The aim of this study was to evaluate c-erbB-2 overexpression by means of a quantitative biochemical technique in 488 primary breast cancer patients with long-term follow-up (median,10 years) and its relation to other biochemical prognostic factors(uPA, p53, and epidermal growth factor receptor) and adjuvant therapy.

High levels of c-erbB-2 (>500 IU/mg protein) were associated with estrogen receptor (ER) and progesterone receptor negativity, high histoprognostic SBR grade and high levels of uPA and p53. Univariate analyses showed shorter metastasis-free survival (MFS) and overall survival (OS) in patients whose tumors overexpressed c-erbB-2 in the overall population, in subgroups defined by ER and uPA status, and in patients with positive pathological nodal status, SBR grade II, progesterone receptor,and p53-negative tumors. Patients with ER-positive,c-erbB-2-positive tumors had a shorter MFS and OS than those patients with c-erbB-2-negative tumors. No difference was observed between adjuvant-treated and untreated patients(chemotherapy and/or hormone therapy) in the c-erbB-2-negative subgroup. There was a trend toward a longer short-term MFS in c-erbB-2-positive patients treated with chemotherapy, whereas an opposite effect was observed with hormone therapy.

Cox multivariate analyses showed that high levels of c-erbB-2 negatively influenced MFS in the overall population as well as in node-positive patients and in tamoxifen-treated patients, along with pN and uPA. Results for OS were comparable with those obtained for MFS. These results suggest that c-erbB-2 overexpression in breast cancer may be a better predictor of the response to tamoxifen than is ER status alone.

Breast cancer is a heterogeneous disease with highly variable biological and clinical behavior. Besides the classical prognostic factors used in clinical practice, c-erbB-2 has generated numerous studies on its predictive value for survival and on response to adjuvant therapy in breast cancer.

The techniques used to assess c-erbB-2 status in breast cancer have included gene-based assays such as Southern blotting,PCR-based methods, in situ hybridization (1),and, more recently, real-time reverse transcription-PCR (2). Qualitative and quantitative c-erbB-2 protein measurements have been performed by means of IHC3(on frozen and archived tissues), Western blotting, and ELISAs (1, 3). Most published data on c-erbB-2 were obtained with semiquantitative immunohistochemical methods on paraffin-embedded material. The results of breast cancer outcome studies have tended to conflict, and discrepancies between the results of different methods for detecting c-erbB-2 abnormalities have been described (3). Nonetheless, nearly all studies reported, to date, have shown a strong negative relationship between c-erbB-2 and steroid receptor status. In most studies, a significantly worse prognosis was noted in c-erbB-2-positive patients, whatever the treatment, suggesting that c-erbB-2 could be a marker of the treatment response (1, 3, 4). The aim of this study was to assess c-erbB-2 overexpression by using a quantitative biochemical technique in 488 primary breast cancer patients with long-term follow-up, and its relation to adjuvant therapy and other biochemical prognostic factors (c-erbB-2, uPA,p53, and EGFR). All these parameters were determined on surgical samples obtained for ER and PR assays.

Patients.

The 488 patients in this study were selected from a prospectively updated database and form a well representative group of breast cancer patients who met in our institution. At that time, the following criteria were known: female; primary, operable, unilateral breast cancer with no metastases and other primary cancers at the diagnosis;assay of the primary tumor for ERs and PRs; and sufficient tumor material for assays of biological factors. Surgery took place between 1980 and 1985 in 20% of cases and between 1986 and 1990 in the remaining 80%. All of the patients were treated at Centre RenéHuguenin de Lutte Contre Le Cancer. The median follow-up at the time of this study was 10 years (mean, 9.5; maximum, 18.0). Two hundred sixty patients underwent modified radical mastectomy, and 228 (46.5%)underwent partial mastectomy, both with lymph node clearance. Postoperative radiation was given to 211 patients (43.2%) as part of their locoregional treatment. Among the 226 node-negative patients, 59 received adjuvant chemotherapy alone, and 11 patients received tamoxifen, either alone (6 patients) or in combination with chemotherapy (5 patients). Chemotherapy for node-negative patients consisted of one course of FAC. Among the 262 node-positive patients,58 received adjuvant chemotherapy alone, consisting of various combinations of CMF and FAC. One hundred eighty-eight patients received tamoxifen, either alone (108 patients) or in combination with chemotherapy (80 patients). Tamoxifen (20 mg/day) was given for at least 2 years. The tamoxifen treatment decision was based on menopausal status rather than ER status. All of the patients underwent clinical,radiological, and biological examinations every 3 months for the first 2 years and yearly thereafter. At the time of this analysis (June 2000), 140 patients had had distant metastases, 68 had had local recurrences, and 113 had died of breast cancer.

Evaluation of Prognostic Factors.

Tumor sampling was always performed by a pathologist, and fat and necrosis were removed. Samples were taken opposite those frozen for intraoperative histological diagnosis and were immediately frozen in liquid nitrogen. The percentage of tumor cells was always checked by the imprint method, and samples containing <20% of tumor cells were eliminated.

The clinical size of the tumor was determined at diagnosis. At surgery,macroscopic tumor size was established and defined as the largest tumor diameter measured by the pathologist; the histological grade and hormone receptor status of the tumor, as well as the number of positive axillary lymph nodes, were also recorded at that time. Histological grading was performed according to Bloom and Richardson (5) and Scarff and Torloni (6).

ERs and PRs were assayed until 1988 by using the Dextran-coated charcoal method (7), and thereafter using a commercial immunoenzymatic method (Abbott ER-EIA-monoclonal kit and Abbott PgR-EIA-monoclonal kit; Abbott Laboratories, North Chicago, IL). The detection cutoff was set at 10 fmol/mg protein until 1988 and then at 15 fmol/mg protein. Quality control was based on regular testing of both internal controls (pooled cytosols) and European Organization for Research and Treatment of Cancer controls (8, 9). For this study, frozen tumor specimens (mean weight, 0.22 g) were homogenized in hormone receptor buffer [10 mm Tris-HCl (pH 7.4), 1.5 mm EDTA, 10 mmNa2MoO4, 0.5 mmdithiotreitol, and 10% glycerol] and centrifuged for 1 h at 105,000 × g. The supernatants were aliquoted and stored at −80°C until analysis (p53 and uPA). The pellets were homogenized in Triton buffer containing 20 mmTris-HCl, 125 mm NaCl, 1% Triton X-100 (pH 8.8),and left for 15 min at 0°C before centrifugation (100,000 × g for 1 h). The supernatants, consisting of membrane extracts, were collected and stored at −80°C until analysis(c-erbB-2 and EGFR). For each parameter, internal controls(pooled cytosols or membrane extracts) were used in each series of tests. Cytosol and membrane extract protein concentrations were determined using the BCA assay (Pierce Chemical Co., Rockford,IL). All assays were done in duplicate.

