A method for aiding in the diagnosis of, and monitoring the progression
of, breast cancer in a patient by measuring the amount of NCA 50/90
in a blood sample, e.g. serum sample, obtained from the patient.
Measurement in a single sample of an amount of NCA 50/90 significantly
higher than the mean amount of NCA 50/90 in the normal population
is an indication of breast cancer in the patient. The progression
of breast cancer can also be monitored by performing a series of
specific immunoassays over time to determine changes in the level
of NCA 50/90 in blood samples. Increases in blood NCA 50/90 levels
over time are indicative of a deteriorating condition whereas decreasing
levels of blood NCA 50/90 over time indicate an improving condition.
What is claimed is:
1. A method for aiding in the diagnosis of breast cancer in . a
patient, comprising the steps of determining the amount of NCA 50/90
in a blood sample obtained from said patient and comparing said
measured amount of NCA 50/90 to the mean amount of NCA 50/90 in
the normal population, whereby the presence of a significantly increased
higher amount of NCA 50/90 in the patient's blood is an indication
of breast cancer in the patient.
2. The method of claim 1 wherein the amount of NCA 50/90 in the
patient's blood sample is measured by performing a sandwich immunoassay
in which at least one of the antibody reagents is specific for NCA
50/90 with no substantial reactivity for CEA, NCA 95 or BGP.
3. The method of claim 2 wherein said NCA 50/90 specific antibody
reagent is a monoclonal antibody reagent.
4. The method of claim 2 wherein said NCA 50/90 specific antibody
reagent is the monoclonal antibody produced by the hybridoma deposited
with the American Type Culture Collection and identified as ATCC
HB11204, or a monoclonal antibody which binds to the same epitope
as the aforesaid monoclonal antibody produced by hybridoma ATCC11204.
5. The method of claim 1 wherein said blood samples is a serum
or plasma sample.
6. A method for monitoring the progression of breast cancer in
a patient, comprising performing a series of specific immunoassays
over time to determine changes in the level of NCA 50/90 in blood
samples obtained from such patient, whereby increases in blood NCA
50/90 levels indicate a deteriorating condition while decreases
in blood NCA 50/90 levels indicate an improving condition.
7. The method of claim 6 wherein the immunoassays performed are
sandwich immunoassays in which at least one of the antibody reagents
is specific for NCA 50/90 with no substantial reactivity for CEA,
NCA 95, BGP or PSG.
8. The method of claim 7 wherein said NCA 50/90 specific antibody
reagent is a monoclonal antibody reagent.
9. The method of claim 7 wherein said NCA 50/90 specific antibody
reagent is the monoclonal antibody produced by the hybridoma deposited
with the American Type Culture Collection and identified as ATCC
HB11204, or a monoclonal antibody which binds to the same epitope
as the aforesaid monoclonal antibody produced by hybridoma ATCC11204.
10. The method of claim 6 wherein said blood samples are serum
or plasma samples.
11. An NCA 50/90 specific antibody reagent comprising the monoclonal
antibody produced by the hybridoma deposited with the American Type
Culture Collection and identified as ATCC HB11204, or a monoclonal
antibody which binds to the same epitope as the aforesaid monoclonal
antibody produced by hybridoma ATCC11204.
12. A sandwich immunoassay test kit for use in the determination
of NCA 50/90 in a blood sample, comprising, as first antibody reagent,
the antibody reagent of claim 11, and a second antibody reagent
capable of binding specifically or nonspecifically with NCA 50/90.
13. The test kit of claim 12 wherein said second antibody reagent
is also monoclonal.
14. The test kit of claim 12 wherein one of said antibody reagents
is labeled with an enzyme.
Breast cancer descriptionBACKGROUND OF THE INVENTION
The present invention relates to monitoring the progression or
stage of disease in breast cancer patients. More particularly, the
invention relates to such monitoring methods based on measurement
of cancer marker blood levels.
