Still Confusion about the Usefulness of PSA-screening.

13 04 2009

Prostate cancer is the most commonly diagnosed cancer affecting older men and second-biggest cancer killer. pc_epid_fig11a

Prostate Specific Antigen (PSA), a protein mainly produced by the prostate gland, is often elevated in prostate cancer – and often proportional to the prostate cancer volume. Since more men are diagnosed with prostate cancer by using PSA screening, middle-aged men have been advised to undergo a simple blood test to determine their blood PSA levels.

Indeed in the 20 years that the PSA test has been used there has been a significant drop in prostate cancer deaths.

However, this may have also resulted from better treatment modalities.

Furthermore, PSA tests are prone to false negative results (prostate cancer present in the complete absence of an elevated PSA level ), or vice versa, false positive results: elevated PSA occurring in non-cancerous prostate diseases, like prostate infection and benign prostatic hyperplasia (BPH). Some detected prostate cancers may also be indolent, never giving any trouble on the long term. Since the further diagnosis methods (biopsy) and treatment methods (irradiation, surgery, hormonal treatment) often have serious side effects (erectile dysfunction, urinary incontinence and bowel problems), there is a clear need to demonstrate whether PSA screening is worth the high risks of overdiagnosis and overtreatment:

Thus, does PSA screening really saves lives?
And what is the trade off between benefits and harms?

A Cochrane Systematic Review from 2006 [5] (also reviewed in EBM-online) concluded that there was no proof of benefit of PSA-screening. Yet absence of proof is not proof of absence. Moreover, both trials on which the review was based had methodological weaknesses.
Therefore, the main conclusion was to wait for the results from two large scale ongoing randomized controlled trials (RCTs).

The first study results of these two large RCT’s,  that many observers hoped would settle the controversy, have appeared in the March issue of the New England Journal of Medicine (NEJM). [1,2] The results are discussed in an accompanying editorial [3] and in a Perspective Roundtable [4] (with a video).

It should be stressed, however, that these are just interim results.

One of these two studies [1], the prostate component of the U.S. National Cancer Institute’s Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) showed no prostate specific mortality reduction over 11 yrs follow-up in 76,705 men by annual PSA screening and DRE (digital rectal exam). However:

  • The cut off is relatively high (4.0 ng per milliliter), which means that some prostate cancers could have been missed (on the other hand lowering the screening criteria might also have led to a higher false negative response)
  • The control group is “highly contaminated”, meaning that many men in the so called nonscreened arm had a PSA-test anyway ((52% in the nonscreened versus 85% in the screened arm).
  • The 11 yr follow up may be too short to show any significant effect. “Only” 0,1% of the men died of prostate cancer. On the long term the differences might become larger.
  • Since there were only 122 prostate cancer deaths in the screening group versus 135 in the control group, the power of the study to find any differences is mortality seems to be rather low.

The European ERSPC study [2] is larger than the PLCO trial (190,000 men), the cut off rate was lower (3.0 µg/L), and there was less contamination of the nonscreened arm. A shortcoming of the trial is that the diagnosis methods varied widely among centers participating in the trial. The follow-up time is 9 years.

The ESPRC trial noticed a difference in mortality between the screened and non-screened arms. Surprisingly the same outcome led to widely different conclusions, especially in the media (see Ben Goldacre on his blog Bad Science [6])

English newspapers concluded that the ERPSC study showed a clear advantage: Prostate cancer screening could cut deaths by 20% said the Guardian. Better cancer screening is every man’s right was the editorial in the Scotsman (see 6). These newspapers didn’t mention the lack of effect in the US study.

But most US newspapers, and scientists, concluded that the benefits didn’t outweigh the risks.

Why this different interpretation?

It is because 20% is the relative risk reduction. This means that the risk of getting prostate cancer is reduced by 20%. This sounds more impressive than it is, because it depends on your baseline risk. It is the absolute reduction that counts.
Suppose you would have a baseline chance of 10% of getting prostate cancer. Reducing this risk by 20% means that the risk is reduced from 10% to 8%. This sounds a lot less impressive.
But in reality your chance of getting prostate cancer comes closer to 0,1%. Then, a risk reduction of 20% becomes even less significant: it means your risk has decreased to 0,08%.

