Table of Contents
- Key Points
- Background: Why This Research Matters
- How the Study Was Conducted
- Who Took Part
- What the Screening Invitation Involved
- What the Researchers Measured
- Key Findings: Prostate Cancer Deaths
- Overall Survival
- Cancer Grade and Stage at Diagnosis
- Prostate Cancer Diagnosis Rates Over Time
- Subgroup and Exploratory Findings
- Adverse Events
- What This Means for Patients
- Limitations of the Study
- Recommendations
- Frequently Asked Questions
- Source Information
Key Points
- In a UK trial of 415,357 men aged 50–69, a single PSA screening invitation reduced prostate cancer deaths at 15 years by about 9 per 10,000 men invited.
- The trial found no significant difference in overall survival between men invited for PSA screening and those not invited.
- Screening increased detection of low-grade and localized prostate cancers but not high-grade or advanced disease.
- Over-diagnosis was estimated at 9.2% for men aged 50–54 and 20.8% for men aged 65–69 in the trial.
- Harms of PSA screening include over-diagnosis, biopsy complications, and treatment side effects on urinary, sexual, and bowel function.
Background: Why This Research Matters
Prostate cancer is one of the most commonly diagnosed cancers in men. In England, the number of men diagnosed with prostate cancer increased by 68% between 2001 and 2019, rising from 28,216 to 47,479 cases. This rise reflects both an aging population and increased use of PSA testing.
In the United States, approximately 3.3 million men currently live with a prostate cancer diagnosis. While low-risk prostate cancer grows slowly and carries a low risk of death, aggressive prostate cancer causes approximately 12,000 deaths in the UK and 34,700 deaths in the U.S. each year.
The goal of PSA screening is to find curable cancer early and thereby reduce deaths. However, uncertainty has remained about whether PSA screening actually lowers long-term mortality. The CAP trial was designed to help answer this question.
The CAP randomized clinical trial previously reported that at a median of 10 years, a single invitation for PSA screening did not reduce prostate cancer mortality. This finding was compared with a control group (0.29% vs 0.30%; rate ratio 0.96; 95% CI 0.85–1.08; p=0.5). The current report describes the pre-specified secondary outcome of prostate cancer mortality at 15-year follow-up.
How the Study Was Conducted
The CAP trial was a cluster randomized clinical trial conducted in England and Wales. In this design, entire primary-care practices — rather than individual men — were randomly assigned to either the screening invitation group or the control group.
Between 2001 and 2007, 785 eligible general practices across eight hospital catchment areas (Birmingham, Bristol, Cambridge, Cardiff, Leeds, Leicester, Newcastle, and Sheffield) were randomized. Randomization used a computerized random number generator and was stratified within groups of 10–12 neighboring practices.
Because allocation happened before practices were invited to take part, it was not possible to conceal which group each practice was assigned to. Of the eligible practices, 573 (73%) agreed to participate — 68% of those randomized to the intervention group and 78% of those randomized to the control group.
Men were enrolled between January 8, 2002, and January 20, 2009. Final follow-up was completed on March 31, 2021. The Derby National Research Ethics Service Committee East Midlands approved the study.
Men who attended PSA testing in the intervention group gave individual written informed consent. For men in the control group and non-responders in the intervention group, individual consent was not sought. Instead, approval for identification and linkage to routine electronic records was obtained under Section 251 of the NHS Act 2006.
Prostate cancer deaths were identified using death certificates from the Office for National Statistics and were reviewed by an independent Cause of Death Evaluation committee. Cancer stage and Gleason grade were obtained from the National Disease Registration Service.
Statistical analysis used random-effects Poisson regression to estimate rate ratios, accounting for clustering within GP practices. Follow-up for each participant was divided into five-year age groups. Results were considered statistically significant if the p-value was less than 0.05.
Who Took Part
The study included 415,357 men aged 50–69 years registered with participating general practices. The mean age was 59.0 years (standard deviation 5.6 years).
Men were assigned to two groups:
- Intervention group: 195,912 men invited for PSA screening
- Control group: 219,445 men who received standard care without a screening invitation
Follow-up data for cancer diagnosis and mortality at a median of 15 years were available for 408,721 men (98% of those eligible). This included 189,326 men (97%) in the intervention group and 219,395 men (more than 99%) in the control group.
Baseline characteristics were similar between the two groups at both the practice level and the individual level. Men with prostate cancer diagnosed on or before the randomization date were excluded, as were men registered with practices on a temporary or emergency basis.
