Publication: integrated signatures define mutational processes in prostate cancer

“Integrated signatures define mutational processes in prostate cancer” is published in Nature. It is the first major publication from the Pan Prostate Cancer Group (PPCG), an international consortium with molecular data from more than 2,000 prostate cancers across nine countries. The study was co-led by Andreas Gruber, André Olsen, Barbara Hernando and Dr. Kevin Cheng (Reimand lab) as co-first authors, with Geoff Macintyre (CNIO), Jüri Reimand (OICR) and Joachim Weischenfeldt (University of Copenhagen and Rigshospitalet) as co-senior authors. Diogo Pellegrina (Reimand lab) also contributed.

Rather than focusing only on cancer genes, the team used whole-genome sequencing of almost 1,000 prostate tumours, the largest whole-genome study of prostate cancer to date, and machine learning to read the mutational “fingerprints” that different biological processes leave across the genome. Key findings:

Eight integrated signatures account for about 85% of prostate cancers.
Four of these were linked to a higher risk of the cancer spreading, particularly in tumours driven by male sex hormones.
Errors during DNA replication appear to underlie about one third of tumours, a much larger role than previously recognised.
Other signatures were associated with age and ancestry, including patterns more common in Black men, who face a higher risk of developing and dying from prostate cancer.
Some signatures were associated with better response to chemotherapy in an independent patient cohort.

The next step is to test whether these fingerprints can classify prostate cancers by how aggressive they are likely to become, and guide treatment decisions once validated in clinical studies. The same approach could also apply to other cancers, such as breast and ovarian cancer.