A number of substances have been determined to be useful markers
in monitoring the course of various cancer types. Some useful markers
that have been identified are oncofetal antigens such as carcinoembryonic
antigen (CEA) and alpha-fetoprotein, tissue-specific antigens such
as prostate-specific antigen (PSA), and mucin antigens such as those
conventionally known as CA-125 and CA-19-9. Immunoassays for antigens
such as these are typically used as confirmatory tests at the time
of diagnosis and subsequently for monitoring patient status. Occasionally,
the use of such tests crosses the boundaries of tumor type (for
example, the use of CEA tests in colon, breast, and lung cancer,
and alpha-fetoprotein in hepatocellular and testicular cancer),
but the utility of each test type is foremost for a single tumor
type (for example, PSA for prostate cancer and CA-125 for ovarian
A family of antigenic proteins have been identified which are genetically
and immunologically related to CEA (Thompson, J. and W. Zimmerman
(1988) Tumor Biol. 9, 63-83; and Barnett, T. and W. Zimmerman (1990)
Tumor Biol. 11, 59-63). Among these are the nonspecific cross-reacting
antigens (NCAs), the trans-membrane antigens designated biliary
glycoprotein (BGP, and sometimes referred to as TM-CEAs), and the
family of pregnancy-specific .beta.-glycoproteins (PSGs) (for a
description of the accepted nomenclature of these genes and their
protein products, reference can be made to: Barnett, T. and W. Zimmerman
(1990) Tumor Biol. 11, 59-63). Molecular cloning of the CEA gene
family has enabled the identification of 22 members, of which 20
are probably expressed (Frangsmyr, L. et al. (1992) Tumor Biol.
13, 98-99; and Hammerstrom, S. et al Tumor Biol. 13, 57). The results
of molecular genetic analysis have given a better understanding
of the complex group of glycoproteins in the CEA gene family.
NCA was originally described as a component of normal tissue which
cross-reacted with antibodies raised to CEA (Mach, J.-P. and G.
Pusztaszeri (1972) Immunochemistry 9, 1031-1034; and von Kleist,
S., Chavenel, G. and P. Burtin (1972) Proc. Natl. Acad. Sci. USA
69, 2492-2494). As such, NCA was considered a potential nontumor
derived interferant in assays for CEA. Molecular cloning identified
one species of NCA of calculated M.sub.r 37,000 designated by one
group as NCA-BT (Barnett, T., Goebel, S. J., Nothdurft, M. A. and
J. J. Elting (1988) Genomics 3, 59-66) to denote the breast tissue
origin of the cloned cDNA, and by others as NCA (Tawaraji, Y. et
al. (1988) Biochem. Biophys. Res. Commun. 150, 89-96; and Neumaier,
M. et al (1988) J. Biol. Chem. 263, 3203-3207). This single NCA
species has since been termed NCA 50/90 (Kolbinger, F., Schwarz,
K., Brombacher, F., von Kleist, S., and Grunert, F. (1989) Biochem.
Biophys. Res. Commun. 161, 1126-1134) because it was now known to
be processed into two mature isoforms of M.sub.r 50,000 and M.sub.r
90,000 which have different degrees of glycosylation. A second and
distinct NCA gene was subsequently identified by molecular cloning
from leukemic cells that codes for an M.sub.r 95,000 glycoprotein
(Kuroki, M. et al (1991) J. Biol. Chem. 266, 11810-11817). This
latter NCA has been termed NCA 95.
Early studies also identified a cross-reacting antigen from adult
stools and from meconium which, for historical reasons, was termed
NCA-2 (Burtin, P., Chavenel, G. and H. Hirsch-Marie (1973) J. Immunol.
111, 1926-1928). The designation of this antigen as NCA is, however,
a misnomer. It has been identified as a proteolytic fragment of
CEA since the first 30 amino acids of the meconium-derived NCA-2
are identical in sequence with CEA (Siepen, D. et al (1987) Biochem.
Biophys. Res. Commun. 174, 212-218). In contrast, cDNAs for NCA
50/90 and NCA 95 have been described and code for distinct and different
amino acid sequences in this region. Indeed, a recent report suggests
that variability in CEA results obtained with different commercial
kits may be due to interference with NCA-2 (O. P. Bormer (1991)
Clin. Chem. 37, 1736-1739).
Given the improved understanding of the CEA gene family resulting
from molecular cloning analysis, monoclonal antibodies can now be
identified which recognize specific family members and do not cross
react with closely related molecules. Previous attempts to raise
antibodies to NCA have been plagued with the problem of cross reactivity
with CEA family members. This may explain why NCA has been considered
a poor serum marker for cancer diagnosis and monitoring (Shively,
J. E., Spayth, V., Chang, F.-F., Metter, G. E., Klein, L., Present,
C. A., and C. W. Todd (1982) Cancer Res. 42, 2502-2513; and Burtin,
P., Chavenel, G., Hendrick, J. C. and N. Frenoy (1986) J. Immunol.