What the Screening Invitation Involved
Men in the intervention group received a single invitation for a PSA test after counseling. If the PSA result was between 3.0 and 19.9 ng/ml, they were offered a 10-core transrectal ultrasound-guided biopsy.
All laboratories participated in the UK National External Quality Assessment Service for PSA testing. Test results that did not meet quality assurance requirements, were lost, or came from ambiguous consent or insufficient blood samples were considered non-valid.
Men in the intervention group who were diagnosed with localized prostate cancer were invited to join a second randomized trial called ProtecT. That trial randomly assigned participants to active monitoring (regular PSA testing and clinical review), radical prostatectomy (surgical removal of the prostate), or radical conformal radiotherapy with hormone treatment.
Men with a PSA of 20 ng/ml or higher were referred to a urologist and received standard care.
Men in the control group received standard NHS management but no formal invitation for PSA testing as part of this study. Based on analysis of a national primary care database, cumulative PSA testing for prostate cancer detection in the control group was estimated at 10% to 15% over 10 years of median follow-up.
Among men in the intervention group, 75,694 (40%) underwent PSA testing, and 64,425 (34%) had a valid test result. Of those with valid results, 6,855 (11%) had a PSA value between 3 and 19.9 ng/ml and were eligible for the ProtecT trial. Of these, 5,848 (85%) had a prostate biopsy.
What the Researchers Measured
The primary outcome of the CAP trial — prostate cancer mortality at 10 years — was reported previously. This report covers the remaining pre-specified secondary outcomes at 15-year follow-up:
- Definite or probable prostate cancer mortality at 15 years
- All-cause mortality at 15 years
- Prostate cancer grade and stage at diagnosis at 15 years
Other secondary outcomes reported earlier included all-cause mortality at 10 years, disease grade and stage at 10 years, cost-effectiveness, and health-related quality of life.
The researchers also conducted exploratory analyses. These included a sensitivity analysis that re-defined the primary outcome to include definite, probable, possible, and treatment-related prostate cancer deaths. They also estimated mean sojourn time — the period during which a tumor is asymptomatic but detectable by screening — and over-diagnosis rates.
Key Findings: Prostate Cancer Deaths
After a median follow-up of 15.4 years (interquartile range 14.2–16.4 years; range 12.2–19.2 years), 1,199 men (0.69%) in the intervention group and 1,451 men (0.78%) in the control group died of prostate cancer.
The rate of prostate cancer death was 0.47 per 1,000 person-years in the intervention group and 0.50 per 1,000 person-years in the control group. The rate ratio was 0.92 (95% CI 0.85–0.99; p=0.03), meaning the screening invitation was associated with an 8% relative reduction in the rate of prostate cancer death.
The absolute reduction in prostate cancer deaths was 0.09% (risk difference −0.09%; 95% CI −0.15 to −0.03; p=0.02). In plain terms, this means that about 9 fewer men out of every 10,000 invited for screening died of prostate cancer over 15 years.
An instrumental variable analysis estimated the effect of actually attending the PSA screening clinic. Among men who attended, the prostate cancer mortality rate ratio was 0.83 (95% CI 0.68–1.00; p=0.053). This result was at the borderline of statistical significance.
Overall Survival
There was no statistically significant difference in overall survival between the two groups. There were 45,084 deaths (23.2%) in the intervention group and 50,336 deaths (23.3%) in the control group.
The rate ratio for all-cause mortality was 0.97 (95% CI 0.94–1.01; p=0.11). This means the difference in overall deaths between the groups could have been due to chance. Other causes of death were similar between the two groups.
Cancer Grade and Stage at Diagnosis
Screening changed the types of prostate cancers that were detected. Compared with the control group, men in the intervention group were more likely to be diagnosed with low-grade and localized cancers. Men in the intervention group were less likely to be diagnosed with high-grade or advanced cancers.
Specific findings over the 15-year follow-up:
- Low-grade cancer (Gleason score 6 or less): 2.2% of men in the intervention group vs 1.6% in the control group (risk difference 0.58%; 95% CI 0.50%–0.67%)
- High-grade cancer (Gleason score 8 or higher): 1.2% vs 1.3% (risk difference −0.15%; 95% CI −0.22% to −0.08%)
- Localized cancer (stage T1/T2): 3.6% vs 3.1% (risk difference 0.56%; 95% CI 0.44%–0.67%)
- Advanced-stage cancer (T4/N1/M1): 0.9% vs 1.1% (risk difference −0.16%; 95% CI −0.22% to −0.10%)
The trend across cancer grades and stages was statistically significant (p for trend less than 0.001 for both). There was no significant difference between groups in intermediate-grade (Gleason score 7), locally-advanced (stage T3), or distally-advanced tumors.
