Frequently Asked Questions - Prostate-Specific Antigen (PSA) Test

The PSA blood test is probably the single most accurate test we have for the detection of prostate cancer. PSA is a protein that’s produced in the prostate gland. Normally, the PSA should be very low in the bloodstream.

Our studies showed that the median PSA level is 0.7 for men in their 40s, 0.9 for men in their 50s and 1.4 for men in their 60s.

If the prostate is diseased – from benign enlargement, inflammation, or prostate cancer – the PSA is elevated.

If it’s diseased, the elevated PSA is like the red light going on in the cockpit of an airplane. It indicates that there is something wrong in the prostate gland. It may not be prostate cancer; it may be a benign condition, but there’s something going on there that needs further evaluation, which usually means a biopsy of the prostate. Nevertheless, the higher the PSA level, the more likely that a man has prostate cancer and the more likely that it will be aggressive and need treatment.

The PSA is not a perfect test. Initially, people were hoping for something like a pregnancy test. If the test was positive, the patient always had cancer, and when it was negative, the patient never had cancer. That’s almost too much to ask of any medical test. When used properly, the PSA test can detect prostate cancer in a curable stage in the great majority of patients.

Rather than consider the PSA range in terms of “normal or abnormal,” it is more useful to consider it as a means of assessing the risk that prostate cancer would be found on a biopsy.

For instance, regardless of age, if a man has a perfectly healthy prostate gland (without inflammation, benign enlargement or cancer) the PSA should be very low – certainly less than 2. Our studies showed than the median PSA level is 0.7 for men in their 40s, 0.9 for men in their 50s and 1.4 for men in their 60s.

If there is cancer, inflammation, or benign enlargement, or a combination of these conditions, the PSA level will be higher.

If the PSA is in the 2.5-4 range, the chances of finding cancer on biopsy are about 25%. If the PSA is 4-10, the chances of finding cancer are 35-40%. If the PSA is higher than 10, the chances of finding cancer are about 60%.

With cancer and benign enlargement, the PSA levels always go up and never come down on their own. With inflammation, PSA levels go up with a flare-up and come back down with resolution of the inflammation.

There are other tests that help discriminate between elevations caused by prostate cancer and those caused by other benign conditions. These include PSA rate of change, PSA indexed to the prostate volume, and measurements of free or complexed PSA.

In general, the more rapidly the PSA rises, the more likely there is cancer, except for very dramatic rises of more than 2ng/ml per year that are more likely to be caused by inflammation in the prostate gland (prostatitis).

Recent studies have caused me to believe that men should have an initial PSA test earlier than I previously recommended.

All men should have an annual PSA test and a digital rectal exam beginning at age 40 to assess their risk of having prostate cancer or developing prostate cancer in the future. And if men have a family history of early age-at-onset prostate cancer, then the PSA should be tested at 35.

I recommend biopsy for a suspicious DRE or a PSA of more than 2.5. Also, I recommend that PSA velocity be considered in biopsy decisions.

Some doctors do not recommend a biopsy if the PSA is over 2.5 or even if it is over 4. (Again, my recommendation is biopsy with a PSA 2.5 or over.) However, if the PSA velocity is higher than 0.5 ng/ml/year in a man whose total PSA is below four or if the PSA velocity is higher than 0.75 in a man with a PSA higher than 4, it is even more important to have a prostate biopsy.

An important caveat is that very dramatic increases in PSA (more than 2 ng/ml/year) are more likely to be caused by prostatitis rather than prostate cancer.

Also, in patients who have had a prior biopsy for a PSA over 2.5 or over 4.0 and the biopsy was negative for cancer, a rising PSA makes a case for repeating the biopsy procedure.

These recommendations are similar to the 2004 National Comprehensive Cancer Center Network guidelines.

Yes. By congressional act, all Medicare and Medicaid patients can receive an annual PSA test as part of their benefits.

The following can cause the PSA to rise: cancer, benign enlargement, and inflammation (prostatitis) of the prostate, or infection of the urinary tract.

With cancer and benign enlargement, the PSA goes up persistently; the slope is steeper with cancer. With infection, the PSA can rise with a flare-up and come back down with resolution of the inflammation.

A first response to an elevated PSA is an initial two-week prescription of CIPRO, the medication for treatment of an infection in the urinary tract or prostate gland. If, after taking CIPRO, the PSA lowers and stays low, then the cause of the elevated PSA was an infection.

Interpreting the PSA pattern is complicated because some patients have two or three conditions at the same time. If CIPRO results in a lower PSA, I would still recommend retesting in 3 months and consider a prostate biopsy if the PSA has risen again or remains high with respect to the median value for the age group.

