[CANCER RESEARCH 48, 4543-4548, August 15. 1988]
Monoclonal Antibody PR92 with Restricted Specificity for Tumor-associated
Antigen of Prostate and Breast Carcinoma
Yung D. Kim,1 Deborah Y. Robinson, and Joseph T. Tornita
Diagnostics Division, Dept. 90C, Abbott Laboratories, North Chicago, Illinois 60064
ABSTRACT
A unique mouse hybridoma, PR92, was obtained using human prostate
adenocarcinoma cell line 1)11-15. The monoclonal antibody produced by
the PR92 clone was reactive with DU145, MCF-7 (breast adenocarci
noma), and CHACO (lung carcinoma), but not with normal cell lines and
16 other cell lines of cancerous origin. A homologous solid-phase sand
wich radioimmunoassay (PR92-RIA) was developed utilizing PR92
monoclonal antibody. The PR92-RIA recognized unique antigen present
in DU 145 cell extract but did not detect 16 other known tumor-associated
markers. In preliminary studies, the PR92-RIA measured greater than 5
units/ml level of PR92 monoclonal antibody-binding antigen in 23 of 31
(74%) serum specimens of active prostate cancer and 27 of 31 (87%)
active breast carcinoma patients. Only 1 of 79 (1%) sex-matched normal
donors and 1 of 57 (2%) benign disease control patients showed the
serum antigen level greater than 5 units/ml. A high degree of correlation
was observed between the PR92 antigen activity and the clinical status
of four prostate and four breast cancer patients during therapeutic treat
ment. Thus, the PR92-RIA detects new tumor-associated antigen which
may be useful in detection and monitoring of prostate and breast carci
noma patients.
INTRODUCTION
Hybridoma technologies have provided means of identifying
many new tumor-associated antigens. These antigens were char
acterized by their immunoreactivity toward specific monoclonal
antibodies and were proposed for use as diagnostic and prog
nostic aids in the clinical management of tumor disease states.
For example, the hybridoma clone developed with human colorectal carcinoma cell line SW1116 and epithelial ovarian
carcinoma cell line OVCA433 led to the discoveries of tumorassociated antigens CA 19-9 (1) and CA 125 (2), respectively.
Prostate carcinoma ranks third in cancer incidence in men in
the United States, following skin cancer and lung carcinoma
(3). PAP2 and PA have been widely used as prostate tumorassociated markers and found to be elevated in sera of prostatic
carcinoma patients, especially those in an advanced disease
state (4-6). However, neither marker is ideal as a serological
marker owing to lack of adequate sensitivity or specificity (5,
7).
Previous investigators prepared monoclonal antibodies
against human prostate cancer cell line DU 145 or PC3 (8-12).
Hybridoma clones were commonly selected based on the obser
vation that MAbs recognize the immunized tumor cells and
cross-react less with normal cells or other benign or malignant
cell lines of different origin. The selected MAbs generally did
not recognize PAP nor PA and exhibited unique staining patReceived 12/22/87; revised 4/25/88; accepted 5/18/88.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1To whom requests for reprints should be addressed.
2 The abbreviations used are: PAP, prostatic acid phosphatase; PA, prostate
antigen; CEA, carcinoembryonic antigen; AFP, o-fetoprotein; PBS, 0.01 M
sodium phosphate-O.I5 M sodium chloride (pH 7.4) solution; OPD solution,
o-phenylenediamine •
2 HC1; MAb, monoclonal antibody; HEK, human embryonic
kidney cell; HRPO, horseradish peroxidase; BSA, bovine serum albumin; RIA,
radioimmunoassay; NED, no evidence of disease.
terns with pathological prostatic tissues in immunohistochemical studies.
We describe in this paper development of a new hybridoma
cell line PR92 which produced MAb with restricted specificity
in recognizing tumor cell lines and, in addition, a high degree
of specificity and sensitivity in a serological radioimmunoassay,
in identifying prostate and breast carcinomas.
