Download Clusterin as a possible predictor for biochemical recurrence

Prostate Cancer and Prostatic Diseases (2004) 7, 243–248
& 2004 Nature Publishing Group All rights reserved 1365-7852/04 $30.00
www.nature.com/pcan
Clusterin as a possible predictor
for biochemical recurrence of
prostate cancer following radical
prostatectomy with intermediate
Gleason scores: a preliminary
report
MR Pins1, JE Fiadjoe1,2, F Korley1, M Wong1, AW Rademaker1,
B Jovanovic1, TK Yoo1,3, JM Kozlowski1, A Raji1, XJ Yang1 & C Lee1*
1
Departments of Urology, Preventive Medicine, Pathology, and Robert H. Lurie
Comprehensive Cancer Center, Northwestern University Feinberg School of Medicine,
Chicago, Illinois, USA
Disease recurrence following radical prostatectomy is a major concern in prostate
cancer patients. Gleason scores are useful in predicting recurrence. Low Gleason
scores are usually associated with long disease-free intervals, while high Gleason
scores are suggestive of early recurrence. However, prediction of recurrence has
been difficult with intermediate Gleason scores. Clusterin is a ubiquitous
secretory sulfated glycoprotein. It is also an antiapoptotic mediator in prostate
cancer. The objective of the present study is to determine if clusterin can serve as a
predictive biomarker for recurrence of prostate cancer with intermediate Gleason
scores in patients following radical prostatectomy. Prostatic specimens with
Gleason score of 6 (3 þ 3) or 7 (3 þ 4) were obtained from the archival bank. Three
groups of specimens were investigated. The first group was from nine patients
who developed recurrent disease according to a persistent rise of serum PSA
within 3 years following radical prostatectomy. Those in the second group and the
third group were from patients who showed no evidence of disease recurrence for
at least 5 y (11 patients) and 10 y (eight patients), respectively following the
surgery. Histological sections were subjected to immunohistochemical staining
using a monoclonal antibody specific for clusterin. The staining intensity was
scored as 0, 1, 2, and 3, with 0 being no staining, 1 showing less than 25% positive
staining, 2 being 25–50% positive, and 3 showing greater than 75% positive
staining. One-way ANOVA with Bonferroni correction was used for statistical
analysis. Evaluation of the scores of clusterin staining was carried out according to
four specific areas in each specimen. They were (a) benign epithelial cells, (b)
malignant epithelial cells (cancer epithelia), (c) stromal cells surrounding benign
*Correspondence: C Lee, Department of Urology, Tarry Building,
Room 16-733, Northwestern University Feinberg School of Medicine,
303 East Chicago Avenue, Chicago, IL 60611, USA.
E-mail: [email protected]
An abstract of this report was presented at the NCI sponsored
prostate cancer SPORE workshop, Westfield Conference Center,
Chantilly, Virginia, July 14–16, 2002.
2
Current address: Department of Anesthesiology, Downstate Medical
Center, 450 Clarkson Avenue, P.O. Box 6, Brooklyn, NY 11210, USA
3
Current address: Department of Urology, Eul Ji Hospital, 280-1,
Hagye 1 Dong, Nowon Gu, Seoul 139-711, South Korea
An abstract of this report was presented at the NCI sponsored
prostate cancer SPORE workshop, Westfield Conference Center,
Chantilly, Virginia, July 14–16, 2002.
Received 20 November 2003; revised 20 January 2004; accepted 10
February 2004
Clusterin and recurrent prostate cancer
MR Pins et al
244
cells, and (d) stromal cells surrounding malignant cells (cancer stroma). Staining
score in prostatic epithelial cells, benign as well as malignant, showed no
significant relationship among the three patient groups. However, when staining
scores in stromal cells were compared, there was a significant difference between
patients with recurrent disease and those showed no evidence of disease
recurrence for at least 10 y. Results of this preliminary study support the important
role of clusterin in the stromal component for prostate cancer progression.
Clusterin immunostaining may be useful to aid the prediction of chance of disease
recurrence in patients with Gleason score 6 or 7 prostate cancer following radical
prostatectomy. Further studies with a large number of cases are warranted to verify
this preliminary finding.
