be used at this stage to influence the men’s treatment decisions. But the researchers will compare these decisions to what doctors would have decided if they had used the AI report, and see whether the tool would have helped them in those challenging cases.

The team will also check that the AI tool can be used effectively in clinical settings without causing delays to men receiving their biopsy results or disrupting hospital systems. They'll also investigate any issues that might prevent the uptake of AI in the NHS and how both clinicians and patients respond to this new technology.

Throughout the project, the researchers will gather feedback from men and their clinicians to ensure the AI technology is having a positive impact, and that people are supportive of it. They'll also create training materials and educational information to support the future roll-out of AI within the NHS.

How will this benefit men?

By finding out whether AI can provide more detailed and precise information about a man's cancer, this project could help doctors make better decisions, such as whether to monitor or treat it, and if so, how tailoring the decision to each man's specific needs.

Testing the AI in NHS settings will also help the team make sure that it works effectively in ‘real life’, paving the way for its broader adoption. Ultimately, this could lead to better outcomes for men with prostate cancer, improving their quality of life and potentially saving lives.