MATERIALS AND METHODS
Cell Lines. The following cell lines were obtained from American
Type Culture Collection (Rockville, MD) and grown in the media as
recommended by the supplier: DU 145; PC3; T24; J82; HT 1197;
HT1376; RT4; MIA; DLD-1; HT29; MCF-7; MDA-MB; BT-20;
C33A; SKUT-1B; and ACHN. DU 145 cell line was isolated originally
by Stone et al. (13) from a lesion in the brain of a patient with metastatic
carcinoma of the prostate and a 3-yr history of lymphocytic leukemia.
In addition, the cell line 647V from Dr. B. Liu (University of California,
Los Angeles, CA) and CHAGO from Dr. J. Braatz (National Cancer
Institute, Frederick, MD) were used in this study. Normal human cells
of blood type ABH and lymphocytes were prepared from the blood of
healthy normal donors. Primary human embryonic kidney cells, cul
tured and supplied by Dr. Tribby (Abbott Labs, North Chicago, IL),
were also used.
Serum Specimens. Clinical serum specimens were provided by Dr.
H. Pritsche at M. D. Anderson Hospital (Houston, TX), Dr. A. Malkin
at Sunnybrook Medical Center (Toronto, Canada), and Dr. R. Vessella
at the V. A. Hospital (Minneapolis, MN). The patient history and the
clinical data at the time of sampling were provided by the attending
physicians at the institutions. Serum specimens of normal healthy
donors were obtained from laboratory workers at Abbott Labs and also
at South Chicago Community Hospital (Chicago, IL). Informed con
sent was obtained from each specimen donor at the respective institu
tions. All specimens were aliquoted and stored at —
20°Cuntil tested.
Production of Hybridomas. Female BALB/c mice were given injec
tions of DU 145 cells following the procedure described by Starling et
al. (10). The spleens of the mice were removed for subsequent hybridi
zation. The serum samples from each mouse were assayed for antibodies
to DU 145 cell using a microtiter plate that had been previously coated
with DU 145 cell fragments (see the following section). The mouse
spleen cells were fused with cells from mouse myeloma cell line SP2/
0-Agl4 according to the procedure of De St. Groth and Scheidegger
(14). Hybridoma clones which produced antibodies exhibiting strong
specificity to DU 145 cell line were selected for further propagation to
obtain pure clones for amplification. Following the subcloning process,
supernatant solutions were screened against DU 145 cell fragments,
HEK extract, purified CEA, and PAP (Abbott Labs.). Sixteen hybridomas showing strong binding with DU 145 cell line but essentially no
binding with HEK extract, CEA, and PAP were selected. Pristaneprimed female BALB/c mice were given injections of 45 x 10' hybrid
oma cells. The ascitic fluid suspensions were passed in vivo every 7 to
10 days. Antibodies produced by each of the 16 clones were further
tested for binding specificity using the plates that had been coated with
cell fragments of normal cells or various tumor cell lines.
Preparation of Cell-coated Plate. Viable cells of 10 to 20 x IO6cells/
ml in Hanks' balanced salt solution (Gibco, Grand Island, NY) con
taining 2 mM phenylmethylsulfonyl fluoride and 20 ^il/ml of trasylol
were sonicated for 2'/2 min to disrupt the cells into fragments. The
fragments, equivalent to 5 x IO4 cells/70 p\, were added to each well
of a microtiter plate (Dynatech, Alexandria, VA), and the plate was
centrifuged for 3 min at 1400 x g on a Beckman J6-B centrifuge
4543
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1988 American Association for Cancer Research.
PROSTATE AND BREAST TUMOR-ASSOCIATED ANTIGENS
(Beckman Instruments, Palo Alto, CA) at 20°C.Glutaraldehyde solu
RESULTS
tion (Sigma Chemicals, St. Louis, MO), 0.05% in PBS, was added, and
the cell fragments were allowed to fix to the well for 15 min at room
temperature. The wells were then rinsed 3 times with PBS and filled
with PBS-0.05% gelatin-0.0005% thimerasol.