Prostate Cancer and Prostatic Diseases (2004) 7, 243–248. doi:10.1038/sj.pcan.4500722
Keywords:
recurrence
clusterin immunostaining; stromal cells; Gleason scores; PSA
Introduction
Radical prostatectomy is the treatment of choice with an
intention to provide a cure for patients with organconfined prostate cancer. When the postoperative serum
prostate-specific antigen (PSA) is undetectable, it is
suggestive of a cure. The resurgence of a persistently
detectable serum PSA is indicative of biochemical
recurrence of prostate cancer. In at least 30–50% of
patients, who enjoy an initial disease free state; recur at
various postoperative intervals.1 Although the exact
factors controlling recurrence in prostate cancer remains
unclear, clinical experience dictates that Gleason scores
have been useful in predicting disease recurrence.2–5
Cases with high Gleason scores have a tendency of
recurring shortly after radical prostatectomy; while those
with low Gleason scores often enjoy a prolonged diseasefree interval. The dilemma lies in those with intermediate
Gleason scores, when prediction of disease recurrence
has been difficult. A tumor biomarker that can aid the
prediction of recurrence would be valuable in the
management of this special group of patients.
Clusterin, also known as SGP-2, CLI, Apo-J, and
TRPM-2, is a ubiquitous secretory sulfated glycoprotein.6–10 It is a multifunctional protein, which has been
implicated in diverse processes as cell aggregation,6
inhibition of complement-mediated cytotoxicity,8 and
lipid transport.11 Clusterin has been shown to play an
antiapoptotic role in prostate cancer cells.12,13 The level of
clusterin expression has been correlated with Gleason
scores in prostate cancer.14 Clusterin overexpression has
resulted in the development of resistance to androgen
ablation.15 Recently, the use of antisense oligodeoxynucleotide targeting the clusterin gene enhanced apoptosis
induced by either radiation or chemotherapeutic agents
further support the importance of clusterin expression in
tumor progression.16–18 The objective of this preliminary
study is to determine if clusterin can be used as a
predictor of disease recurrence in prostate cancer with
intermediate Gleason scores following radical prostatectomy.
identify patients who had radical prostatectomy and
subsequent clinical follow-up data from 1989 to 2001. A
series of exclusion criteria was applied. In order to test
our hypothesis, we assembled a cohort of patients who
had undergone prostatectomy and had regularly (annually or semiannually) documented postoperative
serum PSA levels. The cohort was limited to patients
who had undetectable serum PSA levels (defined as
o0.5 ng/ml) at the time of their first postoperative PSA
determination and who had no adjuvant therapy prior to
surgery. Next, we searched clinical specimens corresponding to the cohort from the archival bank of the
Department of Pathology of Northwestern University
Medical School. We selected three groups of patients
under three broad categories. First, we limited the
specimens to be included into this study only with
Gleason scores 6 (3 þ 3) and 7 (3 þ 4). This inclusion
criterion defines the scope of the present study, as
prediction of disease recurrence with extreme Gleason
scores has been well established.3–5 Another inclusion
criterion is the reduction of serum PSA levels to an
undetectable level, defined as less than 0.5 ng/ml; 3–6
months following surgery in all patients. Patients with a
detectable serum PSA level immediately following
radical prostatectomy were excluded from this study.
Finally, patients showing positive surgical margins in
their surgical specimens were also excluded from this
study. We restricted a group of patients with recurrence
within 3 y of surgery (Group A). Biochemical recurrence
was defined as patients who had elevated serum PSA
levels above the baseline for at least two consecutive
office visits. We then selected two additional groups of
patients. One group was those who remained diseasefree with a 5-y follow-up (Group B) and another group
was those who were disease-free with a 10-y follow-up
(Group C). The assignment of patient groups and the
number of study subjects in each group are indicated in
the following chart:
Patient data
Group A consisted of patients who recurred within 3 y after
prostatectomy (nine patients)
Group B consisted of patients without recurrence after 5 y of followup (11 patients)
Group C consisted of patients without recurrence after 10 y of
follow-up (eight patients)
After receiving Institutional Review Board approval, a
search of the hospital tumor registry was performed to
There were a total of 28 patients. In many cases, where
more than one specimen per patient was obtained, each
Materials and methods
Prostate Cancer and Prostatic Diseases
Clusterin and recurrent prostate cancer
MR Pins et al
specimen was considered as one determination and was
analyzed toward the overall mean and variation for that
patient.
