Quick answer
Free PSA and PSA density were both designed to answer the same question: does a high PSA come from cancer or from a benign prostate? Free PSA measures the share of PSA that circulates unbound in the blood, and the lower that share, the higher the chance of cancer.2 Density divides the PSA by the volume of the prostate, in grams.3 Both help, but today a prostate MRI answers that question more directly.1,7 In my practice, I do not order free PSA routinely, and density comes in after the MRI, as one of the parameters.
What free PSA is
Some of the PSA in the blood is bound to other proteins, and some circulates on its own. The standard test measures the total. Free PSA measures only the unbound part, and the ratio shows how much of the total it makes up.
The study that set out how to use the test was published in 1998. It included 773 men aged 50 to 75, with a normal rectal exam and a PSA between 4 and 10 ng/mL, all of whom had a biopsy.2 The lower the ratio, the higher the risk of cancer, which ranged from 8% to 56%. Doing a biopsy only in men with 25% free PSA or less found 95% of cancers and avoided 20% of unnecessary biopsies.2 The cancers missed by that cutoff were more common in older men and were generally smaller and of lower grade.2
The test has limits. The study applied to a PSA between 4 and 10 ng/mL with a normal rectal exam, and it measured the chance of any cancer. Today the question that matters most is whether there is a significant cancer, one that needs treatment. In the comparison made by the European guideline, PHI and the 4Kscore performed better than free PSA at detecting cancer.1
What PSA density is
A large prostate makes more PSA, even without cancer. Density corrects for that. It was described in 1992, in a small study of 61 men. The average was 0.044 in benign enlargement and 0.581 in cancer.3
The math is simple. A PSA of 6 ng/mL in a 40-gram prostate gives 0.15. The same PSA in a 60-gram prostate gives 0.10, and in a 25-gram prostate, 0.24. The volume comes from an imaging test, and today almost always from the MRI itself.
In the studies, the cutoffs most often linked to cancer sit between 0.10 and 0.15, and 0.20 is the cutoff used in the European guideline's recommendations.1 These are the figures it cites:
| Situation | Chance of significant cancer |
|---|---|
| Density below 0.09 | 4% |
| PI-RADS 1 or 2 and density below 0.10 | 3% |
| PI-RADS 1 or 2 and density above 0.20 | 18% |
| Negative MRI and density above 0.15 to 0.20 | 27% to 40%, depending on the study |
Density grades the risk, and the figures in the table show it says more when it comes with an MRI.1
What the European guideline says
The European guideline is the one I respect most. For men without symptoms and a PSA between 3 and 20 ng/mL, it advises using a tool before recommending a biopsy. Prostate MRI carries a strong recommendation. A well-calibrated risk calculator or a blood or urine biomarker carries a weak one.1
Density comes in alongside the MRI. The guideline accepts skipping the biopsy and monitoring the PSA when the PI-RADS score is 2 or lower, the density is below 0.20 and there is no prostate cancer in the family. With PI-RADS 3 and a very low suspicion, meaning a density below 0.10 and no family history, it also accepts monitoring without a biopsy.1 Both recommendations are weak and meant to be decided together with the patient.
Density also works the other way. When the MRI is negative but suspicion stays high, for example with a high density, the guideline notes that a systematic biopsy is still needed.1
This is where my practice differs: with PI-RADS 3 or higher, I always recommend a biopsy. I explain the scores in MRI and PI-RADS.
PHI, 4Kscore and Stockholm3
After free PSA came blood tests that combine several markers. All of them estimate the risk of significant cancer to help decide on a biopsy:
| Test | What it combines | What the main study showed |
|---|---|---|
| PHI | Total PSA, free PSA and a precursor form of PSA, proPSA | In 892 men with a PSA between 2 and 10 ng/mL, it separated cancer better than PSA and free PSA4 |
| 4Kscore | Four proteins from the PSA family plus clinical data | In 1,012 men, it could allow 30% to 58% fewer biopsies, with a delayed diagnosis in 1.3% to 4.7% of significant cancers, depending on the cutoff5 |
| Stockholm3 | Blood proteins, genetic variants and clinical data | At the same sensitivity as a PSA of 3 ng/mL or more, it could allow 32% fewer biopsies and avoid 44% of biopsies without cancer6 |
Despite these figures, the European guideline considers that the efficacy of these tests has not been validated and that none of them can be recommended for clinical practice at present.1 It notes that Stockholm3, used together with MRI, reduced the share of clinically insignificant cancers.1
Why MRI changed their role
Free PSA, density and the biomarkers give a risk. MRI also grades the risk, and it shows whether there is a lesion and where it is. That lets the biopsy be aimed at the suspicious area.1
In the PROMIS study, 576 men had an MRI and a mapping biopsy of the whole prostate. MRI picked up 93% of significant cancers. Standard biopsy through the rectum, without MRI, found 48%. Used before biopsy, MRI could let 27% of men avoid a first biopsy.7
MRI is not perfect either. In the same study, it was less specific than biopsy: many areas flagged as suspicious had no significant cancer.7 That is why a biopsy is still the test that confirms the diagnosis.
The MRI also measures the volume of the prostate. With the PSA at hand, the density comes from that number, with no separate test.
How I approach it
In my practice, I do not order free PSA routinely, or PSA density before the MRI. To me, MRI has made these tests redundant, because it answers the question more directly. I always look at the density after the MRI, calculated from the prostate volume: with PI-RADS 1 or 2, it is one of the parameters I use, but not the only one, and I weigh everything about the patient before deciding on a biopsy.
