close

Вход

Забыли?

вход по аккаунту

?

532

код для вставкиСкачать
90
The Ratio of Free to Total Serum Prostate Specific
Antigen and Its Use in Differential Diagnosis of
Prostate Carcinoma in Japan
Shin Egawa, M.D.1
Shigehiro Soh, M.D.1
Makoto Ohori, M.D.1
Toyoaki Uchida, M.D.1
Kazuo Gohji, M.D.2
Akio Fujii, M.D.2
Sadahito Kuwao, M.D.3
Ken Koshiba, M.D.1
BACKGROUND. To improve the clinical usefulness of prostate specific antigen (PSA),
unique methods have been proposed. The percentage of free PSA in serum facilitates the distinction between benign histologic conditions and prostate carcinoma
while retaining high sensitivity.
METHODS. Using monoclonal antibodies, an AIA total PSA assay system was established that recognized PSA equally in free or complex form. The clinical usefulness
of the ratio of two different molecular forms of PSA was investigated using 268
archival serum samples.
RESULTS. Men with prostate carcinoma had significantly lower ratios of free to
1
Department of Urology, Kitasato University
School of Medicine, Sagamihara, Japan.
2
Department of Urology, Hyogo Medical Center
for Adults, Akashi, Japan.
3
Department of Pathology, Kitasato University
School of Medicine, Sagamihara, Japan.
total PSA than those with benign prostatic hyperplasia (BPH) (P Å 0.0001). At total
PSA levels between 2.1 and 10 ng/mL, medians of total PSA were not significantly
different between men with prostate carcinoma and men with BPH. Differences
in median percentages of free PSA for these two groups were statistically significant
(P Å 0.0001). The ratio of free to total PSA was useful for identifying prostate
carcinoma in palpably benign glands with total PSA of 2.1–10 ng/mL (P Å 0.0001),
whereas total PSA was not useful for such identification. When calculated for low
total PSA levels between 2.1 and 4 ng/mL, sensitivity and specificity were 91.7%
and 72.2%, respectively, with a cutoff value of 17%. This ratio of free to total PSA
was as useful as PSA density in receiver-operating curve analysis.
CONCLUSIONS. The use of the ratio of free to total PSA renders total PSA of greater
use for distinguishing prostate carcinoma and is applicable to patients with low
total PSA. Elderly men with a clinical diagnosis of BPH who are scheduled for
surgery may benefit from the determination of this ratio. Cancer 1997; 79:90–8.
q 1997 American Cancer Society.
KEYWORDS: Prostate carcinoma, free prostate specific antigen, total prostate specific antigen, free to total prostate specific antigen ratio, a1-antichymotrypsin,
nonpalpable cancer.
Supported in part by a grant from the Ministry
of Health and Welfare of Japan (7-42).
The authors thank Mr. Kazuhiro Shimazu and
Dr. Naoko Maruo of TOSOH Corporation for
providing excellent technical support.
Address for reprints: Shin Egawa, M.D., Department of Urology, Kitasato University School of
Medicine,1-15-1 Kitasato, Sagamihara, Kanagawa 228, Japan.
Received June 21, 1996; revision received September 20, 1996; accepted September 20, 1996.
P
rostate specific antigen (PSA) is a 33 kDa single chain glycoprotein
produced in the secretory epithelium of the prostate gland. Its
primary structure shows considerable homology to kallikrein,1 though
this proteolytic enzyme is unlikely to be present in the serum in an
active form. The dominating portion of PSA in the serum forms a
complex with the protease inhibitor a1-antichymotrypsin(ACT). Only
a minor fraction is present as uncombined (free) PSA.2 Though extremely useful for diagnosis and for determining the management of
prostate carcinoma, PSA alone is not sufficiently sensitive or specific.
The need for increased clinical specificity and sensitivity has become
more urgent with the expanded use of PSA. The proportion of free
PSA has reportedly been significantly higher in benign prostatic hy-
q 1997 American Cancer Society
/ 7b48$$0809
12-10-96 11:16:30
cana
W: Cancer
Free to Total PSA Ratio in Prostate Carcinoma/Egawa et al.
perplasia (BPH) than in prostate carcinoma.3 – 6 The
percentage of free PSA in serum facilitates the distinguishing of benign histologic conditions from cancer
while retaining high sensitivity, especially in the diagnostic gray zone of total PSA that exists at 4.1 – 10 ng/
mL.
The clinical usefulness of this ratio of two different
molecular forms of PSA in an Asian male population
has yet to be adequately demonstrated.6 The quantity
of PSA in serum is smaller in Japanese men than in
Caucasian men of the same age.7 – 10 A given PSA value
may thus have different clinical meaning for patients
of different races. Prostate carcinoma occurs in 9.7%
of Japanese males with serum PSA below 4 ng/mL in
a screened population.7 Refinement in testing is thus
essential, especially at lower PSA levels, to have fewer
negative biopsies without missing tumors of significant biologic potential. The diagnosis of nonpalpable
prostate carcinoma is also complicated.11 There are
now many noninvasive alternatives to transurethral
resection of the prostate for BPH, and these developments have resulted in a lack of prostatic tissue for
histologic examination. Many patients with BPH have
mildly elevated total PSA. Better testing is thus needed
in patients with clinical diagnosis of BPH prior to surgery. The clinical significance of the ratio of free to
total PSA in the differential diagnosis of prostate carcinoma in Japanese males was assessed in the current
study.