Test of MAb Using the Cell-coated Plate. The cell-coated plate was
overcoated, prior to use, with 25% calf serum (Gibco) in PBS-thimerasol for 15 min at room temperature. A sample (serum, ascitic fluid, or
purified MAb), prediluted in 10% calf serum-PBS-thimerasol, was
incubated for 2 h at 37°C,and the well was rinsed 4 times with PBS.
HRPO-conjugated IgG fraction of goat anti-mouse immunoglobulins
(Cappel Labs., Malvern, PA) was added and incubated for 45 min at
37°C.After washing 4 times with PBS solution, substrate solution
(orthophenylene diamine; Abbott Labs.) was added. Subsequently, the
reaction was quenched by the addition of l N HC1 solution, and the
resulting color formation was measured by EIA Autoreader (Bio-Tech
Instrument, Burlington, VT).
Isolation and Isotyping of Monoclonal Antibodies. Mouse ascitic fluid
containing monoclonal antibodies was fractionated by precipitation in
50% saturated ammonium sulfate solution. The precipitate was dis
solved with PBS and dialyzed against the same buffer overnight at 4°C.
This ^-fraction was used for most of the study. Alternatively, the
monoclonal antibodies were isolated by a Protein A-coupled Sepharose
4B (Pharmacia, Piscataway, NJ) column. The bound immunoglobulins
were eluted by applying a linear pH gradient with a mixture of 0.2 M
sodium phosphate and 0.1 M citrate between pH 8.0 and 3.0. Three
elution peaks emerged corresponding to pH values of 6.0, 4.6, and 3.5,
respectively.
Purified MAb was incubated in the DU 145 cell-coated well. After
rinsing, isotyping reagents (Cappel Labs) in the form of goat antimouse immunoglobulins were added to the well. Finally, HRPO-rabbit
anti-goat IgG (Cappel Labs.) was added as a probe antibody. Reaction
with substrate (OPD) solution identified the isotype of both the heavy
and light chain of the MAb.
Solid-Phase Sandwich Radioimmunoassay. A microtitration plate
(Removawell strips; Dynatech) was coated by allowing 40 Mg/ml of a
7-fraction of ascitic fluid or 10 Mg/ml of purified MAb in 0.01 M Tris,
pH 8.0, to contact the well surface for l h or longer. The entire
procedure was carried out at room temperature. The wells were then
overcoated with 1% BSA in 0.01 M Tris, pH 8.0, for l h. A series of
standard reagent or prediluted specimen in 1% BSA-PBS-0.1% Tween
20 (specimen buffer) was added to each well and incubated for 3 h.
Each sample was run in duplicate. After rinsing the wells twice with
PBS-Tween buffer, '"I-labeled MAb was added to each well, and the
reaction proceeded for 2 h. The probe MAb had been '"I-labeled
according to chloramine-T method (15). Following the second incuba
tion, the wells were rinsed 3 times with PBS-Tween buffer, and each
well was separated by breaking the bridge connecting the Removawells.
Each well was dropped into a disposable culture tube, and the yradiation activity was measured on a gamma counter (1.KM, Bromma,
Sweden).
To assess immunological specificity of PR92-RIA, some of the
recognized tumor-associated markers were tested for a dose-response
phenomenon. The standard reagents included in the commercially
available immunoassay kits were used as a source of PAP, CEA, AFP,
and ferritin (Abbott Labs.); PA (Yang Labs., Bellevue, WA); and CA
19-9, CA 125, and CA 15-3 (Centocor, Malvern, PA), respectively.
Human actin, fibrinogen, and ft-microglobulin were obtained from
Calbiochem (La Jolla, CA). Human blood group substances and human
colonie mucin extract were gifts of Dr. G. Hass and J. Slota (Abbott
Labs.).
Molecular Weight Determination. Molecular weight of PR92 MAbbinding protein was estimated by high-performance liquid chromatography (HPLC unit: Perkin-Elmer, Norwalk, CT) with a Bio-Sil TSK
250 gel filtration column (Bio-Rad, Richmond, CA), using Bio-Rad gel
filtration standard (Bio-Rad) as the molecular weight marker. The
column eluent was monitored by PR92-RIA to measure the PR92 MAb
binding-protein level in each fraction.