Absolute numbers are more meaningful. In the ESPRC trial[2], the estimated absolute reduction in prostate-cancer mortality was about 7 deaths per 10,000 men after 9 years of follow-up. This is not a tremendous effect. However the costs are high: to prevent one death from prostate cancer 1410 men would need to be screened and 48 additional cases of prostate cancer would need to be treated.

Overdiagnosis and overtreatment are probably the most important adverse effects of prostate-cancer screening and are vastly more common than in screening for breast, colorectal, or cervical cancer.

It is difficult to realize the impact of a false negative diagnosis. People tend to think that saving any live is worth any cost. But that isn’t the case.

This quote says a lot (from Ray Sahelian)

A few years ago my dad was found to have a high PSA test. He was 74 at the time. He underwent multiple visits to the doctor over the next few months with repeated PSA tests and exams, and eventually a biopsy indicated he had a small prostate cancer. I remember my dad calling me several times a month during that period constantly asking my thoughts on how he should proceed with radiation or other treatments for his cancer. My dad had a preexisting heart condition known as atrial fibrillation. I suggested he not undergo any treatment for the small cancer but just to follow the PSA levels. His doctor had agreed with my opinion. His PSA test stayed relatively the same over the next few years and the prostate cancer did not grow larger. My dad died at 78 from a heart rhythm problem. Ever since the discovery of the high PSA level, he was constantly worried about this prostate gland. What good did it do to have this PSA test at his age? It only led to more doctor visits, a painful prostate gland biopsy, and constant worry. Maybe the constant worry even made his heart weaker.

Indeed more men die with prostate cancer than of it.It’s estimated that appr 30% of American men over age 60 have small, harmless prostate cancers.

Although still hypothetical, non-invasive tests that would discriminate between low- and high risk prostate cancer could be a real solution to the problem. One such candidate might be the recently discovered urine test for sarcosine [7]

In conclusion
PSA-screening is associated with an earlier diagnosis of prostate cancer, but the present evidence shows at the most a slight reduction in prostate related mortality. Since screening and subsequent testing do have serious side effects, there seems a trade off between uncertain benefits and known harms. However, definite conclusions can only be drawn after complete follow-up and analyses of these and other studies [1,2,3]


  1. ResearchBlogging.orgAndriole, G., Grubb, R., Buys, S., Chia, D., Church, T., Fouad, M., Gelmann, E., Kvale, P., Reding, D., Weissfeld, J., Yokochi, L., Crawford, E., O’Brien, B., Clapp, J., Rathmell, J., Riley, T., Hayes, R., Kramer, B., Izmirlian, G., Miller, A., Pinsky, P., Prorok, P., Gohagan, J., Berg, C., & , . (2009). Mortality Results from a Randomized Prostate-Cancer Screening Trial New England Journal of Medicine DOI: 10.1056/NEJMoa0810696
  2. Schroder, F., Hugosson, J., Roobol, M., Tammela, T., Ciatto, S., Nelen, V., Kwiatkowski, M., Lujan, M., Lilja, H., Zappa, M., Denis, L., Recker, F., Berenguer, A., Maattanen, L., Bangma, C., Aus, G., Villers, A., Rebillard, X., van der Kwast, T., Blijenberg, B., Moss, S., de Koning, H., Auvinen, A., & , . (2009). Screening and Prostate-Cancer Mortality in a Randomized European Study New England Journal of Medicine DOI: 10.1056/NEJMoa0810084
  3. Barry, M. (2009). Screening for Prostate Cancer — The Controversy That Refuses to Die New England Journal of Medicine, 360 (13), 1351-1354 DOI: 10.1056/NEJMe0901166
  4. Lee, T., Kantoff, P., & McNaughton-Collins, M. (2009). Screening for Prostate Cancer New England Journal of Medicine, 360 (13) DOI: 10.1056/NEJMp0901825
  5. Ilic D, O’Connor D, Green S, Wilt T. Screening for prostate cancer. Cochrane Database Syst Rev. 2006;3:CD004720.[Medline]
  6. Goldacre, Ben (2009) Bad Science: (2009/03/), also Published in The Guardian, 21 March 2009
  7. Sreekumar A et al. (2009) Metabolomic profiles delineate potential role for sarcosine in prostate cancer progression Nature 457 (7231): 910-914 DOI: 10.1038/nature07762