ORIGINAL ARTICLE
Immunohistochemical Staining of Precursor Forms
of Prostate-specific Antigen (proPSA) in Metastatic
Prostate Cancer
Anil V. Parwani, MD, PhD,* Cameron Marlow, BS,w Angelo M. Demarzo, MD, PhD,w
Stephen D. Mikolajczyk, PhD,z Harry G. Rittenhouse, PhD,z Robert W. Veltri, MD,w
and Theresa Y. Chan, MDw
Abstract: Precursors of prostate-specific antigen (proPSA) have
been previously shown to be more concentrated in prostate
cancer tissue. This study characterizes the immunohistochemical
staining (IHS) of proPSA forms in metastatic prostate cancer
compared with prostate specific antigen (PSA) and prostatic
acid phosphatase (PAP). A tissue microarray, consisting of 74
cases of metastatic prostate carcinoma and control tissues, was
used. IHS, using monoclonal antibodies against proPSA with a
truncated proleader peptide containing 2 amino acids
([  2]pPSA), native ([  5/  7]pPSA), PSA, and PAP, was
analyzed. The monoclonal antibodies were specific for both
benign and malignant prostatic glandular tissue. IHS with [  5/
 7]pPSA showed the least number of cases with negative
staining (3%), and the most number of cases with moderate or
strong staining (76%). In the 60 cases where all 4 stains could be
evaluated, none of them were negative for proPSA and positive
for PSA or PAP, and all 7 cases that were negative for both PSA
and PAP showed IHS to proPSA. [  5/  7]pPSA (native
proPSA) may be a better marker than PSA and PAP in
characterizing metastatic prostate adenocarcinoma, with most
of the cases showing positivity for the marker. Even cases that
were negative for PSA and PAP, were reactive for proPSA. Such
enhanced detection is particularly important in poorly differ-
entiated carcinomas involving metastatic sites where prostate
carcinoma is a consideration. A panel of markers, including
proPSA, should be performed when metastatic prostate
carcinoma is in the differential diagnosis.
Key Words: proPSA, PSA, metastatic, prostate, carcinoma,
immunohistochemical, stain
(Am J Surg Pathol 2006;30:1231–1236)
From the *Univeristy of Pittsburgh Medical Center; wJohns Hopkins
Hospital, Baltimore, MD; and zHybritech Incorporated, a sub-
sidiary of Beckman Coulter, Inc, San Diego, CA.
Reprints: Anil V. Parwani, MD, PhD, Division of Pathology
Informatics and Genito-Urinary Pathology, UPMC Shadyside
Hospital, Department of Pathology, 5230 Centre Avenue, Room
WG 02.10, Pittsburgh, PA 15232 (e-mail: parwaniav@upmc.edu).
Copyright r 2006 by Lippincott Williams & Wilkins
Am J Surg Pathol  Volume 30, Number 10, October 2006
P rostate carcinoma is the most common carcinoma
affecting American men and is the second leading
cause of cancer-related deaths in the United States.11
Widespread use of prostate-specific antigen (PSA) as a
serum tumor marker has revolutionized the management
of patients with prostate cancer. PSA is a serine protease
produced by both prostate epithelial and cancer cells.
Multiple molecular forms of PSA exist including proPSA,
which is a precursor form of the enzyme and has been
shown to be a more cancer-specific marker. ProPSA was
previously shown to be more concentrated in prostate
cancer tissue extracts than in benign tissue.2,22,18,24
In a previous study, we analyzed immunohisto-
chemical staining (IHS) of formalin-fixed, paraffin-
embedded sections using monoclonal antibodies (mABs)
against proPSA with a truncated proleader peptide
containing 2 amino acids ([  2]pPSA), and against native
proPSA ([  5/  7]pPSA). We demonstrated that mABs
to proPSA are specific for benign and malignant prostatic
tissue, and proPSA remained uniform among the different
tumor grades.7
Previous studies recognized a strong inverse corre-
lation between PSA staining intensities and Gleason
score, and it is known that metastatic prostate carcinoma
may show negative staining.1,7,27 The objective of this
study is to compare the IHS patterns of proPSA with that
of PSA and prostatic acid phosphate (PAP) on a series of
metastatic prostate carcinoma cases to explore the utility
of proPSA in these lesions.
