NIHR DC Discover

NIHR Signal Single routine offer of a blood test for prostate cancer did not save lives

Published on 12 June 2018

doi: 10.3310/signal-000602

Offering all men aged 50 to 69 a single, screening prostate-specific antigen (PSA) blood test did not prevent deaths from prostate cancer.

This large trial included 573 UK general practices and over 400,000 men. It found that men who were invited to have a PSA test were 19% more likely to be diagnosed with prostate cancer, but no less likely to die from the condition, over an average 10 years of follow up. Forty per cent of men took up the offer.

Controversy over PSA testing has persisted for many years. Two previous trials have had conflicting findings about whether repeated PSA testing reduces prostate cancer deaths. In addition, concerns about test accuracy, over-diagnosis and over-treatment have meant that PSA testing is not routinely offered on the NHS to men with no symptoms.  

The current study supports the idea that one-off PSA screening in this age group is not likely to be of benefit.

Share your views on the research.

Why was this study needed?

Around 47,000 men are diagnosed with prostate cancer in the UK each year, making it the most common cancer among men. Currently, the only biomarker screening test available is a blood test for PSA.

While high PSA levels can be a sign of prostate cancer, PSA can also be raised for other reasons and some of the cancers detected would never have caused the man problems in his lifetime (over diagnosis). Therefore, some men who screen positive receive unnecessary investigations and treatment, which can have both physical and psychological harms. In addition, not all prostate cancers are picked up by the PSA test.

A previous European study of repeated PSA tests in men without symptoms found a reduction in prostate cancer deaths (in contrast to a similar US study) but also discovered that over-diagnosis was common. This trial was funded to investigate whether offering a single PSA test to healthy men could avoid over-diagnosis while still reducing prostate cancer deaths.

What did this study do?

Researchers carrying out the Cluster Randomised Trial of PSA Testing for Prostate Cancer (CAP) randomly assigned GP practices to screening or control groups, before inviting participation. After randomisation, more practices accepted invitations to be in the control group (302; 78% of those invited) than in the screening group (271; 68% of those invited).

Men registered at the screening group practices aged 50 to 69 who had no history of prostate cancer (189,386 men) were invited to have a PSA test. Those who had a PSA level of 3ng/mL or over were offered a transrectal ultrasound-guided biopsy, followed by treatment (including in a separate trial) if cancerous cells were found.

The 219,439 eligible men in the control group were not routinely provided with information about the PSA test but could request one if they wished.

This is the largest ever trial of PSA screening. It used consistent methods of diagnosis and treatment and had a high level of follow up.

What did it find?

  • Of the men offered screening, 40% attended the PSA testing clinic, and 36% had a PSA test. Of the men who were tested and had a valid result, 11% screened positive, and 85% of these men had a prostate biopsy. About 10 to 15% of men in the control group were estimated to have had a PSA test over the study period.
  • There was no difference in rates of prostate cancer death between the men offered screening and the men not offered screening, after 10 years. Prostate cancer mortality was 0.30 per 1,000 person-years for the intervention group and 0.31 for the control group (rate difference -0.013 per 1,000 person-years, 95% confidence interval [CI] -0.047 to +0.022; rate ratio [RR] 0.96, 95% CI 0.85 to 1.08).   
  • More men in the screening group were diagnosed with prostate cancer: 8,054 men in the screening group (4.3%; 4.45 per 1,000 person-years), compared with 7,853 (3.6%; 3.80 per 1,000 person-years) of those in the control group (rate difference 0.65 per 1,000 person-years, 95% CI 0.52 to 0.78; RR 1.19, 95% CI 1.14 to 1.25).
  • The cancers diagnosed in the screening group tended to be smaller and lower grade than those diagnosed in the control group. Tumours in the intervention group were less likely to be high grade (based on Gleason grade; odds ratio [OR] 0.68, 95% CI 0.64 to 0.73) or advanced (based on TNM staging; OR 0.68, 95% CI 0.62 to 0.75).
  • Some eventually-fatal prostate cancers among men in the screening group were not picked up by screening.
    Of the 549 men in the screening group who died of prostate cancer, 188 had attended the PSA testing clinic but only just under a third of them (59 men; 31%) had cancer detected.
    Of 129 men who did not have cancer detected, 42 (33%) did not have the PSA test, while 68 (53%) had the test but screened negative (had low PSA levels) so did not have a follow-up biopsy or treatment. In addition, 15 (12%) screened positive but did not have a biopsy, and four (3%) had a biopsy that showed no cancerous cells.