Selection of Hybridomas. Among 356 hybridomas generated
by the initial fusion experiment, 16 hybridomas were chosen
based on reactivity of their antibodies with DU 145 cell and
nonreactivity with CEA, PAP, AFP, and HEK cell extract.
Following subcloning, the MAbs of the 16 hybridomas were
tested for their specificity in reaction with 24 different cell lines.
Four clones, H29C138, H65C137, H92C149, and H92C152,
produced antibodies that did not react with normal cells but
displayed a restrictive specificity to tumor cell lines. The cellbinding profile and isotypes of the four monoclonal antibodies
are summarized in Table 1. Using a combination of the four
monoclonal antibodies, as immobilized catcher MAb and/or
125I-labeled probe MAb, we developed solid-phase sandwich
radioimmunoassays. To test each RIA system, DU 145 cells
were treated according to the procedure of Masuko et al. (16),
and the soluble extract was used as a source of antigens and
also as a positive control sample. A serum panel was then
assembled and used repeatedly to evaluate the clinical specific-ity
of each RIA system. The serum specimens had been prepared
for this particular purpose by pooling individual serum speci
mens representing normal, benign disease, and malignant disTable 1 Binding of four monoclonal antibodies to human cell lines
A mouse ascitic fluid containing monoclonal antibodies was tested using a
microtiter plate coated with respective cell fragments. The results were developed
using enzyme-conjugated goat anti-mouse immunoglobulin and substrate solu
tion. The color formation, measured by an EIA reader, was judged positive if the
absorbance at 490 nni was greater than 2 times that of a negative control (normal
mouse ascitic fluid).
Hybridoma clone
H92C149
Cell line
7lA°
H92C152
H29C138
H65C137
W*
Prostate cancer
DU145
PCS
Bladder cancer
T24
J82
HT1197
HT1376
647
RT4
Pancreatic cancer
MIA
Lung cancer
A549
CHACO
Colon cancer
DLD-1
HT29
Breast cancer
MCF7
MDA-MB-175 VIII
BT20
Cervical cancer
C33A
Renal cancer
ACHN
Uterine cancer
SKUT-IB
Normal cell
ABH
Lymphocytes
Human fetal kidney
' Heavy-chain and light-chain isotypes, respectively.
4544
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1988 American Association for Cancer Research.
PROSTATE AND BREAST TUMOR-ASSOCIATED ANTIGENS
ease patient populations. This serum panel was used to compare
the different RIA systems in relation to each other. The overall
results (not shown) indicated that the M Ab of H92C149 in
combination with itself (homologous) has superior sensitivity
and specificity to other combinations of MAb in RIA systems
in identifying the serum specimens from prostate and breast
cancer patients. Thus, we designated this monoclonal antibody,
PR92 MAb, and investigated its molecular characteristics and
explored potential application in developing a clinical test.
Production and Purification of PR92 MAb. In vitro culture of
H92C149 clone to produce PR92 MAb has not been successful.
Therefore, H92C149 hybridoma cells were passed in vivo using
primed mice, and PR92 MAb was isolated from mouse ascitic
fluid. In addition to PR92 MAb of the IgGl subtype, a small
but significant amount of nonspecific IgG2a and IgG3 was
present in the ascitic fluids. Thus, PR92 MAb was isolated
from the other nonspecific immunoglobulins, in Protein A
column procedure, by elution at pH 5.5.
Evaluation of PR92-RIA. A typical standard curve of PR92RIA is shown in Fig. 1. One batch of soluble DU145 cell extract
was assigned as the primary standard reagent containing an
arbitrary 100 units/ml of the PR92 MAb-binding antigen, as
assayed by the PR92-RIA. To determine the immunological
specificity, the PR92-RIA was tested with a number of recog
nized tumor-associated antigens, including PAP, PA, CEA, and
AFP. The results presented in Table 2 indicate that the PR92RIA specifically detects a unique antigen present in the DU 145
cell extract.