Enzyme Immunoassay for c-erbB-2.

The Triton c-erbB-2 Tissue Extract EIA kit (Ciba Corning Diagnostics, Alameda, CA) is a double monoclonal antibody-based assay. The two monoclonal antibodies, designated TAB 257 and TAB 259, are specific for different epitopes of the external domain of the c-erbB-2 molecule (10).

Assays for uPA, p53, and EGFR.

uPA and p53 in cytosol were assayed by using a luminometric immunoassay(LIA-mat uPA and LIA-mat p53; Byk Sangtec Diagnostica, Bromma, Sweden)based on a monoclonal antibody two-site incubation (11). The Triton EGFR Tissue Extract EIA kit (Ciba Corning Diagnostics) is a double monoclonal antibody-based assay. The two monoclonal antibodies used in this assay are specific for the extracellular domains of the protein (12).

Statistical Methods.

MFS was defined as the time between diagnosis and the occurrence of the first distant metastasis or the end of the study. Patients who died of causes unrelated to breast cancer were considered as censored at the time of death. Hereafter, “death” refers to breast cancer-related death for OS.

Differences in the distribution of characteristics among patient subgroups were analyzed using the χ2 test. The optimal cutoff for dichotomized variables to discriminate low-risk and high-risk patients were determined using log-rank statistics corrected according to Hilsenbeck and Clark (13). Actuarial MFS and OS rates were computed using the Kaplan-Meier method (14)and compared using the log-rank test (15). A stepwise selection procedure based on the Cox proportional hazards model (16) was used to assess the relative importance of the prognostic factors. The multivariate analysis took into account all of the variables cited in Table 1 and also adjuvant hormone therapy and adjuvant chemotherapy.

Clinical, histological and biological characteristics of the overall population are shown in Table 1. The optimal cutoff value for c-erbB-2 was optimized at 500 IU/mg protein, which selected 13.3% of the patients as high risk. The mean c-erbB-2 level was 327 (+/−433.1) IU/mg protein (range, 0–2640 IU/mg protein), and the median value was 170 IU/mg protein.

Relationship between c-erbB-2 and the Other Variables.

The relationships between c-erbB-2 and classical and biological variables are presented in Table 1. c-erbB-2 status was not related to age, clinical or histological tumor size,pathological nodal, or EGFR status. ER-negative tumors were more likely to be c-erbB-2 positive than were ER-positive tumors (28.1%versus 8%). Similarly, PR-negative tumors were more likely to be c-erbB-2 positive than were PR-positive tumors (21%versus 7.5%). High levels of c-erbB-2 correlated with high histological SBR grade, elevated uPA, and, to a lesser extent, high levels of p53.

Clinical Variables.

Shorter MFS was observed in patients with large tumors (clinical and histological tumor sizes) compared with those with small tumors(P < 0.001). Likewise, MFS was shorter in patients with high SBR grade (P < 0.001). Patients with more than three involved axillary lymph nodes had shorter MFS than those with three or fewer involved axillary lymph nodes (P <0.001). Log rank test analyses of OS gave results similar to those of MFS.

Biological Variables.

ER status did not influence MFS. Patients with negative PR had lower MFS (P < 0.01) than did those with positive PR status. Shorter MFS was observed in patients with high levels of uPA than in those with low levels of uPA (P < 0.0001). Shorter MFS was observed in patients whose tumors contained high concentrations of p53 (P < 0.01). With respect to EGFR, patients whose tumors contained low or high levels had shorter MFS than did those with intermediate levels (P <0.01). Log rank test analyses of OS gave results similar to those of MFS, except for ER-negative patients who had shorter OS than ER-positive patients (P = 0.0043).

c-erbB-2 in the Overall Population.

Patients whose tumors contained high levels of c-erbB-2 had lower MFS (P < 0.0005) than did those with low levels of c-erbB-2 (Fig. 1). The 5-year MFS rate was 62.6% in patients with high c-erbB-2 levels and 84.2% in patients with low c-erbB-2 levels. Patients whose tumors contained high c-erbB-2 levels had shorter OS than those with low c-erbB-2 levels (P < 0.0001): the respective 5-year OS rates were 67.7% and 88.4%.

c-erbB-2 in Subgroups Defined by Classical Prognostic Factors.

We studied combinations of known prognostic factors and c-erbB-2 status for their influence on MFS and OS. The 5-year MFS and OS rates and 95% CIs are presented in Table 2.

Age and c-erbB-2 Status.

Shorter MFS and OS were observed in patients of >50 years of age whose tumors overexpressed c-erbB-2 than in those with low c-erbB-2 expression (P < 0.001). Among patients under 50 years of age, no significant differences in MFS and OS were observed between those with low and high c-erbB-2 expression.

Pathological Tumor Size and c-erbB-2 Status.

For MFS and OS, c-erbB-2 status added prognostic information to large tumor size (>20 mm; P < 0.001), but not to small tumor size.

Lymph Node Status and c-erbB-2 Status.

Among the 226 node-negative patients, MFS was not significantly different between those with low c-erbB-2 expression and those with high c-erbB-2 expression, whereas slightly shorter OS was observed in patients with high c-erbB-2 expression (P = 0.032). In the node-positive group, MFS and OS were significantly longer in patients with low c-erbB-2 expression than in those with high c-erbB-2 expression (P < 0.001).

Histological Grade and c-erbB-2 Status.

For MFS and OS, c-erbB-2 added prognostic information to SBR grade II patients (P = 0.035 and 0.004, respectively). For OS, c-erbB-2 added also prognostic information to SBR grade III patients (P = 0.018).

Hormone Receptor and c-erbB-2 Status.

c-erbB-2 added prognostic value to ER status (Fig. 2,a and b). Among ER-positive patients, we observed a lower MFS and OS in c-erbB-2-positive patients than in c-erbB-2-negative patients (P =0.025 and 0.00063, respectively). MFS and OS were significantly shorter in ER-negative, c-erbB-2-positive patients than in ER-negative, c-erbB-2-negative patients (P =0.04 and 0.03, respectively). Among PR-positive patients, no statistical difference was observed for MFS, whereas a shorter OS was observed for c-erbB-2-positive patients (P =0.005). Among PR-negative patients, MFS and OS were shorter in c-erbB-2-positive patients than in c-erbB-2-negative patients (P = 0.02 and 0.0054, respectively).

uPA and c-erbB-2 Status.

c-erbB-2 status added prognostic value to uPA status (Fig. 3,a and b). Among patients with low levels of uPA,c-erbB-2-positive patients had shorter MFS and OS than c-erbB-2-negative patients (P = 0.04 and 0.0096, respectively). Among patients with high levels of uPA (>1.14 ng/mg protein), MFS and OS were shorter in c-erbB-2-positive patients than in c-erbB-2-negative patients(P = 0.02 and 0.0018, respectively).

p53 and c-erbB-2 Status.

c-erbB-2 status added prognostic value to low p53 values(Fig. 4,a), but not to high p53 values (Fig. 4 b). MFS and OS were significantly shorter in the low-p53 c-erbB-2-negative group than in the low-p53 c-erbB-2-positive group (P < 0.001). No statistical difference was observed in the high p53 patients group.