What does current guidance say on this issue?

The UK National Screening Committee does not currently recommend universal screening for prostate cancer using the PSA test.

The Prostate Cancer Risk Management Programme says: “The PSA test is not accurate enough to meet the requirements of a national screening programme.” It goes on to state that “Any asymptomatic man aged 50 and over can make an appointment with their GP to discuss having the PSA test. GPs should not proactively raise the issue with asymptomatic men.”

What are the implications?

These results add weight to the argument against using a single PSA test to screen asymptomatic men for prostate cancer in the UK. It is possible that repeated testing or screening that was taken up by more men could have benefits, though this study does not offer hope of that.

Further longer-term results from this study are expected and may change the conclusions regarding long-term outcomes. 

Citation and Funding

Martin RM, Donovan JL, Turner EL, et al. Effect of a low-intensity PSA-based screening intervention on prostate cancer mortality. The CAP randomized clinical trial. JAMA. 2018;319(9):883-95.

This study was partially funded by the National Institute for Health Research and by grants C11043/A4286, C18281/A8145, C18281/A11326, and C18281/A15064 from Cancer Research UK.

Bibliography

Cancer Research UK. Types and grades. London: Cancer Research UK; reviewed 2016.

Cancer Research UK. TNM staging. London: Cancer Research UK; reviewed 2016.

MJ Barry. Screening for prostate cancer. Is the third trial the charm? JAMA. 2018;319(9):868-69.

Public Health England. Prostate cancer risk management programme: overview. London: Public Health England; 2015.

UK NSC. The UK NSC recommendation on prostate cancer screening/PSA testing in men over the age of 50. London: UK National Screening Committee; 2016.

Why was this study needed?

Around 47,000 men are diagnosed with prostate cancer in the UK each year, making it the most common cancer among men. Currently, the only biomarker screening test available is a blood test for PSA.

While high PSA levels can be a sign of prostate cancer, PSA can also be raised for other reasons and some of the cancers detected would never have caused the man problems in his lifetime (over diagnosis). Therefore, some men who screen positive receive unnecessary investigations and treatment, which can have both physical and psychological harms. In addition, not all prostate cancers are picked up by the PSA test.

A previous European study of repeated PSA tests in men without symptoms found a reduction in prostate cancer deaths (in contrast to a similar US study) but also discovered that over-diagnosis was common. This trial was funded to investigate whether offering a single PSA test to healthy men could avoid over-diagnosis while still reducing prostate cancer deaths.

What did this study do?

Researchers carrying out the Cluster Randomised Trial of PSA Testing for Prostate Cancer (CAP) randomly assigned GP practices to screening or control groups, before inviting participation. After randomisation, more practices accepted invitations to be in the control group (302; 78% of those invited) than in the screening group (271; 68% of those invited).

Men registered at the screening group practices aged 50 to 69 who had no history of prostate cancer (189,386 men) were invited to have a PSA test. Those who had a PSA level of 3ng/mL or over were offered a transrectal ultrasound-guided biopsy, followed by treatment (including in a separate trial) if cancerous cells were found.

The 219,439 eligible men in the control group were not routinely provided with information about the PSA test but could request one if they wished.

This is the largest ever trial of PSA screening. It used consistent methods of diagnosis and treatment and had a high level of follow up.

What did it find?

