Complex PSA: A More Specific Marker for Prostate Cancer
Dr. Michael Brawer

It is clear that prostate specific antigen (PSA) has revolutionized the diagnosis of prostate cancer. PSA, the most important tumor marker in all of oncology, provides earlier detection than digital rectal examination and allows us to identify men whose cancer is not palpable (Stage T1C). The widespread utilization of PSA has resulted in a peaking of prostate cancer incidence in most Westernized nations. Moreover, PSA testing may be one factor associated with the decreasing prostate cancer mortality recently observed in the United States.

Despite these important advances, PSA is not perfect. It both lacks sensitivity (some men have cancer with a normal PSA), but more importantly it lacks specificity. That is to say, between 2/3 and 3/4 of men with an abnormal PSA will not be shown to have carcinoma at least on their initial biopsy. In an effort to enhance specificity of PSA, investigators have utilized a variety of so-called PSA derivatives including age specific PSA cutoffs, PSA velocity, and PSA density. Unfortunately none of these have demonstrated to benefit in general clinical practice.

The recognition that PSA circulates in the systemic circulation in a variety of molecular forms has provided an important advance. The majority of PSA is complexed with protease inhibitors, the most important being alpha-1-antichymotrypsin. Recognized initially by Scandinavian investigators and confirmed throughout the world, PSA complexed with alpha-1-antichymotrypsin occurs to a greater proportion of the total PSA in men with cancer. Conversely, the free form of PSA (the form that is present in the ejaculate) occurs to a greater extent in men without malignancy. These observations have led a number of clinicians to adopt the use of the ratio of the free-to-total PSA in clinical testing. The definitive trial investigating the ratio of free-to-total PSA demonstrates at a sensitivity of 95%, the ratio of the free-to-total PSA, resulted in a 20% enhancement of specificity. ¹

While measurement of the ratio of the free-to-total PSA does seem to enhance specificity in multiple trials, several problems exist. Obviously this requires the measurement of 2 analytes, effectively doubling the cost of PSA testing. Moreover, because in part on the absence of international standards for PSA testing, significant bias can occur which is magnified when one number is divided into another. Thus it is impossible to utilize a published cutoff of PSA ratio unless the same manufacturers’ assays are utilized. We have previously reported that even when only the free PSA assay is varied with one total PSA as the denominator, significant bias in the ratio exists. ² Even within one manufacturer’s assays for both the free and total, significant differences in the free/total ratio and test performance on the same patient serum are observed ³

It should be noted that the development of the free-to-total PSA assay was actually an approximation of what was desired: a quantitation of the ACT-complexed PSA. This is the form as noted that occurs to a greater proportion in men with cancer. The hurdles in establishing a specific assay for the ACT-complexed form of PSA however were considerable. Recently the Bayer Corporation has developed a specific assay for the complex form of cPSA ⁴ which has been approved for the monitoring of patients with prostatic carcinoma. This assay has been submitted to the U.S. F.D.A. for a screening indication.

We evaluated this assay in 300 men who underwent ultrasound guided prostate needle biopsy ⁵. We selected patients from our archival bank such that 25% had carcinoma. Serum had been stored at –80º with no freeze thaw. We compared the Bayer Immuno-1 cPSA assay (Bayer Corporation, Tarrytown, NY) with the Hybritech Tandem R assays for both total and free PSA (Hybritech Incorporated, San Diego, CA). The foremost characteristics with respect to sensitivity analysis are shown in Table 1. At a clinically relevant sensitivity of 95%, the complex PSA provided enhanced specificity to both the total PSA as well as the free-to-total PSA ratio.

Stamey and associates ⁶ also carried out an investigation with the Bayer complex PSA assay. In these investigators’ hands, the complex PSA provided no enhanced performance relative to total PSA. Several differences between the two studies exist. The Stanford series involved 160 men. To be considered malignant the subjects had to and men had to have at least 5 mm of cancer on the biopsy. Moreover, the 90 men with negative biopsies had to have 2 sets of negative biopsies. While these constraints certainly offer assurance of accuracy of diagnosis, they do depart from our entry criteria. Perhaps the biggest difference between the Stanford series and our own is that while in our series (as in most clinical settings) the total PSA was greater in men with malignancy compared to those without, this was not found in the Stanford series. We believe this latter observation is evidence of considerable difference between the populations in these 2 investigations.

In an effort to future elucidate the utility of complex PSA, we have expanded our Seattle experience by including patients evaluated at the Johns Hopkins University as well as other sites. ⁷ In this series, 385 men without malignancy and 272 with cancer were studied. The significant findings may be found in Table 2. These findings confirm now in a multi-institutional setting enhanced specificity with the complex PSA rivaling that of the free-to-total PSA ratio.

Enhanced specificity of complex PSA compared with the free-to-total observed in our initial study ⁵ may have been attributed to the instability even at –80º of the free analyte of PSA with long term storage. Degradation of the free component would result in decreased specificity with the free-to-total PSA ratio.

We believe that the complex assay for PSA provides substantially equivalent enhancement in specificity compared with the free-to-total PSA ratio. The ability to achieve this with a single analyte determination not only is an economic advantage but obviates the assay variability and quotient bias noted above. As this assay has equivalent utility in monitoring patients with established malignancy as well as staging and prognostic information, clinicians can substitute this assay for all PSA testing

Table 1: Specificity of the cut-off values of the different PSA assays at selected sensitivities.

% Sensitivity	Total PSA		Complexed PSA		Free/Total PSA
	Cut-off (ng/mL)	% Specificity	Cut-off (ng/mL)	% Specificity	Cut-off (%)	% Specificity
80	4.11	35.6	3.98	51.6	19	46.2
85	3.86	31.1	3.34	38.7	22	32.4
90	3.4	25.3	2.94	33.8	24	26.2
95	3.06	21.8	2.52	26.7	28	15.6

Table 2: Specificity of the cut-off values of the different PSA assays at selected sensitivities.

% Sensitivity	Total PSA		Complexed PSA		Free/Total PSA
	Cut-off (ng/mL)	% Specificity	Cut-off (ng/mL)	% Specificity	Cut-off (%)	% Specificity
80	4.64	41	4.09	46	17.1	52*1
85	4.33	33	3.79	41*	18.9	42*
90	3.99	28	3.40	32	20.9	31
95	3.06	18	2.75	24*	23.9	23*

^* Significantly different from total PSA.
¹ Significantly different from complexed PSA.

References

1. 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 279(19):1542-1547, 1998.

2. Nixon RG, Gold MH, Blase AB, Meyer GE, Brawer MK: Comparison of three investigative assays for the free form of prostate-specific antigen. J Urol 160:420-425, 1998.

3. Roth HJ, Christensen-Stewart S, Brawer MK: A comparison of three free and total PSA assays. PCPD 1(6):326-331, 1998.

4. Allard WJ, Zhou Z, Yeung KK: Novel immunoassay for the measurement of complexed prostate-specific antigen in serum. Clin Chem 44(6):1216-1223, 1998.

5. Brawer MK, Meyer GE, Letran JL, et al.: Measurement of complexed PSA improves specificity for early detection of prostate cancer. Urology 52(3):372-378, 1998.

6. Stamey TA, Yemoto CE: Examination of the 3 molecular forms of serum prostate specific antigen for distinguishing negative from positive biopsy: Relationship to transition zone volume. J Urol 163(1):119-126, 2000.

7. Brawer MK, Cheli CD, Neaman IE, et al.: Complexed prostate specific antigen provides significant enhancement of specificity compared with total prostate specific antigen for detecting prostate cancer. J Urol 2000 (in press).