MATERIALS AND METHODS
Between January 1992 and August 1995, pretreatment
serum samples of patients who underwent therapy for
prostate carcinoma or surgery for clinically diagnosed
BPH were systematically frozen and kept at -80 7C at
2 different facilities (Kitasato University Hospital and
Hyogo Medical Center for Adults). A total of 268 samples were used, 155 from patients with BPH and 113
from patients with prostate carcinoma. The mean period of storage was 14.7 months (range, 1 – 44). Histology was confirmed by six-sextant biopsy and/or transurethral resection in all cases. An arbitrarily chosen
lower PSA cutoff value of 2 ng/mL in Abbott IMx assay
was used at both hospitals as indication of biopsy instead of that recommended by the manufacturer(0 –
4 ng/mL).7 Some patients, even some with small prostate volumes, underwent surgery owing to urinary obstructive symptoms. Twenty-one frozen serum samples from patients with prostate carcinoma were obtained from Hyogo Medical Center for Adult Disease.
This was done prior to any prostatic manipulation,
such as digital rectal examination (DRE) or transrectal
ultrasonography. All patients with prostate carcinoma
were thoroughly examined for disease classification;
/ 7b48$$0809
12-10-96 11:16:30
cana
91
the examination included DRE, prostatic acid phosphatase, intravenous urography, magnetic resonance
imaging, and bone scanning supplemented by radiographs of abnormal areas. Rectal examination was
conducted by three of the authors (S.E., K.G., and A.F.),
and disease was clinically classified according to the
1992 revision of the TNM classification.12 Briefly, tumors classified as T1a were incidental histologic findings in 5% or less of resected tissue, and tumors classified as T1b were incidental histologic findings in more
than 5% of resected tissue. Patients with positive biopsy cores and palpably benign glands were classified
as T1c when sonographic findings were unremarkable
for malignancy. Tumors classified as T2a and b were
palpable and confined within the prostate, involving
a portion of or the entire lobe; T2c tumors were palpable, were confined within the prostate, and involved
both lobes; T3a and b tumors were palpable and extended through the prostatic capsule with unilateral
or bilateral extension, respectively; T3c tumors were
palpable, extended through the prostatic capsule, and
involved the seminal vesicles; and T4 tumors either
were fixed or invaded adjacent structures other than
the seminal vesicles. Distant metastasis was classified
as M1 disease.
Radical prostatectomy specimens were examined
by the whole organ step-section technique, as previously described.7 Specimens were fixed in 10% formalin and sectioned at 5 mm intervals in a plane perpendicular to the long axis of the gland, from the prostatic apex to the tips of the seminal vesicles. Each
specimen was stained with hematoxylin and eosin.
The depth of capsular penetration was determined.
Searches for and assessments of seminal vesicle invasion and lymph node involvement were made. All histology slides were examined and reviewed by a single
pathologist (S.K.), and tumor grade was assigned
based on the Gleason grading system. Total and free
serum PSA were determined by the fully automated
enzyme immunoassay analyzer system, AIA-1200(TOSOH Corporation, Tokyo, Japan). The performance
characteristics of this system have been reported previously.13 The results of this assay for total PSA were
closely correlated and may be considered identical to
those of Hybritech Tandem-R and thus the same as
the Abbott IMx assay.7,9 The AIA total PSA assay utilizes
a pair of monoclonal antibodies and thus recognizes
PSA equally as free or as part of a complex formed
with ACT. The amounts of free PSA in serum samples
were measured by investigational immunoenzymetric
assay, using a pair of monoclonal antibodies obtained
from CanAg Diagnostics, Gothenburg, Sweden (PSA
19 and PSA 66).14 Free PSA assay uses the anti-PSA
monoclonal antibody as capture and alkaline phos-
W: Cancer
92
CANCER January 1, 1997 / Volume 79 / Number 1
TABLE 1
Patient Characteristics (Age, Prostate Volume, and Different Molecular Forms of Prostate Specific Antigen)
Characteristics
BPH (median { SE (range)
Prostate carcinoma
median { SE (range)
P value
No. of patients
Age (yrs)
Prostate volume (cm3)
Total PSA (ng/mL)
Free PSA (ng/mL)
Free to total PSA ratio (%)
155
68 { 0.6 (50–89)
31.2 { 1.7 (9.1–150)
3.8 { 0.5 (0.36–35.6)
0.8 { 0.1 (0.06–11)
22.6 { 1 (4.8–82.8)
113
67 { 0.7 (48–85)
25.6 { 2.2 (9.5–180)
12.2 { 41 (0.43–4202.28)
1.6 { 7.1 (0.15–678)
10.9 { 1.3 (2.8–99.2)
0.2127
0.0001
0.0001
0.0001
0.0001
BPH: benign prostatic hyperplasia; SE: standard error; PSA: prostate specific antigen.