It is important to realize that PSA levels may fluctuate from 10-15% between readings. With prostate cancer, PSA levels usually increase about 20% per year, depending upon the growth rate of the tumor.

The rate of rise in PSA level is called PSA Velocity and we now think it could be one of the most important indicators of prostate cancer and the aggressiveness of the prostate cancer.

A persistently rising PSA level can be an indication of prostate cancer, even if the absolute level is within the “normal” range. On the other hand, very dramatic rises in PSA of greater than 2ng/ml per year are more likely to be caused by prostatitis.

PSA exists in the blood in two forms: one is free floating (free) and the other is attached to proteins.

In patients with PSA elevations due to cancer, more of the PSA is in the attached form and less is in the free form.

In patients with PSA elevations due to benign conditions such as benign enlargement or inflammation, more PSA is in the free form.

Thus, in patients whose PSA values are in the gray zone (2.5-10) in terms of prostate cancer risk, the percentage of PSA that is free and that is attached helps to evaluate the risk of prostate cancer.

For example, if the total PSA is between 4 and 10 and the % free PSA is over 25%, there is only an 8% chance that a biopsy would show cancer. On the other hand, if the % free PSA is less than 10%, the chances for prostate cancer would be almost 60%.

One of the problems with PSA blood tests is that if the PSA level is elevated, about 25 to 60 percent (depending on how high the PSA level is) of those men are actually found to have prostate cancer. That means roughly 40% to 80% of men who have elevated PSA levels don’t have prostate cancer and they go through the biopsy unnecessarily.

The free PSA blood test can eliminate at least 20 percent of these biopsies and still detect 95 percent of these cancers.

Yes. Usually, the larger the prostate gland, the higher the PSA. The ratio of serum PSA to prostate volume is called PSA density. If the density is greater than 0.1, there is a suspicion of cancer. A lower density is most consistent with benign enlargement but does not absolutely rule out prostate cancer.

After the prostate gland has been completely removed, any PSA in the blood is produced by prostate cancer cells that have left the prostate gland and spread to other areas of the body. However, sometimes the PSA can appear to be elevated because of a laboratory error.

All elevated PSA levels after surgery should be re-checked to rule out a laboratory error.

For practical purposes, it is very difficult to detect PSA in tests when the levels are less than 0.2 ng/ml; therefore, any PSA value less than 0.2 is considered negligible. Hence, a value of 0.2 ng/ml or 0.1 ng/ml would be considered essentially zero. (Other prostate cancer treatments might have different time-frames for post-treatment PSA values, but at some point, the PSA value of more than 0.2 indicates a possibility if recurrence.

Clearly, early detection of prostate cancer is going to have more potential benefit for men in their 50s and early 60s because of their life expectancy. It is harder to show a benefit in older men who are more likely to die of another cause first. But with age expectancy rising, the definition of older men is also changing.

In addition, many older men develop aggressive prostate cancer that ultimately kills them in a very painful way. Early detection and some form of treatment can help prevent that outcome as well.

Statistical models for overdetection do not take into consideration the clinical judgment of a well-informed doctor and patient regarding the risk-benefit ratio of screening and treatment for prostate cancer.

Knowledge is useful, and it is usually beneficial to know whether cancer is likely to be present and whether there are signs of progression, such as a rising PSA level, so appropriate intervention can take place, if indicated. Some “overdetection” takes place with any cancer-screening program, but good clinical judgment can usually prevent this occurrence from harming the patient.

It is easier to call overdetection in patients with a short life expectancy than in patients with a long life expectancy. In a young patient with a very long life expectancy, it is not possible to say with certainty that a prostate cancer does not have or will not acquire the capacity to cause disability and/or death.

When the total PSA is between 2.5 and 4, freePSA results still work in assisting diagnosis, but they are less robust in avoiding unnecessary biopsies.
Accordingly, if the % free PSA is quite low (10% or less), it is worrisome for prostate cancer; and if it is very high (25% or more), it suggests that benign hyperplasia is the main cause for the elevated PSA.

Changes in percent freePSA over time can also be helpful in some cases. But the correlations are not so strong as they are in the 4-10 ng/ml PSA range.

They do elevate the PSA for about 24 hours and could have an effect on the PSA results.

