| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Experimental Medicine and
Pathology*
and the Geriatric Oncology
Unit,
University "La Sapienza", Rome; the
Department of Experimental Oncology,
European
Institute of Oncology, Milan; and the Department of Oncology and
Neurosciences,§
University "G.
D’Annunzio", Chieti, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ErbB2 alterations in breast carcinoma occur at early clinical stages and persist during disease progression.5,9,10 Circulating proteolytic cleavage fragments of the ErbB2 protein become detectable in serum as a possible consequence of tumor burden. Consistently, their appearance in vivo correlates with ErbB2 overexpression and disease progression or tumor size in human malignancy or animal models, respectively.11-18 Furthermore, it is well documented that some cancer patients harboring ErbB2 overexpressing tumors, may spontaneously develop an autologous anti-ErbB2 response in the course of their disease.19-22
Mechanistically, at least a subset of B cells is amenable to positive selection, generation and maintenance of autoreactivity by self-antigens.23,24 Consistently, normal ErbB2 epitopes can elicit syngeneic immunity to self-ErbB2 protein,25 providing a functional basis for the development of ErbB2 immunity in breast cancer. ErbB2 in the latter typically lacks mutations that activate the rat homologue in chemically induced neuroblastomas.26 While pathophysiological roles of shed ErbB2 extracellular domain and native anti-ErbB2 immunity have not unequivocally been defined, their elucidation might hold important clues for disease monitoring and design of immunotherapeutic strategies. To this purpose, we evaluated in the present study occurrence of both ErbB2 specific immunity and soluble ErbB2 extracellular domain in sera of breast cancer patients. An association of ErbB2 immune response with ErbB2 ectodomain serum levels was assessed and correlated with clinicopathological disease parameters.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Sera from 71 breast cancer patients and 31 healthy female donors were obtained with informed consent and stored in aliquots at -20°C. Tumor patient sera were collected before adjuvant therapy either the day of surgery or the day before. Tumor types included 62 infiltrating ductal, 8 infiltrating lobular, and 1 medullar carcinoma. The age ranges were 23 to 83 years for patients and 25 to 60 years for normal donors.
Cell Lines and Antibodies
NIH3T3 transfectants with human ErbB227 and NIH3T3 controls were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% calf serum. Polyclonal rabbit antiserum M6 raised against a cytoplasmic epitope of human ErbB2 and monoclonal antibody E2–1 specifically recognizing the ErbB2 extracellular domain have previously been described.28 Purified mouse myeloma immunoglobulins (MOPC21; Cappel, Organon Teknika Corp., West Chester, PA) and normal rabbit serum served as negative controls.
Immunoprecipitation and Immunoblotting
Immunoprecipitation and immunoblot analysis for specific detection of human anti-ErbB2 antibodies were conducted essentially as described.22 For direct immunoblotting, 100 µg protein lysate of LTR-ErbB2 or control NIH3T3 transfectants were electrophoresed in denaturing 8% Tris-glycine polyacrylamide gels (sodium dodecyl sulfate-polyacrylamide gel electrophoresis), and proteins were transferred to a nitrocellulose membrane at 40 V for 1 hour. Membranes were reacted with patient sera (1:100), undiluted culture supernatants of immortalized PBLs, anti-ErbB2 specific (1:400), or preimmune rabbit serum (1:400). A dilution of 1:100 for patient sera was determined after titration to be the highest concentration lacking substantial background reactivity. After washing, the membranes were incubated with alkaline-phosphatase-conjugated goat anti-human IgG and IgM or goat anti-rabbit antiserum (Gibco BRL, Gaithersburg, MD). Bound antibody was visualized using 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and Nitro blue tetrazolium (NBT; Kirkegaard and Perry Laboratories, Inc., Gaithersburg, MD). For immunoprecipitation, 250 µg of LTR-ErbB2 transfectant lysate were precipitated in Staph A buffer using 3 µg of monoclonal antibody (mAb) E2–1 or the negative control immunoglobulin MOPC-21 and 20 µl of protein G sepharose. Following three washes in Staph A buffer, immunoprecipitates were processed for immunoblotting as described above.