MATERIALS AND METHODS
Tissue Microarray Design
A tissue microarray design (TMA) of 74 cases of
metastatic prostate carcinoma [soft tissue/bone (n = 40);
lymph nodes (n = 34)] was created from specimens
retrieved from the surgical pathology files of The Johns
Hopkins Hospital. The specimens were from patients with
nonhormone refractory tumors, and 3 cores of each
specimen were used to construct the TMA. Details of
TMA construction is previously described.28 All the cases
included in the TMA were either metastatic to bone or
soft tissue (n = 40), or were metastatic to lymph nodes
1231
Parwani et al Am J Surg Pathol  Volume 30, Number 10, October 2006

TABLE 1. IHS Results for all Cases of Metastatic Prostate Carcinoma That Could be Evaluated
IHS PSA (n = 60) PAP (n = 60) [  2]pPSA (n = 69) [  5/  7]pPSA (n = 68)
Negative 11 (18%) 10 (17%) 8 (12%) 2 (3%)
Weak 17 (28%) 24 (40%) 23 (33%) 14 (21%)
Moderate 23 (38%) 24 (40%) 35 (51%) 37 (54%)
Strong 9 (16%) 2 (3%) 3 (4%) 15 (22%)

(n = 34). All the metastatic tumors were of higher grade, the 3 spots representing that specimen. In addition,
poorly differentiated carcinomas. Control tissues includ- 9 cases could not be evaluated for PSA and PAP stains
ing primary prostate adenocarcinoma with Gleason owing to a lack of tissue on those stains, and only the
scores ranging from 6 to 7 (n = 20), normal prostatic ProPSA mABs could be studied. All 4 mABs were
tissue (n = 20), and normal tissues from brain, skin, evaluated in a total of 60 cases.
colon, thyroid, kidney, bladder, lymph node, stomach,
smooth muscle, pancreas, testes, ovary, thymus, liver,
tonsils, spleen, and salivary gland were also in the TMA. RESULTS
The IHS results are shown in Table 1. [  5/
Immunohistochemistry  7]pPSA showed the most number of cases with
Immunohistochemistry (IHS) was performed on moderate or strong staining (76%), and the least number
deparaffinized TMA slides using a standard streptavidin of cases with negative staining (3%). [  2]pPSA staining
horseradish-peroxidase conjugate.4 A purified mouse was comparable with PSA and PAP (55% moderate to
immunoglobulin G mAB, PS2P446, which recognizes strong compared with 54% and 43%, respectively; and
proPSA with both the 7 aa and 5 aa leader peptides ([  5/ 12% negative compared with 18% and 17%, respec-
 7]pPSA, Beckman Coulter, San Diego, CA), was tively). There were minor differences between the
applied at a working concentration of 1:320 (6.31 mg/mL) 4 antibodies in extent of staining. [  5/  7]pPSA and
to the TMA. A purified mouse immunoglobulin G PSA showed the most number of cases with diffuse
mAB, PS2  373.3, which recognizes proPSA with a staining (75%, 69%), and the least number of cases with
2 aa leader peptide ([  2]pPSA, Beckman Coulter), was patchy staining (25% and 31%). [  2]pPSA and PAP
applied at a working concentration of 1:80 (4.64 mg/mL) showed diffuse staining in 60% and 56% of cases,
to the specimens. These proPSA mABs had less than respectively.
0.2% cross-reactivity to PSA or to each other. The In the 60 cases where all 4 stains could be evaluated,
Beckman Coulter proPSA mABs are intended for all cases were positive for either [  5/  7]pPSA or
research use only and not for diagnostic use. The ProPSA [  2]pPSA. [  5/  7]pPSA showed the most number of
antibodies are currently not available for commercial use positive cases (98%) compared with [  2]pPSA (80%),
and they may require dilution on site when they become PSA (81%), PAP (83%) (Table 2). The majority of cases
available. For PSA and PAP (DAKO, Carpinteria, CA), (63%) stained with all 4 mABs (Fig. 1). Of these 38 cases
the mABs were applied at a dilution of 1:500 and where all 4 stains were positive, strong or moderate
1:50,000, respectively. staining was seen in 82% [  5/  7]pPSA, 76% PSA, 39%
The intensity and extent of IHS were evaluated in PAP, and 29% [  2]pPSA (Fig. 2).
