Int Urol Nephrol (2008) 40:85–89

DOI 10.1007/s11255-007-9236-4

ORIGINAL ARTICLE

Diagnostic efficacy of free to total ratio of prostate-specific

antigen and prostate-specific antigen velocity, singly
and in combination, in detecting prostate cancer in patients
with total serum prostate-specific antigen
between 4 and 10 ng/ml
Shingo Yamamoto Æ Takuo Maruyama Æ Nobuyuki Kondoh Æ Michio Nojima Æ
Hidekazu Takiuchi Æ Seiichi Hirota Æ Hiroki Shima

Received: 30 March 2007 / Accepted: 10 May 2007 / Published online: 6 July 2007

Springer Science+Business Media B.V. 2007

Abstract To assess the diagnostic efficacy of
prostate-specific antigen (PSA)-related parameters,
using a total of 226 patients with gray zone PSA who
underwent prostate biopsy, various cutoff points of
free to total ratio of PSA (f/t PSA) and PSA velocity
(PSAV) were evaluated. Higher cutoff points of f/t
PSA resulted in high sensitivity and negative predic-
tive value (NPV): at f/t PSA <15%, sensitivity was
82.0% (41/50) and NPV 84.7% (50/59), and at f/t
PSA <20%, 96.0% (48/50) and 92.3% (24/26).
Lowering cutoff points also resulted in higher
sensitivity and NPV: at PSAV 0.75 ng/ml per year,
sensitivity was 71.4% (15/21) and NPV 82.4%
(28/34), and at PSAV 0.40 ng/ml per year, 95.2%
(20/21) and 95.2% (20/21). Further, among the
patients with both of these parameters available, both
sensitivity and NPV achieved 100% (10/10 and 7/7)
when the indication for biopsy was determined as f/t
PSA <15% or PSAV 0.40 ng/ml per year. Our
results showed that unnecessary prostate biopsies
could be more effectively avoided among patients
S. Yamamoto (&)
T. Maruyama
N. Kondoh

M. Nojima
H. Takiuchi
H. Shima
Department of Urology, Hyogo College of Medicine,
1-1 Mukogawacho, Nishinomiya, Hyogo 663-8501, Japan
e-mail: shingoy@hyo-med.ac.jp
S. Hirota
Department of Surgical Pathology, Hyogo College
of Medicine, Nishinomiya, Hyogo, Japan
with ‘‘gray zone PSA’’ by combination of f/t PSA
and PSAV than single usage of these indexes.
Keywords Free to total ratio of prostate-specific
antigen (f/t PSA)
Prostate-specific antigen (PSA)

Prostate-specific antigen velocity (PSAV)
Introduction
Serum prostate-specific antigen (PSA) has been
widely regarded as the most clinically useful marker
for the detection of prostate cancer (PCa) [1].
However, the elevation of PSA levels is not unique
to PCa, and expression of PSA at 4–10 ng/ml (‘‘gray
zone PSA’’) can also be observed in cases of benign
prostate hyperplasia (BPH), prostatitis, and prostatic
intraepithelial neoplasia [2–4]. This lack of specific-
ity often leads to the performance of unnecessary
diagnostic procedures, including invasive and costly
prostate biopsy, on patients with ‘‘gray zone PSA’’.
Among PSA-related parameters, PSA density
(PSAD), PSA velocity (PSAV), and free to total
ratio of PSA (f/t PSA) have been demonstrated to be
useful clinically in terms of sensitivity and specificity
of PCa detection [5–10].
To detect early PCa accurately and minimize
unnecessary prostate biopsy, it is desirable to estab-
lish a screening method with high sensitivity for PCa
and high negative predictive value (NPV) for non-

123
86
PCa [11]. In this study, we analyzed f/t PSA and
PSAV, singly and in combination, for sensitivity and
NPV in differential diagnosis of PCa and benign
diseases in patients with ‘‘gray zone PSA’’
Patients and methods
A total of 226 with PSA levels between 4 ng/ml and
10 ng/ml underwent systemic transperineal 8-core
prostate biopsy under transrectal ultrasound (TRUS)
guidance in our clinic between January 1999 and
December 2004. Of the 226 patients, 146 were
measured for f/t PSA. Serum total PSA and free PSA
were measured by equimolar kits according to
Tandem-R1. Eighty-one patients were measured for
PSA multiple times for 6 months to 2 years before
biopsy, allowing to determine PSAV, the rate of
change of serum PSA levels per year. Both f/t PSA
and PSAV were available with 46 patients.
Various cutoff points of f/t PSA (10–20%) and
PSAV (1.00–0.40 ng/ml per year) were assessed in
terms of sensitivity, specificity, positive predictive
value (PPV), and NPV for PCa detection.
Statistical comparisons between groups were
assessed by unpaired Student’s t test.
Results
Biopsy detected PCa in 78 (34.5%) of 226 patients
with PSA levels between 4 ng/ml and 10 ng/ml. Total
PSA and f/t PSA, but not PSAV, were significantly
different between patients with PCa and benign
diseases (Table 1).
Sensitivity and NPV were determined at several
cutoff points of f/t PSA. The results showed that both
sensitivity and NPV increased by increasing the f/t
PSA cutoff point (Fig. 1). Thus, at f/t PSA <10%,
sensitivity was 38.0% (19/50) and NPV 81.0%
Table 1 Characteristics of the 226 patients who underwent prostate biopsy with PSA levels between 4 ng/ml and 10 ng/ml
Outcomes of biopsy PCa (78)
Age 70.47 ± 7.65
Total PSA 7.08 ± 1.76
f/t PSA 11.83 ± 6.36 (50)
PSAV 1.67 ± 1.39 (21)
Numbers of patients are shown in parenthesis
123
Int Urol Nephrol (2008) 40:85–89
(76/107), at f/t PSA <15%, 82.0% (41/50) and
84.7% (50/59), and at f/t PSA <20%, 96.0% (48/50)
and 92.3% (24/26) (Table 2). On the other hand, PPV
was not significantly improved by lowering the cutoff
point of f/t PSA: at f/t PSA <10%, PPV was 48.7%
(19/39), at f/t PSA <15%, 47.1% (41/87), and at f/t
PSA <20%, 40.0% (48/120).
In the case of PSAV, lowering cutoff points also
resulted in higher sensitivity and NPV (Fig. 2): at
PSAV 1.00 ng/ml per year, sensitivity was 57.1%
(12/21) and NPV 77.5% (31/40), at PSAV 0.75 ng/ml
per year, 71.4% (15/21) and 82.4% (28/34), and at
PSAV 0.40 ng/ml per year, 95.2% (20/21) and
95.2% (20/21) (Table 3). PPV was approximately
30% at all PSAV cutoff points examined.
When diagnostic efficacy of f/t PSA and PSAV
were analyzed again using 46 patients with whom
both of these parameters were available, f/t PSA was
significantly lower in patients with PCa than in those
without PCa (11.10 ± 4.41 vs 17.44 ± 9.18%,
P = 0.03812), while PSAV was not significantly
different between those with and without PCa
(1.32 ± 0.08 vs 0.95 ± 2.17 ng/ml per year,
P = 0.60173). In this group, sensitivity and NPV
were 80% (8/10) and 91.3% (21/23) when cutoff
point was singly set at f/t PSA <15%, while set singly
at PSAV 0.4 ng/ml per year, 90% (9/10) and 93.8%
(15/16). Of note, sensitivity and NPV yielded 100%
(10/10 and 7/7, respectively) when the indication for
biopsy was determined as f/t PSA <15% or PSAV
0.40 ng/ml per year (Fig. 3).
Discussion
PSA is considered to be the best serum marker for
PCa, but its specificity is not sufficiently high at
4–10 ng/ml, the level observed with about 80% of
PCa patients. In addition, it is estimated that 30–50%
Benign (148) P values
71.07 ± 7.54 0.57204
6.37 ± 1.58 0.00228
15.57 ± 7.48 (96) 0.00306
0.98 ± 1.87 (60) 0.12623
Int Urol Nephrol (2008) 40:85–89 87

Fig. 1 Sensitivity, (% )
specificity, and positive and 100
Sensitivity
negative predictive values Specificity
at various cutoff levels of f/t PPV
PSA in 146 patients with 80
NPV
PSA levels between 4 ng/ml
and 10 ng/ml
60

40

20

0
10 12 14 16 18 20
f/t PSA (%)

Table 2 Sensitvity,
f/t PSA cutoff levels Sensitivity Specificity PPV NPV
specificity, and positive and
negative predictive values <10.0 38.0 (19/50) 79.2 (76/96) 48.7 (19/39) 81.0 (76/107)
(%) at various cutoff levels
of f/t PSA in 146 patients <11.0 48.0 (24/50) 75.0 (72/96) 50.0 (24/48) 73.5 (72/98)
with PSA levels between <12.0 56.0 (28/50) 66.7 (64/96) 46.7 (28/60) 74.4 (64/86)
4 ng/ml and 10 ng/ml <13.0 62.0 (31/50) 62.5 (60/96) 39.7 (31/67) 75.9 (60/79)
<14.0 74.0 (37/50) 57.3 (55/96) 47.4 (37/78) 80.9 (55/68)
<15.0 82.0 (41/50) 52.1 (50/96) 47.1 (41/87) 84.7 (50/59)
<16.0 84.0 (42/50) 47.9 (46/96) 45.7 (42/92) 85.2 (46/54)
<17.0 86.0 (43/50) 43.8 (42/96) 46.2 (43/93) 79.2 (42/53)
<18.0 92.0 (46/50) 32.3 (31/96) 41.4 (46/111) 88.6 (31/35)
<19.0 94.0 (47/50) 28.1 (27/96) 40.5 (47/116) 90.0 (27/30)
Numbers of patients are <20.0 96.0 (48/50) 25.0 (24/96) 40.0 (48/120) 92.3 (24/26)
shown in parenthesis

Fig. 2 Sensitivity, (% )
specificity, and positive and 100
Sensitivity
negative predictive values Specificity
at various cutoff levels of PPV
80
PSAV in 81 patients with NPV

PSA levels between 4 ng/ml

and 10 ng/ml
60

40

20

0
1.0 0.9 0.8 0.7 0.6 0.5 0.4
PSAV (ng/ml per year)

of patients with BPH express PSA exceeding 4 ng/ml discomfort, several risks and at high cost. To improve
[12], with a significant number of these BPH patients sensitivity and specificity of PCa detection, various
subjected to unnecessary prostate biopsy with PSA-related parameters including f/t PSA [5–7],

123
88 Int Urol Nephrol (2008) 40:85–89

Table 3 Sensitivity,
PSAV cutoff levels Sensitivity Specificity PPV NPV
specificity, and positive and
negative predective values 1.00 57.1 (12/21) 51.7 (31/60) 29.3 (12/41) 77.5 (31/40)
(%) at various cutoff levels
of PSAV in 81 patients with 0.95 57.1 (12/21) 51.7 (31/60) 29.3 (12/41) 77.5 (31/40)
PSA levels between 4 ng/ 0.90 57.1 (12/21) 50.0 (30/60) 28.6 (12/42) 76.9 (30/39)
ml and 10 ng/ml 0.85 61.9 (13/21) 50.0 (30/60) 30.2 (13/43) 78.9 (30/38)
0.80 71.4 (15/21) 50.0 (30/60) 33.3 (15/45) 83.3 (30/36)
0.75 71.4 (15/21) 46.7 (28/60) 31.9 (15/47) 82.4 (28/34)
0.70 71.4 (15/21) 45.0 (27/60) 31.3 (15/48) 81.8 (27/33)
0.65 81.0 (17/21) 43.3 (26/60) 33.3 (17/51) 86.7 (26/30)
0.60 81.0 (17/21) 40.0 (24/60) 32.1 (17/53) 85.7 (24/28)
0.65 81.0 (17/21) 40.0 (24/60) 32.1 (17/53) 85.7 (24/28)
0.50 85.7 (18/21) 36.7 (22/60) 32.1 (18/56) 88.0 (22/25)
0.45 85.7 (18/21) 36.7 (22/60) 32.1 (18/56) 88.0 (22/25)
Numbers of patients are 0.40 95.2 (20/21) 33.3 (20/60) 33.3 (20/60) 95.2 (20/21)
shown in parenthesis

50 comparison of PSAV and other PSA-related param-

× PCa Benign eters. In the present study, therefore, we focused on
40 sensitivity and NPV of f/t PSA and PSAV to
maximize early PCa detection and minimize unnec-
30 essary biopsy, respectively, in patients with ‘‘gray
f/t PSA (%)

20
15.0
10
0 high cutoff points of f/t PSA resulted in high
- 2 .0 0 .0 2.0 4 .0
PSAV (ng/ml per year)
0.4
Fig. 3 Distribution of f/t PSA and PSAV of 46 patients with
PCa (·) and without PCa (d). Sensitivity and NPV yielded
100% (10/10 and 7/7, respectively) when the indication for
biopsy was determined as f/t PSA <15% or PSAV 0.40 ng/ml
per year
PSAD [8], transitional zone PSA density (PSAD-TZ)
[13], and PSAV [9], have been assessed.
To obtain PSAD and PSAD-TZ, patients must
undergo TRUS and therefore are subject to the
technical limitations of the TRUS procedure. On the
other hand, PSAV and f/t PSA can be obtained
readily and objectively, although multiple yearly
measurements of PSA are required for PSAV, and the
assays of the total and free PSA using methods of the
same manufacture are required for f/t PSA. Several
papers have compared prognostic efficacy of these
PSA-related parameters such as f/t PSA, PSAD, and
PSATZ [14–16], but few have reported on the
zone PSA’’.
When f/t PSA was analyzed among the 146
patients with whom this PSA-related parameter was
available, those with PCa showed significantly low
f/t PSA levels than those without PCa. The use of
6.0 8 .0
sensitivity and NPV. When the cutoff point of f/t
PSA was set at <15%, sensitivity and NPV were
82.0% and 84.7%, respectively, while these values
increased to 96.0% and 92.3%, respectively, at the
cutoff point <20%, indicating that f/t PSA would
be useful to improve sensitivity and NPV to be
able to predict the outcome of prostate biopsy.
Among the 81 patients with whom PSAV was
available, those with PCa failed to show signifi-
cantly higher PSAV when compared with those
without PCa. However, sensitivity and NPV were
improved by setting the cutoff point of PSAV low;
PSAV 1.00 ng/ml per year, sensitivity was 57.1%
and NPV 77.5%, at PSAV 0.75 ng/ml per year,
71.4% and 82.4%, and at PSAV 0.40 ng/ml per
year, 95.2% and 95.2%.
Carter et al. reported first that, when the cutoff
point of PSAV was set at 0.75 ng/ml per year in 18
men with BPH and 20 men with PCa, patients with
PSA level <10 ng/ml, sensitivity and specificity were
72% and 95%, respectively [17]. Mettlin et al.

123
Int Urol Nephrol (2008) 40:85–89
reported that sensitivity was as low as 54.8% at cutoff
point of PSAV 0.75 ng/ml in 1,473 men without
PCa and 84 men with PCa, concluding that PSAV may
not be useful for early detection of PCa [18]. On the
other hand, Smith and Catalona showed that, for
subjects with PSA levels of 4.0 ng/ml, PSAV was
demonstrated to be an excellent index when the cutoff
point was set lower, at 0.4 ng/ml per year [19].
In this regard, our results showed that high
sensitivity and NPV exceeding 90% were obtained
at PSAV 0.4 ng/ml per year, indicating that
PSAV could be a comparable index to other PSA-
related parameters. Further, among the patients with
whom both of these parameters were available,
both sensitivity and NPV achieved 100% when the
indication for biopsy was determined as f/t PSA
<15% or PSAV 0.40 ng/ml per year whereas they
were 80% and 91.3% by single cutoff of f/t PSA
<15%, suggesting that combination of f/t PSA and
PSAV may produce better predictive value than
single usage of these indexes.
In summary, our results indicated that unneces-
sary prostate biopsies might be reduced more
effectively by setting cutoff levels using combina-
tion of f/t PSA and PSAV. However, it should be
mentioned that this study was conducted retrospec-
tively in a single institute from Japan and conse-
quently the patients analyzed may not reflect the
general population. A prospective multi-center
study including a larger number of patients is
desirable to confirm the utility of the approach
presented in this study.
References
1. Partin AW, Oesteling JE (1994) The clinical usefulness
of prostate specific antigen: update. J Urol 152:1358–
1368
2. Stamey TA, Yang N, Hay AR et al (1987) Prostate-specific
antigen as a serum marker for adenoxarcinoma of the
prostate. N Engl J Med 317:909–916
3. Catalona WJ, Smith DS, Ratliff TL et al (1991)
Measurement of prostate-specific antigen in serum as a
screening test for prostate cancer. N Engl J Med
324:1156–1161
4. Catalona WJ, Richie JP, Ahmann FR et al (1994) Com-
parison of digital rectal examination and serum prostate
specific antigen in the early detection of prostate cancer:
results of a multicenter clinical trial of 6630 men. J Urol
151:1283–1290
89
5. Partin AW, Oesterling JE (1996) The clinical usefulness of
percent free prostate-specific antigen. Urology 48(6A
Suppl):1–87
6. Catalona WJ, Partin AW, Slawin KM et al (1998) Use of
the percentage of free prostate-specific antigen to enhance
differentiation of prostate cancer from benign prostate
disease: a prospective multicenter clinical trial. JAMA
279:1542–1547
7. Catalona WJ, Southwick PC, Slawin KM et al (2000)
Comparison of percent free PSA, PSA density, and age-
specific PSA cutoffs for prostate cancer detection and
staging. Urology 56:255–260
8. Beduschi MC, Oesteling JE (1997) Prostate-specific anti-
gen density. Urol Clin North Am 24:323–332
9. Carter HB, Pearson JD (1997) Prostate-specific antigen
velocity and repeated measures of prostate-specific anti-
gen. Urol Clin North Am 24:333–338
10. Djavan B, Remzi M, Zlotta AR et al (1999) Combination
and multivariate analysis of PSA-based parameters for
prostate cancer prediction. Tech Urol 5:71–76
11. Djavan B, Remzi M, Zlotta A et al (2002) Novel artificial
neural network for early detection of prostate cancer. J Clin
Oncol 20:921–929
12. Beduschi MC, Oesteling JE (1998) Percent free prostate-
specific antigen: the next frontier in prostate-specific
antigen testing. Urology 51(5A Suppl):98–109
13. Djavan B, Zlotta AR, Byttebier G et al (1998) Prostate
specific antigen density of the transition zone for early
detection of prostate cancer. J Urol 160:411–419
14. Bangma CH, Rietbergen JB, Kranse R et al (1997) The
free-to-total prostate specific antigen ratio improves the
specificity of prostate specific antigen in screening for
prostate cancer in the general population. J Urol
157:2191–2196
15. Morote J, Raventos CX, Lorente JA et al (1997) Com-
parison of percent free prostate specific antigen and
prostate specific antigen density as methods to enhance
prostate specific antigen specificity in early prostate
cancer detection in men with normal rectal examination
and prostate specific antigen between 4.1 and 10 ng/ml.
J Urol 158:502–504
16. Zlotta AR, Djavan B, Petein M et al (1998) Prostate spe-
cific antigen density of the transition zone for predicting
pathological stage of localized prostate cancer in patients
with serum prostate specific antigen less than 10 ng/ml.
J Urol 160:2089–2095
17. Carter HB, Pearson JD, Metter EJ et al (1992) Longitudinal
evaluation of prostate-specific antigen levels in men with
and without prostate disease. JAMA 267:2215–2220
18. Mettlin C, Littrup PJ, Kane RA et al (1994) Relative
sensitivity and specificity of serum prostate specific anti-
gen (PSA) level compared with age-referenced PSA, PSA
density, and PSA change. Data from the American Cancer
Society National Prostate Cancer Detection Project Cancer
74:1615–1620
19. Smith DS, Catalona WJ (1994) Rate of change in serum
prostate specific antigen levels as a method for prostate
cancer detection. J Urol 152:1163–1167
123