  • Of the men offered screening, 40% attended the PSA testing clinic, and 36% had a PSA test. Of the men who were tested and had a valid result, 11% screened positive, and 85% of these men had a prostate biopsy. About 10 to 15% of men in the control group were estimated to have had a PSA test over the study period.
  • There was no difference in rates of prostate cancer death between the men offered screening and the men not offered screening, after 10 years. Prostate cancer mortality was 0.30 per 1,000 person-years for the intervention group and 0.31 for the control group (rate difference -0.013 per 1,000 person-years, 95% confidence interval [CI] -0.047 to +0.022; rate ratio [RR] 0.96, 95% CI 0.85 to 1.08).   
  • More men in the screening group were diagnosed with prostate cancer: 8,054 men in the screening group (4.3%; 4.45 per 1,000 person-years), compared with 7,853 (3.6%; 3.80 per 1,000 person-years) of those in the control group (rate difference 0.65 per 1,000 person-years, 95% CI 0.52 to 0.78; RR 1.19, 95% CI 1.14 to 1.25).
  • The cancers diagnosed in the screening group tended to be smaller and lower grade than those diagnosed in the control group. Tumours in the intervention group were less likely to be high grade (based on Gleason grade; odds ratio [OR] 0.68, 95% CI 0.64 to 0.73) or advanced (based on TNM staging; OR 0.68, 95% CI 0.62 to 0.75).
  • Some eventually-fatal prostate cancers among men in the screening group were not picked up by screening.
    Of the 549 men in the screening group who died of prostate cancer, 188 had attended the PSA testing clinic but only just under a third of them (59 men; 31%) had cancer detected.
    Of 129 men who did not have cancer detected, 42 (33%) did not have the PSA test, while 68 (53%) had the test but screened negative (had low PSA levels) so did not have a follow-up biopsy or treatment. In addition, 15 (12%) screened positive but did not have a biopsy, and four (3%) had a biopsy that showed no cancerous cells.

What does current guidance say on this issue?

The UK National Screening Committee does not currently recommend universal screening for prostate cancer using the PSA test.

The Prostate Cancer Risk Management Programme says: “The PSA test is not accurate enough to meet the requirements of a national screening programme.” It goes on to state that “Any asymptomatic man aged 50 and over can make an appointment with their GP to discuss having the PSA test. GPs should not proactively raise the issue with asymptomatic men.”

What are the implications?

These results add weight to the argument against using a single PSA test to screen asymptomatic men for prostate cancer in the UK. It is possible that repeated testing or screening that was taken up by more men could have benefits, though this study does not offer hope of that.

Further longer-term results from this study are expected and may change the conclusions regarding long-term outcomes. 

Citation and Funding

Martin RM, Donovan JL, Turner EL, et al. Effect of a low-intensity PSA-based screening intervention on prostate cancer mortality. The CAP randomized clinical trial. JAMA. 2018;319(9):883-95.

This study was partially funded by the National Institute for Health Research and by grants C11043/A4286, C18281/A8145, C18281/A11326, and C18281/A15064 from Cancer Research UK.

Bibliography

Cancer Research UK. Types and grades. London: Cancer Research UK; reviewed 2016.

Cancer Research UK. TNM staging. London: Cancer Research UK; reviewed 2016.

MJ Barry. Screening for prostate cancer. Is the third trial the charm? JAMA. 2018;319(9):868-69.

Public Health England. Prostate cancer risk management programme: overview. London: Public Health England; 2015.

UK NSC. The UK NSC recommendation on prostate cancer screening/PSA testing in men over the age of 50. London: UK National Screening Committee; 2016.

Effect of a Low-Intensity PSA-Based Screening Intervention on Prostate Cancer Mortality: The CAP Randomized Clinical Trial

Published on 7 March 2018

Martin, R. M.,Donovan, J. L.,Turner, E. L.,Metcalfe, C.,Young, G. J.,Walsh, E. I.,Lane, J. A.,Noble, S.,Oliver, S. E.,Evans, S.,Sterne, J. A. C.,Holding, P.,Ben-Shlomo, Y.,Brindle, P.,Williams, N. J.,Hill, E. M.,Ng, S. Y.,Toole, J.,Tazewell, M. K.,Hughes, L. J.,Davies, C. F.,Thorn, J. C.,Down, E.,Davey Smith, G.,Neal, D. E.,Hamdy, F. C.

