How to improve the diagnosis of prostate cancer

A scientist inside a droplet of blood shines a torch revealing RNA strands and circles denoting molecules
A scientist inside a droplet of blood shines a torch revealing RNA strands and circles denoting molecules

Credit: Sèbastien Thibault

The search for a new way to diagnose prostate cancer preoccupies many urologists — and for good reason. Screening currently relies on a blood test that looks for the prostate-specific antigen (PSA), but there are serious doubts about the test’s accuracy.

A systematic review and meta-analysis in 2018 looked at data gathered from more than 700,000 men across 5 separate trials, and tried to assess the impact of screening for prostate cancer using a PSA test1. It found a large number of false-negative results: roughly 15% of men who had a negative PSA test were subsequently diagnosed with prostate cancer within 7 years. Prostate tumours are typically slow growing, so this probably means that many cancers are not being detected by the PSA test, squandering the opportunity to start managing the disease early.

The review found even higher rates of false positives: around two-thirds of men with an initially positive PSA test went on to have a negative tissue biopsy. Other studies have shown similar error rates. This is one of the biggest concerns about the current system of screening.

Prostate biopsies are the natural next step after a positive PSA test, but they are invasive and can cause a host of complications. “We usually take 32 tissue cores for a biopsy,” explains Nathan Lawrentschuk, a urological surgeon and oncologist at the Royal Melbourne Hospital in Australia. “That means you stick a needle in 32 times and each core has one to two centimetres of tissue. You can imagine that’s not pleasant.”

There are two main schools of thought on how to replace PSA screening. The first is to look for better biomarkers in the blood or urine, and the second eschews testing samples altogether in favour of sophisticated imaging techniques. Whichever approach wins out, better screening with fewer false positives should mean fewer patients undergo needless biopsies.

Researchers say that tests should also try to stratify risk. “What you want to do is catch the aggressive cases, but there’s no good way of doing that,” says Kevin Koo, head of assay development at biotechnology company Xing Technologies in Brisbane, Australia, which specializes in diagnostics. Many prostate cancers could be safely left alone, but with a confirmed diagnosis comes the temptation to treat — and as with a biopsy, these treatments are not to be entered into lightly. Studies have shown that 17% of men who undergo surgery for prostate cancer experience urinary incontinence, and 83% suffer from erectile dysfunction1. For those treated with radiation, the figures are 4% and 73%, respectively1.

“The specificity of PSA is shocking,” Koo says. “You cast a big net and hope for the best.” A better biomarker that could both accurately confirm a diagnosis of prostate cancer and tell doctors something about the risk of it progressing could help to prevent overtreatment and preserve people’s quality of life.

Biomarker hunting

As scientists look for alternatives to PSA, they are keen to ensure they don’t lose one of PSA’s few advantages: its simplicity. No one wants to replace a non-invasive blood test with something more uncomfortable for patients, more difficult to perform or more likely to cause side effects. That’s why the bulk of these efforts are focused on finding biomarkers in samples of blood or urine.

Exosomes are one such candidate. These nano-sized capsules are secreted by almost every type of cell in the body and travel in the extracellular matrix. Their cargo is often distinct to the cell they originated from, meaning that exosomes produced by prostate cancer cells could be easily identified.

When exosomes were discovered in the 1980s, scientists were unsure what to make of them. “It was first thought they were basically garbage excretions, but we now know exosomes can be a cell-to-cell communication tool,” says Ayuko Hoshino, a molecular biologist at the Tokyo Institute of Technology who studies how exosomes can influence the spread of a cancer.

A coloured TEM of cancer-derived exosomes

Cancer-derived exosomes (pink).Credit: A. Nasiri, T. Ogawa and A. Hoshino

Cells manufacture and export exosomes for a variety of reasons, one of which is in response to inflammation — they allow immune cells to talk to each other and coordinate their defence against infections. Long-term inflammation can contribute to the formation of prostate cancer, which further bolsters the credentials of exosomes as a potential biomarker.