137, 839-845). It has been further speculated that NCA-specific
monoclonal antibodies such as are now widely accepted for CEA and
other antigens would be very difficult to develop (Burtin, P. et
In addition, it is now clear that members of the CEA gene family
are differentially expressed by various tumor types. For example,
it is well known that CEA is expressed in most if not all colorectal
carcinomas, while expression is limited to a minority of breast
carcinomas. Prior to the generation of specific monoclonal antibodies,
attempts to quantitate NCA levels in the serum of cancer patients
were confounded by the presence of other CEA gene family members
that cross reacted with the antibodies being used. However, because
of the successful production of monoclonal antibodies specific to
NCA 50/90, it is now possible to determine the incidence of elevated
NCA 50/90 protein in different cancer types.
Although there have been reports of monoclonal antibodies specific
for NCA 50/90 (Chavenel, G., Frenoy, N., Escribano, M. J. and P.
Burtin (1983) Oncodev. Biol. and Med. 4, 209-217; and Yeung, M.,
M.-W. Hammerstrom, M. L., Baranov, V. and S. Hammerstrom (1992)
Tumor Biol. 9, 119), there have been no reports of a monoclonal
antibody which binds to NCA 50/90 but does not recognize any other
CEA family members including CEA, NCA 95, NCA 2, BGP or PSG. Similarly,
several reports have suggested that NCA may be elevated in the serum
of cancer patients (von Kleist, S., Troupel, S., King, M. and P.
Burtin (1977) Br. J. Cancer 35, 875-880; and Wahren, B., Gahrton,
G., Ruden, U. and S. Hammerstrom (1982) Int. J. Cancer 29, 133-1.37;
and Harlozinska, A., Rachel, F., Gawlikowski, W., Richter, R. and
J. Kolodziej (1991) Eur. J. Surg. Oncol. 17, 59-64; and Reck:, W.,
Daniel, S., Nagel, G., Hirn, M., von Kleist, S., and F. Grunert
(1992) Tumor Biol. 13, 110-111), but these measurements utilized
antibodies that have not been shown to recognize NCA 50/90 to the
exclusion of other CEA-related molecules. In addition, there have
been no reports of a correlation between blood NCA levels and the
clinical status of any particular cancer patients.
Previous attempts to quantitate the level of NCA 50/90 in the serum
have been hampered by the lack of a suitable standard. Measurements
of NCA in blood have shown mean values in serum from normal individuals
of from 30 ng/ml (Harlozinska, A., et al. supra) to 130 ng/ml (von
Kleist, S., Troupel, S., King, M. and P. Burtin (1977) Br. J. Cancer
35, 875-880). This is due to the use of biochemically purified NCA
as a standard to calibrate immunoassay measurements of NCA in blood
and blood fluids. Just as the monoclonal antibodies have not been
demonstrated to specifically recognize NCA 50/90, neither has the
purity of the NCA standard preparations been determined.
SUMMARY OF THE INVENTION
It has now been found that NCA 50/90 can be significantly elevated
in the blood of patients with breast cancer. Accordingly, the present
invention provides a method for aiding in the diagnosis of breast
cancer in a patient, comprising the steps of determining the amount
of NCA 50/90 in a blood sample obtained from said patient, and comparing
such measured amount of NCA 50/90 to the mean amount of NCA 50/90
in the normal population, whereby the presence of a significantly
increased higher amount of NCA 50/90 in the patient's blood is an
indication of breast cancer in the patient. The present method also
provides a means for monitoring the progression of the disease.
Increases in blood NCA 50/90 levels measured by performing a series
of specific immunoassays over time indicate a deteriorating condition
in a significant number of patients while decreases in blood NCA
50/90 levels indicate an improving condition in such patients.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a graph showing the specificity of the particular immunoassay
described in the Examples below for NCA 50/90 compared to CEA and
FIG. 2 is a graph showing the specificity of the particular immunoassay
described in the Examples below for NCA 50/90 compared to recombinant
NCA 95 and NCA purified from the spleen which was shown by amino
acid sequencing to be identical to NCA 50/90.
FIGS. 3-8 are graphs showing a comparison of the correlations of
NCA 50/90 and CEA blood levels with the stage of breast cancer in
patients under treatment.
FIG. 9 is a reproduction of a Western blot showing that the 228.2
monoclonal antibody binds to NCA 50/90, but does not recognize NCA
95, CEA, BGPs or NCA 2. For each of the three blots presented, the
indicated monoclonal antibodies were blotted against CEA family
members as follows: lane 1, CEA; lane 2, BGP; lane 3, NCA 50/90;
lane 4, NCA 95; lane 5, NCA 2. The control blots in this figure
also show that the preparation of NCA 50/90 used as a standard in
the NCA 50/90 ELISA contains only NCA 50/90.
