15 Percent Men with PSA Less Than 4 ng/mL is Having Prostate Cancer

Should Men With Serum PSA Levels of 4.0 ng/mL or Less Undergo a Prostate Biopsy?

J Brantley Thrasher; William L Valk

Posted: 12/21/2004; Nat Clin Pract Urol. 2004;1(1) © 2004 Nature Publishing Group



There is a general consensus among clinicians that a serum prostate-specific antigen (PSA) level ≥4.0 ng/ml is predictive of prostate cancer, and necessitates a biopsy. But there is a dearth of prospective data on the predictive value of PSA levels below this threshold.


To analyse the prevalence of prostate cancer in men with a serum PSA level ≤4.0 ng/ml.

Design and Intervention

Assessments of serum PSA levels and digital rectal examinations were performed annually for 7 years on the 18,882 participants in the Prostate Cancer Prevention Trial. This randomized, double-blind, placebo-controlled Phase III trial was designed to test the efficacy of daily finasteride treatment at reducing the prevalence of prostate cancer. A subgroup analysis was performed on the trial’s placebo cohort to assess the predictive value of PSA levels ≤4.0 ng/ml. Men from the placebo group whose PSA had never exceeded 4.0 ng/ml, who had never previously undergone a prostate biopsy or transurethral prostate resection, and who had consistently normal results from digital rectal examinations were included in the subgroup analysis. Biopsies were performed and serum PSA levels measured at the end of the trial period.

Outcome Measures

The predictive value of serum PSA level for the detection of prostate cancer was calculated at study completion. This value was defined as the probability of detecting disease during end-of-study biopsy if serum PSA levels fell within a prespecified range.


The 2950 men that satisfied the selection criteria and consented to an end-of-study biopsy were aged between 62 and 91 years.

  • Prostate cancer was detected in 449 (15%) of these subjects. Men that had developed cancer had higher serum PSA levels than cancer-free men (means 1.78 vs 1.34 ng/ml, P <0.001).
  • The risk of developing cancer increased with rising PSA level, from 6.6% in men with a PSA value ≤0.5 ng/ml to 26.9% in men with 3.1-4.0 ng/ml PSA (odds ratio [OR] for developing prostate cancer, 1.66 per unit increase in serum PSA; 95% CI 1.50 to 1.85).
  • Men whose father, brother(s) or son(s) were, or had been, affected by prostate cancer were significantly more likely to develop the disease (P = 0.004).
  • All staged cancers that developed during the trial were stage T1, with Gleason scores of between 2 and 9. Serum PSA level was positively correlated with high-grade disease (OR for Gleason scores ≥7, 2.10 per unit increase in serum PSA; 95% CI 1.66 to 2.65).


Prostate cancer is surprisingly prevalent in men with serum PSA levels that are considered by many clinicians to be in the normal range (i.e. ≤4.0 ng/ml). Levels <4.0 ng/ml can be predictive of the risk of developing prostate cancer, and of the likelihood of malignancies being high-grade.


The article by Thompson et al. has caused significant uproar with regard to the utility of PSA for detecting prostate cancer. A few comments and cautions are warranted.

Given what is known from autopsy studies about the incidence of prostate cancer and the disparity between a man’s lifetime risk of developing clinically significant disease and dying from the disease,these new data are not a complete surprise. As men age, their risk of developing prostate cancer increases. But the lifetime risk of disease diagnosis is 16.6% (1 in 6) and the risk of disease-associated death 3-4%. This indicates that many detected cancers might not be clinically important nor potentially lethal.

We have long suspected that significantly more cancers are present in the ‘screenable’ population (otherwise healthy men aged 55-75 years) than are detected. The findings of Thompson and co-workers have prompted some to dismiss PSA as useless for identification of prostate cancer and to suggest biopsy of all men aged 55-75 years – approximately 30 million in the US alone. The Prostate Cancer Prevention Trial reported the prevalence of disease in untreated men who underwent biopsy to be 24.4%.[3] So, more than 7 million men would probably be diagnosed with prostate cancer if 30 million were biopsied.

The CaPSURE database indicates that 90% of men diagnosed with prostate cancer will elect to be treated. Medicare data and the 1999 Patterns of Care Study indicate that 35% of patients in the US undergo radical prostatectomy (3% with neo-adjuvant hormonal therapy), 50% radiation therapy (50-76% with hormonal therapy), 10% watchful waiting, and 5% hormonal therapy only. The total estimated cost of treating more than 7 million men (not including follow-up, inflation, or additional intervention) would be US$104.6-116.5 billion. Of the approximately 2.6 million patients (35%) that would undergo radical prostatectomy, 18,500 could be lost (based on a 1999 US mortality rate for this procedure of 0.5-0.7%) and 43,000 could have a major cardiovascular complication. So, radical prostatectomy alone could kill two-thirds of the total number of men that currently die of prostate cancerand cause significant morbidities in double that number.

The report from Thompson et al. will obviously change the way we counsel our patients, but:

  • We should not rashly dismiss PSA.
  • PSA is an imperfect marker, but it is the best marker we have.
  • We can increase its sensitivity and specificity by measuring PSA velocity, free and total PSA ratios, Pro-PSA and other isoforms.

The true value of PSA will not be known until results of studies such as the Prostate, Lung, Colorectal, and Ovarian Screening Trial of approximately 148,000 patients become available in about 5 years. In the meantime, clinicians must recognize that the lack of a definitive PSA threshold prevents us from assuring patients that they do not have prostate cancer. We must individualize our diagnostic approach and provide more counseling about management strategies. Furthermore, research to hone PSA sensitivity and specificity, and to identify better serum and tissue markers, must continue.

Original Article

Thompson IM et al. (2004) Prevalence of prostate cancer among men with a prostate-specific antigen level ≤4.0 ng per milliliter. N Engl J Med 350: 2239-2246

Practice Point

The 4.0 ng/ml PSA threshold has been challenged; more frequent use of age-specific PSA reference ranges and isoforms should be encouraged, and the need for biopsy determined on a case-by-case basis

For more detailed information I recommend you to visit directly to the sources of this article:

Prostate Biopsy


  1. http://www.medscape.com/viewarticle/494379
  2. http://www.medscape.com/viewpublication/1171

About yuan ade sukma, MD

I'm an Indonesian doctor. I Believe that science and knowledge do not belong to anyone in the world. Science and knowledge is meant to be shared to make the world a better place to live. But if you find any material posted here is violating any copyrights, feel free to contact me and I will delete that. I believe that someday my blog can change the world. I BELIEVE in the POWER of WORDS! Do you..?
This entry was posted in Uncategorized and tagged , , , , , . Bookmark the permalink.

3 Responses to 15 Percent Men with PSA Less Than 4 ng/mL is Having Prostate Cancer

  1. Wow that was odd. I just wrote an really long comment but after I clicked submit my comment didn’t appear. Grrrr… well I’m not writing all that over again. Anyhow, just wanted to say fantastic blog!

  2. Dale Manzi says:

    Never knew this, thankyou for letting me know.

  3. Merely wanna admit that this is extremely helpful, Thanks for taking your time to write this.

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