Determination of Soluble ErbB2 in Patient and Healthy Sera
Quantitation of ErbB2 extracellular domain in human sera was performed by an enzymatic immunocapture assay using two anti-ErbB2 mAbs (Bender MedSystems, Boehringer Ingelheim Group, Vienna, Austria). According to manufacturer’s recommendation, 96-well microtiter plates coated with anti-ErbB2, were incubated with human sera (diluted 1:20) along with ErbB2 extracellular domain protein standards. Bound ErbB2 protein was revealed by a second anti-ErbB2 mAb conjugated to horseradish peroxidase. Absorbance was measured at 450 nm, and ErbB2 serum concentrations were extrapolated from the standard curve. Serum levels above 6 ng/ml were considered positive, as indicated by the manufacturer.
Epstein-Barr Virus (EBV) Transformation of Patient Peripheral Blood Leukocytes (PBL)
Heparinized blood was collected before adjuvant therapy from breast cancer patients with informed consent and diluted with equal volume of RPMI 1640 medium (Gibco BRL). Human PBL were isolated by density gradient separation through Ficoll/Hypaque (Pharmacia, Uppsala, Sweden) and washed four times in PBS. Approximately 3 x 106 cells were immediately processed for transformation; the rest was frozen and stored in liquid nitrogen as previously described.29 For transformation, PBL were pelleted and resuspended in 1 ml of RPMI 1640 medium. Two hundred microliters of concentrated EBV suspension derived from marmoset leukocyte supernatant were added and incubated for 2 hours at 37°C. Cells were washed twice in RPMI 1640 medium, and the pellet was resuspended at 5 x 104 cells/ml in Yssel’s modified Iscove’s medium supplemented with 15% heat-inactivated FCS, 2 mmol/L L-glutamine, and 50 µg/ml gentamicin. EBV-transformed PBL were mixed with irradiated (7000 rad) Ltk-L transfectants (105/ml) recombinantly expressing the human ligand for CD40 (CD40 L).29,30 One hundred microliters of cell mixture were then seeded and cultured in growth medium. At days 12 and 19, culture supernatants were tested for the presence of ErbB2 antibodies by enzyme-linked immunosorbent assay (ELISA). Following expansion and subcloning at ~10 cells/well, supernatants were characterized by ELISA and Western blotting.
Detection of ErbB2-Reactive Immunoglobulins by ELISA
LTR-ErbB2 and NIH3T3 cells were seeded at 5 x 105 cells/well on 96-well plates and allowed to adhere overnight for detection of anti-ErbB2 immunoglobulins by ELISA. Nonspecific binding was blocked by incubation with 5% bovine serum albumin (BSA) in phosphate buffered saline (PBS) for 1 hour. Subsequently, cells were incubated with undiluted tissue culture supernatant for 1 hour at 37°C. After washing, horseradish peroxidase-conjugated goat anti-human IgG+IgM (Gibco BRL) was added, and bound antibody detected with o-phenyl-diamine dihydrochloride (OPD; Sigma Chemical Co., St. Louis, MO). The reaction was stopped by addition of 25 µl of 4N H2SO4, and the absorbance was measured at 492 nm with an automatic ELISA reader.
Statistical Analysis
Statistical significance of associations between anti-ErbB2
humoral response, shed receptor, and clinicopathological parameters was
evaluated using Fisher’s exact test. P values 
0.05
were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Employing a recombinant expression model, sera of 71 breast cancer patients and 31 healthy donors were screened by immunoblotting for specific immunoreactivity to human ErbB2 essentially as reported.22 Analysis included 13 breast cancer and 21 control donor sera, which had previously been analyzed.22 Native anti-ErbB2 receptor antibodies as detected by immunoblot analysis were observed in 14 of 71 (20%) breast cancer patients and 1 of 31 controls, the latter observation confirming the known positivity of a single normal donor serum. Statistical analysis demonstrated significant association of an anti-ErbB2 serum response with breast cancer patients as compared to normal donor sera (P = 0.0342).
