Mass screening and stage migration: a UK analysis


A recent publication has attempted to determine the impact of mass, population-based screening for prostate cancer in the UK, using PSA testing and subsequent biopsies as appropriate.

Between 2001 and 2008, the Prostate Testing for Cancer and Treatment (ProtecT) trial recruited men aged 50-69 years from nine cities in the UK and from randomly selected practices of general practitioners (primary care physicians). All patients were offered PSA testing; those with a PSA level of > 3 ng/ml were offered a prostate biopsy.

Moore et al. have now compared the age, PSA level, clinical stage, and grade at diagnosis of participants found to have prostate cancer in the ProtecT study with data from contemporaneous patients with incident cases of prostate cancer aged 50-69 years registered with the UK’s Eastern Cancer Registration and Information Centre (ECRIC).

The results of this analysis are as follows:

  • Data from the ProtecT study
    • 94,427 men (50 percent of those contacted) agreed to have a PSA test.
    • 8,807/94,427 of the participants (9.3 percent) had a raised PSA level.
    • 2,022/8,807 (23.0 percent) had prostate cancer.
    • 229/2,022 had locally advanced (T3 or T4) or metastatic cancers; the rest had clinically localized (T1c or T2) disease.
  • Data from ECRIC
    • 12,661 prostate cancer cases were recorded over the same period.
    • 3,714/12,661 were in men aged 50-69 years at diagnosis.
  • Men identified with cancer as a consequence of participating in the ProtecT study had a lower age distribution and PSA level than those identified incidentally and reported to ECRIC.
  • The cancers identified in participants in the ProtecT study were of lower stage and grade than those identified incidentally and reported to ERIC (P < 0.001 for all comparisons).
  • If mass, population-based PSA testing were introduced in the UK, approximately 2,660/100,000 men aged 50-69 years would be found to have prostate cancer, compared to current rates of approximately 130/100,000.
  • If half of the male population of the UK accepted PSA testing, approximately 160,000 cancers would be found each year, compared to the 30,000 diagnosed each year at present.

The authors conclude that, “Population-based PSA testing resulted in a significant downward stage and grade migration, and most such cancers were of low stage and grade, which could lead to risks of over-treatment for some men.”

This is an important study because it provides a quantifiable assessment of the distinction between the number of histologically significant cases of prostate cancer potentially identifiable each year in a specific national population and compares it to the likely number of clinically significant, incident  cases.

Of course the question that this analysis does not resolve is the potential benefit of identifying patients with prostate cancer when they are younger, with an earlier stage of disease (and therefore have greater potential for curative therapy) as compared to the potential for harm to all the men who have histologically identifiable but not clinically significant prostate cancer that will never have real impact if untreated. 

What this study does do, however, is identify the real clinical value of a test that could actually discriminate with accuracy, early on between those patients at real risk for clinically significant disease (apparently about 30,000 per annum in the UK) and the 130,000 who might be identified with histologically identifiable but clinically non-significant disease. That’s an awful lot of men who could benefit from knowing early on what their real risk might be for clinically significant disease — one way or the other!

One Response

  1. The data might be represented as

    • Number of men who took the PSA test: 94,427
    • Number of men with elevated PSA: 8,807 = 9% of men tested
    • Number of men diagnosed: 2,022 = 23% of men with elevated PSA and 2% of men tested
    • Number of men with “advanced” disease requiring early treatment: 229 = 11% of men diagnosed, 3% of men with elevated PSA, and 0.25% of men tested

    As I recall there was a paper (possibly authored by Stamey) demonstrating that if all men were subjected to biopsy there would be a similar outcome to that reported if only men with an elevated PSA were subject to this procedure.

    If that was correct (and if my memory is correct) then we might expect 23% of men with a “normal” PSA to be diagnosed – that is to say 19,693 men, of whom some 2,230 (11%) would have “advanced” disease requiring early treatment.

    Isn’t this why it is said that it is important to find a disease-specific test to replace PSA – so that men who are missed by the PSA test are identified? This test might also distinguish between disease that requires attention and separate those cases from the vast majority that will not require attention.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: