Prostate Cancer Risk Calculator

Dr. Mark Wickman headshot Medical review by: Dr. Mark Wickman

Educational tool, not a diagnosis or screening recommendation. PSA results describe probabilities and patterns in groups. They cannot establish or exclude prostate cancer in one person. Discuss testing and follow-up with a qualified clinician who knows your history, examination findings and medicines.

Introduction to PSA risk indicators

Prostate-specific antigen, usually shortened to PSA, is made by prostate tissue rather than by cancer alone. A raised blood result can occur with prostate cancer, but it can also reflect benign prostate enlargement, inflammation, infection, urinary retention or recent instrumentation. Cycling, ejaculation and differences between laboratory assays may also affect a reading. This is why one PSA number should not be treated as a diagnosis.

This calculator places total PSA beside four commonly discussed indicators. It compares the result with an age-specific reference limit, divides PSA by prostate volume when an imaging measurement is available, estimates the annual rate of change when a prior result is supplied and looks up a published biopsy probability when percent free PSA falls within the population studied by Catalona and colleagues.

The indicators remain separate in the result. Adding their flags together would imply that they form a validated score, which they do not. A family history and digital rectal examination result are also recorded, but the page does not invent numerical coefficients for them.

How to use this PSA assessment

Enter age in years and total PSA in ng/mL. Those are the only required values. Prostate volume, measured in cubic centimetres or cc, normally comes from an MRI or transrectal ultrasound report. Percent free PSA comes from a laboratory free-and-total PSA panel. If the laboratory gives free PSA as a concentration rather than a percentage, divide free PSA by total PSA and multiply by 100.

A previous PSA and the number of months since that test allow the calculator to estimate velocity. Ideally, compare at least three tests obtained over 18 to 24 months using the same laboratory assay. A velocity based on only two draws can be strongly affected by ordinary biological and measurement variation.

Select the digital rectal examination result accurately. The free-PSA probability table used here was measured in men aged 50 to 75 whose total PSA was 4.0 to 10.0 ng/mL and whose examination was normal. The calculator withholds that probability when those conditions are not met instead of extrapolating beyond the study.

Formulas behind PSA density, velocity and free PSA

PSA density adjusts the total blood result for the amount of prostate tissue producing it. With prostate volume $V$ in cc, the calculation is:

PSAD = \frac{PSA}{V}

The result has units of ng/mL per cc. A value of 0.15 or higher is a long-standing clinical reference point, not an automatic biopsy rule. A larger gland may produce more PSA through benign enlargement, so density can add context that total PSA alone lacks.

PSA velocity annualises the difference between the current result and a prior reading ${PSA}_{0}$ obtained $m$ months earlier:

PSAV = \frac{PSA - {PSA}_{0}}{m / 12}

The unit is ng/mL per year. The classic comparison point is 0.75 ng/mL per year, although modern practice generally gives short-term velocity less weight because PSA fluctuates.

Percent free PSA describes how much circulating PSA is unbound:

% fPSA = 100 \cdot \frac{{PSA}_{free}}{{PSA}_{total}}

A lower free fraction was associated with a higher chance of finding cancer on biopsy in the study population. Even the most reassuring band was not zero risk, and the percentages should not be used outside the study conditions.

Reference thresholds used for the PSA result

Age-specific PSA upper reference limits
Age	Upper limit	Context
40–49	2.5 ng/mL	Lower limit than the historical flat 4.0 threshold
50–59	3.5 ng/mL	Accounts partly for age-related gland growth
60–69	4.5 ng/mL	Slightly above the historical threshold
70–79	6.5 ng/mL	More permissive for benign enlargement

Published probability of cancer on biopsy by percent free PSA
Free PSA	Probability	Study-band interpretation
0–10%	56%	Highest table band
Above 10–15%	28%	Intermediate
Above 15–20%	20%	Intermediate
Above 20–25%	16%	Lower
Above 25%	8%	Lowest table band, not zero

The positive predictive value formula

Sensitivity is the proportion of people with the condition who test positive. Specificity is the proportion without it who test negative. Neither number alone tells you what a positive result means. Positive predictive value also depends on prevalence, or the base rate $p$ , in the tested group:

PPV = \frac{p \cdot Se}{p \cdot Se + (1 - p) \cdot (1 - Sp)}

The cohort calculations used in the optional game start with a population of size N. The expected number of true-positive or found cases is:

TP = N \cdot p \cdot Se

Cases that are present but missed by the selected threshold are calculated from the complement of sensitivity:

FN = N \cdot p \cdot (1 - Se)

False alarms occur among people without the condition. Their expected count depends on one minus specificity:

FP = N \cdot (1 - p) \cdot (1 - Sp)

Correctly cleared results are the people without the condition who receive a negative test result:

TN = N \cdot (1 - p) \cdot Sp

Negative predictive value describes the proportion of negative results that are true negatives. Like positive predictive value, it changes when prevalence changes, even if sensitivity and specificity remain fixed:

NPV = \frac{(1 - p) \cdot Sp}{(1 - p) \cdot Sp + p \cdot (1 - Se)}

These screening formulas describe hypothetical groups, not an individual diagnosis. They explain why the same PSA threshold can produce a different mix of genuine positives and false alarms when it is used in populations with different underlying prevalence.