Detection specificity of ErbB2 by direct immunoblotting was further
substantiated on ErbB2 immunoprecipitates for 9 positive and 2 negative
human sera. Figure 1
representatively
illustrates specific detection of an anti-ErbB2 response in human serum
by direct immunoblotting (Figure 1A)
or following immunoprecipitation
(Figure 1B)
. Identical results were obtained by both procedures in the
11 cases simultaneously subjected to immunoprecipitation, confirming
adequate specificity and sensitivity of detection. Development of
separate immunoblots using IgM- and IgG-specific secondary antibodies
revealed IgM-specific anti-ErbB2 immunoglobulins in 3 patients, while
all positive sera contained ErbB2-specific antibodies of the IgG
isotype. The latter observation implicated T helper cell activity in
the development of an anti-ErbB2 response.
|
In direct serum analysis, an existent anti-ErbB2 response could
have escaped detection due to a low serum titer in vivo. In
search of independent confirmation for existent ErbB2 immunity, we
screened patients’ B lymphocytes for evidence of ErbB2-specific
priming. To this purpose, peripheral blood leukocytes (PBL) were
isolated from heparinized blood, activated by CD40 ligand and
immortalized by EBV in vitro. PBL of 15 patients, including
7 positive and 8 negative for anti-ErbB2 serum antibodies, were
selected for analysis. Tissue culture supernatants of at least 3
x 106
transformed PBL from each patient were
tested for ErbB2-specific immunoreactivity by ELISA on LTR-ErbB2 and
NIH3T3 negative controls. Supernatants from B lymphocytes with the
relatively highest ErbB2-specific ELISA values (Table 1)
were then subjected to immunoblotting
using the recombinant expression model. Figure 2
depicts immunoblot analysis of two
positive and two negative cases. Supernatants of transformed PBL from
patients 1 and 2 recognized a single ~185-kd protein species in
LTR-ErbB2, but not control NIH3T3, comigrating with ErbB2 protein
visualized by a receptor-specific antiserum. By contrast, PBL
supernatants of patients 10 through 15 were negative in ErbB2 detection
by ELISA and immunoblot analysis indicating that an ErbB2 response was
not a general consequence of in vitro manipulation, but
rather reflected ErbB2 immunological properties inherent to PBL
in vivo (Table 1)
|
|
lists for each patient the maximal ErbB2-specific ELISA value
determined from supernatants of immortalized B cells and reactivity of
these supernatants in immunoblot analysis in comparison with native
ErbB2 antibody serum response in vivo. ErbB2-specific
immunoreactivity by Western blot analysis was observed in supernatants
of B cells derived from 7 patients, whereas 8 patients scored negative
(Table 1)
. Thirteen of 15 patients, including 6 positive and 7 negative
for ErbB2 serum antibodies, scored similarly in direct serum analysis
and following PBL activation in vitro. Furthermore,
activated PBL of patient 6 displayed ErbB2 reactivity in the absence of
detectable ErbB2 serum antibodies implying that antibody titer in
vivo was below the threshold of detection. On the other hand, PBL
supernatant of patient 9 exhibiting ErbB2 serum antibodies in
vivo, failed to reveal ErbB2 reactivity by immunoblotting (Figure 2)
. Given the relatively low albeit specific reactivity in ELISA (Table 1)
, it is conceivable that antibody titer of this PBL supernatant was
below detection in immunoblot analysis. All of these observations
suggested rather sensitive detection of ErbB2 serum antibodies in
vivo, as merely 1 of 8 serum-negative patients revealed ErbB2
immunity following in vitro activation of her lymphocytes.
