Artificial Intelligence Identified 2 Evolutionary Pathways of Prostate Cancer

ReachMD Healthcare Image

03/20/2024

According to Cancer Research UK, prostate cancer affects one in six men during their lifetime and is the second most common cause of cancer deaths, with just over 12,000 deaths every year from 2017 to 2019 in the UK. Thus, early and accurate diagnosis of prostate cancer is important to reduce the number of cancer deaths. In research published in February 2024 in Cell Genomics, Woodcock and colleagues provided a breakthrough in the classification of prostate cancer. Through the utilisation of artificial intelligence (AI), the team found that prostate cancer could be divided into two different subtypes, known as ‘evotypes,’ based on differences in how their tumours evolved. This gives researchers a new method for describing prostate cancer epidemiology, and this discovery has the potential to lead to novel developments in prostate cancer’s diagnosis and treatment based on differing characteristics of the two evotypes. Additionally, the publication illustrates the positive potential use of AI in diagnosing patients for prostate cancer.

The study, funded by Cancer Research UK and Prostate Cancer Research, analysed the genetic information of 159 patients who were treated in hospitals located in London, Cambridge, Oxford, and Shanghai. Biological samples of the patients’ prostate tumours were collected, and their genetic information was analysed via whole genome sequencing alongside matched controls. The AI method, known as a neural network, was then used to classify the tumours of the patient based on their different evolutionary pathways. Of the 159 tumors, 125 were classified as ‘canonical’ tumors while only 34 were classified as ‘alternative’ tumors. These results were then further validated by two other classification methods that assessed other aspects of how the tumours evolved. The fundamental difference between the two evotypes relates to a protein called androgen receptor whereby in the alternative evotype, genes for this protein are mutated, while for the canonical evotype they are not.

As mentioned, the categorisation of prostate cancer by evotype holds great promise in aiding its diagnosis and treatment, as evotypes may be associated with different characteristics and disease outcomes. For example, the publication inferred those of non-white ethnicity had increased incidence of tumours displaying characteristics associated with the alternative evotype, and therefore suggested those of non-white ethnicity may be more likely to develop this evotype compared to the canonical evotype. It is well known that wide ethnic disparities exist regarding prostate cancer incidence, with those of Black ethnicity being especially susceptible to the disease. Therefore, acknowledgement of these evotypes may prove useful in bridging the gap of this health inequality.

Furthermore, the identification of new methods for understanding prostate cancer is becoming ever more crucial as prostate cancer incidence increases. Prostate cancer cases are steadily increasing in the UK, and GlobalData epidemiologists forecast the incident cases for men aged 30 years and older living in the UK will increase from 57,000 cases in 2024 to about 60,000

in 2028. Additionally, GlobalData forecasts estimate the five-year diagnosed prevalent cases will increase from almost 200,000 cases in 2024 to just over 215,000 cases in 2028. Therefore, as the burden of prostate cancer will continue to rise, AI’s contribution could be crucial in the diagnosis and treatment of prostate cancer.

How well do you really know your competitors?

Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.

Company Profile – free sample

Thank you!

Your download email will arrive shortly

Not ready to buy yet? Download a free sample

We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form

By GlobalData
Register

We're glad to see you're enjoying Rheumatology Academy…
but how about a more personalized experience?

Register for free