For example, consider 1000 hypothetical men in a group where 15% have biopsy-detectable cancer. At a 4.1 ng/mL threshold, published trial figures give 20.5% sensitivity and 93.8% specificity. About $150 \times 0.205 \approx 31$ of the 150 affected men test positive. About $850 \times 0.062 \approx 53$ unaffected men also test positive. Only 31 of 84 positive results are therefore genuine in that illustration, a positive predictive value near 37%.

Worked example: PSA 6.2 at age 62

Suppose a 62-year-old has total PSA of 6.2 ng/mL, prostate volume of 52 cc, percent free PSA of 19%, a normal rectal examination and a prior PSA of 4.9 ng/mL from 18 months earlier. The age-specific upper limit is 4.5 ng/mL, so the total result is above that reference.

PSAD = \frac{6.2}{52} = 0.12 ng/mL per cc

The density is below 0.15, which offers some context for the raised total PSA because the gland is enlarged. Velocity points in the other direction:

PSAV = \frac{6.2 - 4.9}{18 / 12} = \frac{1.3}{1.5} = 0.87 ng/mL per year

That exceeds the classic 0.75 reference, but it is based on two readings and should be interpreted cautiously. A free PSA of 19% falls in the 15–20% band, corresponding to a published biopsy probability of 20% under the study conditions. These mixed findings illustrate why the page reports separate indicators instead of producing one dramatic score.

Limitations of this PSA interpretation

This assessment does not include ancestry, detailed family-history patterns, urinary symptoms, MRI findings, genetic variants, prior biopsy results or medicine adjustments. Finasteride and dutasteride can reduce measured PSA by roughly half, so treatment status matters. Infection, prostatitis, urinary retention, catheterisation and recent prostate procedures may temporarily raise PSA.

Age-specific limits are reference percentiles, not boundaries between healthy and ill. PSA density depends on an accurate volume estimate. Velocity is unstable over short intervals. Free-PSA probabilities apply only to the population studied. None of these measures distinguishes reliably between an indolent tumour and clinically significant high-grade disease.

A clinician may recommend repeating a result after transient causes have settled, reviewing medicines, performing an examination, obtaining MRI or using a validated multivariable tool such as the Prostate Biopsy Collaborative Group or ERSPC calculator. The appropriate next step depends on personal values as well as clinical evidence.

Sources: Oesterling et al., JAMA 1993, for age-specific ranges; Catalona et al., JAMA 1998, for percent free PSA bands; Thompson et al., JAMA 2005, for PSA operating characteristics; and the USPSTF prostate cancer screening recommendation for shared decision-making context.

Questions about PSA calculations

Why is there no single overall risk percentage?

The four indicators came from different studies and were not validated as one combined score. A reliable overall probability requires a fitted multivariable model and the clinical details that model expects.

Is a PSA of 4.0 automatically abnormal?

No. PSA risk is continuous, while reference limits vary by age and clinical context. Cancer can occur below 4.0, and benign conditions can produce values above it.

What can raise PSA besides cancer?

Common explanations include benign enlargement, prostatitis, urinary infection, retention, recent catheterisation or biopsy, ejaculation and vigorous cycling. Timing and medication use should be discussed before interpreting a result.

Enter an age and total PSA to compare the result with published reference indicators.

Assess a result to plot it against the age-specific reference limits.

Cohort Lens: the PSA threshold challenge

Cohort Lens turns sensitivity, specificity and prevalence into a short statistical strategy game. Estimate how 1000 hypothetical people divide into found cases, missed cases, false alarms and correctly cleared results. It teaches screening arithmetic only and never evaluates a player’s health.

FoundMissedFalse alarmCleared

Six changing PSA scenarios await. Lower thresholds find more cases but usually create more false alarms.

Score

Time

75s

Streak

Progress

0 / 6

Best

Phase

Ready

Cleared

700

Total miss

—

Round score

—

Mission: resolve the cohort

You have 75 seconds and six PSA thresholds. Drag the Found, Missed and False alarm bars toward your estimate; Cleared fills automatically. Press Lock estimate to reveal the cohort. Accurate rounds build a streak.

Controls: tap or drag a bar. Keyboard users can choose bars with ← and →, adjust with ↑ and ↓, and submit with Enter.

Click to play, then estimate the four outcomes before time runs out.

Educational takeaway: sensitivity and specificity describe test performance, but prevalence strongly changes the chance that a positive result is genuine.