Moreover, in vitro activation of PBL independently
confirmed a natural ErbB2-specific immune response in the majority of
these patients. Comparison of ErbB2 Immune Response and Clinical Disease Parameters
To explore whether an ErbB2 immune response in vivo was
associated with a particular disease condition, correlation with
clinical disease characteristics were evaluated including tumor
histology, grade, stage, lymph node, steroid receptor and menopausal
status. ErbB2 immunoreactivity was preferentially detected in ductal
breast carcinoma representing the majority of cases
(n = 62) under analysis. No significant
association was observed with grade of differentiation (G1-G3) taking
into account that only few cases of G1 (n = 3)
were analyzed. Furthermore, there was no significant correlation
between ErbB2 immune response and steroid receptor or menopausal
status. While detected at all stages of breast carcinoma, ErbB2
immunity correlated with disease progression as reflected by lymph node
status and disease stage (Table 2)
. Three
of 45 lymph node-negative patients and 12 of 26 lymph node-positive
patients revealed anti-ErbB2 serum antibodies indicating a
significantly higher prevalence following lymph node metastasis
(P = 0.0002). Similarly, 1 positive of 28 stage
I tumor patients as compared to 14 of 43 stage II-IV patients
demonstrated significant association of ErbB2 serum antibodies with an
advanced disease stage (P = 0.0030). Statistical
significance increased to P < 0.0001 when comparing
frequency of ErbB2 immune responses in stages I and II
(n = 58) versus stages III and IV
(n = 13) tumor patients, thus suggesting an
association with overall disease progression rather than lymph node
status itself (Table 2)
.
|
Experimental animal models established that serum concentration of
soluble ErbB2 extracellular domain correlates with tumor burden of
ErbB2-overexpressing malignancies.11
We therefore
quantified soluble ErbB2 extracellular domain in serum of all patients
and normal donors using a commercial double-sandwich ELISA.
Considering a cutoff threshold of 6 ng/ml as recommended by the
manufacturer, 16 of 71 breast cancer patients revealed elevated serum
levels of ErbB2 ectodomain. In addition, 1 of 31 normal donor sera
scored marginally positive (6.5 ng/ml). Prevalence of soluble ErbB2 in
breast tumor patients was significant when compared to normal donors
(P = 0.019). Mean soluble ErbB2 serum levels
were significantly higher in metastatic disease when compared to stage
I patients or normal donors (P < 0.0001)
projecting a trend for increase of circulating ErbB2 antigen level with
disease progression (Figure 3A)
.
|
, 12 of 15 patients with in vivo
or in vitro evidence for existent ErbB2 immunity harbored
elevated serum levels for soluble ErbB2
(ab+/ag+) ranging from 7.0
to 58.4 ng/ml (mean, 14.8 ng/ml). By contrast, only 4 of 56
antibody-negative sera displayed elevated ErbB2 antigen levels
(ab-/ag+) in the range of
6.2 to 7.6 ng/ml. Conversely, 3 of 55 patients negative for soluble
ErbB2 provided evidence for existent ErbB2 immunity
(ab+/ag-), whereas 52
patients lacked both ErbB2 serum antibodies and circulating
extracellular domain altogether
(ab-/ag-). None of 31
normal donors exhibited simultaneous positivity of ErbB2 serum
antibodies and soluble ErbB2, rendering an association highly selective
for tumor patients (P < 0.0001). Concurrence of
ErbB2 immunity and shed ectodomain associated with metastatic disease
(P < 0.0001) and in addition to the individual
parameters correlated with grade of differentiation (G2,
P = 0.0133; G3, P < 0.0001).
