ON THIS PAGE:
Authors MJ Litchfield, RG Cumming, DP Smith, V Naganathan, DG Le Couteur, LM Waite, FM Blyth, DJ Handelsman
Review Date March 2012
Citation Med J Aust 2012; 196 (6): 395-398
PSA screening for prostate cancer in unselected populations remains controversial due to the limited data on benefits, and significant risks of over-diagnosis and over-treatment. It is generally acknowledged that older men (the US guidelines state over 75 years) will not benefit from screening but it appears that many older Australian are having PSA tests.
PSA levels increase with age such that age-specific reference ranges could help to increase the specificity of the test. However, there are few population-based PSA studies that include old men, or the numbers of men in older age groups are small.
It is also known that PSA levels vary across ethic groups, an issue of importance in the ethnically diverse Australian population.
To describe values of serum prostate-specific antigen (PSA) in older men without diagnosed prostate cancer, categorised by age and country of birth, and to describe self-reported prostate cancer screening.
The Concord Health and Ageing in Men Project (CHAMP) is a cohort study involving a representative sample of 1434 eligible community-dwelling men with no diagnosis of prostate cancer who were aged between 70 and 97 years of age and living in a defined geographic area in Sydney. Baseline data were collected between January 2005 and June 2007.
Serum PSA levels were measured at baseline. Information on self-reported PSA testing and/or digital rectal examination (DRE), age, country of birth, and the International Prostate Symptom Score (IPSS) was collected in a baseline questionnaire that included other demographic and health variables.
Eleven per cent of men (n=155) had a PSA level of ≥ 6.5 ng/mL (the suggested cut-off for further investigation for men aged 70 or over, from an American study), increasing from 7.5% of men aged 70–74 years to 31.4% of men aged ≥ 90 years.
PSA levels varied with country of birth: Australian-born men (n=695) had the highest levels (median, 2.3 ng/mL; 5th–95th percentile, 0.4–10.1 ng/mL), followed by men born in China (n = 42; 2.1 ng/mL; 0.4–12.4 ng/mL), United Kingdom and Ireland (n = 70; 1.9 ng/mL; 0.3–8.9 ng/mL), Greece (n = 59; 1.5 ng/mL; 0.2–6.1 ng/mL), and Italy (n = 293; 1.4 ng/mL; 0.3–7.2 ng/mL).
A PSA test in the previous 2 years was reported by 48% of participants, and a digital rectal examination (DRE) in the previous 2 years by 37%.
Men with moderate to severe lower urinary tract symptoms (defined by IPSS scores) were more likely to have a PSA test while a lower education level and being born in China (compared to Australia) were associated with a lower likelihood of having PSA testing.
A significant number of men aged over 70 years reported recent prostate cancer tests. The PSA level ranges, according to age and country of birth, reported in this cohort will help with interpreting serum PSA level findings in men aged over 70 years.
Points to Note
- This is one of the largest studies of PSA levels in men aged over 70 and the only reported study to include men aged 85 or over.
- Nearly half of men reported PSA tests in the past 2 years.
- The 95th percentile of 9.0 ng/mL in men aged 70-79 is higher than the recommended 6.5 ng/mL cut-off based on one American study with implications for management of men in this age group.
- Australian born men appear to have higher PSA levels than other ethnic groups, especially Italian-born men.
- PSA levels in Chinese-born men were almost as high as Australian-born men, consistent with a higher prevalence of prostate cancer in Chinese men outside of China, compared to those in China.
- Variation in PSA testing by country of birth and education level support findings in other studies and need further investigation to ascertain the reasons why.
- The study was not able to exclude men with undiagnosed prostate cancer, which might have resulted in higher PSA levels than if they had been excluded.
- Men were not asked why they had PSA testing and the results suggest some might have been due to lower urinary tract symptoms rather than prostate cancer screening.