Immunohistochemistry
All specimens were from radical prostatectomy specimens and were of formalin-fixed paraffin-embedded
tissues. Sections of 6 mm thick were cut and subjected to
immunohistochemical staining for clusterin. Sections
were first deparaffinized in xylene and then rehydrated
in graded alcohol solutions. Endogenous peroxidase
activity was quenched using a commercial peroxidase
blocking solution (DAKO, Carpinteria, CA, USA). Antigen retrieval was performed using DAKO Target
retrieval solution in a steamer at 901C. Sections were
then incubated with a 1:200 dilution with a mouse
monoclonal antibody against the clusterin alpha chain
(Upstate Biotech, Lake Placid, NY, USA) for 30 min at
room temperature followed by incubation with a
secondary antibody–polymer–HRP (horseradish peroxidase) conjugate (DAKO’s EnVision þ system) for 30 min
at room temperature. Clusterin expression was visualized by incubation with 3,30 -diaminobenzidine (DAB)
for 2 min followed by counterstaining with Gill’s
hematoxylin. All slides were processed at the same time,
thus eliminating the run-to-run variation. Negative
control slides were processed in a similar manner, with
substitution of the primary antibody with nonimmune
mouse IgG at the same concentration for the primary
antibody. No color reaction was observed in any of the
negative control slides.
Evaluation of staining patterns
Since the intensity of clusterin staining was different
from one cell type to another in each specimen, it is
important that comparisons were made according to the
specific cell type. In an attempt to elucidate characteristics of clusterin staining in specific areas of each
specimen, the following regions were evaluated from
each slide:
(1) Staining status of benign epithelium
(2) Staining status of adenocarcinoma
(3) Staining status of stroma associated with benign
epithelium
(4) Staining status of stroma associated with adenocarcinoma
The staining pattern of all specimens was evaluated by
two authors (MRP and JEF) independently. The evaluators had no knowledge of follow-up information at the
time of the evaluation. Occasional discrepancies were
resolved by re-evaluation and consensus. Results were
scored as 0, 1 þ , 2 þ , and 3 þ , where 0 representing no
staining, 1 þ representing less than 25% of cells stained
positive, 2 þ representing 25–75% of cells stained
positive, and 3 þ representing greater than 75% of cells
stained positive. All comparisons of staining patterns
were made at 40 magnification. The resulting scores
were then tabulated and the patients sorted into the three
groups as described earlier.
245
Statistical analysis
The status of positive clusterin staining was compared
across the three groups: Group A (recurrence within 3 y),
Group B (5 y disease-free), and Group C (10 y diseasefree), using a one-way nested analysis of variance.19,20.The correlation among multiple determinations
on the same patient was taken into account when
calculating the standard error of the mean for each
group. This resulted in an effective sample size between
the number of persons (if the correlation was near 1) and
the number of determinations (if the correlation was near
Table 1 Summary of clusterin immunostaining status in prostate
cancer specimens for all patients
Patient
ID
Replicate Benign
Benign
number epithelia stroma
Cancer
Cancer Total
epithelia stroma
Group A
1
1
1
2
3
3
4
5
5
6
6
7
8
8
9
9
1
2
3
1
1
2
1
1
2
1
2
1
1
2
1
2
1
1
2
2
1
3
2
2
1
1
2
2
1
1
1
1
3
2
3
2
2
3
3
3
2
3
3
2
2
3
3
2
2
1
1
1
2
3
3
3
3
2
2
3
3
2
1
3
3
2
2
1
1
1
1
2
2
1
2
3
3
1
2
2
9
6
8
6
6
10
9
10
8
7
9
10
9
7
7
8
Group B
18
18
19
20
21
22
22
23
23
24
24
25
25
26
26
26
26
27
28
1
2
1
1
1
1
2
1
2
1
2
1
2
1
2
3
4
1
1
1
2
0
1
1
3
1
2
1
2
2
1
2
1
3
3
1
1
1
1
3
1
1
2
2
2
2
3
3
3
3
3
1
3
3
1
2
1
3
3
1
2
1
2
2
1
3
0
3
2
3
3
3
2
3
0
2
2
3
2
1
0
1
1
2
3
2
3
2
3
1
2
1
1
0
1
7
11
4
5
4
8
6
7
10
7
11
8
11
6
11
9
6
3
5