EGFR and c-erbB-2 Status.

For MFS and OS, high c-erbB-2 expression aggravated the prognosis of patients with intermediate EGFR expression(P < 0.001), whereas high c-erbB-2 expression had no effect on the prognostic value of low and high EGFR expression.

c-erbB-2 Status and Adjuvant Treatment.

Among patients with c-erbB-2-negative expression, MFS did not differ between those treated with adjuvant tamoxifen and those who did not receive tamoxifen (Fig. 5,a) or between those treated with chemotherapy and those who did not receive chemotherapy (Fig. 5,b).In patients’c-erbB-2-positive expression MFS curves (Fig. 5,c)showed an adverse effect of tamoxifen during the first 5 years of follow-up, but the difference with untreated patients was not significant. With regard to chemotherapy (Fig. 5 d), patients with c-erbB-2-positive expression benefited from the treatment during the first 5 years of follow-up relative to the untreated group.

Log rank test analyses of OS gave results similar to those of MFS. Among the ER-positive, c-erbB-2-positive subgroup tamoxifen seemed detrimental [P < 0.02 for MFS (Fig. 6,a) and P = 0.009 for OS], whereas no significant difference was observed in ER-positive,c-erbB-2-negative patients (Fig. 6 b).

All of the variables listed in Table 1 were tested in multivariate Cox regression analyses for their relationship with OS and MFS, both in the overall population and in subgroups defined by lymph node status(Table 3) and adjuvant therapy (Table 4).

Overall Population.

A higher uPA value, a larger number of positive lymph nodes, a larger tumor size, and a higher c-erbB-2 expression were significant independent predictors of shorter MFS and OS. For OS, 50 years of age or more was also an independent predictor of shorter OS.

Subgroup Defined by Nodal Status.

In node-negative patients, c-erbB-2 was not found to be an independent factor. For MFS, higher uPA values were the only adverse prognostic selected factor in this subgroup. Larger clinical tumor size was related to a shorter OS.

In node-positive patients, a higher uPA value, a larger number of positive lymph nodes, a larger tumor size, and higher c-erbB-2 expression were significant independent predictors of shorter MFS and OS. For OS, 50 years of age or more was also a predictor of shorter OS.

Patients Receiving Chemotherapy.

Multivariate analysis showed that higher uPA values, a larger tumor size, and a larger number of positive lymph nodes were adverse significant predictors of MFS and OS.

Patients Receiving Tamoxifen.

Multivariate analysis showed that a larger number of positive lymph nodes and higher c-erbB-2 expression were significant adverse predictors of MFS and OS. Higher uPA values were also predictive of shorter OS.

c-erbB-2 gene amplification occurs in about 15–25% of breast cancers, whereas the interstudy range of c-erbB-2 protein overexpression is higher (10–50%). Immunohistochemistry is the most widely used method for c-erbB-2 protein assay, with a variety of polyclonal and monoclonal antibodies and different cutoff values (1, 17). The range of c-erbB-2 protein overexpression falls to 15–20% when gene amplification is also present. For example,Robertson et al.(18), using a fluorescence in situ hybridization method, found that 91% of breast cancers overexpressed c-erbB-2 protein, whereas gene amplification was only observed in 21% of the patients with very high c-erbB-2 protein values. Similar findings have been made with real-time reverse transcription-PCR, which showed that c-erbB-2 overexpression at the mRNA level (17.2% of patients) was present in all cases of gene amplification but uncommon in the absence of gene amplification (2).

In this study we used a simple, reproducible biochemical assay, which is sensitive, quantitative, and appropriate for routine analysis. Another multicenter study using the same assay method showed a comparable c-erbB-2 distribution in terms of the median,geometric mean, and range (19). In the same study, very low c-erbB-2 values (defined as less than the geometric mean minus one SD) were suggested as a negative prognostic factor, but this was not confirmed (2, 20). The cutoff of 500 IU/mg protein was chosen on the basis of so-called “optimal cutoff” corrected as suggested by Hilsenbeck and Clark (13). The optimal cutoff used in our study identified a relatively low percentage of tumors overexpressing c-erbB-2 protein, but was consistent with the rate of c-erbB-2 gene amplification usually found in breast cancer. IHC was performed by using CB11 (Novocastra) monoclonal antibody on 20 paraffin-embedded tumors; we also found a strong correlation between c-erbB-2 expression measured by IHC and biochemical c-erbB-2 concentrations (data not shown).

We observed a strong correlation between c-erbB-2 protein overexpression and steroid receptor negativity, as generally described in the literature, with a variety of techniques (1). We also observed an association between c-erbB-2 overexpression and histoprognostic grade (SBR). Most previous studies have suggested that c-erbB-2 positivity is related to higher SBR grade (21, 22). Significantly higher c-erbB-2 values were also found in tumors with high p53 and uPA protein levels; several previous studies have shown a close correlation between c-erbB-2 and p53 positivity (23), whereas others have not (24). We found no published data on the possible link between c-erbB-2 and uPA.

High c-erbB-2 values were associated with significant shorter MFS and OS in the overall population (1, 3). Regarding MFS and OS, when we analyzed c-erbB-2 status in combination with clinical and biological prognostic factors, we found that c-erbB-2 overexpression added information to nodal status, tumor size, SBR grade, and ER. The same was true when c-erbB-2 was combined with uPA and EGFR. In contrast,c-erbB-2 status defined a group of patients with a poor prognosis among those usually considered to have good prognosis, such as patients with low p53 values. As shown by Sjörgen et al.(25), the combination of newer prognostic markers such as c-erbB-2 with conventional markers can identify new subgroups of patients with a poorer prognosis.

Multivariate analyses applied to the overall population and to node-positive patients identified c-erbB-2 overexpression associated with shorter MFS and OS along with positive nodal status,high levels of uPA, and larger tumor size. In node-negative patients,c-erbB-2 was not a significant predictor for any prognostic criterion for MFS and OS, in keeping with reports by Borg et al.(26) and Carlomagno et al.(27), and we confirmed the strong prognostic value of uPA (11). EGFR and p53 were not selected as major prognostic factors when all other clinical, histological, and biological variables were included in the Cox model.