the metastatic prostate carcinomas, and accompanying Seven cases (12%) were negative for both PSA and
controls by human eye. The intensity of staining was PAP, but showed staining with [  5/  7]pPSA and/or
categorized as weak, moderate, or strong. For all the [  2]pPSA (Fig. 3). Neither PSA nor PAP was the only
cases and controls, the intensity of the staining was positive marker in any of the cases; however, [  5/
evaluated and graded visually. A numerical score was  7]pPSA or [  2]pPSA was the only positive marker in 5
assigned to each spot on the TMA with a score of 0 for
negative, 1 for weak, 2 for moderate, and 3 for strong
staining. Using this score, a mean score was calculated for TABLE 2. Immunohistochemistry Profiles of the 60 Cases
each sample present in triplicate. In evaluating cases Where all 4 Stains Were Evaluated
where no tissue was present in one of the 3 spots, a mean PSA PAP [  5/  7]pPSA [  2]pPSA Total (%)
score was calculated from the intensity scores of the 2 + + + + 38 (63%)
spots. The mean score was then used to assign the final + + +  8 (13%)
intensity pattern ranging from negative to weak, using the  + + + 4 (7%)
  + + 3 (5%)
following scheme: (0 to 0.5 = negative; 0.5 to 1.5 = weak,   +  3 (5%)
1.5 to 2.5 = moderate, 2.5 or greater = strong). The +  + + 2 (3%)
extent of staining was categorized as diffuse (more than    + 1 (2%)
50% of cells) or patchy (less than 50% of cells). +  +  1 (2%)
+ +   0 (0)
Five cases could not be evaluated for any of the     0 (0)
mABs because of the insufficient tissue on more than 2 of

1232 r 2006 Lippincott Williams & Wilkins

Am J Surg Pathol  Volume 30, Number 10, October 2006 Immunohistochemical Staining of proPSA

FIGURE 1. Metastatic adenocarcinoma involving a pelvic lymph node. (H&E stain, A). PSA staining was strong (B), and moderate
staining was seen with PAP (C), [  2]pPSA (D), and [  5/  7]pPSA (E).

and 2% of cases, respectively (Table 2). In these 3 cases The proPSA, PSA, and PAP mABs were specific for
where proPSA was the only positive marker, the intensity both benign and malignant prostatic glandular tissue. No
of staining of [  5/  7]pPSA was moderate in 2 and weak differences were seen in the staining pattern of benign
in 1, and the 1 case of [  2]pPSA showed weak staining. glands in the control tissue when stained with any of the

FIGURE 2. Metastatic carcinoma involving the brain. (H&E stains, A [40  ] and D [200  ]). [  5/  7]pPSA showing strong
staining (C [40  ], and F[200  ]), whereas PSA staining was weak (B [40  ] and E [200  ]).

r 2006 Lippincott Williams & Wilkins 1233

Parwani et al
FIGURE 3. Metastatic adenocarcinoma extensively involving bone and soft tissue of the spine (A, H&E stain). Note the strong
staining with [  5/  7]pPSA (D), whereas PSA (B) and PAP (C) are negative.
mABs. There was no staining in prostatic stromal and
vascular tissue. It must be noted that only prostatic
tumors were evaluated in this study. No other tumor
types were included in this study. Multiple nonprostatic
control tissues including skin, colon, testis, bladder,
ovary, brain, thyroid, and lymph nodes were included in
the TMA. There was only focal weak cytoplasmic
staining (diffuse, ‘‘blush’’ type of staining) with both the
proPSA mABs to control thyroid and kidney tissues, and
no staining with the other control tissues.
