Role of Artificial Intelligence in prostate cancer

Scientists at the University of East Anglia used artificial intelligence (AI) to discover a new type of aggressive prostate cancer that could transform how the disease is identified and treated in the future.

A study funded by Cancer Research UK and released today indicates that prostate cancer, which affects one in every eight men in their lifetime, is classified into two subtypes known as evotypes.

An international team made the discovery by applying AI to DNA data to uncover two distinct prostate subtypes.

The researchers believe that their results will save thousands of lives in the future and change the way prostate cancer is detected and treated. Ultimately, it might provide individualized medicines to each individual patient based on a genetic test, which will also be supplied by artificial intelligence.

The groundbreaking study included experts from the University of Oxford, the University of Manchester, the University of East Anglia, and the Institute of Cancer Research in London, as well as extra funding from Prostate Cancer Research.

Prof. Colin Cooper of the University of East Anglia’s Norwich Medical School stated, “Prostate cancer is the most common cancer in men and kills one man every 45 minutes in the UK. But while prostate cancer is responsible for a large proportion of all male cancer deaths, it is more commonly a disease men die with than from.

“For many years, we have been researching what causes some prostate cancers to become more aggressive than others. But it is only now, with advancements in artificial intelligence, that we have been able to show that there are actually two different subtypes at play.”

This means that men can avoid unneeded treatment, avoiding side effects like incontinence and impotence.

Cooper said, “This is really important because, until now, we thought that prostate cancer was just one type of heterogenous cancer. It explains why some men experience a more aggressive disease. We think that these patients are more likely to have what we have called the ‘alternative form’.

“The two different forms of the disease may be completely different, and this new classification will shape everything in our understanding– from what causes it, to how it is treated.

“We hope that our findings will not only save lives through better diagnosis and tailored treatments in the future, but it may help researchers working in other cancer fields better understand other types of cancer too.”

According to Cancer Research UK, prostate cancer is the most frequent cancer in men in the UK, accounting for approximately 52,000 occurrences per year.

Dr. Rupal Mistry, the charity’s senior science engagement manager, stated, “The work published today by this global consortium of researchers has the potential to make a real difference to people suffering from prostate cancer.”

“The more we understand about cancer, the better chance we have of developing treatments to beat it. We are proud to have helped fund this cutting-edge work, which has laid the foundations for personalised treatments for people with prostate cancer, allowing more people to beat their disease.”

Lead researcher Dr Dan Woodcock of the University of Oxford’s Nuffield Department of Surgical Sciences stated, “Our research reveals that prostate tumors grow along numerous paths, resulting in two distinct disease types.

“This understanding is pivotal as it allows us to classify tumours based on how the cancer evolves rather than solely on individual gene mutations or expression patterns.”

The researchers collaborated as part of The Pan Prostate Cancer Group, an international cooperation founded by scientists from the Institute of Cancer Research (ICR) and the University of East Anglia, to evaluate genetic data from thousands of prostate cancer samples from nine nations.

According to Professor Daniel Brewer of the University of East Anglia’s Norwich Medical School, the partnership brought together scientific skills in mathematics, computing, oncology, clinical care, and histopathology.

He added, “It brought many scientists, maybe 60 scientists or so, together to carry out the research.

“Within that consortium Norwich had a very important role in organising the research and also carrying out computing and mathematical studies on the data.”

Crucially, the team’s cooperation with Cancer Research UK (CRUK) intends to create a genetic test that, when combined with traditional staging and grading, can provide a more exact prognosis for each patient, allowing for more targeted treatment decisions.

The researchers used artificial intelligence (AI) to analyze changes in the DNA of 159 prostate cancer samples (by whole genome sequencing).

They used neural networks, an AI approach, to identify two unique cancer groupings among these patients. These two categories were confirmed using two more mathematical methodologies applied to various features of the data. This conclusion was confirmed in additional independent datasets from Canada and Australia.

They then combined all of the data to create an evolutionary tree that depicts how the two subtypes of prostate cancer develop, eventually resulting in two unique disease kinds known as ‘evotypes’.

Professor David Wedge of the Manchester Cancer Research Centre, who led the study, stated, “This realization is what allows us to identify between disease kinds. This has never been done previously since it is more complex than HER2+ in breast cancer, for example.

“This understanding is pivotal as it allows us to classify tumours based on their evolutionary trajectory rather than solely on individual gene mutations or expression patterns.”

Dr Naomi Elster, Director of Research at Prostate Cancer Research, stated, “We simply don’t know enough about what a prostate cancer diagnosis means at this time – there are many men who have disease that is or may become aggressive, and being able to treat aggressive disease more effectively is critical.” On the other hand, there are far too many guys who suffer from the side effects of cancer treatment that they may not have required.

“These results could be the beginning of us being able to take the same ‘divide and conquer’ approach to prostate cancer that has worked in other diseases, such as breast cancer.”

Professor Ros Eeles, Professor of Oncogenetics at The Institute of Cancer Research in London and Honorary Consultant in Clinical Oncology and Cancer Genetics at The Royal Marsden NHS Foundation Trust, stated, “This study has used the enormous genomic dataset from The Pan Prostate Cancer Group – a powerhouse of information about prostate cancer from around the world.”