The predictive value of c-erbB-2 in patients receiving hormone therapy is still unclear. Some studies indicate that c-erbB-2-positive tumors exhibit a poor response to hormone therapy (21, 26, 27, 28, 29). Carlomagno et al.(27) first studied the interaction between c-erbB-2 and hormone treatment in a randomized trial of adjuvant tamoxifen and showed that tamoxifen shortened survival in node-negative patients whose tumors overexpressed c-erbB-2. On the contrary, a recent study by Elledge et al.(30) in ER-positive metastatic breast cancer showed that c-erbB-2 overexpression was not associated with a poorer response to tamoxifen or a more aggressive clinical course. In our study, multivariate analysis showed that high levels of c-erbB-2 were associated with shorter MFS and OS in tamoxifen-treated patients (45.5% of whom were ER positive). Moreover,in tamoxifen-treated patients, when classical factors alone were tested by multivariate analysis (i.e., without c-erbB-2), ER was the first variable to be selected (data not shown). When c-erbB-2, uPA, EGFR, and p53 were also tested, c-erbB-2 replaced ER. These results suggest that c-erbB-2 overexpression may be better than ER status as a predictor of the response to tamoxifen and that the effect of c-erbB-2 may not be solely mediated by ER down-regulation. Indeed, tamoxifen seemed to be detrimental in the small group of ER-positive, c-erbB-2-positive patients when compared with their untreated counterparts.

Data on the relation between c-erbB-2 and the response to chemotherapy are conflicting. In some cases, c-erbB-2 overexpression is associated with resistance to chemotherapy consisting of conventional doses of cyclophosphamide, methotrexate, and fluorouracil (CMF regimen; Refs. 31 and 32). However, Muss et al.(33) showed that in patients treated with different doses (low, moderate, and high) of FAC adjuvant chemotherapy (fluorouracil, doxorubicine, and cyclophosphamide), c-erbB-2 overexpression was predictive of increased tumor sensitivity to high-dose FAC. The same authors (34) confirmed this finding in a second set of patients from the same trial. Paik et al.(35), in a retrospective study including patients with axillary lymph node-positive, hormone receptor-negative breast cancer randomly treated with either l-phenylalanine mustard plus 5-fluorouracil or a combination of l-phenylalanine mustard, 5-fluorouracil, and doxorubicine, observed a clinical benefit of doxorubicine(l-phenylalanine mustard, 5-fluorouracil, and doxorubicin versus 5-fluorouracil) in patients with c-erbB-2-positive tumors. In our study, c-erbB-2 was not selected as an independent factor in patients receiving different chemotherapy protocols, including various combinations of FAC and CMF. The most consistent feature of chemotherapy was the presence of 5-fluorouracil and an anthracycline, the dose of which varied. The only group receiving uniform chemotherapy in this study consisted of 61 node-negative patients who received one course of FAC. In this subgroup, a trend toward longer MFS was found in patients with high c-erbB-2 levels, but the difference was not statistically significant (data not shown). This is in keeping with the results of Muss et al.(33), Thor et al.(34), and Paik et al.(35), who observed a benefit when patients with c-erbB-2-positive tumors were treated with an anthracycline (30, 31, 32).

In conclusion, we confirm the absence of significance of c-erbB-2 expression in node-negative patients using multivariate analysis. Regarding hormone therapy (tamoxifen),c-erbB-2 overexpression provided additional information relative to ER. A combination of c-erbB-2 and biological prognostic factors could be of clinical value by defining subgroups that might benefit from more aggressive treatment, particularly in vaccine therapy approaches based on anti-c-erbB-2 antibodies. Randomized trials or trials stratified according to c-erbB-2 status are required to determine the precise predictive value of c-erbB-2 expression on the breast tumor response to classical therapy.

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.

        
1

The Ligue contre le cancer des Yvelines, de l’Essonne et des Hauts de Seine supported this work.

                
3

The abbreviations used are: IHC,immunohistochemistry; EGFR, epidermal growth factor receptor; ER,estrogen receptor; PR, progesterone receptor; CMF, cyclophosphamide,methotrexate, and 5-fluorouracil; MFS, metastasis-free survival; OS,overall survival; CI, confidence interval; FAC, fluorouracil,doxorubicine, and cyclophosphamide; SBR, Scarff, Bloom, and Richardson; EIA, enzyme immunoassay.

Table 1

Relationships between c-erbB-2 and classical variables (χ2 test)

VariablesNumber of patients (%)c-erbB-2 (IU/mg protein)P
≤500 (n = 423)>500 (n = 65)
Age <50 yr 119 (24.4) 99 20  
 ≥50 yr 369 (75.6) 324 45 nsa 
Clinical tumor size ≤25 mm 181 (73.1) 163 18  
 >25 mm 307 (62.9) 260 47 ns 
Histologic tumor size (pT) ≤20 mm 207 (42.4) 186 21  
 >20 mm 281 (57.6) 237 44 ns 
Lymph node status (pN) N = 0 226 (46.3) 200 26  
 1≤N≤3 175 (35.9) 154 21  
 N>3 87 (17.8) 69 18 ns 
SBR grade 47 (9.6) 46  
 II 304 (62.3) 270 34  
 III 137 (28.1) 107 30 <0.001 
Estrogen receptors (fmol/mg of protein) Negative 127 (26) 92 36  
 Positive 361 (74) 331 29 <0.001 
PRs (fmol/mg protein) Negative 207 (42.4) 165 44  
 Positive 281 (57.6) 258 21 <0.001 
uPA (ng/mg protein) ≤0.20 134 (27.5) 125  
 0.20<uPA≤1.14 304 (62.3) 263 41  
 >1.14 50 (10.2) 35 15 <0.001 
p53 (ng/ml protein) ≤0.37 351 (71.9) 311 40  
 >0.37 137 (28.1) 112 25 0.05 
EGFR (fmol/mg protein) ≤6 191 (39.1) 162 29  
 6<EGFR≤17.5 256 (52.5) 228 28  
 >17.5 41 (8.4) 33 ns 
VariablesNumber of patients (%)c-erbB-2 (IU/mg protein)P
≤500 (n = 423)>500 (n = 65)
Age <50 yr 119 (24.4) 99 20  
 ≥50 yr 369 (75.6) 324 45 nsa 
Clinical tumor size ≤25 mm 181 (73.1) 163 18  
 >25 mm 307 (62.9) 260 47 ns 
Histologic tumor size (pT) ≤20 mm 207 (42.4) 186 21  
 >20 mm 281 (57.6) 237 44 ns 
Lymph node status (pN) N = 0 226 (46.3) 200 26  
 1≤N≤3 175 (35.9) 154 21  
 N>3 87 (17.8) 69 18 ns 
SBR grade 47 (9.6) 46  
 II 304 (62.3) 270 34  
 III 137 (28.1) 107 30 <0.001 
Estrogen receptors (fmol/mg of protein) Negative 127 (26) 92 36  
 Positive 361 (74) 331 29 <0.001 
PRs (fmol/mg protein) Negative 207 (42.4) 165 44  
 Positive 281 (57.6) 258 21 <0.001 
uPA (ng/mg protein) ≤0.20 134 (27.5) 125  
 0.20<uPA≤1.14 304 (62.3) 263 41  
 >1.14 50 (10.2) 35 15 <0.001 
p53 (ng/ml protein) ≤0.37 351 (71.9) 311 40  
 >0.37 137 (28.1) 112 25 0.05 
EGFR (fmol/mg protein) ≤6 191 (39.1) 162 29  
 6<EGFR≤17.5 256 (52.5) 228 28  
 >17.5 41 (8.4) 33 ns 

a ns, not significant.