DISCUSSION
Our group was the first to demonstrate that mABs
to proPSA can be used as specific IHS for benign and
malignant prostatic tissue.7 We have previously demon-
strated the staining characteristics of mABs to proPSA in
high-grade prostatic intraepithelial neoplasia, benign
tissue, and adenocarcinoma.7 In the present study, we
evaluated the immunostaining characteristics of mABs to
proPSA for their use as diagnostic markers in the
detection of metastatic prostatic adenocarcinoma by
comparing them with the currently used immunohisto-
chemical markers, PSA, and PAP. In our previous
1234
Am J Surg Pathol  Volume 30, Number 10, October 2006
studies, [  2] proPSA appeared to be preferentially more
concentrated in cancer tissue than in benign glands. We
also showed that using immunohistochemistry, proPSA
remained uniform among the different tumor grades.7
In the current study, no difference was seen in the
staining pattern of benign glands when all 4 mABs were
compared. However, when the staining intensity of PSA
and PAP was compared with proPSA [  5/  7], there
was a marked difference. proPSA [  5/  7] showed the
most number of cases with moderate or strong staining
(76%) as compared with PSA (56%), PAP (43%) and
proPSA [  2] (55%). This is an important observation
and confirms our earlier studies in which we had shown
that PSA shows a decrease in expression from benign
epithelium to high-grade prostatic intraepithelial neopla-
sia to carcinoma.7 The current study also confirms our
previous observations that proPSA has stronger staining
in cancer glands compared with benign glands.7
In the control specimens from the current study,
there was no difference in staining intensities using the
proPSA mABs amongst cases representing different
Gleason scores. Previous studies have shown that PSA
has a strong inverse correlation between staining intensity
and Gleason score with decreased staining in higher
r 2006 Lippincott Williams & Wilkins
Am J Surg Pathol  Volume 30, Number 10, October 2006
Gleason scores.1,8,27 We have also confirmed this observa-
tion in our previous study and in the evaluation of controls
from the current study. proPSA mAB staining intensity in
contrast, does not show this inverse relationship and
retains similar staining intensity across different Gleason
scores.7 This becomes particularly important when a
pathologist is evaluating a case of poorly differentiated
carcinoma from an unknown primary and uses PSA or
PAP as a diagnostic marker to exclude prostate as a
source. In that situation, because of the staining variability
of PSA and PAP, we cannot confidently eliminate the
possibility of prostate carcinoma.
The proPSA antibodies described in this report are
currently not available for commercial use and they may
require dilution on site when they become available. Once
these antibodies become freely available, the data from
our studies suggest that they will serve as useful
diagnostic markers. On the basis of our current studies
on tissue specimens and the metastatic prostatic carcino-
ma cases, it seems that  5/  7 proPSA (native proPSA)
may be a better marker than PSA and PAP in
characterizing metastatic prostate adenocarcinoma, with
only 2 cases (3%) being negative for the marker. In
addition, 10% of cases were reactive for proPSA, whereas
PSA and PAP were negative.
PSA is a serine protease and a member of the tissue
kallikrein family. It is a major protein in seminal fluid,
where its function is to cleave semenogelins in the seminal
coagulum.2 PSA is produced by both prostate epithelial
cells and prostate cancer and is secreted into prostatic
ducts as an inactive 244-amino acid proenzyme (proPSA)
which is activated by cleavage of 7 N-terminal amino
acids.9,10
Upon its entry into the circulation, intact PSA
becomes bound by protease inhibitors such as a-1-
antichymotrypsin. A small fraction of the PSA is
inactivated in the lumen by proteolysis and circulates in
the serum as free PSA. The intact precursor form [  7]
proPSA and a slightly truncated form [  5] proPSA are
recognized by a mAB [  5/  7] proPSA.2 There
are additional more truncated forms of proPSA such as
[  2] and [  4]-proPSA which differ from the [  5/  7]
proPSA in their biochemical properties which have been
previously described.13–24
Serum total PSA levels are increased in prostate
carcinoma and other benign conditions. Measurement of
free versus total PSA can increase specificity for prostate
carcinoma and multiple immunoassays and molecular-
based methods are being developed for the diagnosis and
management of prostate carcinoma. Studies have indi-
cated that there may be additional roles for PSA in the
