Artificial intelligence reveals prostate cancer is not just one disease

Key Points:

  • Prostate cancer has two different subtypes, or evotypes, that evolve along multiple pathways.
  • The use of artificial intelligence (AI) helped identify these distinct cancer groups.
  • The findings could revolutionize how prostate cancer is diagnosed and treated in the future, potentially leading to better tailored treatments.


Research published in Cell Genomics has unveiled a groundbreaking discovery about prostate cancer, a disease affecting one in six men in the UK during their lifetime. This study, conducted by an international consortium including researchers from the University of Oxford, the University of Manchester, the University of East Anglia, and the Institute of Cancer Research, London, utilized artificial intelligence (AI) to reveal the presence of two distinct subtypes of prostate cancer―dubbed evotypes. Lead researcher Dr. Dan Woodcock emphasized that understanding these evolutionary pathways is crucial to classifying tumors and developing targeted treatments, moving away from solely focusing on individual gene mutations.


By analyzing the DNA of prostate cancer samples from 159 patients through whole-genome sequencing and employing AI techniques such as neural networks, researchers were able to identify and confirm the existence of two separate cancer groups. This groundbreaking finding is supported by subsequent mathematical analysis and has been validated in independent datasets from Canada and Australia. The culmination of this research is an evolutionary tree illustrating the development of the two evotypes of prostate cancer, converging into distinct subtypes.


This study signifies a paradigm shift in our understanding of prostate cancer, challenging the belief that it is a homogenous disease. Professor Colin Cooper from UEA’s Norwich Medical School highlighted the significance of this discovery, emphasizing that the application of artificial intelligence has unveiled the existence of these previously unrecognized subtypes. The implications of this research extend beyond prostate cancer, offering insights that could potentially benefit research in other cancer fields as well.


Looking ahead, these findings have the potential to revolutionize the diagnosis and treatment of prostate cancer. By collaborating with Cancer Research UK, the research team aims to develop a genetic test that, in conjunction with conventional diagnostic methods, can provide personalized prognoses for patients, informing tailored treatment strategies. Ultimately, this research paves the way for a more precise and targeted approach to managing prostate cancer, with the promise of improving outcomes and saving lives in the future.



Prompt Engineering Guides



©2024 The Horizon