Serum specimens from the patients diagnosed to have cancer
of the kidney, ovary, bladder, liver, thyroid, colorectum, esoph
agus, stomach, brain, head, neck and cervix, as well as the sera
from patients with melanoma and leukemia, were subjected to
14
Table 2 Detection of tumor-associated markers using PR92 MAb
radioimmunoassay
Each tumor-associated marker was tested for a dose-response phenomenon by
the PR92-RIA. With the exception of DU 145 cell extract, all other markers
exhibited negative dose-response.
Marker
DU145cellextract
Prostatic acid phosphatase
Prostate specific antigen
Carcinoembryonic antigen
a-Fetoprotein
Ferritin
Human blood group substance
A. B. H, T
CA 19-9
CA 125
CA 15-3
Human actin
Fibrinogen
Human colonie mucin
02-Microglobulin
Detection
Highest
concentration
tested
Not applicable
30 ng/ml
50 ng/ml
60 ng/ml
40 ng/ml
400 ng/ml
Erythrocyte extract
120 units/ml
500 units/ml
200 units/ml
1 ng/ml
1 ng/ml
2 ng/ml
the PR92-RIA testing. The clinical status of a cancer patient
was determined by the attending physician as progressive (ad
vance of disease), remission (diminution of disease symptoms),
response (improvement resulting from therapy), stable (no
change in status), or NED. For the purpose of analysis, the
serum specimens of cancer patients were grouped, in this study,
as progressive status or nonprogressive (remission, stable,
quiescent, NED) status. The results are shown in Fig. 2.
Serum specimens from 58 normal males over 50 yr old, 29
specimens from patients with benign prostate disease, and 30
from prostate cancer patients at clinically progressive status
were assayed by the PR92-RIA. The results are shown in Fig.
3. One hundred ten serum specimens, obtained from normal
female subjects and patients with benign or malignant breast
disease, were also assayed by the PR92-RIA, and the results
are shown in Fig. 4. Specimens listed in Figs. 2 to 4, with values
greater than 50, 11, and 30 units/ml, respectively, could not be
reassayed due to limited specimen volume.
Serum specimens of four prostate cancer patients at Stage D
were collected over a period ranging from 108 to 719 days,
stored at -20°C, and analyzed by the PR92-RIA. The results
were correlated to clinical evaluations of the disease state of the
patients at the time of sampling. The resulting data are set forth
in Fig. 5. In a similar manner, serum specimens from four
patients with breast cancer at Stage IV, taken at clinically
relevant times, were assayed by the PR92-RIA, and the results
are presented in Fig. 6.
The molecular weight distribution of the PR92 MAb-binding
protein (PR92 antigen) was investigated by subjecting DU 145
cell extract to high-performance liquid chromatography. Analy
sis of the column eluent by PR92-RIA revealed that PR92
antigen eluted as a single peak of which the molecular weight
corresponded to about 470,000.
12
I
G
DISCUSSION
PR92 Antigen (Unit/Ml)
Fig. 1. Standard curve for PR92-RIA. Points, arilhmatic mean of duplicate
determinations with the average intraassay coefficient of variation of 2%. (See
"Materials and Methods" for details.)
Many investigators have generated hybridomas using estab
lished tumor cell lines (8, 9, 10, 12, 17, 18) and tested the
antibodies produced by the hybridomas against a variety of
normal, benign, or malignant cells. The hybridomas whose
MAb displayed specific affinity to certain types of cancer cells
were selected for detailed investigation, such as immunohistochemistry. The objective of the present study was to search for
hybridomas, producing monoclonal antibodies that can specif
ically recognize prostate tumor-associated antigen(s) that may
4545
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1988 American Association for Cancer Research.
PROSTATE AND BREAST TUMOR-ASSOCIATED
ANTIGENS
50
Ã
; ®â€¢-odo
-.••Colon*Da
,A
40JfcooBrain,
50
40
Fig. 2. Serum PR92 antigen level of pa
tients with benign and malignant disease. •,
progressive cancer; O, nonprogressive cancer;
D, newly diagnosed cancer; A, benign disease.