Group C
10
11
12
13
13
13
14
14
15
15
16
17
1
1
1
1
2
3
1
2
1
2
1
1
1
1
2
2
2
1
1
1
1
1
1
1
1
2
1
3
3
3
1
1
1
1
0
0
2
0
1
1
2
1
1
0
1
2
3
0
0
1
1
1
2
0
0
1
0
1
1
0
4
4
5
7
9
5
3
3
3
5
5
1
Prostate Cancer and Prostatic Diseases
Clusterin and recurrent prostate cancer
MR Pins et al
246
Figure 1 Clusterin immunostaining in benign epithelial cells, prostate cancer cells, and in stromal cells. (a–c) Examples of clusterin staining in benign
prostatic epithelial cells in prostatectomy specimens: (a) An example of staining score of 1, in which less than 25% of epithelial cells showed positive staining
for clusterin. (b) An example of staining score of 2, in which 25–75% of epithelial cells showed positive staining for clusterin. (c) An example of staining score
of 3, in which more than 75% of epithelial cells showed positive staining for clusterin. (d–f) Examples of clusterin staining in malignant prostatic epithelial
(prostate cancer) cells: (d) An example of staining score of 1, in which less than 25% of cancer cells showed positive staining for clusterin. (e) An example of
staining score of 2, in which 25–75% of cancer cells showed positive staining for clusterin. (f) An example of staining score of 3, in which more than 75% of
cancer cells showed positive staining for clusterin. (g–i) Examples of clusterin staining in stromal cells of prostate cancer specimens: (g) An example of
staining score of 1, in which less than 25% of stromal cells showed positive staining for clusterin. (h) An example of staining score of 2, in which 25–75% of
stromal cells showed positive staining for clusterin. (i) An example of staining score of 3, in which more than 75% of stromal cells showed positive staining
for clusterin.
zero). An overall analysis of variance P-value for group
was determined and Po0.05 was used as the criterion for
significance for this overall P-value. To control the
overall chance of making a Type I error at 5% for the
three pairwise comparisons (Group A vs Group B, Group
A vs Group C, and Group B vs Group C), these
comparisons among groups were made using t-tests
with the Bonferroni correction,20 so that a P-value
o0.0167 ¼ 0.05/3 (two-tail test) was needed for statistical
significance. Statistical analyses were performed using
PROC MIXED in SAS (SAS Institute Inc., Cary, NC,
USA).
Results
Typical patterns of immunohistochemical staining for
clusterin are shown in Figure 1. It is known that clusterin
is expressed in both benign and malignant prostatic
epithelial cells.14,21 Clusterin can also be expressed in
prostatic stromal cells.22 Therefore, the staining intensity
was evaluated separately in following four distinct types
of cells: (a) benign epithelium, (b) cancerous epithelium,
(c) stroma associated with benign epithelium, and (d)
stroma associated with cancerous epithelium. Table 1
shows results of clusterin staining in the above four
types of cells for each slide.
Prostate Cancer and Prostatic Diseases
The staining intensity of clusterin expression was
compared across the three groups using one-way
analysis of variance,19 taking into account that there
were multiple determinations per individual. An overall
ANOVA P-value was determined. Table 2 shows the
result of statistical analysis comparing each type of cells
and the cumulative score. There was no significant
difference in staining intensity when only the benign
prostatic epithelium was subjected for comparison
(P ¼ 0.41). A statistically significant difference was
observed, when the staining intensity in the stromal
cells adjacent to the benign prostatic epithelium
(P ¼ 0.006) the staining intensity of cancerous epithelium
(P ¼ 0.03), or the staining intensity of stromal cells
adjacent to the cancerous prostatic epithelium was
evaluated (P ¼ 0.008). Significant differences were observed between patients who had recurred within 3 y of
prostatectomy and those who had no recurrence in 10 y.
