Prostate cancer genomes reveal drivers beyond protein-coding genes
Whole-genome analysis of 961 primary prostate tumours identified recurrent changes in coding genes and the regions that regulate them. Some DNA rearrangements brought regulatory elements close to cancer-promoting genes, while genomic groupings were associated with different disease outcomes. The findings expand candidate mechanisms and treatment targets, but do not demonstrate that choosing therapies from these patterns improves patients’ health.
Science··Night
Regulatory DNA broadens the driver map
Changes that help a prostate tumour develop need not sit inside a protein-coding gene. The analysis in Nature Communications examined 961 primary tumours across their whole genomes, combining several kinds of DNA alteration. At least one recurrent driver or candidate driver appeared in 97 per cent of the samples. OICR, which co-led the research, describes the same genome-wide search.[1], [2]
The team prioritised variants supported by two computational workflows and identified 53 recurrent alterations present in at least one per cent of samples. Copy-number changes and structural rearrangements expanded the map beyond ordinary gene panels. Different changes affecting the same gene could nevertheless have different biological effects, so a gene’s name alone did not capture the full pattern.[1]
Rearrangements can bring genetic switches near cancer genes
One candidate mechanism involved regulatory elements normally associated with solute-carrier genes. Rearrangements around SLC30A4 and SLC45A3 placed these elements near cancer-related genes. Gene-expression and chromatin measurements supported increased activity in some of those genes. The combined evidence narrowed possible mechanisms, while leaving individual candidate drivers to be tested more fully through functional experiments.[1]
Patterns across the alterations also separated four genomic tumour groups associated with different clinical outcomes. Such grouping may help researchers study why some cancers spread. It remains an association drawn from sequencing and follow-up, rather than proof that changing treatment according to the groups improves survival or avoids unnecessary treatment.[1], [2]
Drug matches remain candidates for validation
Comparisons with drug databases identified potential treatment targets, including evidence from other cancer types. Those matches do not establish efficacy or approval for prostate cancer. Missing clinical annotations, variation within tumours and a cohort dominated by European ancestry constrain interpretation. Independent testing across broader populations is needed before the proposed drivers or genomic groups can reliably guide routine clinical decisions.[1]