pathogenesis of prostate carcinoma. Investigators have
also identified a specific molecular form of clipped-free
PSA which is called BPSA.26 The latter form is increased
within the prostatic TZ of patients with nodular Benign
prostatic hyperplasia. The current hypothesis is that a
proportion of the serum percent free PSA found in
patients with BPH may be composed of BPSA released
into the serum.19
r 2006 Lippincott Williams & Wilkins
Immunohistochemical Staining of proPSA
Studies have indicated that proPSA significantly
increases the specificity for prostate carcinoma, particu-
larly in the 2 to 4 ng/mL PSA range.5,13,14 A study by the
Catalona et al6 indicated that percent proPSA was not
only better than percent free and calculated complexed
PSA for detecting prostate carcinoma in the PSA range of
2 to 10 ng/mL but also it had selectivity for detecting
more aggressive cancers, such as those with Gleason score
7 or greater and/or extracapsular tumor extension.6
Prostate cancer is the most common type of cancer
found in American men, other than skin cancer.9,11 Even
though significant advances in the treatment of prostatic
carcinoma have resulted from the widespread use of PSA
as a serum marker for cancer, PSA as a serum marker has
limitations.10 In the diagnostic range of 4 to 10 ng/mL
PSA has limited specificity for distinguishing early
prostatic adenocarcinoma from benign prostatic hyper-
plasia.9 There is significant controversy in defining
the optimal upper limit of normal for serum PSA for
prostate carcinoma screening. There is a dire need for
new biomarkers of disease which are better predictors
of disease prognosis.10 Newer and more robust
tumor markers such as proPSA are being evaluated
with variable results, for use in screening and
diagnosis.3,12,13,23,25
In summary, enhanced and improved detection of
prostatic carcinoma is particulary important in poorly
differentiated carcinomas involving metastatic sites in
which prostate carcinoma is a consideration and there is
poor or no immunostaining using PSA or PAP. On the
basis of our study results, consideration should be given
to include proPSA in a panel of markers where the
differential diagnosis includes metastatic prostate carci-
noma. Additional studies are underway to further extend
the use of proPSA as an immunohistochemical marker in
larger number of specimens from variety of sources
including cytology specimens in where there is only
limited amount of diagnostic material available to the
pathologist for diagnosis.
ACKNOWLEDGMENTS
The radical prostatectomy tissue microarrays were
produced by Helen Fedor from the surgical pathology files
of The Johns Hopkins Hospital through the Pathology
Core (P. I., Dr Angelo DeMarzo) of The Johns Hopkins
University Prostate Cancer Specialized Programs of
Research Excellence (SPORE) Grant. Please note that
the tPSA (PSM 773) and pPSA (PS2P446.4) monoclonal
antibodies were research-use-only antibodies and kindly
provided as a gift by Beckman Counter, Inc, San Diego, CA.
REFERENCES
1. Aihara M, Lebovitz RM, Wheeler TM, et al. Prostate specific
antigen and gleason grade: an immunohistochemical study of
prostate cancer. J Urol. 1994;151:1558–1564.
2. Balk SP, Ko YJ, Bubley GJ. Biology of prostate-specific antigen.
J Clin Oncol. 2003;21:383–391.
3. Bangma CH, Wildhagen MF, Yurdakul G, et al. The value of
(  7,  5)pro-prostate-specific antigen and human kallikrein-2 as
serum markers for grading prostate cancer. BJU Int. 2004;93:720–724.
1235
Parwani et al
4. Cartun RW, Pedersen CA. An immunocytochemical technique
offering increased sensitivity and lowered cost with a streptavidin-
horseradish peroxidase conjugate. J Histotechnol. 1989;12:
273–277.
5. Catalona WJ, Bartsch G, Rittenhouse HG, et al. Serum pro prostate
specific antigen improves cancer detection compared to free and
complexed prostate specific antigen in men with prostate specific
antigen 2 to 4 ng/ml. J Urol. 2003;170:2181–2185.
6. Catalona WJ, Bartsch G, Rittenhouse HG, et al. Serum pro-prostate
specific antigen preferentially detects aggressive prostate cancers in
men with 2 to 4 ng/ml prostate specific antigen. J Urol. 2004;
171:2239–2244.
7. Chan TY, Mikolajczyk SD, Lecksell K, et al. Immunohistochemical
staining of prostate cancer with monoclonal antibodies to the
precursor of prostate-specific antigen. Urology. 2003;62:177–181.
8. Goldstein NS. Immunophenotypic characterization of 225 prostate
adenocarcinomas with intermediate or high Gleason scores. Am J
Clin Pathol. 2002;117:471–477.
9. Hernandez J, Thompson IM. Diagnosis and treatment of prostate
cancer. Med Clin North Am. 2004;88:267–279, ix.
