Exciting new findings from a landmark international collaboration co-led by OICR were published in Nature.
Prostate cancer leaves behind genetic “fingerprints” that provide clues about how aggressive it will become and could help personalize cancer treatment in the future, according to groundbreaking new research co-led by the Ontario Institute for Cancer Research (OICR).
About one in eight men in Canada will be diagnosed with prostate cancer in their lifetime. While some prostate cancers are relatively harmless and patients can live many years without symptoms, others will spread aggressively and become life-threatening.
In a study published today in Nature, an international team of researchers analyzed samples from nearly 1,000 men with prostate cancer. Rather than focusing only on cancer-causing genes, the researchers used high-powered whole-genome sequencing and machine learning tools to examine genetic mutations across the genome. Different biological processes leave different patterns of mutations, creating distinct genetic “fingerprints”. They then determined how those fingerprints were associated with tumour characteristics.
They discovered eight distinct mutational ‘fingerprints’ left behind by different cancers that helped explain how those cancers developed. Four of those fingerprints were linked to a high risk of cancer spreading.
“Knowing which tumours are most likely to spread and become deadly would be a significant advantage when planning treatment for prostate cancer,” says Dr. Jüri Reimand, an OICR Principal Investigator, Associate Professor in the University of Toronto’s Departments of Molecular Genetics and Medical Biophysics and one of the senior authors of the study. “Patients with the highest-risk tumours could get more intensive treatments, while others could be spared from potentially unnecessary treatments and side effects.”
The study is the first publication from the Pan Prostate Cancer Group (PPCG), a large-scale international collaboration co-founded by The Institute of Cancer Research and the University of East Anglia in the UK, that is studying molecular data from 2,000 prostate cancer patients across nine countries.
The Nature publication was co-led by Dr. Kevin Cheng of Reimand’s lab, which develops and applies machine learning methods to analyze large cancer datasets. Reimand and Cheng collaborated with Dr. Joachim Weischenfeldt of the University of Copenhagen and Rigshospitalet in Denmark, Dr. Geoff Macintyre of the Centro Nacional de Investigaciones Oncológicas in Spain, Dr. Andreas Gruber of the Universität Konstanz in Germany and colleagues from around the world.
“We have effectively created a map of the biological processes that drive prostate cancer,” says Dr. Joachim Weischenfeldt of University of Copenhagen and Rigshospitalet, another of the study’s senior authors. “This opens an entirely new field for understanding the underlying biology and for developing new treatments.”
Researchers will now investigate whether these mutational fingerprints can be used to classify prostate cancers based on how aggressive they are likely to become. If the classification system is validated in clinical studies, it could one day become a valuable tool to shape treatment decisions for patients with prostate cancer.
The study also has broader implications. Researchers also found that some fingerprints were associated with improved response to chemotherapy in an independent patient cohort. The same approach could potentially be applied to other cancers as well, helping identify the most aggressive breast, ovarian and other tumour types.
“This new way of understanding how cancer genomes evolve gives us another layer of information that can potentially be used to make sure each cancer patient gets the right treatment at the right time,” Reimand says.
This story informed in part by publications by the The Institute of Cancer Research and the University of Copenhagen and Rigshospitalet