For cancer stromal cells, there was also a difference for
those who had no recurrence in 5 and 10 y. When the
staining intensity of the cumulative scores was subjected
to analysis, a similar degree of significant difference was
noted (Po0.001).
In addition to the above statistics, we performed
comparisons using different calculation parameters.
Table 2 also shows that when the scores of two types of
stroma cells (associated with benign and malignant
Clusterin and recurrent prostate cancer
MR Pins et al
247
Table 2 Statistical analysis of clusterin staining in human prostate cancer specimens
Group
No. of
patients
A
B
C
P-value
9
11
8
No. of
specimens
16
19
12
Benign
epithelia
Benign
stroma
Cancer
epithelia
Cancer
stroma
Total
stroma
Total
score
1.4870.16
1.5970.14
1.2770.19
0.41a
2.5470.24
2.0570.22
1.2670.27
0.006b
2.2170.26
1.9870.24
1.1670.29
0.03b
1.7970.23
1.6070.21
0.6570.26
0.008c
4.3270.39
3.62+0.36
1.89+0.44
0.001c
8.0770.56
7.2070.52
4.4270.64
0.001c
% Total
score45.0
10070
58711
9713
0.001d
Each value represents mean and standard error of the mean (mean7s.e.m.). In each designated area, the intensity of the staining was scored from 0 to 3, with 0
being no staining and 3 showing the strongest staining.
Group A ¼ patients who recurred within 3 y after prostatectomy (nine patients).
Group B ¼ patients without recurrence after 5 y of follow-up (11 patients).
Group C ¼ patients without recurrence after 10 y of follow-up (eight patients).
a
No significant pairwise comparisons.
b
Significant pairwise comparisons: recurrence (Group A) vs 10-y (Group C).
c
Significant pairwise comparisons: recurrence (Group A) vs 10-y (Group C), 5-y (Group B) vs 10-y (Group C).
d
Significant pairwise comparisons: recurrence (Group A) vs 5-y (Group B), recurrence (Group A) vs 10-y (Group C), 5-y (Group B) vs 10-y (Group C).
epithelia) were combined, the differences among different groups were also significant (Po0.001). The best
significance was obtained when the cumulative total
scores were calculated as the percentage of the total
scores that were greater than 5, it was highly significant
(Po0.001) among all three groups (last column in
Table 2).
Discussion
Findings of this preliminary study have indicated that
clusterin staining in stromal cells may be a significant
factor in predicting disease recurrence in prostate cancer
patients following radical prostatectomy irrespective of
whether or not the stromal cells are associated with the
cancerous tissues. As prediction of recurrence in prostate
cancer with extreme Gleason scores has been well
characterized, the present study has focused on cases
with intermediate Gleason scores (6 or 7). In earlier
studies, clusterin immunostaining has correlated with
Gleason scores in prostate cancer specimens.14 Since
clusterin is an antiapoptotic mediator,12 prostate cancer
cells expressing more clusterin would be endowed with
a greater ability for survival. This is consistent with
cancer cells with high Gleason scores with a greater
potential for survival. However, results of the present
study seem to suggest that the intensity of clusterin
staining in the cancer tissue is unable to provide a
prediction for disease recurrence among cases with
Gleason score 6 and 7.
The exact mechanism for disease recurrence in
prostate cancer patients following a successful radical
prostatectomy remains a medical mystery. Radical
prostatectomy is intended to cure, if the malignancy is
confined to the prostate. Theoretically, if postoperative
serum PSA remains undetectable, it is suggestive that the
disease has been cured. However, the high frequency of
disease recurrence from our clinical experience is
inconsistent with the possibility that there might be the
presence of occasional extraprostatic carcinoma cells at
the time of radical prostatectomy, especially in the face of
a negative finding from surgical margins and from pelvic
lymph nodes. The release of carcinoma cells during the
course of radical prostatectomy can be considered as one
of the possibilities for disease recurrence, unless the host
environment is hostile to the released cancer cells.