Men in the intervention group were diagnosed at a mean age 1.22 years younger than men in the control group (95% CI 1.02–1.42; p<0.001).
Prostate Cancer Diagnosis Rates Over Time
Over the full 15-year follow-up, 12,013 men (4.88 per 1,000 person-years; cumulative risk 7.1%) were diagnosed with prostate cancer in the intervention group, compared with 12,958 men (4.60 per 1,000 person-years; cumulative risk 6.9%) in the control group.
The difference in diagnosis rates between the groups changed markedly over time. The cumulative risk difference per 1,000 men was:
- 12.23 (95% CI 11.63–12.84) at 18 months
- 4.80 (95% CI 3.53–6.07) at 10 years
- 1.38 (95% CI −0.38 to 3.14) at 15 years
- 0.86 (95% CI −1.80 to 3.53) at 18 years
This means screening caused a large spike in diagnoses early on, but the difference between groups narrowed over time as cancers were also detected in the control group.
The researchers also estimated mean sojourn time and over-diagnosis rates. For men aged 50–54, the mean sojourn time was 12.1 years and over-diagnosis was 9.2%. For men aged 65–69, the mean sojourn time was 15.3 years and over-diagnosis was 20.8%.
Subgroup and Exploratory Findings
The mortality results were similar when the researchers included prostate cancer deaths judged as "possible" by the Cause of Death Evaluation committee. The results were also similar when they restricted the analysis to deaths judged as "definite" prostate cancer deaths.
There was little evidence that the effect of screening differed by age group or socioeconomic status. The p-values for interaction were 0.46 or higher, meaning no statistically significant differences were found across these subgroups.
Adverse Events
Among the deaths due to prostate cancer, 8 deaths (0.7%) in the intervention group and 7 deaths (0.5%) in the control group were related to a diagnostic biopsy or prostate cancer treatment. Other adverse events were reported in previous publications.
What This Means for Patients
This trial found that a single invitation for PSA screening reduced prostate cancer deaths at 15 years, but the absolute benefit was small. The reduction was 0.09% — about 9 fewer deaths per 10,000 men invited.
The researchers noted that this reduction was smaller than the effect size they had defined in advance as important for clinical and public health benefit. There was no effect on overall survival.
The harms of PSA testing include over-diagnosis (finding cancers that would never have caused symptoms), biopsy complications, and adverse treatment effects on urinary, sexual, and bowel function. There is also the potential to miss an aggressive prostate cancer.
The CAP trial's single-invitation approach was designed to minimize over-diagnosis and over-treatment compared with other screening trials. However, over-diagnosis was still observed after 15 years of median follow-up.
The researchers emphasized that policy-makers considering prostate cancer screening should weigh this small reduction in deaths against the potential adverse effects associated with over-diagnosis and over-treatment.
The European Randomized Study of Prostate Cancer Screening (ERSPC) enrolled 162,243 men across 7 centers. ERSPC reported that PSA screening conducted every 2–4 years (a mean of 1.4 tests per participant) reduced prostate cancer mortality. The CAP trial used a different approach with only a single invitation.
Limitations of the Study
The study had several limitations that patients should understand.
First, the absolute reduction in prostate cancer deaths was very small. The number of men who would need to be invited for screening to prevent one prostate cancer death is large.
Second, there was no improvement in overall survival. This means that even though fewer men died of prostate cancer in the screening group, the total number of deaths was not significantly different between the groups.
Third, the trial used a single invitation for PSA screening rather than a repeated screening program. This design may not reflect how PSA screening is typically performed in practice, where men may be tested multiple times over many years.
Fourth, some men in the control group did undergo PSA testing outside the trial. An estimated 10% to 15% of control-group men had PSA testing over 10 years. This could reduce the apparent difference between the two groups.
Fifth, the study did not adjust p-values for multiple comparisons. The researchers explained that conventional adjustments assume statistical independence between estimates, which was not the case for analyses of the same outcome at 10 and 15 years.
Sixth, some data were missing. Among men who developed prostate cancer in the intervention group, 9.4% were missing cancer stage data and 10.4% were missing Gleason grade data. In the control group, 7.8% were missing stage data and 11.2% were missing grade data.