I recommend starting PSA testing at 40 for all men. I have seen many patients aged 40 to 45 with high-risk prostate cancer. When the PSA is normal, I repeat it once a year. Before the test, I ask for at least three days of rest for the prostate: no ejaculation, no cycling and no rectal exam. What else can change the result is covered in what affects PSA.
When the PSA is abnormal, I go straight to a prostate MRI without repeating the test first. The European guideline advises repeating the PSA when it is between 3 and 10 ng/mL, before any other test.1 With PI-RADS 3 or higher, I always recommend a transperineal biopsy, done under sedation.
I keep testing PSA for as long as a man is healthy and has a life expectancy of more than 10 years.
When to see a urologist
- If your PSA is abnormal, even without symptoms
- If your report shows a low free PSA or a high density
- If your PSA goes up from one test to the next
- From age 40, to start PSA monitoring
Expected values for each age group are in PSA by age, and when to begin is in when to start PSA testing.
Frequently asked questions
What is the free-to-total PSA ratio?+
It is the percentage of PSA that circulates unbound in the blood, not attached to other proteins. The lower the percentage, the higher the chance of cancer. In the study that set out how to use the test, in men with a PSA between 4 and 10 ng/mL and a normal rectal exam, the risk of cancer ranged from 8% to 56% depending on the ratio.2
What is a normal free PSA?+
There is no single value that separates normal from abnormal. The best-known cutoff is 25%: in the 1998 study, doing a biopsy in men with 25% free PSA or less found 95% of cancers and avoided 20% of unnecessary biopsies.2 The result only applies to men with a PSA between 4 and 10 ng/mL.2
How is PSA density calculated?+
The PSA value is divided by the volume of the prostate, in grams, measured on an imaging test.3 A PSA of 6 ng/mL in a 40-gram prostate gives a density of 0.15. The higher the density, the higher the chance of a significant cancer.1
Does a PSA density above 0.15 mean cancer?+
Not necessarily. It points to a higher risk. Several studies found a density above 0.10 to 0.15 to be predictive of prostate cancer.1 With a normal MRI and a density above 0.15 to 0.20, the chance of significant cancer was 27% to 40% in the studies cited by the European guideline.1
Can a normal MRI and a low density mean no biopsy?+
The European guideline accepts skipping the biopsy and monitoring the PSA when the PI-RADS score is 2 or lower, the density is below 0.20 and there is no prostate cancer in the family.1 It is a weak recommendation, meant for shared decision-making. In my practice, with PI-RADS 3 or higher, I always recommend a biopsy. With PI-RADS 1 or 2, I always look at the density, but it is not the only parameter: I weigh everything about the patient before deciding.
Do PHI, the 4Kscore or Stockholm3 replace a biopsy?+
No. They are blood tests that estimate the risk of significant cancer and can reduce the number of biopsies.4,5,6 The European guideline considers that their efficacy has not been validated and that no specific test can be recommended for clinical practice at present.1 When one of them points to risk, a biopsy is still what confirms the diagnosis.
Should I ask for a free PSA if my PSA came back high?+
In my practice, no. When the PSA is abnormal, I go straight to a prostate MRI. The European guideline places MRI first among the tools for deciding on a biopsy, with a strong recommendation, and biomarkers and risk calculators as alternatives, with a weak one.1 In the PROMIS study, MRI picked up 93% of significant cancers.7
Who can review free PSA and PSA density results in São Paulo?+
I see men with an abnormal PSA in São Paulo. I am Dr. Petronio Melo, a urologist and robotic surgeon, CRM-SP 157.598 and RQE 70.275. I hold a PhD from the University of São Paulo Medical School, with award-winning research on artificial intelligence in prostate cancer diagnosis, and I have performed more than 1,500 robotic and laparoscopic surgeries, more than 500 of them robotic. When the PSA is abnormal, I order a prostate MRI and, with PI-RADS 3 or higher, I recommend a transperineal biopsy, always under sedation. My office is in Vila Mariana, and the first consultation can be in person or online, in English.
References
- EAU-EANM-ESTRO-ESUR-ISUP-SIOG Guidelines on Prostate Cancer. European Association of Urology; 2026. EAU
- Catalona WJ, Partin AW, Slawin KM, et al. Use of the percentage of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. JAMA. 1998;279(19):1542-1547. PubMed
- Benson MC, Whang IS, Pantuck A, et al. Prostate specific antigen density: a means of distinguishing benign prostatic hypertrophy and prostate cancer. J Urol. 1992;147(3 Pt 2):815-816. PubMed
- Catalona WJ, Partin AW, Sanda MG, et al. A multicenter study of [-2]pro-prostate specific antigen combined with prostate specific antigen and free prostate specific antigen for prostate cancer detection in the 2.0 to 10.0 ng/ml prostate specific antigen range. J Urol. 2011;185(5):1650-1655. PubMed
- Parekh DJ, Punnen S, Sjoberg DD, et al. A multi-institutional prospective trial in the USA confirms that the 4Kscore accurately identifies men with high-grade prostate cancer. Eur Urol. 2015;68(3):464-470. PubMed
- Grönberg H, Adolfsson J, Aly M, et al. Prostate cancer screening in men aged 50-69 years (STHLM3): a prospective population-based diagnostic study. Lancet Oncol. 2015;16(16):1667-1676. PubMed
- Ahmed HU, El-Shater Bosaily A, Brown LC, et al. Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study. Lancet. 2017;389(10071):815-822. PubMed
Would you like to talk about your case?
In the consultation, I go over your tests and your priorities.
Book a consultation
