Prostate test 'predicts death risk'

Behind the Headlines

Wednesday September 15 2010

PSA levels are raised by many factors, not just a tumour (orange)

A blood test at the age of 60 could identify men likely to develop fatal prostate cancer, BBC News has reported.

The well-conducted research behind this news looked at a group of 60-year-old men who were given blood tests to measure their levels of prostate specific antigen (PSA), a protein that can sometimes indicate prostate cancer. Researchers then followed the men for 25 years to look at whether PSA levels were associated with the chance of developing prostate cancer and fatal prostate cancer. Most prostate cancer deaths were in men with the highest PSA levels at 60 years of age, although only a minority of men with raised PSA developed fatal prostate cancer. Men with the lowest PSA concentrations had a lower chance of developing life-threatening prostate cancer.

Importantly, the study’s authors did not conclude that all men should receive PSA screening at the age of 60. There are many issues to consider for any screening test, and PSA screening carries the risk of unnecessary investigation and treatment of small cancers that would not necessarily affect a man’s health or lifespan. Whether PSA screening can save lives is the important question, but further research is needed to provide the answer.


Where did the story come from?

The study was carried out by researchers from Memorial Sloan-Kettering Cancer Center, New York, and other institutions in the US and Sweden. Funding was provided by a number of institutions, including the US National Cancer Institute, the Swedish Cancer Society, the Swedish Research Council and the Sidney Kimmel Center for Prostate and Urologic Cancers. The study was published in the peer-reviewed British Medical Journal.

The media generally reported the findings of this research accurately, but did not examine the bigger issue surrounding PSA screening, namely the uncertain balance of risks and benefits involved. This research contributes to the debate but cannot provide an answer.

The BBC emphasised that a positive test might identify those likely to die from prostate cancer, whereas The Independent also highlighted the researchers’ other conclusion, that a negative test at the age of 60 may identify those men at negligible risk of dying from prostate cancer.


What kind of research was this?

Prostate-specific antigen (PSA) is a protein made by a man’s prostate gland. While there are normally low levels of PSA in a man’s blood, these can be raised by prostate cancer. However, raised PSA levels are more often due to other factors such as inflammation, infection or the harmless enlargement of the prostate seen with age. This means that raised PSA levels do not necessarily indicate cancer.

To further complicate the issue, prostate cancer is not always harmful, and many cases do not have any impact on health or lifespan. Given the uncertainties of PSA test results and the invasive, worrying explorations needed to examine potential prostate cancers, the use of the PSA test is the subject of much debate.

This was a case-control study which examined the relationship between concentrations of PSA in men aged 60 and diagnosis of “clinically relevant” prostate cancer. Clinically relevant means that the prostate cancer could be expected to cause symptoms or shorten a man’s life.

The men were all enrolled in a larger cohort study that had followed them for up to 25 years. In the current study, the researchers carried out a nested case-control analysis, in which they identified men who had been diagnosed with prostate cancer during follow-up and compared them with a sample of men from the larger cohort who had not developed the disease.


What did the research involve?

The research involved a subset of men from a larger cohort study called the Malmo Preventive Project, which sourced participants from the general population of Sweden. This newly reported case-control study involved 1,167 men (71% of this cohort) who provided blood samples and completed medical and lifestyle questionnaires in 1981 at the age of 60 years. They were then followed up to the age of 85.

The aim of this study was not to use the PSA test to screen for prostate cancer. Rather, it aimed to determine a reasonable threshold for the PSA test, evaluating if PSA levels could be used to differentiate between high- and low-risk groups of men who could benefit from closer monitoring or screening.

Diagnoses of prostate cancer were identified through the Cancer Registry at the Swedish National Board of Health and Welfare. The main outcomes that the researchers were interested in were diagnoses of cancer, metastatic prostate cancer (advanced prostate cancer that had spread to distant parts of the body) or deaths from prostate cancer (identified from the Cause of Death Registry).

The nested case-control involved matching three random age-matched control subjects without cancer to each person with one of the three study outcomes (cancer, metastatic cancer or fatal cancer). The researchers looked at the relationship between PSA levels in the men at the age of 60 and the risk of each of the three outcomes.


What were the basic results?