Fig. 1.

MFS according to c-erbB-2 status in the overall population (cutoff value, 500 IU/mg protein).

Fig. 1.

MFS according to c-erbB-2 status in the overall population (cutoff value, 500 IU/mg protein).

Close modal
Table 2

Comparison of c-erbB-2 expression with subgroups defined by classical prognostic factors: MFS and OS rates at 5 years

Variables5 year MFS rate [95% CI] c-erbB-2 (IU/mg protein)5 year OS rate [95% CI] c-erbB-2 (IU/mg protein)
≤500 (n = 423)>500 (n = 65)≤500 (n = 423)>500 (n = 65)
Age <50 yr 81.8% [73.1–81.9] 65.0% [43.3–81.9] 90.9% [83.6–95.1] 74.7% [52.6–88.7] 
 ≥50 yr 85.0% [80.6–88.5] 61.4% [46.6–74.3] 91.8% [88.3–94.3] 66.7% [52.1–78.6] 
Histologic tumor size (pT) ≤20 mm 90.2% [85.1–93.7] 81.0% [60.0–92.3] 96.2% [92.3–98.1] 95.2% [77.3–99.1] 
 >20 mm 79.5% [73.8–84.1] 53.7% [39.1–67.7] 88.0% [83.2–91.6] 56.8% [42.2–70.3] 
Lymph node status (pN) N = 0 87.9% [82.6–91.7] 69.2% [50.0–83.5] 94.4% [90.3–96.9] 80.6% [61.8–91.4] 
 1 ≤ N ≤ 3 86.8% [80.4–91.3] 66.7% [45.4–82.8] 94.7% [89.9–97.3] 76.2% [54.9–89.3] 
 N>3 67.6% [55.8–77.6] 46.0% [24.8–68.7] 76.3% [64.9–84.9] 44.4% [24.6–66.3] 
SBR grade 93.4% [82.4–97.7] a 100% a 
 II 86.1% [81.4–89.7] 70.6% [53.8–83.2] 93.6% [89.9–95.9] 73.5% [56.9–85.6] 
 III 75.5% [66.5–82.7] 51.1% [33.7–69.3] 83.0% [74.7–88.9] 63.3% [45.5–78.1] 
Estrogen receptors (fmol/mg of protein) Negative 79.1% [69.9–86.2] 60.4% [43.9–74.8] 84.6% [75.8–90.6] 61.1% [44.9–75.2] 
 Positive 85.6% [81.4–89.0] 62.5% [47.3–80.1] 93.6% [90.3–95.7] 79.2% [61.4–90.1] 
PRs (fmol/mg protein) Negative 77.7% [70.7–83.5] 58.2% [43.3–71.7] 87.6% [81.6–91.8] 63.6% [48.8–76.1] 
 Positive 88.3% [83.7–91.7] 71.4% [50.5–86.2] 94.1% [90.5–96.4] 81.0% [60.0–92.3] 
uPA (ng/mg protein) ≤1.14 86.4% [82.6–89.5] 71.7% [57.9–82.3] 92.9% [89.9–95.1] 82.0% [69.2–90.2] 
 >1.14 59.3% [42.7–74.0] 32.0% [14.0–57.7] 76.5% [60.0–87.6] 26.7% [10.9–51.9] 
p53 (ng/mg protein) ≤0.37 87.3% [83.1–90.5] 62.5% [47.0–75.8] 93.8% [90.5–96.0] 75.0% [59.8–85.8] 
 >0.37 75.7% [67.0–82.7] 63.5% [43.8–79.5] 85.6% [77.8–90.9] 60.0% [40.7–76.6] 
EGFR (fmol/mg protein) ≤6 78.8% [71.9–84.4] 68.4% [49.9–82.4] 88.8% [82.9–92.8] 69.0% [50.8–82.7] 
 6 < EGFR ≤ 17.5 90.2% [85.6–93.4] 60.0% [41.5–76.0] 94.6% [90.8–96.9] 75.0% [56.6–87.3] 
 >17.5 69.3% [51.2–82.4] 50.0% [21.5–78.5] 84.8% [69.1–93.3] 50.0% [21.5–78.5] 
Variables5 year MFS rate [95% CI] c-erbB-2 (IU/mg protein)5 year OS rate [95% CI] c-erbB-2 (IU/mg protein)
≤500 (n = 423)>500 (n = 65)≤500 (n = 423)>500 (n = 65)
Age <50 yr 81.8% [73.1–81.9] 65.0% [43.3–81.9] 90.9% [83.6–95.1] 74.7% [52.6–88.7] 
 ≥50 yr 85.0% [80.6–88.5] 61.4% [46.6–74.3] 91.8% [88.3–94.3] 66.7% [52.1–78.6] 
Histologic tumor size (pT) ≤20 mm 90.2% [85.1–93.7] 81.0% [60.0–92.3] 96.2% [92.3–98.1] 95.2% [77.3–99.1] 
 >20 mm 79.5% [73.8–84.1] 53.7% [39.1–67.7] 88.0% [83.2–91.6] 56.8% [42.2–70.3] 
Lymph node status (pN) N = 0 87.9% [82.6–91.7] 69.2% [50.0–83.5] 94.4% [90.3–96.9] 80.6% [61.8–91.4] 
 1 ≤ N ≤ 3 86.8% [80.4–91.3] 66.7% [45.4–82.8] 94.7% [89.9–97.3] 76.2% [54.9–89.3] 
 N>3 67.6% [55.8–77.6] 46.0% [24.8–68.7] 76.3% [64.9–84.9] 44.4% [24.6–66.3] 
SBR grade 93.4% [82.4–97.7] a 100% a 
 II 86.1% [81.4–89.7] 70.6% [53.8–83.2] 93.6% [89.9–95.9] 73.5% [56.9–85.6] 
 III 75.5% [66.5–82.7] 51.1% [33.7–69.3] 83.0% [74.7–88.9] 63.3% [45.5–78.1] 
Estrogen receptors (fmol/mg of protein) Negative 79.1% [69.9–86.2] 60.4% [43.9–74.8] 84.6% [75.8–90.6] 61.1% [44.9–75.2] 
 Positive 85.6% [81.4–89.0] 62.5% [47.3–80.1] 93.6% [90.3–95.7] 79.2% [61.4–90.1] 
PRs (fmol/mg protein) Negative 77.7% [70.7–83.5] 58.2% [43.3–71.7] 87.6% [81.6–91.8] 63.6% [48.8–76.1] 
 Positive 88.3% [83.7–91.7] 71.4% [50.5–86.2] 94.1% [90.5–96.4] 81.0% [60.0–92.3] 
uPA (ng/mg protein) ≤1.14 86.4% [82.6–89.5] 71.7% [57.9–82.3] 92.9% [89.9–95.1] 82.0% [69.2–90.2] 
 >1.14 59.3% [42.7–74.0] 32.0% [14.0–57.7] 76.5% [60.0–87.6] 26.7% [10.9–51.9] 
p53 (ng/mg protein) ≤0.37 87.3% [83.1–90.5] 62.5% [47.0–75.8] 93.8% [90.5–96.0] 75.0% [59.8–85.8] 
 >0.37 75.7% [67.0–82.7] 63.5% [43.8–79.5] 85.6% [77.8–90.9] 60.0% [40.7–76.6] 
EGFR (fmol/mg protein) ≤6 78.8% [71.9–84.4] 68.4% [49.9–82.4] 88.8% [82.9–92.8] 69.0% [50.8–82.7] 
 6 < EGFR ≤ 17.5 90.2% [85.6–93.4] 60.0% [41.5–76.0] 94.6% [90.8–96.9] 75.0% [56.6–87.3] 
 >17.5 69.3% [51.2–82.4] 50.0% [21.5–78.5] 84.8% [69.1–93.3] 50.0% [21.5–78.5] 

