Making sense of Free-to-total PSA ratios

Article under construction

The above graph (Catalona WJ, Beiser JA, and Smith DS. Serum Free PSA and PSA Density Measurements for Predicting Cancer in Men with Prior Negative Prostatic Biopsies. J Urol 158,2162-2167, December 1997) demonstrates %free PSA and total PSA in men with and without cancer on follow-up biopsy. Percent free PSA cutoff of 28% eliminates 13% of negative biopsies. Note that benign repeat biopsies can be obtained with a %PSA as low as 7% and malignant biopsies can be obtained with men with %PSA as high as 40%.

********************************************

Comments of Dr. Boccon-Gibbod:

PSA age reference ranges have not been shown in screening series to be vastly superior to the usual PSA cut-off levels.

-PSA density (the ratio of PSA to prostate volume as determined by TRUS) is too highly operator-dependent, as evidenced by the variable cut-off values even in the literature, to be transferable from one institution to another. Furthermore, although the ratio of PSA to the volume of the transition zone popularised by Professor Claude Schulman seems extremely promising, this approach has yet to be validated by other groups.

-PSA velocity needs to be calculated precisely, requiring a significant number of values over time. This approach is therefore of limited value in a patient who presents with an abnormal PSA and requires urgent attention.

-The ratio of free to total PSA (f/tPSA) is the PSA derivative generating most recent interest. This approach is based on the fact that BPH-produced PSA is predominantly in the free form, whereas tumour-derived PSA is mainly complexed. Although this approach is probably intellectually the most satisfactory, a number of caveats have to be expressed:

-In order to be meaningful, free and total PSA should be always measured using the same assay.

-Published cut-off levels vary between institutions and assays, and are probably not transferable from the screening studies to the case finding setting.

-As usual, the trade-off is between sensitivity and specificity. In other words, what is the lowest acceptable level of undiagnosed prostate cancers that will unavoidably occur with the use of this technique, and how can we be sure that these undiagnosed prostate cancers will still be organ-confined and amenable to local radical treatment if and when they are diagnosed in the future?

It appears likely that PSA derivatives and, more specifically, analysis of the various forms of PSA detectable in the blood stream will certainly play a role in the more precise identification of the indications for prostate biopsies in the future. However, at present, it must be acknowledged that systematic random TRUS-guided prostate biopsy is still the optimum approach for the clinical management of patients with abnormal PSA in the intermediate range (4 to 10 ng/ml) and a life expectancy of more than 10-15 years, unless the f/tPSA ratio far exceeds normal cut-off values (25 to 30%) indicating with almost 100% certainty that the prostate is benign.

********************************************

While finasteride decreases serum PSA levels to about 50% of their normal values, it does not appear to have any affect on the free-to-total PSA ratio (Influence of Finasteride On Free and Total Serum Prostate Specific Antigen Levels in Men with Benign Prostatic Hyperplasia. J Pannek, L Marks, J Pearson, et al. J Urol 159(2) Feb 1998)