Imma Perfetto
Cosmos science journalist
Globally prostate cancer was the second most frequent cancer, and the fifth leading cause of cancer death, among men in 2020.
It was the most commonly diagnosed cancer in Australia in 2018 (and is estimated to remain so in 2022); a man has a 1 in 6 (or 17%) risk of being diagnosed with prostate cancer by the age of 85.
But not everyone on Earth is similarly affected by the disease, in fact there are significant differences in the severity of prostate cancer across different ethnicities – particularly across sub-Saharan Africa, where mortality rates are 2.7 times higher than global averages.
But is it ancestry, geography, or a combination of the two, that’s causing this variation? To address this question, researchers sequenced the genetics of prostate cancer tumours from South African, Brazilian, and Australian donors.
The results, which have been published in two new studies in Nature and Genome Medicine, identified new prostate cancer subtypes and cancer drivers that can distinguish a patient’s ancestry and predict whether the cancer might become life-threatening.
“Our understanding of prostate cancer has been severely limited by a research focus on Western populations,” says senior author, Professor Vanessa Hayes, genomicist and Petre Chair of Prostate Cancer Research at the University of Sydney’s Charles Perkins Centre and Faculty of Medicine and Health in Australia.
“We found Africans to be impacted by a greater number and spectrum of acquired (including cancer driver) genetic alterations, with significant implications for ancestral consideration when managing and treating prostate cancer,” Hayes says.
Africans’ prostate cancer tumours have more mutations
The researchers sequenced the genomes of untreated prostate cancer samples collected from 183 patients – including 123 South African, 53 Australian, and 7 Brazilian individuals – and were able to identify around 2 million genetic variants (mutations) involved in the cancer.
“What was unique about this study is that we sequenced – it means we read the entire DNA sequence of the tumour and blood – from the Africans and Australians in the exact same batch, everything was done in Australia,” explains Hayes.
“And that was really important, because that meant the samples went through one technical pipeline and one analysis pipeline.”
This was necessary so that the genomic data of all the patients in the study, whether from Australia, Brazil, or South Africa, could be compared – like apples to apples.
“Most Australian men, nine out of ten actually, will die with prostate cancer rather than from prostate cancer. We have no idea what distinguishes that one of the ten Australians on the line-up, so we actually have to look away from Australia to try and understand the context,” says Hayes.
Read more: Prostate cancer breakthrough
And they found significant differences between the tumours of people with African ancestry compared to those from Europe. In Africans, the tumours were more mutated – they had a higher tumour mutational burden.
According to Hayes, this is important because small mutational events are usually not as common in prostate cancer, like they are in melanoma or lung cancer. And, unlike UV exposure with melanoma or smoking with lung cancer, there is no known carcinogenic driver for prostate cancer.
“What we saw in Africans is that the burden of these small changes was higher than in Australians, which raises the idea: is there some carcinogen, some environmental exposure within Africa, which is contributing to aggressive prostate cancer in the region?
“So, if we can identify it, then maybe that is what that one of the ten Australian men were also exposed to in their lifetime.”
New ways to classify prostate cancer subtypes
Using computational data science, the team was able to classify the prostate cancers into four different subtypes called global mutational subtypes (GMS).
“Combining our unique dataset with the largest public data source of European and Chinese cancer genomes allowed us, for the first time, to place the African prostate cancer genomic landscape into a global context,” says Dr Weerachai Jaratlerdsiri, a computational biologist from the University of Sydney and first author on the Nature paper.
Because the patients’ genomes had been sequenced from samples of their blood as well as the tumours, the researchers were able to define their genetic ancestries. Hayes says it’s like doing Ancestry.com but on steroids, because while “Ancestry.com only looks at 600,000 letters across the DNA, we looked at 7 million.”
They identified two cancer subtypes – GMS-B and GMS-D – that were only found in the populations with African ancestry.
They also identified the universal GMS-A subtype (which occurred in all ethnicities) and the GMS-C subtype – seen in people with African ancestry and people with European ancestry. Those with the GMS-C subtype were significantly more likely to die from prostate cancer than the other subtypes, and clinicians will now be able to use this finding as a prognostic marker to determine whether someone might experience poor clinical outcomes.
Five of the South Africans included in the study had European ancestry, but their families had lived in South Africa for multiple generations. Interestingly, one of them had a tumour categorised as a GMS-D subtype, despite this otherwise only having been seen in patients with African ancestry.
The team have now received funding to look at a further 100 Africans with European ancestry, but whose ancestors had lived in Africa for generations, to see whether there is a geographical, environmental aspect that might be contributing to the accumulation of these types of mutations.
Opening up new avenues for treatment
The second paper, published in Genome Medicine, focused on the large and dramatic changes to the genome, called “structural variations”, that prostate cancer is prone to. For instance, parts of the chromosome break off, delete, or insert themselves somewhere else, or the chromosomes shatter and come back together again causing rearrangements.
These are difficult to locate in the genome because scientists have to use computational methods to infer whether these mutations are there or not. But by using multiple different computational tools the researchers were able to identify brand new mutational drivers of prostate cancer – genes not previously known to be involved in prostate cancer.
This opens up new opportunities for treatment, because knowing these drivers allows scientists to design new therapeutic targets or repurpose existing drugs that may already be used to target these genes in other diseases.