FIG. 10 is a graph showing the standard curve obtained in the NCA
FIG. 11 is a graph showing that the NCA 50/90 ELISA has no significant
cross reactivity with CEA, NCA 95 or BGP.
FIG. 12 is a graph showing that the NCA 50/90 ELISA has no significant
cross reactivity with .alpha.-1-antichymotrypsin, .alpha.-1-antitrypsin,
.alpha.-2-macroglobuln, .alpha.-2-antiplasmin and antithrombin III.
FIG. 13 is a graph showing that the NCA 50/90 ELISA has no significant
reactivity with serum from pregnant women which demonstrates a lack
of reactivity with PSG.
FIG. 14 is a graph which demonstrates that the level of NCA 50/90
is elevated in the serum of patients with breast cancer.
FIG. 15 is a graph which shows a lack of correlation between blood
levels of CEA and NCA 50/90 in patients under treatment for breast
FIGS. 16-20 are a series of graphs which illustrate NCA 50/90 levels
in longitudinal serum samples of individual patients with breast
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Essentially any method may be employed in the measurement of blood
(e.g., serum or plasma) NCA 50/90 levels. Typically, such measurement
will be performed by sandwich immunoassay using two antibody reagents,
one of which is specific for NCA 50/90 (showing no substantial reactivity
with other CEA family member antigens, particularly CEA, NCA 95
and BGP), while the second antibody can bind specifically to NCA
50/90 or can be raised against a different immunogen but crossreacts
with NCA 50/90. Assay format and methods for the preparation of
the required antibody reagents can be selected by the skilled worker
in the field. Suitable antibody reagents can be labeled, e.g., enzyme-labeled,
or immobilized, e.g., coated onto a microtiter plate, bound to plastic
or magnetic beads or particles, and can be comprised of whole immunoglobulins,
e.g., IgG or IgM, or fragments, e.g., Fab, Fab', and F(ab').sub.2
fragments, or aggregates thereof.
Preferably, the NCA 50/90 specific antibody reagent is prepared
by immunization of a host animal with a suitable immunogen such
as an NCA 50/90 containing immunogen mixture, e.g., a purified extract
of spleen or tumor cells; NCA 50/90-expressing transfectant cell
lines (see European Patent Publication 346,702); an immunogen conjugate
comprising a synthetically prepared peptide coupled to a conventional
immunogenic carrier molecule, where the peptide has an amino acid
sequence encompassing an epitope of NCA 50/90; and the like as will
be understood in the art.
Antibody reagents comprising monoclonal antibodies will be generally
preferred. Particularly preferred NCA 50/90 specific monoclonal
antibodies are those which bind to substantially the same epitope
as that produced by the hybridoma that was deposited on Nov. 18,
1992, with the American Type Culture Collection, 12301 Parklawn
Drive, Rockville, Md. 20852 and which has been given deposit accession
number ATCC HB 11204.
It will be understood that a number of standard methods can be
used in order to determine whether a particular monoclonal antibody
binds to substantially the same epitope as the above-mentioned antibody
whose hybridoma has been deposited with the ATCC. A particularly
useful method is competitive binding, wherein the ability of the
antibody of interest to bind to NCA 50/90 in the presence of the
reference antibody is measured. Substantial inability of both antibodies
to bind simultaneously indicates that substantially the same epitope
It will be understood that, similar to other types of accepted
disease monitoring methods, the present method will not be useful
on every patient diagnosed with breast cancer. Rather, the physician
will use NCA 50/90 blood values in combination with other diagnostic
values and clinical observations to diagnose the onset of breast
cancer, thereby providing a method for screening a select population
for the disease, and further, to develop a course of treatment and
therapy for each individual patient. It is also contemplated that
monitoring blood levels of NCA 50/90 will provide a means for monitoring
the progress of a cause of therapy for an individual patient.
The present invention will now be illustrated, but is not intended
to be limited by, the following examples.
Mab 228.2--BALB/C mice were immunized with 50 .mu.g of an emulsion
of NCA purified from human spleen (von Kleist, S. and P. Burtin
(1969) Cancer Res. 29:1961-1964), and Freund's complete adjuvant.