a Number of patients too small(n = 1).

Fig. 2.

MFS according to c-erbB-2 status in ER-positive (a; n = 361) and ER-negative (b; n = 127) breast cancer patients, using a cutoff of 500 IU/mg protein for c-erbB-2 overexpression.

Fig. 2.

MFS according to c-erbB-2 status in ER-positive (a; n = 361) and ER-negative (b; n = 127) breast cancer patients, using a cutoff of 500 IU/mg protein for c-erbB-2 overexpression.

Close modal
Fig. 3.

MFS according to c-erbB-2 status in breast cancer patients with low-uPA (a; ≤1.14 ng/mg protein; n = 438) and high uPA levels(b; n = 50), using a cutoff of 500 IU/mg protein for c-erbB-2 overexpression.

Fig. 3.

MFS according to c-erbB-2 status in breast cancer patients with low-uPA (a; ≤1.14 ng/mg protein; n = 438) and high uPA levels(b; n = 50), using a cutoff of 500 IU/mg protein for c-erbB-2 overexpression.

Close modal
Fig. 4.

MFS according to c-erbB-2 status in breast cancer patients with low-p53 (a; ≤0.37 ng/mg protein; n = 351) and high p53 levels(b; n = 137), using a cutoff of 500 IU/mg protein for c-erbB-2 overexpression.

Fig. 4.

MFS according to c-erbB-2 status in breast cancer patients with low-p53 (a; ≤0.37 ng/mg protein; n = 351) and high p53 levels(b; n = 137), using a cutoff of 500 IU/mg protein for c-erbB-2 overexpression.

Close modal
Fig. 5.

MFS according to adjuvant therapy: hormone therapy (tamoxifen) versus no hormone therapy in c-erbB-2-negative (a; ≤500 IU/mg protein) and in c-erbB-2-positive (c)breast cancer patients and chemotherapy versus no chemotherapy in c-erbB-2-negative (b;≤500 IU/mg protein) and in c-erbB-2-positive(d) breast cancer patients.

Fig. 5.

MFS according to adjuvant therapy: hormone therapy (tamoxifen) versus no hormone therapy in c-erbB-2-negative (a; ≤500 IU/mg protein) and in c-erbB-2-positive (c)breast cancer patients and chemotherapy versus no chemotherapy in c-erbB-2-negative (b;≤500 IU/mg protein) and in c-erbB-2-positive(d) breast cancer patients.

Close modal
Fig. 6.

MFS according to hormone therapy (tamoxifen; yes versus no) in ER-positive,c-erbB-2-negative (a) and in ER-positive,c-erbB-2-positive (b) breast cancer patients.

Fig. 6.

MFS according to hormone therapy (tamoxifen; yes versus no) in ER-positive,c-erbB-2-negative (a) and in ER-positive,c-erbB-2-positive (b) breast cancer patients.

Close modal
Table 3

Cox multivariate analyses of OS and MFS in the 488 patients and in subgroups defined by lymph node status