From the entire cohort, 126 men were diagnosed with prostate cancer. Of these, 43 had metastatic cancer. Most of the cancers were diagnosed by urinary symptoms and none had been picked up by routine screening as this is not recommended practice in Sweden. Just over half of those who were diagnosed received some form of treatment for their cancer.

The researchers noted the outcomes in the original cohort of 1,167 participants at the age of 85:

  • 360 (31%) were alive and did not have prostate cancer.
  • 38 (3%) were alive but had prostate cancer.
  • 0 were living with metastatic prostate cancer.
  • 35 (3%) had died from prostate cancer.
  • 668 (57%) had died without prostate cancer.
  • 53 (5%) had had prostate cancer but died from another cause. Eight of these cancers were metastatic.

Concentration of PSA at the age of 60 was related to prostate cancer diagnosis by the age of 85, metastatic prostate cancer and death from prostate cancer. The average PSA concentration across the entire cohort at the age of 60 was 1.06ng per ml of blood. Ninety per cent of the 35 deaths from prostate cancer occurred in men who had the highest concentrations of PSA at the age of 60 (over 2ng/ml). Conversely, men with the lowest PSA levels at age 60 (1ng/ml or less) were unlikely to develop clinically relevant prostate cancer (cancer that spread to other body sites or that ultimately killed them).


How did the researchers interpret the results?

The researchers concluded that concentration of prostate specific antigen at the age of 60 predicts lifetime risk of metastatic prostate cancer and death from prostate cancer. Men aged 60 with low concentrations of PSA are unlikely to develop life-threatening prostate cancer, while men with higher concentrations may be at higher risk.



This case-control study examined how PSA levels at the age of 60 are related to prostate cancer, metastatic prostate cancer and death from prostate cancer up to the age of 85. The study has several strengths, including taking a large representative sample of the general population at the age of 60, matching control patients from the same cohort, following up participants for a long time and using accurate registries to detect cases of cancer and related deaths.

While the aim of this study was to determine a reasonable threshold for the PSA test, the study did not conclude that all men should receive PSA screening at the age of 60. All screening tests have benefits and risks. The risks mostly relate to the chance of producing test results that are false positives (indicating cancer where there is none) and false negatives (missing the presence of a cancer). Raised PSA levels do not necessarily mean that a man has cancer as they can be caused by benign enlargement of the prostate, inflammation or infection. For these reasons, PSA screening is not routinely offered in the UK. 

The results of this study should be interpreted in the correct context:

  • Though the original cohort included 1,167 men, relatively small numbers had prostate cancer (126) and metastatic disease at diagnosis or later (43), or died from it (35). The researchers calculated risk figures for these outcomes in relation to one of four initial PSA readings. The small numbers of people in these subgroups could have decreased the accuracy of the results.
  • The study examined how PSA levels at the age of 60 were linked to development of prostate cancer, metastatic prostate cancer or death from the cancer. PSA is not, by itself, a diagnostic test for cancer and all men who developed prostate cancer were diagnosed when they went to their doctor with symptoms.

Case-control studies can help researchers determine the appropriate threshold to apply to test results, but are not the ideal method for examining screening strategies. There are many other issues to consider for any screening test. PSA screening carries the risk of unnecessary further investigations, detecting small cancers that would not necessarily cause significant symptoms or affect a man’s lifespan, as well as unnecessary treatments and their associated complications.

Whether PSA screening could save lives is the important question, and one which this research alone is unable to answer. Instead, randomised trials are needed for their validity. As the authors themselves conclude: “screening is associated with considerable overdiagnosis, and many men need to be screened to save one life.” Research in this area is likely to continue.

Analysis by Bazian

Edited by NHS Choices

Links to the headlines

'One-off' prostate cancer tests backed for 60-year-olds. BBC News, September 15 2010

Scientists hail one-off test for prostate cancer. The Independent, September 15 2010

Links to the science

Vickers AJ, Cronin AM, Björk T et al. Prostate specific antigen concentration at age 60 and death or metastasis from prostate cancer: case-control study. BMJ, 341:c4521


How helpful is this page?

Average rating

Based on 0 ratings

All ratings

0  ratings
0  ratings
0  ratings
0  ratings
0  ratings

Add your rating