Spleens from hyperimmune animals were removed from euthanized animals
and the splenocytes were fused with AG8 mouse myeloma cells (ATCC
CRL 1580). The resulting hybridomas were screened for anti-NCA antibody
production by sandwich ELISA. Positive clones were subsequently
screened for anti-CEA and anti-BGP (TM-CEA) activity (see Barnett,
T. and W. Zimmerman supra; and see Barnett, T. R., Kretschmer, A.,
Austen, D.A., Goebel, S. J., Hart, J. T., Elting, J. J., and M.
E. Kamarck (1989) J. Cell Biol. 108:267-276) by sandwich ELISA assay.
Those clones specific for NCA 50/90 were recloned and rechecked
for cross reactivity with CEA and BGP (TM-CEA) by ELISA and again
by FACS analysis using recombinant mouse cell lines expressing CEA,
NCA 50/90 or BGP on their plasma membranes (see European Patent
Publication No. 346,702). The result of this screening process was
identification of Mab 228.2 (deposited with the ATCC, supra) which
is specific for NCA 50/90 with no detectable reactivity with CEA,
NCA 95 or BGP by ELISA.
Mab 176.7.5--BALB/C mice were immunized with BGP as described in
U.S. patent application Ser. No. 480,428, filed Feb. 15, 1990. The
hybridoma that produces the selected antibody has been deposited
with the American Type Culture Collection, Rockville, Maryland,
USA, and given the designation ATCC HB-10411. This antibody reacts
with both NCA 50/90 and BGP but not with CEA. This antibody was
crosslinked to calf intestine alkaline phosphatase (Biozyme Corp.,
San Diego, Calif., USA) after thiolation of the antibody with 2-iminothiolane
using the heterobifunctional cross-linking reagent sulfo-SMPB (Pierce
Chem. Co., Rockford, Ill., USA). This conjugate is hereafter designated
NCA Calibrator--NCA purified from human spleen (von Kleist, S.
et al, supra) was used as the calibrator for the NCA 50/90 assay.
The amount of antigen was determined by amino acid analysis of a
hydrolyzed sample of the purified NCA. Assay calibrators were made
by diluting purified NCA into PBST-BSA (see below).
NCA-Specific Immunoassay--A sandwich ELISA was configured using
NCA-specific Mab 228.2 as the solid phase capture antibody and Mab
176.7.5-A.P. as the reporter antibody. Mab 228.2 was coated in 100
.mu.l of 0.1M Na.sub.2 CO.sub.3 /NaHCO.sub.3, pH 9.0, onto the wells
of microtiter ELISA plates (product ID#25801, Corning Glass Works,
Corning, N.Y., USA) for 1 hour at 37.degree. C. The wells were blocked
with 300 .mu.l 0.1% Tween-20 (polyoxyethylenesorbitan monolaureate)
in 0.05M Na phosphate buffer, pH 7.2, 0.1M NaCl and 0.01% Thimerosal
(PBST) for 1 hour at 37.degree. C. After washing the plate 5 times
with PBST, 50 .mu.l of antigen solution [calibrators in PBST with
5% bovine serum albumin (PBST-BSA) or serum unknowns] were added
and incubated for 2 hours at 37.degree. C. After washing 5 times
with PBST, 100 .mu.l of 176.7.5-A.P. (15 .mu.g/ml in PBST) was added
to each well and incubated for 1 hour at 37.degree. C. After washing,
200 .mu.l of p-nitrophenyl phosphate (in 1.0M diethylamine buffer,
pH 9.8) was added to each well and incubated for 30 minutes at room
temperature. Absorbance at 405 nm was then determined and the amount
of antigen in the test serum samples was determined from the calibrator
Patient serum samples--Serum samples from clinically confirmed
breast cancer patients were obtained from the University of Texas
M.D. Anderson Cancer Center, Houston, Tex., USA. Samples from clinically
normal individuals were obtained from M.D. Anderson and from in-house
The antigenic specificity of the immunoassay for NCA 50/90 is established
by the data shown in FIGS. 1 and 2. FIG. 1 shows the lack of cross-reactivity
with CEA and with BGP. Testing with NCA 50/90 and NCA 95 purified
from a recombinant expression vector containing cloned gene sequences
(FIG. 2) shows that the assay is specific for NCA 50/90 (NCA-spleen
is NCA purified from human spleen as described in von Kleist, S.
et al, supra).
NCA 50/90 levels in the serum of clinically normal females were
determined and the cutoff (951 h percentile confidence level) found
to be 143 ng/ml (Table 1 ). The mean serum concentration of 202
clinically diagnosed breast cancer patients was 358 ng/ml.