phatase-labeled antifree PSA specific antibody as
tracer. The cross-reactivity of these antibodies with
PSA-ACT complex was specified as less than 1%.14 It
was found to be 0.96% at our own laboratory. These
antibodies in excess are used in sandwich assay. During a 40-minute incubation, the antigen forms a sandwich with both antibodies. The amount of enzyme
activity bound to the bead is directly proportional to
the analyte concentration. Alkaline phosphatase converts 4-methyl-umbelliferyl phosphate to 4-methylumbelliferone, which is fluorescent. The fluorescent
signal is measured every second, starting at 10 seconds. The rate of increase in fluorescence is expressed
in terms of enzymatic activity after the addition of
substrate over a period as long as 100 seconds. Analytic
sensitivity was 0.1 ng/mL for total PSA and 0.024 ng/
mL for free PSA. These values represented 2 standard
deviations over the mean optical density of the assay
zero calibrator. The inter- and intra-assay coefficients
of variation were less than 10%. The assay was linear
over the measuring range (0 – 100 ng/mL for total PSA,
0 – 20 ng/mL for free PSA, respectively). Nineteen serum specimens (7 from patients with prostate carcinoma and 12 from benign lesions) subjected to 6
freeze-thaw cycles showed a mean ({ standard error)
recovery of 104.1% ({ 6.7%) and 109.1%({ 4.5%) of
baseline values for total and free PSA, respectively.
Both total and free PSA immunoreactivity remained
stable for up to 8 days in serum (n Å 19 patients)
stored at 4C.
The PSA density (PSAD) was calculated by dividing
serum PSA by prostate volume, according to transrectal ultrasonography. Prostate volume was determined
with the formula for a prolate ellipsoid, Prostate
Length 1 Width 1 Height 1 (p/6).
Mann – Whitney U or Kruskal – Wallis tests were
conducted for comparison of patients’ ages, serum
PSA levels, and prostatic volumes, with P values õ0.05
considered significant. The significance of total PSA,
free to total PSA ratio, and PSAD for predicting pros-
/ 7b48$$0809
12-10-96 11:16:30
cana
tate carcinoma were assessed based on receiver-operating characteristic (ROC) curve analysis, using a statistics graphic data measurement software program
(STATA computing resource, Los Angeles, CA). Logistic
regression analysis was conducted to assess the importance of the free to total PSA ratio for predicting prostate carcinoma while controlling for other variables,
including age, rectal findings, PSAD, and prostate volume.
RESULTS
The median subject age was 67 years (range 49 – 89;
mean, 67.4). For cancer patients, the median age was
67 years; for men with BPH, it was 68 years (P ú 0.05,
Table 1). Prostate volume was measured in all but two
patients with prostate carcinoma. Patients with BPH
had significantly larger prostate glands than prostate
carcinoma patients (31.2 vs. 25.6 cm3 , P Å 0.0001).
Patients with prostate carcinoma had significantly
higher levels of total and free PSA than those with BPH
(P Å 0.0001). Median total PSA ({ standard error) for
patients with BPH was 3.8 { 0.5 ng/mL, as compared
with 12.2 { 41 ng/mL for prostate carcinoma patients.
Median free PSA for BPH and prostate carcinoma patients were 0.8 { 0.1 and 1.6 { 7.1 ng/mL, respectively.
Men with prostate carcinoma had significantly lower
free to total PSA ratios than those with BPH (P Å
0.0001). The median values for patients with prostate
carcinoma versus BPH were 10.9 { 1.3% versus 22.6
{ 1%.
The 113 cancers were classified clinically as T1a
in 1 patient, T1b in 2, T1c in 22, T2a in 15, T2b in 19,
T2c in 15, T3a in 4, T3b in 8, T3c in 23, and T4 in 4.
Nonpalpable cancer was found in 26 patients (23%),
including one case of T2b disease. Distant metastasis
(classified as M1) was present in 27 patients (23.9%).
The distribution of tumor grade was as follows: Gleason score 2 – 4 in 16.8% of patients (19 of 113), 5 and
6 in 23% (26 of 113), 7 in 35.4% (40 of 113), and 8 – 10
in 24.8% (28 of 113).
W: Cancer
Free to Total PSA Ratio in Prostate Carcinoma/Egawa et al.
93
TABLE 2
Distribution of Free to Total PSA Ratio by Total PSA in Patients with BPH and Prostate Carcinoma
Total PSA (ng/mL)
Free to total
PSA ratio
(%)
BPH
Prostate Carcinoma
°2 No. (%)
2.1–4 No. (%)
4.1–10 No. (%)
¢10.1 No. (%)
°2 No. (%)
2.1–4 No. (%)
4.1–10 No. (%)
¢10.1 No. (%)
°10
10.1–15
15.1–20
20.1–25
25.1–30
¢30.1
1 (2.3)
5 (11.4)
4 (9.1)
5 (11.4)
12 (27.2)
17 (38.6)
2 (5.6)
6 (16.7)
10 (27.7)
10 (27.7)
2 (5.6)
6 (16.7)
3 (6)
7 (14)
10 (20)
10 (20)
9 (18)
11 (22)
6 (24)
2 (8)
3 (12)
6 (24)
5 (20)
3 (12)
0 (0)
3 (50)
0 (0)
0 (0)
0 (0)
3 (50)
3 (25)
6 (50)
2 (16.7)
1 (8.3)
0 (0)
0 (0)
12 (42.9)
6 (21.4)
6 (21.4)
1 (3.6)
1 (3.6)
2 (7.1)
35 (52.2)
14 (20.9)
7 (10.4)
3 (4.5)
2 (3)
6 (9)
Total
44 (100)
36 (100)
50 (100)
25 (100)
6 (100)
12 (100)
28 (100)
67 (100)
PSA: prostate specific antigen; BPH; benign prostatic hyperplasia; No.: no. of patients.
TABLE 3
Distribution of Free to Total PSA Ratio by Prostate Volume in Patients with BPH and Prostate Carcinoma
Prostate volume (cm3)
Free to total
PSA ratio
(%)
BPH
Prostate carcinoma
°20 No. (%)
20.1–30 No. (%)
¢30.1 No. (%)
°20 No. (%)
20.1–30 No. (%)
¢30.1 No. (%)
°10
10.1–15
15.1–20
20.1–25
25.1–30
¢30.1
0 (0)
7 (35)
3 (15)
2 (10)
5 (25)
3 (15)
3 (5.8)
7 (13.5)
11 (21.2)
10 (19.2)
6 (11.5)
15 (28.8)
9 (10.8)
6 (7.2)
13 (15.7)
19 (22.9)
17 (20.5)
19 (22.9)
17 (51.5)
7 (21.2)
3 (9.1)
2 (6.1)
0 (0)
4 (12.1)
24 (51.1)
10 (21.2)
5 (10.6)
2 (4.3)
2 (4.3)
4 (8.5)
8 (25.8)
11 (35.5)
7 (22.6)
1 (3.2)
1 (3.2)
3 (9.7)
Total
20 (100)
52 (100)
83 (100)
33 (100)
47 (100)
31 (100)
PSA: prostate specific antigen; BPH: benign prostatic hyperplasia; No.: no. of patients.
Sixty-six of 113 patients underwent radical prostatectomy for resectable prostate carcinoma (°T3). Extraprostatic spread was confirmed pathologically in 44
patients (¢pT3), including 27 with seminal vesicle
involvement and 8 with lymph node involvement. Median total PSA for patients with extraprostatic versus
pathologically organ-confined disease was 11.8 { 6.4
ng/mL versus 5.1 { 1.1 ng/mL. Median free to total
PSA ratios were 8.6 { 2.3% versus 11.4 { 2.7% for these
patients. Total PSA, but not the percentage of free PSA,
was associated with pathologically advanced disease
(P Å 0.0003 and 0.28, respectively). Though total PSA
showed significant correlation with the Gleason score
of the tumor, the percentage of free PSA showed no
such correlation (P Å 0.004 and 0.729, respectively).
The distributions of the free to total PSA ratio according to total PSA and gland volume are presented
in Tables 2 and 3, respectively. Figure 1 illustrates the
/ 7b48$$0809
12-10-96 11:16:30
cana
distribution of the free to total PSA ratio according
to total PSA and prostate volume in the subgroup of
patients with total PSA amounting to 2.1 – 10 ng/mL.
The free to total PSA ratio appeared to be smaller in
patients with prostate carcinoma, irrespective of serum total PSA or prostate volume. Logistic regression
analysis indicated this and prostate volume to be the
only significant predictors for distinguishing cancer
from benign conditions when total PSA amounted to
2.1 – 10 ng/mL after controlling for other variables, including PSAD (P Å 0.001 and 0.002, respectively). This
ratio was the only significant covariate for predicting
prostate carcinoma in other patient subgroups (P õ
0.01).
When total PSA was restricted to 2.1 – 10 ng/mL (n
Å 126), its medians were not significantly different for
prostate carcinoma and BPH patients (P Å 0.123, Table
4). Differences in median percentages of free PSA for
W: Cancer
94
CANCER January 1, 1997 / Volume 79 / Number 1
nificantly better performance of free to total PSA ratio
compared with total PSA in distinguishing prostate
carcinoma was evident in all patient subgroups (P õ
0.05). No significant advantage of free to total PSA ratio
over PSAD could be found (P ú 0.05).
In patients with total PSA between 2.1 and 4 ng/
mL, 87.5% (7 of 8) of the cancers were detected by free
PSA of 17% or less, while 50% (2 of 4) of the biopsies
for prostate volumes less than 20 cm3 were eliminated
(Table 5). In men with larger glands, this cutoff value
was used to detect 100% of cancers (4 of 4), eliminating
75% (24 of 32) of unnecessary biopsies. The sensitivity
of free to total PSA ratio at 17% was 100% (4 of 4),
85.7% (6 of 7), and 100 %(1 of 1) for patients age 59
years or younger, age 60 – 69 years, and age 70 years
or older, respectively. The specificity for these patients
was 75% (3 of 4), 77.8% (14 of 18), and 64.3% (9 of 14).
The median age of this subgroup of 48 patients was
66.5 { 1 years. The median volume of the prostate
gland was 26.6 { 1.7 cm3 , essentially the same as for
the entire group (P ú 0.05).
DISCUSSION
FIGURE 1. Distribution of free to total prostate specific antigen (PSA)
ratio is shown for patients with total PSA levels of 2.1–10 ng/mL. (a)
Data are plotted against total PSA levels. (b) Data are plotted against
prostate volumes.
these two groups were statistically significant (P Å
0.0001). With the free to total PSA ratio, it was possible
to differentiate prostate carcinoma in palpably benign
glands of all PSA ranges, but this was not possible with
total PSA (n Å 181, P Å 0.0001 vs. 0.052). This was also
the case with palpably benign glands and total PSA
between 2.1 and 10 ng/mL (n Å 99, P Å 0.0001 vs.
0.408). When data were limited to 48 patients with
total PSA between 2.1 and 4 ng/mL, all 12 cancers had
free to total PSA ratios of 25% or less. The sensitivity
and specificity at this cutoff value were 100% (12 of
12) and 22.2% (8 of 36), respectively.
To evaluate the performance of total PSA further,
the sensitivities and specificities of free to total PSA
ratio and PSAD were calculated for different cutoff
values. The relation between sensitivity and specificity
in these tests was illustrated objectively by ROC curves
(Fig. 2). The sensitivity and specificity of the free to
total PSA ratio in subgroups of patients with total PSA
between 2.1 and 10 ng/mL, patients with both palpably benign glands and total PSA between 2.1 and 10
ng/mL, and patients with PSA between 2.1 and 4 ng/
mL were 80% and 73.3%, 92.3% and 73.3%, and 91.7%
and 72.2%, respectively, at a cutoff value of 17%. Sig-
/ 7b48$$0809
12-10-96 11:16:30
cana
Serum PSA is presently the best means for identifying,
classifying, and monitoring prostate carcinoma and is
being investigated extensively.15 While PSA may detect
more cancers than DRE and these cancers may be
clinically significant by pathologic criteria,7 this tumor
marker is by no means perfect. Cancer is found in only
10 – 30% of men with intermediate PSA between 4.1
and 10 ng/mL, and some cancers require detection by
repeated biopsy.7,10,15 – 17 To enhance the clinical usefulness of serum PSA for detecting early prostate carcinoma, especially in this gray zone, PSA density,18 PSA
velocity,19 age specific reference ranges,8 and free to
total PSA ratio20,21 have come to be used, each with its
own sensitivity and specificity, attended with failure
to detect a substantial proportion of curable cancers
or a high rate of false-positive results.
Christensson et al. reported that the specificity for
prostate carcinoma significantly increased without
loss of sensitivity at PSA levels of 4 – 20 ng/mL, using
a cutoff value for the free to total PSA ratio of 18%.3
Specificity was 95% at a sensitivity of 71%. Luderer et
al. examined the clinical usefulness of free PSA in the
differential diagnosis of prostate carcinoma in patients
with intermediate PSA levels.4 They found that the ratio of free to total PSA could significantly differentiate
between benign and malignant histologic conditions,
whereas total PSA could not. The sensitivity and specificity of the free to total PSA ratio of 4 – 10 ng/mL were
88% and 50% at a cutoff value of 20%. Catalona et
al. evaluated the free to total PSA ratio in prostate
carcinoma screening of men with serum PSA between
W: Cancer
Free to Total PSA Ratio in Prostate Carcinoma/Egawa et al.
95
TABLE 4
Descriptive Statistics for Total PSA and Free to Total PSA Ratio for Subgroups of Patients
Total PSA 2.1–10 ng/mL
Total PSA (ng/mL)
Mean
Median
Standard error
95% CI for mean
Free to total PSA ratio (%)
Mean
Median
Standard error
95% CI for mean
Normal DRE
Total PSA (ng/mL)
Mean
Median
Standard error
95% CI for mean
Free to total PSA ratio (%)
Mean
Median
Standard error
95% CI for mean
Total PSA 2.1–10 ng/mL
and normal DRE
Total PSA (ng/mL)
Mean
Median
Standard error
95% CI for mean
Free to total PSA ratio (%)
Mean
Median
Standard error
95% CI for mean
Total PSA 2.1–4 ng/ml
Total PSA (ng/mL)
Mean
Median
Standard error
95% CI for mean
Free to total PSA ratio (%)
Mean
Median
Standard error
95% CI for mean
BPH
Prostate carcinoma
n Å 86
4.96
4.46
0.24
4.47–5.44
n Å 86
23.73
21.1
1.37
21.02–26.45
n Å 40
5.55
5.72
2.05
4.89–6.2
n Å 40
13
11.75
1.23
10.51–15.49
n Å 155
5.96
3.82
0.52
4.93–6.99
n Å 155
25.05
22.6
1.37
22.99–27.12
n Å 26
46.18
5.91
36.19
028.35–120.72
n Å 26
14.69
11.95
1.79
10.7–18.09
n Å 86
4.96
4.46
0.24
4.47–5.44
n Å 86
23.73
21.1
1.37
21.02–26.45
n Å 13
4.84
4.76
0.47
3.82–5.85
n Å 13
11.45
11.1
1.19
8.86–14.03
n Å 36
2.87
2.81
0.1
2.66–3.08
n Å 36
23.11
19.5
2.03
18.99–27.24
n Å 12
3.18
3.18
0.19
2.77–3.59
n Å 12
12.37
11.75
1.53
9–15.74
P value
0.123
0.0001
0.052
0.0001
0.408
0.0001
0.146
0.0001
BPH: benign prostatic hyperplasia; PSA: prostate specific antigen; DRE: digital rectal examination; CI: confidence interval.
4.1 and 10 ng/mL.5 Measurement of the percentage of
free serum PSA improved the specificity of prostate
carcinoma screening. In men with prostate volumes
less than 40 cm3 , free PSA of 13.7% or less detected
90% of the cancers while eliminating 76.2% of the biopsies. In men with larger glands, a cutoff value of
20.5% or less was required in this study to detect at
least 90% of cancers while eliminating 38.1% of unnec-
/ 7b48$$0809
12-10-96 11:16:30
cana
essary biopsies. All of these studies indicate significant
improvement in specificity with the percentage of free
PSA and the retention of high sensitivity.
It is yet unclear why the free form of PSA exists in
the presence of excess ACT in human serum. It may represent a nicked form of the original molecule, with internal proteolytic cleavage having lost enzymatic activity and
hence unreactive toward ACT.22 Free PSA may remain to
W: Cancer
96
CANCER January 1, 1997 / Volume 79 / Number 1
FIGURE 2. Receiver-operating characteristic curve analysis compares total prostate specific antigen (PSA), free to total PSA ratio, and PSA density.
Data are shown for (a) all 268 patients in this study, (b) 126 patients with total PSA levels of 2.1–10 ng/mL, (c) 99 patients with total PSA levels of
2.1–10 ng/mL and palpably benign prostate glands, and (d) 48 patients with total PSA levels of 2.1–4 ng/mL.
TABLE 5
Results with Free to Total PSA Ratio at 17% in a 48-Patient Subgroup with Total PSA Levels of 2.1–4 ng/mL,
According to Prostate Volume and Patient Age
No. of patients
No. with cancer
Sensitivity (%)
Specificity (%)
Prostate volume (cm3)
¢20
õ20
36
12
4
8
100 (4/4)
87.5 (7/8)
75 (24/32)
50 (2/4)
Patient age (yrs)
°59
60–69
¢70
8
25
15
4
7
1
100 (4/4)
85.7 (6/7)
100 (1/1)
75 (3/4)
77.8 (14/18)
64.3 (9/14)
Total
48
12
91.7 (11/12)
72.2 (26/36)
PSA: prostate specific antigen.
some extent, as a result of incomplete reaction of PSA
with ACT caused by slow binding.22 Bjork et al. found the
same prostatic epithelial cells that produce PSA to contain
ACT.23 There was a lack of ACT production in benign
hyperplastic nodules containing PSA. In contrast, the production of ACT and PSA was evident in malignant tissue.
/ 7b48$$0809
12-10-96 11:16:30
cana
The significance of this finding is not clear, because serum
PSA mainly forms a complex with the hepatocyte-derived
serine protease inhibitor ACT.20–23
In clinical application, the use of age specific reference ranges, PSAD, or free to total PSA ratio may be
more practical than PSA velocity, due to inherent bio-
W: Cancer
Free to Total PSA Ratio in Prostate Carcinoma/Egawa et al.
logic and laboratory variability.24 Age specific PSA reference ranges for Japanese elderly males with normal
rectal and sonographic findings are less than those for
Caucasians.8,9 The normal upper limits, determined by
the 95th percentile range in AIA-1200 assay (TOSOH
Corporation), are 2.1 ng/mL for men age 40 – 49 years,
2.9 ng/mL for those age 50 – 59 years, 4 ng/mL for men
age 60 – 69 years, 5.2 ng/mL for men age 70 – 79 years,
and 5.9 ng/mL for patients age 80 – 89 years.9 For 126
patients with PSA levels of 2.1 – 10 ng/mL, the sensitivity and specificity for distinguishing cancer from benign conditions were 67.5% (27 of 40) and 46.5% (40
of 86), respectively. Overall, 32.5% of tumors (13 of
40) would have been overlooked if age specific PSA
reference ranges had been used. The sensitivity and
specificity of PSAD of 0.15 or greater in this patient
group were 80% (32 of 40) and 66.3% (57 of 86), respectively. Eight of 40 tumors (20%) would have been overlooked by PSAD of 0.15 or greater. At free PSA cutoff
values of 17% or lower for this patient group, sensitivity and specificity for detecting cancer were 80% (32
of 40) and 73.3% (63 of 86), respectively (Fig. 2). Free
to total PSA ratio and PSAD are thus better detectors
than age specific reference ranges. Though ROC curve
analysis failed to indicate a significant difference between PSAD and the free to total PSA ratio in detecting
prostate carcinoma, the logistic model indicated the
ratio to be more accurate than PSAD for all patient
subgroups.