Recommendations
Based on this study, men and their doctors should consider the following points when deciding about PSA screening:
- Understand the small absolute benefit. A single PSA screening invitation reduced prostate cancer deaths by about 9 per 10,000 men over 15 years. This is a real but modest benefit.
- Consider the risks of over-diagnosis. Screening finds many low-grade, slow-growing cancers that may never have caused harm. Over-diagnosis rates ranged from 9.2% in men aged 50–54 to 20.8% in men aged 65–69.
- Discuss the potential harms. These include biopsy complications and treatment side effects affecting urinary, sexual, and bowel function.
- Recognize that overall survival was not improved. The study found no significant difference in deaths from any cause between screened and unscreened men.
- Talk with your doctor about your individual risk. The decision to undergo PSA testing should be personalized, taking into account your age, family history, and overall health.
- Be aware that this study used a single screening invitation. Other screening approaches, such as repeated testing every 2–4 years, may have different benefits and harms.
The researchers concluded that policy-makers should consider this small reduction in prostate cancer deaths. Policy-makers should weigh this reduction against the potential adverse effects of over-diagnosis and over-treatment when making decisions about prostate cancer screening programs.
Frequently Asked Questions
Who was invited for PSA screening in this trial?
In a UK trial, 415,357 men aged 50 to 69 registered with participating general practices were assigned to two groups. One group of 195,912 men received a single invitation for a PSA test after counseling. The other group of 219,445 men received standard care without a screening invitation. Men already diagnosed with prostate cancer were excluded.
What did the screening invitation involve?
Men in the invitation group had a single PSA test after counseling. If the result was between 3.0 and 19.9 ng/ml, they were offered a 10-core transrectal ultrasound-guided biopsy. Men with a PSA of 20 ng/ml or higher were referred to a urologist for standard care. Men diagnosed with localized cancer were invited to join a separate treatment trial.
How much did a single PSA screening invitation reduce prostate cancer deaths?
After a median follow-up of 15.4 years, 0.69% of men in the invitation group died of prostate cancer versus 0.78% in the control group. This was an 8% relative reduction and an absolute reduction of 0.09%, meaning about 9 fewer deaths per 10,000 men invited. The benefit was small.
Did the screening invitation improve overall survival?
No. In the trial, 23.2% of men in the invitation group died from any cause compared with 23.3% in the control group. The rate ratio was 0.97, and the difference was not statistically significant. This means the difference in overall deaths between the groups could have been due to chance.
What are the risks of PSA screening?
The harms include over-diagnosis, which means finding cancers that would never have caused symptoms, biopsy complications, and treatment side effects affecting urinary, sexual, and bowel function. In the trial, over-diagnosis was estimated at 9.2% for men aged 50 to 54 and 20.8% for men aged 65 to 69.
What does over-diagnosis mean in this trial?
Over-diagnosis means detecting a prostate cancer that would never have caused symptoms or death during a man's lifetime. In the trial, over-diagnosis was estimated at 9.2% for men aged 50 to 54 and 20.8% for men aged 65 to 69. Screening increased detection of low-grade, localized cancers but not high-grade or advanced disease.
Should I have a PSA test?
The decision should be personalized with your doctor, considering your age, family history, and overall health. In the trial, a single PSA screening invitation reduced prostate cancer deaths by about 9 per 10,000 men over 15 years, but overall survival was not improved. Discuss the small absolute benefit and the risks of over-diagnosis and treatment side effects.
Should I get a second opinion before deciding about PSA screening after a single invitation?
A single PSA screening invitation reduced prostate cancer deaths by about 9 per 10,000 men over 15 years, with no difference in overall survival, and over-diagnosis ranged from 9.2% in men aged 50–54 to 20.8% in men aged 65–69. Because the decision depends on age, family history, and overall health, and because repeated testing every 2–4 years may carry different benefits and harms, a second opinion can help clarify whether screening or biopsy is appropriate for you. Diagnostic Detectives Network provides independent expert second opinions.
Source Information
Original article title: Prostate-Specific Antigen Screening and 15-Year Prostate Cancer Mortality
Authors: Richard Michael Martin, Emma L. Turner, Grace Young, Chris D. Metcalfe, Eleanor I. Walsh, J. Athene Lane, et al.
Lead institution: Bristol Medical School, Department of Population Health Sciences, University of Bristol, Bristol, UK
Trial registration: ISRCTN92187251
Note: This patient-friendly article is based on peer-reviewed research. The original article was published in a peer-reviewed medical journal. Patients should discuss any questions about prostate cancer screening with their healthcare provider.