TABLE 1 ______________________________________ Normals Mean 95th
%ile ______________________________________ Female 64.9 ng/ml 135
ng/ml nonsmokers (n = 50) Female 82.3 ng/ml 161 ng/ml smokers (n
= 25) Total Females (n = 75) 70.1 ng/ml 143 ng/ml Breast Cancer
Patients Percent showing 358.1 ng/ml 1756 ng/ml elevated NCA 50/90
(p < 0.0001) levels = 66% ______________________________________
Changes in the serum levels of NCA 50/90 were found to correlate
with the stage of disease in a significant number of patients diagnosed
and under treatment for breast cancer. Serial serum samples were
obtained from clinically confirmed breast cancer patients. These
serial samples had a mean value of 209 ng/ml of NCA 50/90. Six examples
of the correlation of serum CEA and NCA 50/90 levels and the clinical
status of these patients are shown in FIGS. 3-8.
The graphs each show:
(a) the levels of NCA 50/90 as measured by the above-described
sandwich immunoassay method,
(b) the levels of serum CEA that had been measured in a clinical
laboratory and shown on the patient's chart,
(c) the 95th percentile cut-off value for normal NCA 50/90 (i.e.,
143 ng/ml), and
(d) across the top of the graph, a summary of the disease progression
using the following abbreviations:
"Surg"=surgical procedure to remove the cancer,
"Chemo"=period during which chemotherapy was administered,
"Stable" an assessment by the attending physical that
the patient's breast cancer has stabilized,
"Met.uparw."=an assessment by the attending physician
that metastasis had occurred or that the breast cancer was otherwise
"Prog"=an assessment by the attending physician that
the breast cancer was progressing,
"+"=death of the patient,
"Met.dwnarw."=an assessment by the attending physician
that metastasis had been arrested, and
"NED"=an assessment by the attending physician that there
was no longer any evidence of disease.
In these examples, not only does serum NCA 50/90 show correlation
with the clinical regression, stability, and progression of the
disease, but also such correlation is more significant than the
correlation with serum CEA levels (a commonly used serum marker
for monitoring breast cancer).
The high incidence of elevated NCA 50/90 levels in breast cancer
sera, and the significant correlation of clinical status with changes
in the NCA 50/90 levels shows that the present method is useful
as a means for monitoring the progression of breast cancer in patients.
MAb 228.2--Monoclonal Antibody 228.2 specifically recognizes NCA
50/90, and was purified from mouse ascites fluids exactly as described
for Study 1.
Biotinylation of Goat Antibody to CEA--An affinity purified polyclonal
goat antiserum raised to CEA was purchased from BiosPacific, Emeryville,
Calif., USA (Lot No. 015-B4302) and placed into 1.3 ml of 0.1M NaHCO.sub.3,
pH 8.5 at a concentration of 1.0 mg/ml. To this was added 18.5 .mu.l
of a 10 mg/ml solution of NHS-LC-biotin (Pierce, Rockford, Ill.
USA, Catalog No. 21335) in deionized water to give a 50/1 molar
excess of biotin to antibody. After incubation at 0.degree. C. for
4 hours the biotinylated antibody was passed over a buffer-exchange
column using 10 mM phosphate, pH 7.4/150 mM NaCl and stored at 4.degree.
C. with 0.1% thimerosal as preservative.
NCA 50/90 Calibrator--A cDNA corresponding to NCA 50/90 was derived
from the breast tumor cell line BT-20 as described previously (Barnett,
T., Goebel, S. J., Nothdurft, M. A. and J. J. Elting (1988) Genomics
3, 59-66). The coding region for the NCA 50/90 gene was modified
by the elimination of the C-terminal hydrophobic region which signals
replacement by a phosphoinositol glycan linkage, and the addition
of a stretch of six histidine residues, also at the carboxyl terminus
of the molecule (Drake, L. and Barnett, T. (1992) Biotechniques
12, 645-649). This construct was cloned into pVL1393 by PCR and
expressed using recombinant baculovirus phage to infect Spodoptera
frugiperda (Sf9) cells. NCA 50/90 was affinity purified from Sf9
supernatant fluids using a zinc-imidoacetate-Sepharose.RTM. column
as described (Drake and Barnett, supra). The concentration of NCA
50/90 was determined by the BCA protein assay (Pierce, Cat. No.
23225G). For use as a calibrator in the NCA 50/90 ELISA, purified
recombinant NCA 50/90 was diluted in TBST/5 % BSA as described below.
NCA 50/90-Specific Immunoassay--A sandwich ELISA was configured
using the 228.2 monoclonal antibody as the solid phase capture antibody,
and the biotinylated polyclonal anti-CEA as the reporter antibody.
