PSA is a serine protease, a glycoprotein which is a member of the kallikrein family. Its function is to liquefy semen. It is produced by the epithelial cells lining the ducts and acini.
It is found in high volumes in the semen, but also circulates in the serum both bound to protein and free.
PSA is organ specific, but not cancer specific.
What can affect serum PSA levels?
- Age
- Prostate volume
- Prostate disease – cancer, BPH, prostatitis
- Androgen levels – PSA is lower in hypogonadism
- 5-alpha reductase inhibitors will reduce PSA by about half after 6 – 12 months
- Prostatic massage
- Instrumentation or catheterisation
- Ejaculation
- Laboratory / assay used
There can be up to 20 % physiological variation in a man’s PSA level checked day by day.
What is a normal PSA level?
There is no normal PSA level – it is dependent on patient’s age, prostate volume, previous treatments, etc.
The risk of prostate cancer increases as the level of serum PSA increases but there is no threshold.
A cutoff of 4 ng / mL has used in the past – however 2/3 of men with a PSA above 4 will have a negative biopsy (i.e. PPV only about 30 %)
A man presents with an isolated PSA test above the median
PCFA guidelines:
For men aged 50 – 69 with PSA greater than 3.0 ng / mL – repeat the PSA within 1 – 3 months
For those with PSA 3.0 – 5.5 ng / mL – repeat with a free:total ratio
EAU guidelines:
In asymptomatic men with PSA between 3 – 10 ng / mL and normal DRE, repeat the PSA test prior to further investigations
Lower risk of clinically significant prostate cancer if repeat PSA is > 20 % lower
“Empiric use of antibiotics in an asymptomatic patient in order to lower the PSA should not be undertaken”
PSA adjuncts:
- Free to total PSA / Free %
- PSA density
- PSA kinetics / velocity
- PSA doubling time
- Phi
Free to total ratio
Men with prostate cancer have a higher percentage of PSA bound or complexed to protein in the serum; and therefore lower percentage of free PSA.
Free % has only been shown to be useful in men with total PSA 4 – 10 ng / mL – utility is unproven outside that range.
It has no role in men with known prostate cancer.
It can be affected by dialysis (total PSA is not) and may vary in different assays.
The index study for free % showed prostate cancer in 56 % of men with free % < 10, but only 8 % of men with free % > 25 (in men with total PSA 4 – 10).
PCFA recommendation – Free to total ratio should be used in assessment of men with PSA > 3. Biopsy should be offered to men with PSA between 3.0 and 5.5, if free % is less than 25 %.
PSA density
PSA (ng / mL) divided by prostate volume.
Higher PSA density = higher likelihood of clinically significant cancer, especially in smaller prostates with a PSA density > 0.15
Men with a PSA density < 0.1 are less likely to have clinically significant cancer.
PSA density may be useful as a “tie-breaker” or decision aid in men with PIRADs 2 or 3 MRIs.
PSA kinetics
PSA velocity has been suggested in some studies to be an aid to decision making for biopsies. A traditional threshold of > 0.75 ng / mL increase per year was worrisome (published 1992).
PCFA guidelines do not recommend using PSA velocity to increase specificity for PSA > 3.0, and their review found it did not appear to improve diagnostic performance.
In the setting of a negative biopsy – there may be association between PSA kinetics and risk of undiagnosed cancer / cancer found on repeat biopsy.
I agree with EAU guidelines that PSA kinetics in the diagnostic workup has limited use due to “background noise” and I find other adjuncts such as density and free %, as well as MRI are more helpful for decision making.
Other settings:
- PSA velocity of > 1 ng / mL increase / year predicted progression on active surveillance.
- It may be more useful in post-treatment PSA monitoring or watchful waiting of prostate cancer.
PSA doubling time
PSA doubling time can be calculated using a variety of statistical methods or calculators which may differ in their methods.
It has little utility in the diagnosis of early screen-detected prostate cancer.
There is conflicting evidence about PSAdt use in active surveillance, with some studies suggesting doubling time < 3 years associated with higher rates of progression after treatment.
It can be a useful measure in advanced prostate cancer (non metastatic and metastatic, including castrate resistant).
Prostate health index
PHI uses a combination of total PSA, free PSA and proPSA to give a score which may be associated with risk of prostate cancer (separated into low, intermediate and high risk).
PSA is initially secreted from the epithelium of the prostatic duct lumen as a precursor – proPSA.
PHI is approved in the USA for use in men with total PSA 4 – 10 ng / mL.
PCFA recommendation – do not use the PHI test to increase specificity of a PSA test greater than 3, except in the context of research.
Other biomarkers:
- 4K
- PCA3
- SelectMDx
4K
Blood test – 4 kallikrein score uses total PSA, free PSA, intact PSA and human kallikrein 2.
Also assesses age, prior biopsies and DRE and puts all information into an algorithm to give a score aiming to assess risk of detecting clinically significant prostate cancer.
PCA3
Post prostate massage urine sample.
Prostate cells in the voided urine are analysed for PCA3 (prostate cancer gene 3) which is an mRNA overexpressed in prostate cancer. Higher levels suggest higher prostate cancer risk.
Approved in USA for men with previous negative biopsy.
SelectMDx
Post prostate massage urine sample.
Assesses for RNA levels of DLX1 and HOXC6 and then generates a reading including algorithm incorporating age, PSA, prostate volume, DRE and family history. Predicts likelihood of prostate cancer and clinically significant prostate cancer.
Stockholm3
Prediction model based on several clinical variables (age, family history, previous biopsy) and blood markers (PSA, free PSA, free PSA %, human kallikrein 2, MSMB and macrophage inhibitory cytokine-1) -> polygenic risk score to determine likelihood of Gleason 7+ prostate cancer.
Shown to reduce the percent of clinically insignificant prostate cancers when used in combination with MRI in a PSA screening population
EAU guidelines:
Some biomarkers could help in discriminating between aggressive and non aggressive tumours…however upfront MRI is also likely to affect the utility of above-mentioned biomarkers
The role of PHI, PCA3 and SelectMDx in deciding whether to take a repeat biopsy in men with previous negative biopsy is uncertain and probably not cost-effective.