To assess a direct association of ErbB2 immunity and release of
ectodomain in individual tumor patients, we applied Fisher’s exact
test. As shown in Table 3
, existent ErbB2
immunity significantly coincided with elevated levels of ErbB2
ectodomain, yielding a two-sided P value of <0.0001 (Table 3)
. The observed correlation between ErbB2 serum immunity and level of
soluble extracellular domain supports the possibility that ErbB2
fragmentation and/or solubilization play a role in the development of
an anti-ErbB2 response in breast cancer patients. Alternatively,
concurrent segregation with an advanced disease condition might reflect
a similar dependence on tumor burden of both ErbB2 serum immunity as
well as release of soluble ErbB2 extracellular domain.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Evidence derived from animal models and human breast cancer implicates level of ErbB2 overexpression, tumor aggression, and tumor size in the detection of soluble ErbB2 extracellular domain.11-18 Accordingly, serum levels of shed ErbB2 ectodomain can serve as indirect markers of tumor burden in ErbB2-positive malignancy. Determination of circulating ErbB2 ectodomain in the present study confirms previous observations in terms of its frequency, amount, and prevalence in advanced metastatic breast cancer.12-18 Statistical significance of an association with the presence of ErbB2 immunity supports the possibility that the amount and/or fragmentation of normal ErbB2 protein may play a role in the development of a syngeneic immune response in breast cancer patients. In this case, ErbB2 immunity is not necessarily directed against its extracellular domain. Detection of elevated ectodomain levels in serum might merely reflect overall ErbB2 self-antigen load of the organism as well as the degree of proteolytic ErbB2 fragmentation owing to growth rate, necrosis, and cell degeneration. Antigen fragmentation could be critical, since ErbB2 peptides, but not full length protein, gave rise to self-antigen immunity in animals.25
ErbB2 gene amplification and overexpression represent early molecular lesions of spontaneous mammary carcinogenesis maintained during disease progression. Consequently, a similar incidence of these aberrations subsists in primary and metastatic tumors.5,9,31 On the contrary, both ErbB2 immune response and shed extracellular domain correlate with a progressed disease state, inferring a potential utility of these parameters for disease monitoring and design of immunotherapeutic approaches in ErbB2-associated malignancy. Particular relevance can be envisaged in the context of passive immunotherapy trials of advanced metastatic breast cancer involving humanized monoclonal antibodies targeting the extracellular domain of ErbB2.
Several studies have previously established ErbB2-specific immune responses in breast cancer patients.19-22 One prior study performed systematic analysis of different clinical stages and reported 2 of 26 stage I, 10 of 69 stage II, and none of 12 stage IV patients positive.21 Observations from the present study are comparable for stages I and II, with 1 of 28 and 4 of 30 patients displaying ErbB2-specific immunity, respectively. However, 5 of 8 stage IV patients scored positive here, in clear contrast to the preceding observation.21 Although small patient numbers may account for some differences, alternative possibilities cannot be excluded. ErbB2 antigen source in the prior study was SK-BR-3 in an ELISA capture assay, whereas we used denatured human recombinant ErbB2 protein expressed by NIH3T3 cells. SK-BR-3 in addition to an overexpressed ErbB2 protein harbors elevated ErbB3 and EGFR levels engaging ErbB2 in functionally active heterodimers.28,32 Thus, under conditions of incomplete denaturation certain ErbB2 epitopes might be masked. Furthermore, soluble ErbB2 extracellular domain levels increase with disease progression, as reported here and demonstrated previously,12-18 eventually neutralizing certain ErbB2 antibodies present in serum. This possibility rose with increasing disease stage and was more likely to occur under nondenaturing assay conditions.
In view of the possibility that high levels of ErbB2 ectodomain biased detection of ErbB2 immunity in serum analysis, we sought independent confirmation of existing ErbB2 immune responses by immortalization of PBL. Analysis included all 8 patients of stage IV, confirming the results of serum analysis in 7 patients including 3 negative and 4 positive cases. In addition, one patient (patient 6, stage IV) negative for ErbB2 serum antibodies in Western blot analysis had ErbB2-specific immunity associated with her B lymphocytes. It has not been resolved whether low serum titer of ErbB2 antibodies or the elevated ErbB2 ectodomain serum levels observed in this patient, precluded detection of an ErbB2 response in direct serum analysis. In the majority of cases, however, in vitro analysis of immortalized B lymphocytes confirmed the ErbB2 immune status determined in serum. Thirteen of 15 patients, including 6 positive and 7 negative for ErbB2 immunity, showed identical results on in vivo and in vitro analysis, whereas only 2 of 15 patients differed. This observation independently confirmed that in the present study sensitivity and specificity were adequate for serum detection of an existent ErbB2 immune response in vivo.