OutcomeGroupVariableCategoryPRRa95% CI
OS Overall population uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  2.19 1.64–2.94 
   >1.14  4.81 2.67–8.64 
  Nodal status <0.0001 1.00  
   [1–3]  1.68 1.37–2.07 
   >3  2.83 1.87–4.30 
  Clinical tumor size  0.00049 1.02 1.01–1.03 
  c-erbB-2 ≤500 0.002 1.00  
   >500  1.87 1.26–2.78 
  Age <50 0.016 1.00  
   ≥50  1.68 1.10–2.57 
 Node-negative patients uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  2.82 1.72–4.62 
   >1.14  7.94 2.94–21.37 
  Clinical tumor size  0.02 1.02 1.01–1.04 
 Node-positive patients Nodal status [1–3] <0.0001 1.00  
   >3  2.26 1.53–3.35 
  uPA ≤0.20 0.00048 1.00  
   ]0.20–1.14]  1.87 1.32–2.66 
   >1.14  3.50 1.73–7.08 
  Clinical tumor size  0.0085 1.02 1.00–1.03 
  Age <50 0.022 1.00  
   ≥50  1.93 1.10–3.40 
  c-erbB-2 ≤500 0.016 1.00  
   >500  1.82 1.12–2.95 
MFS Overall population uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  1.89 1.41–2.54 
   >1.14  3.59 1.99–6.46 
  Nodal status 0.00039 1.00  
   [1–3]  1.48 1.19–1.84 
   >3  2.19 1.42–3.37 
  Pathological tumor size ≤20 0.0039 1.00  
   >20  1.73 1.19–2.52 
  c-erbB-2 ≤500 0.02 1.00  
   >500  1.65 1.08–2.52 
 Node-negative patient uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  2.63 1.66–4.14 
   >1.14  6.88 2.76–17.17 
 Node-positive patients Nodal status [1–3] <0.0001 1.00  
   >3  2.49 1.63–3.81 
  Pathological tumor size ≤20 0.011 1.00  
   >20  2.06 1.18–3.60 
  uPA ≤0.20 0.021 1.00  
   ]0.20–1.14]  1.53 1.07–2.19 
   >1.14  2.34 1.14–4.80 
  c-erbB-2 ≤500 0.026 1.00  
   >500  1.79 1.07–2.99 
OutcomeGroupVariableCategoryPRRa95% CI
OS Overall population uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  2.19 1.64–2.94 
   >1.14  4.81 2.67–8.64 
  Nodal status <0.0001 1.00  
   [1–3]  1.68 1.37–2.07 
   >3  2.83 1.87–4.30 
  Clinical tumor size  0.00049 1.02 1.01–1.03 
  c-erbB-2 ≤500 0.002 1.00  
   >500  1.87 1.26–2.78 
  Age <50 0.016 1.00  
   ≥50  1.68 1.10–2.57 
 Node-negative patients uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  2.82 1.72–4.62 
   >1.14  7.94 2.94–21.37 
  Clinical tumor size  0.02 1.02 1.01–1.04 
 Node-positive patients Nodal status [1–3] <0.0001 1.00  
   >3  2.26 1.53–3.35 
  uPA ≤0.20 0.00048 1.00  
   ]0.20–1.14]  1.87 1.32–2.66 
   >1.14  3.50 1.73–7.08 
  Clinical tumor size  0.0085 1.02 1.00–1.03 
  Age <50 0.022 1.00  
   ≥50  1.93 1.10–3.40 
  c-erbB-2 ≤500 0.016 1.00  
   >500  1.82 1.12–2.95 
MFS Overall population uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  1.89 1.41–2.54 
   >1.14  3.59 1.99–6.46 
  Nodal status 0.00039 1.00  
   [1–3]  1.48 1.19–1.84 
   >3  2.19 1.42–3.37 
  Pathological tumor size ≤20 0.0039 1.00  
   >20  1.73 1.19–2.52 
  c-erbB-2 ≤500 0.02 1.00  
   >500  1.65 1.08–2.52 
 Node-negative patient uPA ≤0.20 <0.0001 1.00  
   ]0.20–1.14]  2.63 1.66–4.14 
   >1.14  6.88 2.76–17.17 
 Node-positive patients Nodal status [1–3] <0.0001 1.00  
   >3  2.49 1.63–3.81 
  Pathological tumor size ≤20 0.011 1.00  
   >20  2.06 1.18–3.60 
  uPA ≤0.20 0.021 1.00  
   ]0.20–1.14]  1.53 1.07–2.19 
   >1.14  2.34 1.14–4.80 
  c-erbB-2 ≤500 0.026 1.00  
   >500  1.79 1.07–2.99 

a RR, relative risk.

Table 4

Cox multivariate analyses of OS and MFS in patient subgroups defined by adjuvant therapy

OutcomeGroupVariableCategoryPRR95% CI
OS Chemotherapy uPA ≤0.20 0.00016 1.00  
   ]0.20–1.14]  2.73 1.62–4.59 
   >1.14  7.43 2.62–21.09 
  Clinical tumor size  0.000074 1.03 1.02–1.22 
  Nodal status 0.0019 1.00  
   [1–3]  1.85 1.26–2.72 
   >3  3.42 1.58–7.42 
 Hormone therapy Nodal status <0.0001 1.00  
   [1–3]  2.91 1.73–4.90 
   >3  8.47 2.98–24.05 
  c-erbB-2 ≤500 0.00012 1.00  
   >500  3.64 1.88–7.05 
  uPA ≤0.20 0.028 1.00  
   ]0.20–1.14]  1.72 1.06–2.78 
   >1.14  2.95 1.12–7.72 
MFS Chemotherapy uPA ≤0.20 0.00057 1.00  
   ]0.20–1.14]  2.20 1.41–3.45 
   >1.14  4.84 1.98–11.88 
  Clinical tumor size  0.0012 1.00 1.01–2.83 
  Nodal status 0.005 1.00  
   [1–3]  1.62 1.16–2.26 
   >3  2.62 1.34–5.13 
 Hormone therapy Nodal status 0.0013 1.00  
   [1–3]  2.19 1.36–3.52 
   >3  4.78 1.84–12.43 
  c-erbB-2 ≤500 0.021 1.00  
   >500  2.24 1.13–4.45 
OutcomeGroupVariableCategoryPRR95% CI
OS Chemotherapy uPA ≤0.20 0.00016 1.00  
   ]0.20–1.14]  2.73 1.62–4.59 
   >1.14  7.43 2.62–21.09 
  Clinical tumor size  0.000074 1.03 1.02–1.22 
  Nodal status 0.0019 1.00  
   [1–3]  1.85 1.26–2.72 
   >3  3.42 1.58–7.42 
 Hormone therapy Nodal status <0.0001 1.00  
   [1–3]  2.91 1.73–4.90 
   >3  8.47 2.98–24.05 
  c-erbB-2 ≤500 0.00012 1.00  
   >500  3.64 1.88–7.05 
  uPA ≤0.20 0.028 1.00  
   ]0.20–1.14]  1.72 1.06–2.78 
   >1.14  2.95 1.12–7.72 
MFS Chemotherapy uPA ≤0.20 0.00057 1.00  
   ]0.20–1.14]  2.20 1.41–3.45 
   >1.14  4.84 1.98–11.88 
  Clinical tumor size  0.0012 1.00 1.01–2.83 
  Nodal status 0.005 1.00  
   [1–3]  1.62 1.16–2.26 
   >3  2.62 1.34–5.13 
 Hormone therapy Nodal status 0.0013 1.00  
   [1–3]  2.19 1.36–3.52 
   >3  4.78 1.84–12.43 
  c-erbB-2 ≤500 0.021 1.00  
   >500  2.24 1.13–4.45 

We thank all of the clinicians and pathologists at our institution who performed diagnosis, treatment, and follow-up. The help of Sangtec (uPA and p53 luminometric assays) and Triton(c-erbB-2 and EGFR) is gratefully acknowledged.