30
DC
0.
20
10 •
aLiver.
•Renal
Oo0[0pEsopha-
Thyroid Rectum
Cancer«*oo»»0Ovarian
Cancer«OoBladder
Cancer•a0°
Cancer•'Cancer'
*Benign
Head«
Lung.
geal.Stomach
Neck
Leukemia:«Cervical
Cancer'•Ovarian
Liver
DiseaseAADisease
Cancer- Cancer«D
303025201510nf:•
:•...Prostate
i ,••••••;••;;Benign
:
ABenign
•Melanoma
Q«
•.•Breast
*
<Normal
•Breast
••••*••*•Normal•».**.•*•
•*#*»••Benign
MaleOver
SOYrsOld:
Cancer
Prostate
Other Urologica!
and
Breast
Cancer
Dis.-
{Stable. Ned.
Cancer
Quiescent,
(Progressive)
Disease••:|:-:.•*•*•»•*•
Remission)..::..t•••.
Fig. 3. Serum PR92 antigen level of normal male, patients with benign
prostate and urogenital disease, and prostate cancer patients.
Fig. 4. Serum PR92 antigen level of normal female, patients with benign
breast disease, and patients with progressive or nonprogressive breast carcinoma.
be present in serum or urine specimens of the prostate cancer
patients. It was hoped that the MAb could provide a base to
develop an immunodiagnostic test for detection and/or prog
nosis of a prostate neoplasm.
PR92 MAb is distinct from the monoclonal antibodies
D83.21, tt-Pro3, and KR-P8, which were raised against prostate
carcinoma cell lines DU 145 and PC3 (8-10). The other mono
clonal antibodies were reported to bind both DU 145 and PC3
cells as well as some bladder cancer cell lines. The PR92 MAb
bound to DU 145 cells, but did not bind PC3 nor any of the six
bladder carcinoma cell lines tested. Raynor observed that KRP8 MAb strongly interacted with some antigens present in
human seminal plasma (9). The PR92-RIA, however, failed to
detect any seminal antigen.3
3 Unpublished observation.
4546
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1988 American Association for Cancer Research.
PROSTATE AND BREAST TUMOR-ASSOCIATED
20 -
Pro
Pro
15 -
Time —••
Fig. 5. PR92 antigen level of serial serum specimens from prostate cancer
patients. Pro, progressive; St, stable; Res, responsive; NED, no evidence of disease.
#134
20
Pro!
Res
15
10
I
Res
5
Res
Res
Rem
Rem
/Rem
#118
#130
15
15
10
Pro
Res
Res
Time
Fig. 6. PR92 antigen level of serial serum specimens from breast cancer
patients. Pro, progressive; St, stable; Res, responsive; Rem, remission.
ANTIGENS
The PR92-RIA demonstrated immunological specificity by
displaying a positive dose-response phenomenon with the sol
uble antigens in DU 145 cell extract (Fig. 1), while showing
negative responses with 16 other tumor-associated proteins
(Table 2). This observation indicates that the antigenic determinant(s) of the molecules detected by PR92 MAb is distinct
from any of the known tumor-associated proteins. It is note
worthy that the PR92-RIA did not recognize mucin molecules
related to CA 19-9, CA 125, CA 15-3, and a human colonie
mucin preparation. This, however, does not rule out the possi
bility that PR92 antigen may be a glycoprotein derived from
some type of mucins.
Serum specimens of patients with benign or malignant dis
eases were screened by the PR92-RIA. For the purpose of
comparing the serum PR92 antigen level in different groups, 5
units/ml of PR92 antigen were assumed to be the highest value
of normal specimens. The results presented in Fig. 2 revealed
that 1 of 10 specimens of benign disease cases and, among
cancer cases, 6 of 16 ovarian, 1 of 16 bladder, 1 of 6 liver (or
thyroid), 1 of 10 esophageal (or stomach) carcinomas, and 1 of
10 melanoma (or leukemia) cases showed greater than 5 units/
ml of PR92 antigen level. In contrast, the serum specimens
from patients with progressive prostate or breast carcinoma
showed generally higher PR92 antigen activity compared to the
specimens of normal and benign disease controls and nonpro
gressive prostate or breast carcinoma cases. This observation
motivated us to investigate the potential utility of the PR92RIA in detecting progressive prostate or breast carcinoma cases.