10. Hernandez J, Thompson IM. Prostate-specific antigen: a review of
the validation of the most commonly used cancer biomarker.
Cancer. 2004;101:894–904.
11. Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004. CA.
Cancer J Clin. 2004;54:8–29.
12. Khan MA, Partin AW, Rittenhouse HG, et al. Evaluation of
proprostate specific antigen for early detection of prostate cancer in
men with a total prostate specific antigen range of 4.0 to 10.0 ng/ml.
J Urol. 2003;170:723–726.
13. Mikolajczyk SD, Rittenhouse HG. Pro PSA: a more cancer specific
form of prostate specific antigen for the early detection of prostate
cancer. Keio J Med. 2003;52:86–91.
14. Mikolajczyk SD, Rittenhouse HG. Tumor-associated forms of
prostate specific antigen improve the discrimination of prostate
cancer from benign disease. Rinsho Byori. 2004;52:223–230.
15. Mikolajczyk SD, Grauer LS, Millar LS, et al. A precursor form of
PSA (pPSA) is a component of the free PSA in prostate cancer
serum. Urology. 1997;50:710–714.
16. Mikolajczyk SD, Millar LS, Marker KM, et al. Identification of a
novel complex between human kallikrein 2 and protease inhibitor-6
in prostate cancer tissue. Cancer Res. 1999;59:3927–3930.
1236
Am J Surg Pathol  Volume 30, Number 10, October 2006
17. Mikolajczyk SD, Millar LS, Marker KM, et al. Seminal plasma
contains ‘‘BPSA,’’ a molecular form of prostate-specific antigen that
is associated with benign prostatic hyperplasia. Prostate. 2000;
45:271–276.
18. Mikolajczyk SD, Millar LS, Wang TJ, et al. A precursor form of
prostate-specific antigen is more highly elevated in prostate cancer
compared with benign transition zone prostate tissue. Cancer Res.
2000;60:756–759.
19. Mikolajczyk SD, Millar LS, Wang TJ, et al. ‘‘BPSA,’’ a specific
molecular form of free prostate-specific antigen, is found predomi-
nantly in the transition zone of patients with nodular benign
prostatic hyperplasia. Urology. 2000;55:41–45.
20. Mikolajczyk SD, Marker KM, Millar LS, et al. A truncated
precursor form of prostate-specific antigen is a more specific serum
marker of prostate cancer. Cancer Res. 2001;61:6958–6963.
21. Mikolajczyk SD, Marks LS, Partin AW. Free prostate-specific
antigen in serum is becoming more complex. Urology. 2002;
59:797–802.
22. Mikolajczyk SD, Catalona WJ, Evans CL, et al. Proenzyme forms
of prostate-specific antigen in serum improve the detection of
prostate cancer. Clin Chem. 2004;50:1017–1025.
23. Mikolajczyk SD, Song Y, Wong JR, et al. Are multiple markers the
future of prostate cancer diagnostics? Clin Biochem. 2004;37:
519–528.
24. Rittenhouse HG, Finlay JA, Mikolajczyk SD, et al. Human
Kallikrein 2 (hK2) and prostate-specific antigen (PSA): two closely
related, but distinct, kallikreins in the prostate. Crit Rev Clin Lab
Sci. 1998;35:275–368.
25. Sokoll LJ, Chan DW, Mikolajczyk SD, et al. Proenzyme PSA for
the early detection of prostate cancer in the 2.5–4.0 ng/ml total PSA
range: preliminary analysis. Urology. 2003;61:274–276.
26. Wang TJ, Slawin KM, Rittenhouse HG, et al. Benign prostatic
hyperplasia-associated prostate-specific antigen (BPSA) shows
unique immunoreactivity with anti-PSA monoclonal antibodies.
Eur J Biochem. 2000;267:4040–4045.
27. Weir EG, Partin AW, Epstein JI. Correlation of serum prostate
specific antigen and quantitative immunohistochemistry. J Urol.
2000;163:1739–1742.
28. Zha S, Gage WR, Sauvageot J, et al. Cyclooxygenase-2 is
up-regulated in proliferative inflammatory atrophy of the prostate,
but not in prostate carcinoma. Cancer Res. 2001;61:8617–8623.
r 2006 Lippincott Williams & Wilkins