96-well ELISA plates (Iratoulon IV, Dynatech Laboratories, Chantilly,
Va., USA) were coated with 100 .mu.l of 228.2 antibody at 5 .mu.g/ml
in 0.1M NaHCO.sub.3, pH 9.0 and incubated overnight at 4.degree.
C. Wells were emptied and unreacted sites on the plates were quenched
by the addition of 200 .mu.l of 20 mM Tris, pH 7.5/150 mM NaCl/0.05%
Tween 20 (TBST) with 5% bovine albumin (BSA, fraction V, Sigma Chemical
Company, St. Louis, Mo., USA), Catalog No. A-7030) followed by a
1 hour incubation at 37.degree. C. Wells were washed 6 times with
TBST, and 25 .mu.l of either NCA 50/90 calibrators diluted in TBST/5%
BSA or 25 .mu.l of patient sample was added. An equal volume of
50 mM HEPES, pH 7.0/500 mM NaCl/200 .mu.g/ml mouse IgG/5% BSA/50
.mu.g/ml gentamycin/0.1% (w/v) NaN.sub.3 (sample diluent) was added
to each well and the plates were incubated for 2 hours at 37.degree.
C. After washing 6 times, a 100 .mu.l volume of a 0.3 .mu.g/ml solution
of goat anti-CEA-biotin in 50 mM HEPES, pH 7.0/150 mM NaCl/1 mM
MgCl.sub.2.6H.sub.2 O/0.1 mM ZnCl.sub.2 /5 % BSA/50/.mu.g/ml gentamycin/0.1%
NaN.sub.3 (conjugate diluent) was added to all wells and incubated
for 1 hour at 37.degree. C. The wells were washed a further 6 times,
and 100 .mu.l of streptavidin conjugated to alkaline phosphatase
(Pierce, Catalog No. 21324G) diluted 1/5000 in conjugate diluent
was added. After a 1 hour incubation at 37.degree. C., the plates
were washed 12 times with TBST and incubated with 100 .mu.l of p-nitrophenyl
phosphate in DEA substrate buffer (Pierce, Catalog No. 34064) for
30 minutes. The reaction was stopped with 100 .mu.l N NaOH and absorbance
at 405 nm minus absorbance at 490 nm determined using a microplate
reader (Thermo-Max, Molecular Devices Corp., Menlo Park, Calif.,
USA). The amount of NCA 50/90 was determined for each test sample
by comparison with the calibrator standard curve.
Patient Samples--Serum was prepared from blood drawn from normal
healthy volunteers by Hudson Valley Blood Services of Valhalla,
N.Y., USA. Samples from patients with inactive or active breast
cancer were obtained from Dianon Systems of Stratford, Conn., USA,
and from M. D. Anderson Cancer Center, Houston, Tex., USA. Patient
disease status was determined from information supplied by attending
physicians as well as results of testing for the tumor markers CEA,
lipid associated sialic acid (LASA), and CA 15-3.
The antigenic specificity of the 228.2 monoclonal antibody was
determined first by Western blotting and results are shown in FIG.
9. The 228.2 MAb reacts specifically with NCA 50/90 and not with
other proteins related to CEA. The reactivity of the 228.2 MAb with
the high molecular weight band of Mr 110,000 in lane 3 of the 228.2a
blot, probably represents the formation of SDS-stable protein dimers.
Because NCA 2 is a fragment of CEA and the 228.2 antibody does not
bid to CEA, it would be expected that the 228.2 MAb would not recognize
NCA 2. Results in FIG. 1 show that this is indeed the case. The
reactivity of the 228.2 MAb with the NCA 2 preparation is with an
M.sub.r 90,000 protein which does not comigrate with the M.sub.r
160,000 NCA 2 protein, and is likely to represent a low level of
contamination of the NCA 2 preparation with the M.sub.r 90,000 form
of NCA 50/90. The reactivity of the positive control MAb 53.5 with
each of the antigen preparations demonstrates the presence of the
relevant glycoproteins in each preparation. In addition, the 53.5
MAb reacted only with a protein of M.sub.r 50,000 in the NCA 50/90
preparation, which demonstrates the antigenic purity of the NCA
50/90 preparation. The MAb 030-A1101 binds to alpha fetoprotein
and was used as negative control.
The standard curve presented in FIG. 10 demonstrates a nonlinear
increase in absorbance as a function of NCA 50/90 concentration.