Of the specimens from 58 normal males and 29 patients with
benign prostate and urogenital diseases, only one showed PR92
antigen activity greater than 1 unit/ml. In contrast, 23 of 31
specimens from progressive prostate cancer patients displayed
reactivity above 5 units/ml. It should be noted that 21 of 23
specimens from cancer patients at Stage C/D showed an ele
vated PR92 antigen level, while 5 of 7 specimens of Stage A/B
patients yielded the antigen level below 5 units/ml. A similar
trend was observed in the study of breast carcinoma cases. Of
21 serum specimens from normal female volunteers and 28
from patients with benign breast disease, 1 displayed reactivity
above 5 units/ml. Among 61 diagnosed breast cancer patients,
30 were at a clinically nonprogressive state, while 31 were
judged to have progressive cancer. Compared to 27 of 31
specimens from progressive breast cancer patients, only 3 of 30
from nonprogressive cancer patients displayed a PR92 antigen
level greater than 5 units/ml. Considering that the majority of
the progressive breast cancer cases examined here were at Stage
C/D, these results suggest a significant clinical correlation
between the elevation of serum PR92 antigen activity and the
advance of tumor stage in prostate or breast cancer patients.
The potential use of the PR92-RIA as an indicator of clinical
status of a patient was studied with a series of serum specimens.
The results depicted in Figs. 5 and 6 indicate a general increase
in PR92 antigen activity above the normal level with progres
sion of malignant disease and a decrease when a patient re
sponds to clinical treatment. There appears to be a high degree
of correlation between the PR92 antigen level and the clinical
status of the prostate and breast cancer patients. These findings
indicate that the sensitivity of the PR92-RIA is such that it may
be useful as an indicator of a prostate or breast cancer disease
state.
We presented in this paper some evidence that PR92 MAb
specifically recognizes some cancer cell lines. The amount of
the PR92 antigen in serum specimens of prostate and breast
cancer patients appeared to depend upon the tumor stage, but
4547
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1988 American Association for Cancer Research.
PROSTATE AND BREAST TUMOR-ASSOCIATED
more extensive study, including a large number of documented
serum specimens, is necessary to establish this relationship.
Comparison of the PR92-RIA with other recognized prostate
or breast tumor marker-assays may reveal the merit of this new
serum test in detection or monitoring of cancer. Preliminary
data in our laboratory suggest that the PR92 antigen is a
glycoprotein with an approximate molecular weight of 470,000
and may be secreted into body fluids during the development
of malignant disease. An investigation is currently in progress
to elucidate molecular characteristics and to understand any
physiological function of the PR92 antigen.
ACKNOWLEDGMENTS
The authors are indebted to Dr. I. Tribby and S. Willaby for their
excellent assistance in cell culture work.
REFERENCES
1. Koprowski, II Steplewski, Z., and Mitchell. K Coloréela!carcinoma anti
gens detected by hybridoma antibodies. Somatic Cell Genet., 5: 957-972,
1979.
2. Bast, R. C, Jr., Feeney, M., Nadler, L. M., Colvin, R. B., and Knapp, R. C.
Reactivity of a monoclonal antibody with human ovarian carcinoma. J. Clin.
Invest., 68: 1331-1337. 1981.
3. Cancer Facts and Figures—1987, pp. 10-12. New York: American Cancer
Society, 1987.
4. Mohan, D. E., and Doctor, B. P. A radioimmunoassay for human prostatic
acid phosphatase-level in prostatic disease. Clin. Biochem., 12: 12-17, 1979.