The present findings have provided some insights
regarding the role of stromal cells in determining the fate
of prostate cancer cells. It is inevitable that there are
viable prostatic cancer cells released during prostate
removal.23 Although the fate of these intraoperative PSApositive cells remains unclear, some of these cells can be
found later in the circulation.24,25 It is possible that these
released PSA-positive cells attach locally or in a distant
site and eventually develop into viable cancer site. Since
all circulating PSA-positive cells are subjected to the
same host environment, while the majority of released
cancer cells are destined to apoptosis, those endowed
with the ability to survive or those met with a favorable
survival environment will have a growth advantage and
eventually proliferate and develop into clinically significant malignancy. We hypothesize that clusterin
expression in the stromal cells may reflect the overall
ability of the host to produced clusterin into the
circulation may be able to protect these circulating
cancer cells from death and thereby increase the chance
of tumor recurrence after prostatectomy.
A recent report of a downregulation of clusterin
expression in transformed epithelial cells and upregulation in stromal cells in prostate cancer of intermediate
and high Gleason scores supported our observation.22 In
the present study, cases showing weak stromal staining
for clusterin were associated with a prolonged disease
free interval; while those with strong staining for
clusterin were associated with early disease recurrence.
A strong clusterin staining in stromal cells is related to
disease recurrence regardless whether they are associated with cancerous tissues or benign tissues. Since
these stromal cells have been removed from the patients
at the time of prostatectomy, it is not likely that they have
a direct impact on the survival of recurrent cancer cells. It
is likely that these stromal cells reflect the intrinsic
overall ability of the host to produce clusterin, which
provides a protection to released cancer cells. The strong
efficacy of i.p. administration of clusterin antisense
oligonucleotides in inhibiting tumor progression also
supports a systemic role of clusterin in malignancy.16–18
This possibility is plausible and remains to be tested.
In summary, results of this preliminary study support
the notion that high levels of clusterin produced by
Prostate Cancer and Prostatic Diseases
Clusterin and recurrent prostate cancer
MR Pins et al
248
prostatic stromal cells contribute to the biologically
aggressiveness of prostate cancer cells. These results
have indicated that clusterin immunohistochemical
staining in prostatic stromal cells could be used as a
predictor for disease recurrence in patients following
radical prostatectomy, for cases with intermediate Gleason score. The finding that the intensity of clusterin
staining in prostatic stromal cells can determine the
future fate of prostate cancer cells suggests the importance of noncancer cells in the host in disease recurrence.
It is important to point out that the cases targeted in the
present study were those with intermediate Gleason
scores and were otherwise difficult in providing prediction of disease recurrence. Owing to the fact that the
present study followed a stringent exclusion criterion,
we were only able to obtain a total of 28 cases that
satisfied the requirement at Northwestern University.
Further studies with a large number of cases are
warranted to verify this preliminary finding.
9
10
11
12
13
14
15
Acknowledgements
This study was supported in part by NIH Grants
CA80963 and CA90386. We thank for the constructive
comments of Dr Beatrice Knudson of the Fred Hutchinson Cancer Research Center.
16
17
References
1 Croawford ED, Kiker JD. Radical retropubic prostatectomy. J
Urol 1983; 129: 1145–1156.
2 Gleason DF. Classification of prostatic carcinomas. Cancer
Chemother Rep 1966; 50: 125–128.
3 Grayhack JT, Assimos DG. Prognostic significance of tumor
grade and stage in the patient with carcinoma of the prostate.
Prostate 1983; 4: 13–31.
4 Epstein JL, Partin AW, Sauvageot J, Walsh PC. Prediction of
progression following radical prostatectomy. A multivariate
analysis of 721 men with long-term follow-up. Am J Surg Pathol
1996; 20: 286–292.
5 Oefelein MG, Colangelo L, Grayhack J, McVary KT. Survival
after radical retropubic prostatectomy in men with clinically
localized high grade carcinoma of the prostate. Cancer 1995; 76:
2523.