mAbs directed against the ErbB2 extracellular domain have proven effective in preventing or reverting an ErbB2-mediated neoplastic phenotype in vitro and in animal models.33-35 Furthermore, a humanized anti-ErbB2 mAb has prompted objective antitumor effects in some patients with ErbB2-associated metastatic breast cancer.7,8 In light of these encouraging results from clinical trial studies, it appears worthwhile to elucidate distinct pharmacodynamics affecting the other patients, despite comparable ErbB2 expression levels and similar pharmacokinetics of the monoclonal antibody.7,8 On one hand, ErbB2 activity state in distinct tumors might differ owing to heterodimer formation with other ErbB receptors.6,28,36 In this case, a heterodimerizing ErbB receptor or ligand may occlude the binding site for the monoclonal antibody in vivo. On the other hand, there is evidence that pre-existing immunity profiles to ErbB receptors are heterogeneous among patients harboring tumors with ErbB2 overexpression.22 Although soluble ErbB2 extracellular domain might interfere with pharmacokinetics in passive immunotherapy,7 investigation of pre-existent ErbB2 immunity allows assessment of whether a clinical response to ErbB2-mediated immunotherapy is predictable based on endogenous immunity. More importantly, given the heterogeneity of pre-existent anti-ErbB responses in ErbB2-positive tumor patients,22 definition of such responses in breast cancer patients may help to determine whether passive immunotherapy can be improved by either mimicking or avoiding a specific ErbB receptor self-immunity profile of individual patients.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by grants from the Ministero dell’Universita’ e della Ricerca Scientifica (MURST), the Consiglio Nazionale delle Ricerche (project CNR-ACRO), the Associazione Italiana per la Ricerca sul Cancro (AIRC), and the European Community (Program Biomed II).
Accepted for publication December 13, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Robin, J. Andreu-Gallien, M.-H. Schlageter, D. Bengoufa, I. Guillemot, K. Pokorna, C. Robert, J. Larghero, P. Rousselot, E. Raffoux, et al. Frequent antibody production against RAR{alpha} in both APL mice and patients Blood, September 15, 2006; 108(6): 1972 - 1974. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Streckfus and L. Bigler The Use of Soluble, Salivary c-erbB-2 for the Detection and Post-operative Follow-up of Breast Cancer in Women: The Results of a Five-year Translational Research Study Adv. Dent. Res., June 1, 2005; 18(1): 17 - 24. [Full Text] [PDF]
|
|
|
|
 |
|
 W. P. Carney, R. Neumann, A. Lipton, K. Leitzel, S. Ali, and C. P. Price Potential Clinical Utility of Serum HER-2/neu Oncoprotein Concentrations in Patients with Breast Cancer Clin. Chem., October 1, 2003; 49(10): 1579 - 1598. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Carter, D. C. Dillon, L. D. Reynolds, M. W. Retter, G. Fanger, D. A. Molesh, P. R. Sleath, P. D. McNeill, T. S. Vedvick, S. G. Reed, et al. Serum Antibodies to Lipophilin B Detected in Late Stage Breast Cancer Patients Clin. Cancer Res., February 1, 2003; 9(2): 749 - 754. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Piechocki, S. A. Pilon, and W.-Z. Wei Complementary Antitumor Immunity Induced by Plasmid DNA Encoding Secreted and Cytoplasmic Human ErbB-2 J. Immunol., September 15, 2001; 167(6): 3367 - 3374. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-ErbB2 and pre-immune serum, patient serum, and normal human
donor serum was conducted on 100 µg lysate from ErbB2 transfectants
and control NIH3T3
(A) and after
immunoprecipitation of LTR-ErbB2 using ErbB2-specific mAb E2–1 or
nonimmune MOPC-21
(B).