Several studies have reported the low prevalence
of prostate carcinoma (5.3 – 7.2%) in patients with total
PSA levels below 4 ng/mL.25,26 Colberg et al. believed
that this finding did not justify additional invasive diagnostic tests.25 However, 70 – 80% of these cancers are
known to be pathologically organ-confined and thus
potentially curable.27 – 29 Test specificity should thus be
enhanced while sensitivity is also retained for patients
with low PSA. This may be peculiar to Asian males due
to their inherently lower PSA.7,9,10 At a free to total PSA
cutoff ratio of 17% or less in patients with total PSA
between 2.1 and 4 ng/mL, 91.7% of cancers (11 of 12)
were detected and 72.2% of negative biopsies (26 of
36) were eliminated (Table 5). This parameter may
thus be best applicable to patients with low total PSA.
Patients’ prostate volumes and ages did not significantly affect the ability of free to total PSA ratio to
detect cancer in this patient subgroup. Rommel et al.
found PSAD useful for diagnosing prostate carcinoma,
particularly in patients with total PSA levels of 3 – 3.9
ng/mL17 However, with PSAD, there is difficulty in accurately measuring prostate volume. The free to total
PSA ratio may surpass PSAD as a detector of cancer
because it does not depend on prostate volume data.
/ 7b48$$0809
12-10-96 11:16:30
cana
97
The free to total PSA ratio may be useful for patients scheduled for surgery who have a clinical diagnosis of BPH. Many physicians now recommend biopsy in such patients if total PSA exceeds 10 ng/mL,
irrespective of rectal findings. Patients with mildly elevated total PSA (2.1 – 10 ng/ml) may benefit from a
determination of free to total PSA ratio. The sensitivity
and specificity of the free to total PSA ratio for detecting cancer in this study were 92.3% (12 of 13) and
73.3% (63 of 86) at a cutoff value of 17%. It may thus be
possible to detect many cancers while avoiding most
unnecessary biopsies.
Ratios of the three molecular forms of PSA (total
PSA, in a complex formed with ACT, and free PSA)
may serve as reference values for comparative studies
among different countries, provided that similar antibodies that recognize the same epitopes are used. Interestingly, the cutoff value of the free to total PSA
ratio in this study was essentially that for studies conducted in Western countries.3 – 5 Differences in prostate
volumes among males of different races, a major determinant of PSA,10 may have little impact on the free
to total PSA ratio, because each ratio is constant for
all ages.10,30 No meaningful correlation of the free to
total PSA ratio to pathologic stage or Gleason score has
been detected. Whether this applies to all subgroups of
patients remains to be seen.
Only a small number of patients was used in this
study. Bias in selection may have affected the results
more favorably. Nevertheless, the free to total PSA ratio may be useful for detecting cancer more effectively
while avoiding unnecessary biopsies in certain groups
of patients. This parameter is at least as useful as PSAD
and superior to age specific reference ranges. A larger
prospective study is currently underway to confirm the
preliminary findings presented here.
REFERENCES
1.
2.
3.
4.
Watt KWK, Lee P, M’Timkulu T, Chan W, Loor R. Human
prostate-specific antigen: structural and functional similarity with serine proteases. Proc Natl Acad Sci U S A 1986;
83:3166–70.
Lilja H. A kallikrein-like serine protease in prostatic fluid
cleaves the predominant seminal vesicle protein. J Clin Invest 1985;76:1899–903.
Christensson A, Bjork T, Nilsson O, Dahlen U, Matikainen
MT, Cockett AT, et al. Serum prostate-specific antigen complexed to a1-antichymotrypsin as an indicator of prostate
cancer. J Urol 1993;150:100–5.
Luderer AA, Chen Y-T, Soriano TF, Kramp WJ, Carlson G,
Cuny C, et al. Measurement of the proportion of free to total
prostate-specific antigen improves diagnostic performance
of prostate-specific antigen in the diagnostic gray zone of
total prostate-specific antigen. Urology 1995;46:187–94.
W: Cancer
98
CANCER January 1, 1997 / Volume 79 / Number 1
5.
Catalona WJ, Smith DS, Wolfert RL, Wang TJ, Rittenhouse
HG, Ratliff TL, et al. Evaluation of percentage of free serum
prostate-specific antigen to improve specificity of prostate
cancer screening. JAMA 1995;274:1214–20.
Demura T, Watarai Y, Togashi M, Hirano T, Ohashi N, Koyanagi T. Measurement of prostate specific antigen and gseminoprotein ratio: a new means of distinguishing benign
prostatic hyperplasia and prostate cancer. J Urol 1993;
150:1740–5.
Egawa S, Suyama K, Ohori M, Kawakami T, Kuwao S, Hirokado K, et al. Early detection of prostate cancer: results of a
prostate specific antigen–based detection program in Japan.
Cancer 1995;76:463–72.
Oesterling JE, Jacobsen SJ, Chute CG, Guess HA, Girman
CJ, Panser LA, et al. Serum prostate-specific antigen in a
community-based population of healthy men: establishment of age-specific reference ranges. JAMA 1993;270:860–
4.
Imai K, Ichinose Y, Kubota Y, Yamanaka H, Sato J. Diagnostic
significance of prostate specific antigen and the development of a mass screening system for prostate cancer. J Urol
1995;154:1085–9.
