Scientists can predict the chances of prostate cancer returning up to a decade before it does, representing a huge potential increase in survival rates.
Approximately half of men who develop locally advanced prostate cancer will subsequently experience a recurrence of the disease, often with a poor prognosis.
But a new test, which combines genomic sequencing and artificial intelligence (AI), can identify those most at risk at the time of first diagnosis.
Knowing who is most likely to have the disease return means doctors can treat them more aggressively, giving them the best chance of stopping the cancer from coming back.
Experts believe it could be widely used in as little as three years following larger trials, saving the lives of thousands of men in the future.
A new test, combining genomic sequencing and artificial intelligence (AI), can identify men at higher risk of prostate cancer recurrence when they are first diagnosed
In the UK, more than 52,000 men are diagnosed with prostate cancer every year, making it the most common cancer in men. Around 12,000 men die from the disease every year, the equivalent of one every 45 minutes.
Prostate cancer is the most common cancer in men and the second most deadly, responsible for around 12,000 deaths a year in the UK.
There is no test reliable enough for a national screening programme to detect cases early, meaning survival has stagnated.
The disease can be difficult to treat because of its great heterogeneity: significant differences between cancer cells within each tumor and from one patient to another.
It often develops in more than one place within the prostate and produces two or more tumors, making it difficult for doctors to decide on the best treatment.
This can range from a “watch and wait” strategy to radiation therapy, hormone therapy, surgery, or a combination of treatments.
Researchers at The Institute of Cancer Research in London and The Royal Marsden NHS Foundation Trust have developed an artificial intelligence tool to assess biopsies taken at the time of diagnosis.
They used it to analyze 1,923 samples from 250 patients in an earlier clinical trial, focusing on the spatial structure of tissue from cancer biopsies.
They also used a specially designed artificial intelligence technique to perform the Gleason grading, a scoring system that grades cancerous tissue from one to five based on the pattern of its cells.
At the same time, the researchers assessed genetic differences between cells within individual tumors, using 642 biopsy samples from 114 participants in the same radiation therapy trial.
This allowed the scientists to evaluate information about the cells’ genomics and morphology at the same time, to see how both affected the chances of their return.
The long-term nature of the trial meant they could analyse it against patient outcomes over more than a decade, vital in prostate cancer, which can be slow-growing.
They found that the greater the genetic difference between cells (and the greater the difference between the shape, size and structure of cancer cells within a tumor), the greater the likelihood of relapse.
Professor David Dearnaley, emeritus professor at the ICR and retired consultant clinical oncologist at The Royal Marsden, said the findings could have a significant impact on patients with traditionally poor prognoses.
“It’s really very exciting for the future,” he said.
‘We were able to define that the worst 20 percent of patients actually did very poorly compared to the rest, which is exciting because it means we’ve potentially defined the group where treatment can be intensified.
‘There are various forms of treatment that are currently possible and if we can improve the treatments, we can improve the cure rates.
“Until now, we have not been able to separate out patients who are at the highest risk of recurrence, but our new analyses could change this by significantly improving our ability to predict whether the cancer will return.”
The study, to be published in Nature today (Friday), found that the more differences there were between the shape, size and structure of cells, the more likely the cancer was to adapt and survive. This “evolvability” was therefore a strong predictor of recurrence.
The researchers were also able to identify a subgroup of patients who experienced a recurrence of the disease in half the time compared to the rest of the patients.
More than 52,000 men are diagnosed with prostate cancer each year in the UK and 299,000 in the US.
There was also a correlation between the loss of a specific chromosome and a reduced presence of immune cells in the tumor, which may affect the response to certain treatments and guide clinical decisions.
Andrea Sottoriva, professor of Cancer Genomics and Evolution at the ICR at the time of the research and now based in Milan, Italy, said the findings had “never been demonstrated before”.
He added: “By applying a computational approach to multiple data sets, we have been able to decipher some of the dynamics of cancer progression and treatment resistance.
‘This type of research is key to improving our understanding of how and when to treat cancers, including prostate cancer.’
This comes as a landmark prostate cancer screening trial is underway in the UK, likely to be followed by a nationwide screening programme.
Transform will test the effectiveness of diagnostic techniques, including MRIs, against the current standard, the prostate-specific antigen (PSA) blood test.
The Mail has campaigned to improve prostate cancer outcomes for more than two decades.
The researchers will now conduct trials on a larger, more diverse group of men, but anticipate that technology like this will become routine.
Dr Hayley Luxton, Senior Director of Research, Impact and Intelligence at Prostate Cancer UK, said: “It’s devastating to hear that prostate cancer has returned after being given the all clear. Because every man’s cancer is unique, it’s a huge challenge for doctors to predict whether this will happen and to come up with the best treatment plan to prevent it.
‘This project is fascinating because the researchers combined evolutionary studies and AI to develop innovative tools that could be used to identify men whose prostate cancer is likely to return quickly, allowing doctors to combat it by providing the most effective personalized treatment for those men.’
Ends