This issue is in flux at present. Currently, it is thought that a PSAV between 0.3 and 0.5 should be used to recommend a biopsy In men with a PSA lower than 4. New 2007 guidelines will be issued by the National Comprehensive Cancer Center Network and the American urological Association this year, and I believe they will recommend 0.4 and 0.35, respectively. It looks as if prostate cancer should be diagnosed before the PSA is above 4 and the PSA velocity gets above 0.5. Otherwise, a lot of men will be detected with cancer that already has spread to the margins of the prostate gland or beyond. *PSA Velocity is the rate at which PSA rises in a year.

Antibiotics can be a first response to any increasing PSA or to a PSA level that might trigger a biopsy. With cancer, the PSA doesn’t decrease. With infection, it sometimes but not always does.An alternative to antibiotics would be to repeat the PSA after an interval and see if it comes down on its own.Some experts think it is wrong to use antibiotics unless there is a culture test which shows an infection, but I do not agree. My current practice is to prescribe an antibiotic other than Cipro to see if the PSA decreases. It it doesn’t, I wait one month to give the intestinal baceria a chance to return to normal and then perform a biopsy. Alternatively, some patients elect to see whether the PSA will decrease spontaneously without antibiotics and avoid the risk of developing resistant bacteria in their intestines.

Using antibiotics to rule out a false positive PSA test caused by inflammation or infection in the prostate gland is controversial, because in most instances, the inflammation may not be caused by a bacterial infection and because of concerns about irresponsible use of antibiotics contributing to the development of resistant strains of bacteria. Nevertheless, in my experience it has allowed me to avoid performing a biopsy procedure in many patients in my practice. In the situation described in the question where the PSA appeared to have decreased in response to antibiotic therapy, I believe that another trial of antibiotic therapy should be considered. It would also be important for the physician to try to determine whether there is objective evidence for a urinary tract infection by examining the patient’s urine and prostatic fluid. The time course of the PSA returning to normal is extremely variable. Sometimes it occurs within a week and sometimes it takes months. Also, in many cases of prostate inflammation, the PSA will not decrease substantially with antibiotic therapy.

In patients whose total PSA is elevated mainly because of prostate cancer, very little of the PSA is free floating in the blood; most is bound to proteins. Therefore, in a man with a PSA of 6, if the percentage of free-PSA is less than 10%, there is more than a 50-50 chance that he has prostate cancer. In contrast, if another man who also has a total PSA of 6, and the percentage of free-PSA is greater than 25%, there is less than a 1 chance in 10 that the PSA elevation is caused by prostate cancer. However, the free-PSA test is not perfect, so other factors need to be considered, such as the PSA-Velocity, PSA-Density, prior biopsy history, race, family history of prostate cancer, etc. Moreover, in the future, it is likely that the Pro-PSA test (not yet FDA approved) will improve on the accuracy of the free-PSA test.

I tell my patients that there are three main conditions that cause the PSA to increase: prostate cancer, benign prostate enlargement (called BPH), and prostate inflammation. Some men have one of these, some have two, and some have all three. With prostate cancer, the PSA goes up, up, up and does not decrease unless there is treatment. With BPH, it is the same, except the increase in PSA is not so steep. With inflammation, the PSA goes up with a flareup and decreases as inflammation resolves. A fluctuating PSA level indicates that at least part of the PSA elevation is due to inflammation. However, if the PSA does not decrease spontaneously or decrease following a trial of antibiotic therapy to below the median for the age group (0.7 for men in their 40s, 0.9 for men in their 50s, 1.3 for men in their 60s, and 1.7 for men in their 70s), I believe a prostate biopsy should be strongly considered.

The increase in you PSA followed by a decrease suggests that it might have increased because of inflammation in the prostate gland. However, it also could be due to issues related to differences in standardization of different PSA tests. (see article from Winter 2007 Quest, PSA Standardization Dilemma on my website: www.drcatalona.com)

I would recommend that you either take an empirical course of antibiotic therapy to see whether the PSA decreases to the base-line level or re-check the PSA again in one month to see whether it decreased spontaneously. If the PSA does not decrease, I would recommend you have a prostate biopsy procedure.

All elevated PSA levels after surgery should be re-checked to rule out a laboratory error. For practical purposes, it is very difficult to detect PSA when levels are less than 0.1 ng/ml. Therefore, we consider any PSA value less than 0.1 as being negligible.

A value of 0.1ng/ml would be considered essentially zero. PSA 0.2 or above is an indication of cancer recurrence, and treatments for recurrence should begin before the PSA reaches 1.0ng/ml.

Decimal points are important in interpreting PSA tests after surgery. For example, 0.003 is essentially zero; whereas, 0.3 is a level about which concern should begin.

The patient asking this question has a score that is essentially zero. He has no reason to be concerned.