5. Griffiths, J. C. Prostate-specific acid phosphatase: réévaluation
of RIA in
diagnosing prostatic disease. Clin. Chem., 26:433-436, 1980.
ANTIGENS
6. Pontes, J. E., Chu, T. M., Slack, N., Karr, J., and Murphy, G. P. Serum
prostatic antigen measurement in localized prostatic cancer: correlation with
clinical course. J. Urol., 128: 1216-1218, 1982.
7. Kuriyama, M., Wang, M. C., Papsidero, L. D., Kilban, C. S., Shimano, T.,
Valenzuela, L., Nishiura, T. Y., Murphy, G. P., and Chu, T. M. Quantitation
of prostate-specific antigen in serum by a sensitive enzyme immunoassay.
Cancer Res., 40: 4658-4662, 1980.
8. Ware, J. L., Paulson, D. F., Parks, S. F., and Webb, K. S. Production of
monoclonal antibody a-Pro3 recognizing a human prostatic carcinoma an
tigen. Cancer Res., 42: 1215-1222, 1982.
9. Raynor, R. H., Hazra, T. A., Moncure, C. W., and Mohanakumar, T.
Characterization of a monoclonal antibody, KR-P8, that detects a new
prostate-specific marker. J. Nati. Cancer Inst., 73:617-625, 1984.
10. Starling, J. J., Sieg, S. M., Beckett, M. L., Schellhammer, P. F., Ladaga, L.
E., and Wright, G. L., Jr. Monoclonal antibodies to human prostate and
bladder tumor-associated antigens. Cancer Res., 42: 3084-3089, 1982.
11. Wright, G. L., Jr., Beckett, M. L., Starling, J. J., Schellhammer, P. F., Sieg,
S. M., Ladaga, L. E., and Poleskic, S. Immunohistochemical localization of
prostate carcinoma-associated antigens. Cancer Res., 43: 5509-5516, 1983.
12. Carroll, A. M., Zulutsky, M., Schatten, S., Bhan, A., Perry, L. L., Sobotka.
C., Benacerraf, B., and Greene, M. I. Monoclonal antibodies to tissue-specific
cell surface antigens. Clin, luminimi. Immunopathol.,.(.(: 268-281, 1984.
13. Stone, K. R., Mickey, D. D., Wunderli, H., Mickey, G. H., and Paulson, D.
F. Isolation of a human prostate carcinoma cell line (DU145). Int. J. Cancer,
21: 274-281, 1978.
14. De St. Groth, S. F., and Scheidegger, D. Production of monoclonal antibod
ies: strategy and tactics. J. Immunol. Methods, 35: 1-21, 1980.
15. Harwig, J. F., Coleman, R. E., Sherman, S. L., Siegel, B. A., and Welch, M.
J. Highly iodinated fibrinogen, a new thrombus-localization. J. NucÃ-.Med.,
16:756-763, 1975.
16. Masuko, T., Yagita, H., and Hashimoto,, Y. Monoclonal antibodies against
cell surface antigens present in human urinary bladder cancer cells. J. Nati.
Cancer Inst., 72: 523-530, 1984.
17. Dempsey, P. J., Feleppa, F. P., Brown, R. W., de Krestser, T. A., Whitehead,
R. H., and Jose, D. G. Development of monoclonal antibodies to human
breast carcinoma cell line PMC42. J. Nati. Cancer Inst., 77: 1-15, 1986.
18. Pepsidero, L. D., Croghan, G. A., O'Connell, M. J., Valenzuela, L. A.,
Nemoto, T., and Chu, T. M. Monoclonal antibodies (F36/22 and M7/105)
to human breast carcinoma. Cancer Res., 43: 1741-1747, 1983.
4548
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1988 American Association for Cancer Research.
Monoclonal Antibody PR92 with Restricted Specificity for
Tumor-associated Antigen of Prostate and Breast Carcinoma
Yung D. Kim, Deborah Y. Robinson and Joseph T. Tomita
Cancer Res 1988;48:4543-4548.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/48/16/4543
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1988 American Association for Cancer Research.