Wednesday, March 9, 2011

PSA: Not Perfect But Vital

Three letters that many men fear are PSA.  These letters stand for Prostate Specific Antigen, a chemical produced by the prostate which is checked via a blood test to look for prostate cancer.  PSA has been an extremely important tool for doctors in identifying early prostate cancer.  Before PSA, prostate cancer could only be identified by means of a yearly rectal exam.  As prostate cancer is often advanced by the time it can be felt on an exam as a mass, many men at that time were diagnosed with prostate cancer after it had already spread outside of the prostate and could no longer be cured.  As a result, Urology wards in hospitals were often filled with men suffering from painful bone metastases that had no hope of cure.  After the advent of PSA, prostate cancer became a much more manageable disease.  Today, men are often diagnosed when just a few cells of prostate cancer are found in the prostate, making the cancer very curable and avoiding the unnecessary pain and suffering of metastatic disease.

While very beneficial, however, PSA is far from perfect.  First, PSA is not specific for prostate cancer.  As I previously mentioned PSA is a functional enzyme NORMALLY produced by the prostate.  In other words, PSA is not JUST produced by prostate cancer.  As a result, many factors can cause an elevation in the PSA test.  The most common cause of an elevated PSA that is NOT caused by prostate cancer is sexual intercourse.  Men who have any type of sex that results in ejaculation within 2-3 days of giving a blood sample usually demonstrate an elevated PSA.  This elevation of  PSA is caused by natural production and secretion of PSA by the prostate for sex rather than by prostate cancer.  Other causes of an elevated PSA  include a urinary tract infection (that may be asymptomatic), retention of urine, and even riding a motorcycle.  As a Urologist I have countless patients referred to me by their primary doctors for an elevated PSA test. Unless I feel an abnormality of the prostate during a rectal exam, I almost always repeat the PSA test, making sure that the patient does not have sex for 2-3 days prior to the test.  I also have the patient give a urine test at the same time to make sure that there is no infection.  In at least 50% of my patients, a repeat PSA test comes back as normal.  Multiple academic studies have demonstrated the same likelihood of what is called a false positive PSA test.  This means that over half of men with a single elevated PSA could avoid unnecessary procedures and worry by simply repeating the test in the right way!

Another problem with PSA is that sometimes it is too sensitive, meaning that it does too good a job in finding prostate cancer.  Studies have shown that 80% of men over the age of eighty have prostate cancer.  Because prostate cancer is usually slow growing, most of these men will never be symptomatically affected by the disease and so will die with it rather than from it.  They will die from more aggressive diseases such as heart disease or stroke.  To some extent, the same situation can be applied to younger men as well.  For example, a 70 year old man found to have a tiny amount of moderately aggressive prostate cancer may live into his 80s or 90s before that prostate cancer grows to any significant extent.  Many studies both conducted in the United States and in Europe have recently demonstrated that many men are diagnosed and treated for prostate cancer as a result of an elevated PSA test that would otherwise never have been significantly affected by prostate cancer within their lifetimes.

So what do all of these problems with the PSA test mean?  Should we get rid of it altogether?  I definitely think not.  While PSA is not a perfect test, it is vitally important if used the right way.  As I mentioned before, prior to PSA, many more men were needlessly dying of metastatic prostate cancer.  In addition, while identifying prostate cancer in 80 year old men MAY not be useful, finding any prostate cancer in healthy, 50 or 60  year old men while the cancer is still contained is VITAL.  PSA can help doctors identify CLINICALLY SIGNIFICANT prostate cancer that WILL affect and possibly shorten the lives of men who suffer from it. 

The real question then is how do we use PSA appropriately and for the most benefit.  First, as I mentioned previously, care must be taken to make sure that the PSA reading is actually real.  Elevated PSA tests should be repeated while minimizing the effects of other factors causing elevation of the test such as sexual intercourse.  In addition, current recommendations are for men over the age of 75 to stop getting PSA tests altogether.  The thinking behind this recommendation is that, because prostate cancer grows slowly, men in their mid to late seventies are unlikely to live long enough for a small amount of prostate cancer to grow enough to significantly affect them.  I don’t completely agree with this thinking because I treat many men in their 70s who are very healthy and will most likely live for at least 20 years.  As a result, I think the decision of whether to do  PSA testing should be tailored to the individual health and wishes of each patient and should entail a detailed discussion between men and their doctors.  I think that this approach should be applied to all patients who undergo PSA tests.  You and your doctor should take many factors into account when deciding whether to get a PSA test and how to interpret the results of that PSA test including your age, health, and desires.  If used in this patient specific manner rather than a “knee-jerk” fashion the PSA test can help save many lives while avoiding unnecessary procedures, pain, and worry. 

Check out my new Book: 


  1. What are your thoughts on PSA velocity? A well-regarded urologist at a leading academic medical center told me he thinks PSA velocity is more relevant than any absolute PSA number, in deciding who to biopsy. But then I read about studies such as this one that suggest it is less useful:

    Also, do you use PCA-3? Do you find it helpful?

  2. Chamorgadol: I feel that PSA velocity is more reliable than a single PSA but is still limited by the same problem: spurious values. If an abnormal PSA velocity is calculated from a falsely elevated single PSA value that PSA velocity is just as unreliable as that single value. That is why I feel that repeating an abnormal PSA test is so important before acting on it. As for the PCA3 it has potential, especially for men with negative biopsies. However, I have not seen studies that show it correlates with SIGNIFICANT cancer and so its increased sensitivity may not be beneficial as it may just detect cancer that did not need to be treated in the first place. Because it is so new the jury is still out on this test.

  3. Thanks. This is a useful blog. Balanced and good insights. I am interested in the new tests -- beyond psa -- that are being developed to try to diagnose prostate cancer and, more importantly, to try to distinguish between indolent and agressive disease. For example, have you heard anything about the Prostate Px tests, developed by Aureon? Sounds promising, though I have never met anyone who has used it. Or the detailed analysis of biopsy slides offered by Dr. Bonkoff in Germany -- This seems to go well beyond the usual pathology reports -- and even beyond the poloidy analysis you sometimes see --but I wonder how useful this information is in practice, or whether/how it actually affects decision-making. Same with color doppler ultrasound. Would you find all this stuff helpful in trying to help a patient decide what his risk is, and how to proceed?

  4. charmorgadol: thanks for the positive feedback. Although I have not tried these techniques, I have heard of them. Most of these new tests have not been implemented by the majority of urologists because there are not enough positive studies to make them mainstream yet. Some of them are encouraging. The main thing we are all looking for is a test to differentiate significant prostate cancer from an innocent bystander.