Jama Volume 319 , 2018

Importance: Prostate cancer screening remains controversial because potential mortality or quality-of-life benefits may be outweighed by harms from overdetection and overtreatment. Objective: To evaluate the effect of a single prostate-specific antigen (PSA) screening intervention and standardized diagnostic pathway on prostate cancer-specific mortality. Design, Setting, and Participants: The Cluster Randomized Trial of PSA Testing for Prostate Cancer (CAP) included 419582 men aged 50 to 69 years and was conducted at 573 primary care practices across the United Kingdom. Randomization and recruitment of the practices occurred between 2001 and 2009; patient follow-up ended on March 31, 2016. Intervention: An invitation to attend a PSA testing clinic and receive a single PSA test vs standard (unscreened) practice. Main Outcomes and Measures: Primary outcome: prostate cancer-specific mortality at a median follow-up of 10 years. Prespecified secondary outcomes: diagnostic cancer stage and Gleason grade (range, 2-10; higher scores indicate a poorer prognosis) of prostate cancers identified, all-cause mortality, and an instrumental variable analysis estimating the causal effect of attending the PSA screening clinic. Results: Among 415357 randomized men (mean [SD] age, 59.0 [5.6] years), 189386 in the intervention group and 219439 in the control group were included in the analysis (n = 408825; 98%). In the intervention group, 75707 (40%) attended the PSA testing clinic and 67313 (36%) underwent PSA testing. Of 64436 with a valid PSA test result, 6857 (11%) had a PSA level between 3 ng/mL and 19.9 ng/mL, of whom 5850 (85%) had a prostate biopsy. After a median follow-up of 10 years, 549 (0.30 per 1000 person-years) died of prostate cancer in the intervention group vs 647 (0.31 per 1000 person-years) in the control group (rate difference, -0.013 per 1000 person-years [95% CI, -0.047 to 0.022]; rate ratio [RR], 0.96 [95% CI, 0.85 to 1.08]; P = .50). The number diagnosed with prostate cancer was higher in the intervention group (n = 8054; 4.3%) than in the control group (n = 7853; 3.6%) (RR, 1.19 [95% CI, 1.14 to 1.25]; P < .001). More prostate cancer tumors with a Gleason grade of 6 or lower were identified in the intervention group (n = 3263/189386 [1.7%]) than in the control group (n = 2440/219439 [1.1%]) (difference per 1000 men, 6.11 [95% CI, 5.38 to 6.84]; P < .001). In the analysis of all-cause mortality, there were 25459 deaths in the intervention group vs 28306 deaths in the control group (RR, 0.99 [95% CI, 0.94 to 1.03]; P = .49). In the instrumental variable analysis for prostate cancer mortality, the adherence-adjusted causal RR was 0.93 (95% CI, 0.67 to 1.29; P = .66). Conclusions and Relevance: Among practices randomized to a single PSA screening intervention vs standard practice without screening, there was no significant difference in prostate cancer mortality after a median follow-up of 10 years but the detection of low-risk prostate cancer cases increased. Although longer-term follow-up is under way, the findings do not support single PSA testing for population-based screening. Trial Registration: ISRCTN Identifier: ISRCTN92187251.

Gleason score. A grading system used to assess how normal or abnormal the cells in biopsied prostate tissue looks. Grades 1 and 2 indicate that cells look normal, while grades 3 to 5 indicate that cells are cancerous. The higher the grade, the more abnormal the cells look and the poorer the prognosis. An overall Gleason score is obtained by adding together the two most common Gleason grades from the biopsy samples taken. A score of ≤6 is considered low-grade cancer and ≥8 high-grade cancer.

TNM staging. A system used as a measure of how advanced the cancer is based on Tumour size and spread (T), whether it has spread to the lymph Nodes (N) or Metastasised to other parts of the body (M). This staging, as well as Gleason score, helps decide which treatments are appropriate. 

Expert commentary

More men now die from prostate cancer than women do from breast cancer. Screening for prostate cancer, using a simple blood test to detect cancers earlier when they should be easier to cure, is attractive but the results of trials have been inconsistent.

A particular concern is that prostate-specific antigen screening (PSA) leads to over-diagnosis and over-treatment, which can result in complications including impotence. The Cluster Randomised Trial of PSA Testing for Prostate Cancer again shows that screening may increase prostate cancer diagnoses but not reduce deaths. Well-designed trials like this one challenge “advances” that appear common sense but may do harm.

Chris Twelves, Professor of Clinical Cancer Pharmacology & Oncology, University of Leeds; Honorary Consultant Medical Oncologist & Clinical Director, Leeds Teaching Hospitals Trust