Sometimes, patients need to make sure they’re using their basic math skills so they don’t worry needlessly.

But if patients have any questions about postoperative PSA results, they should talk to their doctors.

Neither Cardura nor prescription pain killers have an effect on PSA values. Since Saw Palmetto is not regulated by the FDA, different preparations contain different ingredients. Some preparations can affect PSA levels and some do not, but no one would have a way of knowing without testing individual batches.

It could. We now know that “free” PSA is made up of several components: BPSA and IPSA increase with benign prostatic enlargement (BPH) and proPSA increases with cancer.

So, if the prostate shrinks because of treatment for BPH, the free PSA level could decrease.

To avoid having to repeat the test if it is ordered too early, I usually recommend that the patient wait at least one month.

In patients who have recurrence of cancer after surgery, 50% of the recurrences occur during the first 3 years, 80% during the first 5 years and 99% during the first 10 years.

Thus, some physicians feel that PSA should be monitored more frequently early after surgery and less frequently as time passes.

Too frequent monitoring can cause “PSA anxiety.”

I usually recommend PSA testing every 6 months.

PSA is an excellent marker for use during and after cancer therapy.

Some very high grade cancers do not produce much PSA and a special form of prostate cancer called a “neuroendocrine” tumor does not produce PSA. In these unusual circumstances, the PSA is not a good marker for determining the status of the disease..

But 99% of the time, PSA lets you know how the cancer is responding to the treatment.

With prostate cancer, as with any form of cancer, some “rogue”cells can escape from the prostate before surgery. They are so few in number that you can’t see them on any scans or detect them with blood tests.

The surgeon removes the prostate and the pathologist says it looks as if all the cancer has been removed, and the PSA becomes undetectable. Still, those “rogue” cells will grow and then, later, any PSA they produce will be taken up in the bloodstream.

These “rogue” cells are why it’s really important for any man who’s been treated for prostate cancer to have follow up visits.

I recommend a PSA test every six months for 15 years after the operation.

The waiting time is 48 hours after a rectal exam. A man should wait 6 to 7 weeks after a biopsy.

It provides useful information on the risk of finding prostate cancer on a biopsy in men whose PSA level is between 2.5 and 10. It is a more robust predictor in the PSA range of 4 to 10 than in the lower range.

In the PSA 4-10 range, if the percent free PSA is less than 10%, the likelihood of finding cancer on biopsy is more than 50%; whereas, if the percent free PSA is greater than 25%, the likelihood is less than 10%.

Yes, when the total PSA is much above 10.

They should be only a part of PSA tests used for early detection of prostate cancer. They would not be useful for PSA tests determining effectiveness of treatment for recurrence.

It is reported separately from the total PSA. The % free PSA is calculated as the free PSA divided by the total PSA times 100. (Free/Total x 100)

A %free PSA of more than 25% means that there is about an 8% chance that a biopsy would show prostate cancer; whereas, a %free PSA of less than 10% would mean there is nearly a 60% chance that a biopsy would show cancer.

Also, lower percentages of free PSA are also more likely to be associated with more aggressive forms of prostate cancer.

It is the time required for the PSA level to double. It is dependent to a large extent on the baseline PSA level.

For instance, a PSA level of 0.1 could easily double to 0.2 without much change in the total PSA. However, it would require a much greater increase in PSA for the level to double from 4 to 8.

PSA doubling time is most useful in evaluating the aggressiveness of a cancer that has recurred after radical prostatectomy or hormonal therapy.

Yes, always. Sometimes it can remain elevated for many months too.

Yes, because the DRE can cause a slight elevation in the PSA that can last for a few days.

If they are not allergic to sulfa drugs, I usually begin with trimethaprim-sulfamethoxazole twice daily. If they are allergic to sulfa, I choose another antibiotic – some- times trimethaprim by itself.

Alternatively, sometimes I do not use antibiotics and just repeat the PSA after two or three weeks to see whether the inflammation is resolving on its own.

Conversely, if it is non-aggressive and slow growing, even treatment after symptoms will be successful. And perhaps, if not discovered, no treatment will ever be necessary.

What would you say to this view?

A: It is simply wrong. And the misconception is harmful to men.

The fact is that many cancers are in between these extremes and can be cured if detected early but will be lethal if detected late.

The proof of the pudding is in the decrease in the prostate cancer-specific death rate since early detection screening.

Yes, but because free PSA is a smaller molecule, it is cleared from the blood stream by the kidneys more quickly and thus its levels return to normal more quickly than PSA bound to proteins.