Journal reference: 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 2010;341:c4521 ^[1]

Link: http://dx.doi.org/10.1136/bmj.c4521

Evidence cookie says...

Men aged 60 years with a PSA < 1.0 ng/mL have less than a 0.5% risk of prostate cancer metastasis or prostate cancer death by age 85. It is extremely unlikely that these men (about half at this age) will benefit from a PSA screening program for asymptomatic prostate cancer. Note: prostate cancer screening with PSA is not currently recommended in men of any age in Australian guidelines. [2] [3]

More details:

Article details

Study design:

Study aim:

to determine the relationship between prostate specific antigen (PSA) at age 60 and subsequent diagnosis of clinically relevant prostate cancer

Methods summary:

• Data from the Swedish Malmo Preventive Project (cohort study)
  • 1981-2
  • 60-year old Swedish men provided blood samples as part of cardiovascular studies
  • Cohort in this study: 1167 men born in 1921 → 71% of the eligible population
• Cancer registry at the National Board of Health and Welfare
  • Records of the participants were linked to the cancer registry
  • Identified men diagnosed with prostate cancer up to the end of 2006
• Nested case control design to analyse data from the cohort; separate matches for each of the study events:
  • Clinically diagnosed prostate cancer
  • Prostate cancer metastasis
  • Prostate cancer death
• Laboratory methods:
  • Total PSA was measured from stored anticoagulated blood plasma

Results summary:

• 126 men were diagnosed with prostate cancer
  • 43 men had metastatic disease
  • 35 died from prostate cancer
  • No cancers were detected as the result of screening; most commonly cancer was detected during investigation for lower urinary tract symptoms
  • Curative treatment was rare; only one man underwent radical prostatectomy
• Median PSA concentration for the entire cohort was 1.06 ng/mL
• Prostate cancer metastases
  • PSA < 0.65 ng/mL → nil
  • PSA 0.65-0.99 ng/mL → 7% of cases (0.3% of the entire cohort)
  • PSA 1.00-1.99 ng/mL → 12% of cases  (0.4% of the entire cohort)
  • PSA ≥ 2.00 ng/mL → 81% of cases  (3% of the entire cohort)
• Prostate cancer death
  • PSA < 0.65 ng/mL → nil
  • PSA 0.65-0.99 ng/mL → 3% of cases (0.1% of the entire cohort)
  • PSA 1.00-1.99 ng/mL → 6% of cases (0.2% of the entire cohort)
  • PSA ≥ 2.00 ng/mL → 91% of cases (3% of the entire cohort)
• Other results:
  • Total PSA, free PSA, free:total PSA ratio, and hK2 predicted all three end points
  • Single best marker was total PSA

Study conclusion:

The concentration of prostate specific antigen at age 60 predicts lifetime risk of metastasis and death from prostate cancer. Though men aged 60 with concentrations below the median (≤1 ng/ml) might harbour prostate cancer, it is unlikely to become life threatening. Such men could be exempted from further screening, which should instead focus on men with higher concentrations.

Participants:

• 1167 Swedish men born in 1921 (aged 60 in 1981-2)
• 71% of the eligible population; likely representative of the population
• At time of the baseline tests, did not have diagnosis of prostate cancer
• These men would not have been screened for prostate cancer at the time
• Very few of the participants diagnosed with prostate cancer received curative treatment

Methodological weaknesses

• Very homogenous group of men: Swedish (ethnic Northern European), all born in the same year → results may not be applicable to other populations and potentially threatens its external validity
• Results for men aged 60 are not entirely applicable to questions surrounding screening in younger men
• Results possibly over-estimate risks compared to the present day due to improvements in treatment

Methodological strengths

• Very representative cohort of the population
• Long follow up (age 60 to 85 years)
• The historical absence of prostate cancer screening (with subsequent problems of overdiagnosis, overtreatment and validation bias) in this population allows for a natural experiment examining the association between PSA concentration and subsequent prostate cancer outcomes
• Clinically important outcome measures: prostate cancer metastasis, death and clinically diagnosed prostate cancer

Biases and conflicts of interests

• Lilja H (tenth and final author) holds patents for free PSA and kK2 assays (declared).
• Nil other conflicts of interests declared or seem obvious.

Clinical relevance to primary health care

Prostate cancer screening is a contentious issue in Australian general practice.  Prostate specific antigen is a widely performed test for the purposes of asymptomatic screening.  However, the guidelines for preventative activities from the Royal Australian College of General Practitioners (RACGP) recommend against routine PSA screening ^[2].  Similarly, the guideline from the Urological Society of Australia and New Zealand (USANZ) recommends against population PSA screening ^[3] though their media releases are somewhat ambiguous and can be interpreted as supporting certainly forms of screening ^[4].

It is known that prostate cancer is very common; a third to half of deceased men in their seventies have detectable prostate cancer on autopsy.  Little of this would have been clinically apparent disease.  Overdiagnosis is a serious issue in PSA screening.  A very large European randomised control trial found that although PSA screening decreased prostate cancer mortality by around 20%, it was at the substantial cost of overtreatment; 48 men will need treatment for prostate cancer to prevent one death ^[5].

The results of this study do not explicitly support any specific screening procedure but does suggest a large group of men where screening is extremely unlikely to be beneficial.  Men aged 60 with a PSA < 1.00 ng/mL (about half of men this age) had less than 0.5% risk of prostate cancer metastases or prostate cancer death by age 85 even if they have prostate cancer.

Note: although the evidence is likely applicable and appears concordant with other study findings, caution should be taken given the very homogenous characteristics of the participants (Swedish men born in 1921 only).

References

1. 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 2010;341:c4521
2. Guidelines for preventive activities in general practice (The Red Book) 7th Edition. Royal Australian College of General Practitioners. 2009
3. Urological Society of Australia and New Zealand PSA testing policy 2009. Urological Society of Australia and New Zealand. September 2009
4. Urologists reassure Australian men PSA test is best indicator for prostate cancer [media release]. Urological Society of Australia and New Zealand. 15 March 2010
5. Schroder FH, Hugosson J, Roobol MJ, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med 2009;360:1320-8