A 2019 study2 characterized the exosomes of eight Black men and eight white men with prostate cancer, as well as those of a corresponding control group made up of healthy participants with a similar racial balance. Researchers isolated exosomes from blood samples, quantified them and examined the proteins they contained.

The exosome concentrations in the samples from people with prostate cancer were around 3.5-fold higher than in those without cancer. However, some of their contents differed between Black and white men. Researchers found 7 proteins that were exclusively present in the exosomes of Black men with prostate cancer, and 33 that were only in the exosomes of white men with the disease.

Despite its small size, this study showcases the importance of considering different ethnic groups in the quest to find new ways to screen for prostate cancer. The current PSA test is not particularly accurate for any ethnicity, but it is especially problematic for Black men, because they typically produce significantly lower quantities of PSA and so are less likely to meet the threshold of concern. Black men also have a higher risk of developing prostate cancer and are more than twice as likely to die from the disease than are white men.

Exosomes show promise as a biomarker, but scientists are also investigating microRNA (miRNA). These are small, single-stranded and non-coding genetic sequences that float freely in bodily fluids.

A 2020 study identified distinct miRNA signatures in the blood of people with benign, localized and advanced prostate cancer3. By examining the concentrations of four different miRNAs, the scientists could predict the outcome of biopsies with greater accuracy than by using PSA.

Others are exploring miRNAs found in the urinary system. In another 2020 study4, researchers analysed the urinary sediments of 50 men who were about to undergo a prostate biopsy, 26 of whom would later have a positive result. The scientists looked for 12 different miRNAs that have previously been linked to prostate cancer. Concentrations of two of them led to more accurate predictions than using the current PSA test.

Studies such as these have generated excitement that new blood and urine tests will ultimately solve the PSA problem. But molecular biomarkers such as exosomes and miRNAs aren’t the only game in town.

Imaging alternative

“People think we absolutely need to have biomarkers from blood or urine to fix the PSA problem,” says Lawrentschuk. “But before you get a prostate biopsy in Australia you get an MRI [magnetic resonance imaging] scan, and most people have something show up in the imagery there before they get a biopsy.”

Advances in imaging technology mean that scans can now provide valuable information beyond just the location of a prostate tumour. They can also help to determine the severity of a tumour by showing whether it has started to spread. A 2019 review of 98 studies, for example, found that imaging techniques such as transrectal ultrasonography, computed tomography and MRI can detect prostate cancer even when PSA levels are low5.

Machine learning — the use of computers and algorithms that can accurately extract information from large data sets without explicit instructions — is looking particularly promising. Studies have already shown how the technology could be applied to medical imaging, performing image analysis and combining data from different scanning modalities to accurately detect potential cases of prostate cancer.

“I think the advances in imaging are beating biomarkers at the moment,” says Lawrentschuk. But it’s not really a question of one or the other — it could easily be a combination of imaging and molecular biomarkers that will finally replace PSA.

Before a new diagnostic marker is adopted, however, many more studies will be needed to convince regulators and insurance companies, says Paul Mainwaring, who co-founded Xing Technologies. “Delivering change into the health-care system is a slow and challenging process on many levels.”

That sentiment is echoed by Hoshino. The basic science underpinning exosome-based diagnostics is sound, she says, but testing procedures need to be refined to ensure that they truly offer a better proposition than PSA. “We need to do vigorous testing to look for the best panel of exosome biomarkers that would give us high levels of sensitivity and specificity,” she says. The same is true for the other techniques under development — a critical mass of data is now needed to give scientists and doctors confidence that they are improving on PSA.

Until such a body of evidence materializes, it is likely that any new test will supplement, rather than replace, PSA. No doctor wants to misdiagnose a patient because they neglected to use a conventional test.

But Mainwaring is confident the field will get there. “We’re finally starting to see research funded into this question,” he says. Promising alternatives to PSA could do so much to prevent misdiagnosis, and they are beginning to take shape.

Leave a Reply

Your email address will not be published.