A nonlinear spline curve fit program was used to convert raw patient
data to NCA 50/90 concentrations. The data in FIGS. 11 and 12 demonstrate
that the NCA 50/90 ELISA shows no significant reactivity with CEA,
NCA 95, BGPs, .alpha.-1-antichymotrypsin, .alpha.-1-antitrypsin,
.alpha.-2-macroglobulin, .alpha.-2-antiplasmin and antithrombin
III. The potential for cross reactivity with serine proteinase inhibitors
stems from observations that biochemically purified CEA and NCA
may associate with molecules with amino acid homology to .alpha.-1-antichymotrypsin
and .alpha.-1-antitrypsin (Orjaseter, H. (1976) Acta Path. Microbiol.
Scand. 84, 235-244; and Grunert, F., Abuharfeil, N., Luckenbach,
G. A. and S. von Kleist (1984) Tumor Biol. 5, 221-232). Since the
MAb 228.2 was raised to biochemically purified NCA 50/90 from spleen,
there is also potential for cross reactivity with related proteins.
An additional member of the CEA gene family is pregnancy specific
.beta.-glycoprotein (PSG) which is elevated in the serum of pregnant
women. Reactivity to this protein was tested by examining sera from
15 pregnant women with HCG values ranging from 2,200 to 79,000 (normal
cutoff for HCG=10). As can be seen in FIG. 13, only one patient
showed an NCA 50/90 value above the cutoff value of 18 ng/ml (as
determined below), which demonstrates that the NCA 50/90 ELISA does
not detect PSG.
To establish a cutoff value for normal blood levels of NCA 50/90,
the level of NCA 50/90 was measured in serum from 92 normal healthy
volunteers. A 95% cutoff value was determined to be 18 ng/ml. This
cutoff value differs from that for Study 1 due to differences in
calibrator material. NCA 50/90 values were then measured in plasma
from 31 breast cancer patients undergoing treatment who were clinically
free of cancer, and it was found that 3/31 of the values were above
the cutoff value (FIG. 14). In contrast, 42% of samples from 113
patients with active breast cancer were above the cutoff value,
which demonstrates that NCA 50/90 is elevated above normal levels
in the blood of some patients with breast cancer.
Plasma from 26 breast cancer patients who had active cancer by
clinical examination, but whose levels of CEA and CA 15-3 were below
cutoff for those markers (5 ng/ml and 35 ng/ml, respectively) was
also tested. Of 26 such patients, 7 had NCA 50/90 values in excess
of 18 ng/ml (27%). These results indicate that NCA 50/90 values
can be used to manage patients whose clinical status cannot be monitored
by changes in currently available biomarkers.
Because NCA 50/90 and CEA are both overexpressed in some tumor
tissues, it may be expected that elevations of NCA 50/90 in the
blood of cancer patients would correlate with increases in CEA values.
To test this, we compared the values for NCA 50/90 and CEA in 143
patients under treatment for breast cancer. The results in FIG.
15 demonstrate that the levels of NCA 50/90 and CEA are correlated
only poorly (correlation coefficient, r.sup.2 =0.47). This was unexpected
and demonstrates that increases in blood levels of NCA 50/90 will
occur in a population of cancer patients which is different from
that with increases in blood levels of CEA. This again indicates
that measurement of NCA 50/90 in blood can be of additional clinical
value to currently used biomarkers.
The serum level of NCA 50/90 was found to correlate with the status
of disease in 20 patients diagnosed with and under treatment for
breast cancer. Results obtained with samples from 5 of these patients
are presented in FIGS. 16-20. Patient BS6 had inactive cancer in
the early stages of the study, but developed recurrent disease at
later time points. NCA 50/90 showed elevated values when the clinical
condition worsened from stable (STAB) to progressive disease (PRO),
as determined by the attending physician. Results with Patients
BS3 and BS5 remained below cutoff for both NCA 50/90 and CEA throughout
the course of the study which is in agreement with the clinical
condition of these patients. Results with Patients BS1 and BS19
showed that NCA 50/90 detected cancer recurrence in these patients,
whereas CEA values remained below cutoff. These results again indicate
that NCA 50/90 can be of additional value to current biomarkers
in the management of breast cancer patients. The combined results
with all 20 patients demonstrate that NCA 50/90 values correctly
reflected disease status in 88% of the longitudinal samples. Taken
together, these results demonstrate that NCA 50/90 can be used to
monitor disease status in breast cancer patients under treatment.
The present invention has been particularly described and exemplified
above. Clearly, many other variations and modifications of the invention
can be made without departing from the spirit and scope hereof.