Any man with a PSA above this age adjusted range should consider to be transferred to a specialist Urologist as there is a risk he could have cancer although there are other causes of a raised blood PSA:-
- infections
- recent prostate biopsies
- having a urinary catheter in (a tube to drain urine)
- prostate or bladder surgery
- prolonged exercise, such as long-distance running or cycling
- recent ejaculation
- some drugs
There are ways to make the PSA more reliable, if needed:-
Free versus bound PSA
The PSA circulates in the blood in two forms - the free form and the form bound to a protein. Some laboratory, as well as an absolute level provide the ratio of free versus bound. The benign conditions which elevate the PSA mentioned above produce more free form whilst cancer produces more bound form. Therefore, the greater the ratio (ie more free form) to more likely that it is benign (not cancer). The lower the ratio the higher the likelihood of cancer. Over 25% for example, is unlikely to be cancer related. In practical terms, if the free versus bound ratio is greater than 30% a biopsy could be avoided (or at least a repeat biopsy if the first one is negative).
The PSA doubling time (PSAdt or PSA velocity)
This is the time it take the the blood PSA to double. The PSAdt is much more accurate and sensitive to progression and prognosis that a single value. For example, a trial published by Fowler in 1995 showed that if the PSADT was < 10 months, there is a high probability of metastatic disease.In many cases it hasn't actually doubled but the potential doubling time can be estimated in a number of ways:-
Informal quick method (useful in clinic):
PSAa x t
PSAb - PSAa
Where PSAa is the first value PSAb is the second value and t is the time between them. (e.g. a PSA increases from 2 to 3 over 3 months the PSAdt = 2/1 x 3 = 6 months).
The formal trial method (Traditional). Usually used them measuring the PSAdt in a group of patients within a clinical trial usually calculated by computer program)
PSADT = log2 x dT/(logB-logA)
Where the PSADT will be set to 0 when A = B. This will ensure continuity between cases where A>B and A<B. The traditional method is based on the assumption of a linear increase in log (PSA) over time, ie an exponential increase.
PSA doubling time (linear spline method) This method fits a linear spline model to a graph of log (PSA) against time. It is fitted using the REG procedure in SAS, using the formula:
Log (PSA) = α + β*t + γ*s(t) + e
Where t is time in months, s(t) is a basic linear spline with a knot at the start of treatment, and e is the error. α, β, and γ are fitted parameters.The pre-treatment doubling time is: DT = log (2) / β and the on-treatment doubling time is: DT = log (2) / (β+ γ). The linear spline method fits two straight lines to the graph of Log (PSA) against time. The first line is fitted to all available pre-treatment data, and the second to all post-treatment data up to the timepoint being analysed. The lines intersect at the start of treatment.
