Prostate cancer overdiagnosis risk sharply rises after age 70 – new research

Over the past decade, millions of men without symptoms of prostate cancer have voluntarily undergone a prostate-specific antigen (PSA) test in the UK to find out if they might have prostate cancer.

While research has shown that PSA screening in men aged 50-69 years can reduce cancer deaths, many countries hesitate to recommend or implement formal prostate cancer screening programmes that would offer PSA testing systematically and fairly to all men. The hesitation stems from concern about overdiagnosis and overtreatment.

But our latest research shows that prostate cancer overdiagnosis from PSA screening is mainly a risk for men over the age of 70.

Prostate cancer overdiagnosis occurs when a person is diagnosed with prostate cancer through PSA testing – even though that cancer would not otherwise have been diagnosed within the patient’s lifetime. So had the person not been tested, they might never have known they had prostate cancer.

Overdiagnosis from PSA testing occurs for two main reasons.

The first reason is because PSA tests might identify a cancer that is so slow growing it would never cause problems – even if the man lives to be 100 years old.

The second reason is because a PSA test is able to find prostate cancer a decade or more before it would cause symptoms. Some patients may die from other causes in that time. Had they not been screened, they might have died without ever knowing they had prostate cancer.

Prostate cancer overdiagnosis is a concern because of what follows diagnosis. Subsequent treatment, such as surgery, may lead to harm – including loss of ability to maintain an erection and urinary incontinence.

Had the cancer not been found through screening, the man would not have been treated and would have avoided the side-effects of treatment. Overdiagnosis affects quality of life – and results in costs both to patients and to the healthcare system.

To help men make an informed choice, our research looked at how risk of overdiagnosis changes with age at screening. We found that the risk of prostate cancer overdiagnosis from a PSA test is low in otherwise healthy men in their 50s and early 60s. But this risk sharply increases in men screened from age 70 onward.

First, we looked at long-term data from a large UK trial of more than 400,000 men to examine, over a 15-year period, what proportion of men developed prostate cancer – and whether that proportion differed between those who were screened and those who weren’t.

We found that, on average across all age groups, 12% of prostate cancers were so slow-growing that they would not have caused symptoms or been picked up by a doctor within 15 years of a PSA test. We also found that 88% of prostate cancers detected by PSA tests would, if not treated earlier, cause symptoms and be diagnosed within 15 years – provided the patient lived long enough and did not die of other causes.

We then used national data on men’s deaths in England to understand how many men die from causes other than prostate cancer after a PSA test. Risk of death from other causes within 15 years of a PSA test increases from 10% aged 50, to 49% aged 70 and 89% if aged 80. This steep rise in risk of death drives increased overdiagnosis with old age, because, naturally, the older you are the more likely you are to die from other causes.

Taking these findings together, we projected that there was a 16% chance that the average English man diagnosed with prostate cancer at age 50 from a PSA test would not otherwise have been clinically diagnosed within 15 years. This doubled to 32% for men diagnosed aged 70, and jumped to 58% for men diagnosed at age 80 years.

Essentially, as men age, they are more likely to die from other natural causes before prostate cancer would be detected. For men older than 70 years at screening, screening offers very little, if any, benefit, but carries a high risk of unnecessary harm from overdiagnosis.

It’s also worth noting that health is more than a number based on age. Overdiagnosis risk will be lower for men who are in generally in good health and follow a healthy lifestyle.

It’s important to point out as well that healthcare is evolving. Our findings are based on data from prostate cancer screening done in the UK between 2001 and 2007. Today, doctors use magnetic resonance imaging (MRI) for targeted prostate cancer biopsy in those with an elevated PSA test. This is expected to lower overdiagnosis compared with our estimates by filtering out slower-growing cancers. More significantly, the use of MRI substantially reduces the risk of overtreatment, so the harms of overdiagnosis are smaller than they were 15 years ago.

Two new trials are also evaluating whether such innovations can improve the benefits of screening without increasing the harms.

In the meantime, men without symptoms of prostate cancer who are concerned about their risk have to decide for themselves whether to request a PSA test. For now, our recommendation, as statisticians, is to consider your age before making a decision. But if you do have symptoms, regardless of your age, you should definitely see your GP.

Adam Brentnall receives funding outside this work from Prostate Cancer UK, NIHR, Breast Cancer Research Foundation, MRC, Cancer Research UK. He is a co-applicant on the TRANSFORM trial, and member of the UK National Screening Committee Research and Methodology Group, but this work is independent and not supported by or associated with either role.

Peter Sasieni is a lead investigator on the IMProVE trial, which is investigating whether prostate cancer screening can be improved by combining PSA testing with MRI scanning.

Rhian Gabe receives funding outside this work from Prostate Cancer UK, NIHR, Cancer Research UK and Yorkshire Cancer Research. She is a co-lead on the TRANSFORM trial but this work is independent and not supported by this role.