Oesterling JE. Prostate-specific antigen and detecting early
prostate cancer: the importance of race and patient age. In:
Michael Garraway, editor. Epidemiology of prostate disease.
New York: Springer, 1995:67–81.
Epstein JI, Walsh PC, Carmichael M, Brendler CB. Pathologic
and clinical findings to predict tumor extent of nonpalpable(stage T1c) prostate cancer. JAMA 1994;271:368–74.
International Union Against Cancer. In: Spiessl B, Beahrs
OH, Hermanek P, Hutter RVP, Scheibe O, Sobin LH, et al,
editors. TNM Atlas, 3rd edition, 2nd revision. New York:
Springer-Verlag, 1992:241–50.
Chan DW, Thornton PL. AIA-1200 immunoassay system: immunoassay automation. New York: Academic Press Inc.,
1992:95–115.
Monoclonal antibodies to prostate specific antigen, PSA
[manufacturer’s catalog]. Gothenburg, Sweden: CanAg Diagnostics, 1993;1–8.
Oesterling JE. Prostate-specific antigen: a clinical assessment of the most useful tumor marker for adenocarcinoma
of the prostate. J Urol 1991;145:907–23.
Keetch DW, Catalona WJ, Smith DS. Serial prostatic biopsies
in men with persistently elevated serum prostate specific
antigen values. J Urol 1994;151:1571–4.
Rommel FM, Agusta VE, Breslin JA, Huffnagle HW, Pohl CE,
Sieber PR, et al. The use of prostate specific antigen and
prostate specific antigen density in the diagnosis of prostate
cancer in a community based urology practice. J Urol
1994;151:88–93.
Benson MC, Whang IS, Olsson CA, McMahon DJ, Cooner
WH. The use of prostate specific antigen density to enhance
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
/ 7b48$$0809
12-10-96 11:16:30
cana
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
the predictive value of intermediate levels of serum prostate
specific antigen. J Urol 1992;147:817–21.
Carter HB, Pearson JD, Metter EJ, Brant LJ, Chan DW, Anders
R, et al. Longitudinal evaluation of prostate-specific antigen
levels in men with and without prostate disease. JAMA
1992;267:2215–20.
Stenmann UH, Leinonen J, Alfthan H, Rannikko S, Tuhkanen
K, Alfthan O. A complex between prostate-specific antigen
and alpha-1-antichymotrypsin is the major form of prostatespecific antigen in serum of patients with prostatic cancer:
assay of the complex improves clinical sensitivity for cancer.
Cancer Res 1991;51:222–6.
Lilja H, Christensson A, Dahlen V, Matikainen MT, Nilsson
O, Pettersson K, et al. Prostate-specific antigen in human
serum occurs predominantly in complex with alpha-1 antichymotrypsin. Clin Chem 1991;37:1618-25.
Huber PR, Schmid H-P, Mattarelli G, Strittmatter B, van
Steenbrugge GJ, Maurer A. Serum free prostate specific antigen: isoenzymes in benign hyperplasia and cancer of the
prostate. Prostate 1995;27:212-9.
Bjork T, Hulkko S, Bjartell A, DiSant’agnese A, Abrahamsson
P, Lilja H. Alpha 1-antichymotrypsin production in PSA-producing cells is common in prostate cancer but rare in benign
prostatic hyperplasia. Urology 1994;43:427–34.
Kadmon D, Weinberg AD, Williams RH, Pavlik VN, Cooper P,
Migliore PJ. Pitfalls in interpreting prostate specific antigen
velocity. J Urol 1996;155:1655–7.
Colberg JW, Smith DS, Catalona WJ. Prevalence and pathological extent of prostate cancer in men with prostate specific antigen levels of 2.9 to 4.0ng./ml.. J Urol 1993;149:507–
9.
Labrie F, Dupont A, Suburu R, Cusan L, Tremblay M, Gomez
J-L, et al. Serum prostate specific antigen as pre-screening
test for prostate cancer. J Urol 1992;147:846– 52.
Narayan P, Gajendran V, Taylor SP, Tewari A, Presti JC Jr.,
Leidich R, et al. The role of transrectal ultrasound-guided
biopsy-based staging, preoperative serum prostate-specific
antigen, and biopsy Gleason score in prediction of final
pathologic diagnosis in prostate cancer. Urology 1995;
46:205–12.
Partin AW, Yoo J, Carter HB, Pearson JD, Chan DW, Epstein
JI, et al. The use of prostate specific antigen, clinical stage
and Gleason score to predict pathological stage in men with
localized prostate cancer. J Urol 1993;150:110–4.
Noldus J, Stamey TA. Histological characteristics of radical
prostatectomy specimens in men with a serum prostate specific antigen of 4 ng./ml. or less. J Urol 1996;155:441–3.
Oesterling JE, Jacobsen SJ, Klee GG, Pettersson K, Piironen
T, Abrahamsson P, et al. Free, complexed and total serum
prostate specific antigen: the establishment of appropriate
reference ranges for their concentrations and ratios. J Urol
1995;154:1090–5.
W: Cancer
Документ
Категория
Без категории
Просмотров
3
Размер файла
138 Кб
Теги
532
1/--страниц
Пожаловаться на содержимое документа