6 Fritz IB, Burdzy BK, Setchell B, Blaschuk O. Ram rete testis fluid
contains a protein (clusterin) which influences cell–cell interactions in vitro. Biol Reprod 1983; 28: 1173–1188.
7 de Silva HV et al. Apolipoprotein J: structure and tissue
distribution. Biochemistry 1990; 29: 5380–5389.
8 Jenne DE, Tschopp J. Molecular structure and functional
characterization of a human complement cytolytic inhibitor
found in blood and seminal plasma: identity to sulfated
Prostate Cancer and Prostatic Diseases
18
19
20
21
22
23
24
25
glycoprotein-2, a constituent of rat testis fluid. Proc Natl Acad
Sci USA 1989; 86: 7123–7127.
Jenne DE, Tschopp J. Clusterin: the intriguing guises of a widely
expressed glycoprotein. Trends Biochem Sci 1992; 17: 154–159.
Leger JG, Montpetit ML, Tenniswood MP. Characterization and
cloning of androgen-repressed mRNAs from rat ventral prostate.
Biochem Biophys Res Commun 1987; 147: 196–203.
Sylveste SR, Morales C, Oko R, Griswold MD. Localization of
sulfated glycoprotein-2 (clusterin) on spermatozoa and in the
reproductive tract of the male rat. Biol Reprod 1991; 31: 1087–1101.
Sensibar JA et al. Prevention of cell death induced by tumor
necrosis factor alpha in LNCaP cells by overexpression of
sulfated glycoprotein-2 (clusterin). Cancer Res 1995; 55: 2431–
2437.
Sintich S et al. Cytotoxic sensitivity to tumor necrosis factor-a in
prostate cancer cells is regulated by extracellular levels of SGP-2
(clusterin). Prostate 1999; 39: 87–93.
Steinberg J et al. Intracellular levels of SGP-2 (Clusterin) correlate
with tumor grade in prostate cancer. Clin Cancer Res 1997; 3:
1707–1711.
Miyake H, Nelson C, Rennie PS, Gleave ME. Testosteronerepressed prostate message-2 is an antiapoptotic gene involved
in progression to androgen-independence in prostate cancer.
Cancer Res 2000; 60: 170–176.
Miyake H, Hara I, Kamidono S, Gleave ME. Synergistic
chemosensitivity and inhibition of tumor growth and metastasis
by the antisense oligodeoxynucleotide targeting clusterin gene
in a human bladder cancer model. Clin Cancer Res 2001; 7: 4245–
4252.
Zellweger T et al. Antitumor activity of antisense clusterin
oligonucleotides is improved in vitro and in vivo by incorporation of 20 -O-(2-methoxy)ethyl chemistry. J Pharmacol. Exp Ther
2001; 298: 934–940.
Zellweger T et al. Enhanced radiation sensitivity in prostate
cancer by inhibition of the cell survival protein clusterin. Clin
Cancer Res 2002; 8: 3276–3284.
Rosner B. Fundamentals of Biostatistics, 5th edn. Duxbury Press:
Pacific Grove, CA 2000.
Winer BJ, Brown DR, Michels KM. Statistical Principles in
Experimental Design, 3rd edn. McGraw-Hill: New York, NY
1991; pp 158, 358.
Sensibar JA et al. Prostatic ductal system in rats: regional
variation in localization of an androgen-repressed gene
product, sulfated glycoprotein-2. Endocrinology 1991; 128:
2091–2102.
Bettuzzi S et al. Clusterin gene expression is down-regulated in
transformed epithelial cells but up-regulated in fibroblasts from
prostate cancer. J Urol 2003; 169: 159 (Abstract 617).
Ignatoff JM et al. Prostate specific antigen reverse transcriptasepolymerase chain reaction assay in preoperative staging of
prostate cancer. J Urol 1997; 158: 1870–1874.
Oefelein MG et al. Molecular detection of prostate epithelial cells
from the surgical field and peripheral circulation during radical
prostatectomy. J Urol 1995; 155: 238–242.
Oefelein MG et al. Clinical follow-up after radical retropubic
prostatectomy. J Urol 1999; 162: 293–306.