1
Révillon F., Bonneterre J., Peyrat J. P. ERBB2 oncogene in human breast cancer and its clinical significance.
Eur. J. Cancer
,
34
:
791
-808,  
1998
.
2
Bièche I., Onody P., Laurendeau I., Olivi M., Vidaud D., Lidereau R., Vidaud M. Real-time transcription PCR assay for future management of ERBB2-based clinical applications.
Clin. Chem.
,
45
:
1148
-1156,  
1999
.
3
Ross J. S., Fletcher J. A. The HER-2/neu oncogene in breast cancer: prognostic factor, predictive factor and target for therapy.
Oncologist
,
3
:
237
-252,  
1998
.
4
Klijn J. G., Berns E. M., Foekens J. A. Prognostic factors and response to therapy in breast cancer.
Cancer Surv.
,
18
:
165
-198,  
1993
.
5
Bloom H. J., Richardson W. W. Histological grading and prognosis in breast cancer.
Br. J. Cancer
,
11
:
359
-377,  
1957
.
6
Scarff, R. W., and Torloni, H. Histological Typing of Breast Tumors. Geneva: WHO, 1968.
7
European Organization for Research and Treatment of Cancer (EORTC). Breast cancer cooperative group. Revision of the standards for the assessment of hormone receptors in human breast cancer.
Eur. J. Cancer
,
16
:
1513
-1515,  
1980
.
8
Thorpe S. M., Koenders A. Standardization of steroid receptor assays in human breast cancer.
III. Selection of reference material for intra-and interlaboratory quality control. Eur. J. Cancer Clin. Oncol.
,
22
:
939
-944,  
1986
.
9
Romain S., Spyratos F., Goussard J., Formento J. L., Magdelenat H. Improvement of quality control for steroid receptor measurements: analysis of distributions in more than 40,000 primary breast cancers.
Breast Cancer Res. Treat.
,
41
:
131
-140,  
1996
.
10
Langton B. C., Crenshaw M. C., Chao L. A., Stuart S. G., Akita R. W., Jackson J. E. An antigen immunologically related to the external domain of gp185 is shed from nude mouse overexpressing the c-erbB-2 (HER-2/neu) oncogene.
Cancer Res.
,
51
:
2593
-2598,  
1991
.
11
Broët P., Spyratos F., Romain S., Quillien V., Daver A., Ricolleau G., Rallet A., Toulas C., Asselain B. Prognostic value of uPA and p53 accumulation measured by quantitative biochemical assays in 1245 primary breast cancer patients: a multicentre study.
Br. J. Cancer
,
80
:
536
-545,  
1999
.
12
Iwase H., Kobayashi S., Itoh Y., Kuzushima T., Yamashita H., Iwata H., Naito A., Yamashita T., Itoh K., Masaoka A. Clinical value of enzyme immunoassay of epidermal growth factor receptor in human breast cancer.
Breast Cancer Res. Treat.
,
28
:
215
-221,  
1993
.
13
Hilsenbeck S. G., Clark G. M. Practical p-value adjustment for optimally selected cutpoints.
Stat. Med.
,
5
:
103
-112,  
1996
.
14
Kaplan E. L., Meier P. Nonparametric estimation from incomplete observations.
J. Am. Stat. Assoc.
,
53
:
457
-481,  
1958
.
15
Peto R., Pike M.C. , Design and analysis of randomized clinical trials requiring prolonged observation of each patient.
II. Analysis and examples. Br. J. Cancer
,
35
:
1
-39,  
1977
.
16
Cox D. R. Regression models and life tables.
J. R. Stat. Soc.
,
34
:
187
-220,  
1972
.
17
Press M. F., Hung G., Godolphin W., Slamon D. J. Sensitivity of HER-2/neu antibodies in archival tissue samples: potential source of error in immunohistochemical studies of oncogene expression.
Cancer Res.
,
54
:
2771
-2777,  
1994
.
18
Robertson K. W., Reeves J. R., Smith G., Keith N., Ozanne B. W., Cooke T. G., Stanton P. D. Quantitative estimation of epidermal growth factor receptor and c-erbB-2 in human breast cancer.
Cancer Res.
,
56
:
3823
-3830,  
1996
.
19
Koscielny S., Terrier P., Daver A., Wafflart J., Goussard J., Ricolleau G., Delvincourt C., Delarue J. C. Quantitative determination of c-erbB-2 in human breast tumours: potential prognostic significance of low values.
Eur. J. Cancer
,
34
:
476
-481,  
1998
.
20
Spyratos F., Ferrero-Poüs M., Tubiana-Hulin M., Hacene K. Response—more about: prognostic importance of low c-erbB2 expression in breast tumors.
J. Natl. Cancer Inst.
,
92
:
1443
-1444,  
2000
.
21
Têtu B., Brisson J. Prognostic significance of HER-2/neu oncoprotein expression in node-positive breast cancer.
The influence of the pattern of immunostaining and adjuvant therapy. Cancer (Phila.)
,
73
:
2359
-2365,  
1994
.
22
Rilke F., Colnaghi M. I., Cascinelli N., Andreola S., Baldini M. T., Bufalino R., Della Porta G., Menard S., Pierroti M. A., Testori A. Prognostic significance of HER-2/neu expression in breast cancer and its relationship to other prognostic factors.
Int. J. Cancer
,
49
:
44
-49,  
1991
.
23
Menard S., Casalini P., Pilotti S., Cascinelli N., Rilke F., Colnaghi M. I. No additive impact on patient survival of the double alteration p-53 and c-erbB-2 in breast carcinoma.
J. Natl. Cancer Inst.
,
17
:
1002
-1003,  
1996
.
24
Horne G. M., Anderson J. J., Tiniakos D. G., McIntosh G. G., Thomas M. D., Angus B., Henry J. A., Lennard T. W., Horne C. H. p53 protein as a prognostic indicator in breast carcinoma: a comparison of four antibodies for immunohistochemistry.
Br. J. Cancer
,
73
:
29
-35,  
1996
.
25
Sjörgen S., Inganäs M., Lindgren A., Holmberg L., Bergh J. Prognostic value of c-erbB-2 overexpression in primary breast cancer, alone and in combination with other prognostic markers.
J. Clin. Oncol.
,
16
:
462
-469,  
1998
.
26
Borg A., Baldetorp B., Fernö M., Killander D., Olsson H., Rydén S., Sigurdsson H. ERBB2 amplification is associated with tamoxifen resistance in steroid-receptor positive breast cancer.
Cancer Lett.
,
81
:
137
-144,  
1994
.
27
Carlomagno C., Perrone F., Gallo C., De Laurentiis M., Lauria R., Morabito A., Pettinato G., Panico L., D’Antonio A., Bianco R., De Placido S. c-erb B2 overexpression decreases the benefit of adjuvant tamoxifen in early-stage breast cancer without axillary lymph node metastases.
J. Clin. Oncol.
,
14
:
2702
-2708,  
1996
.
28
ASCO. Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer.
J. Clin. Oncol.
,
14
:
2843
-2877,  
1996
.
29
Wright C., Nicholson S., Angus B., Sainsbury J. R. C., Farndon J., Cairns J., Harris A. L., Horne C. H. Relationship between c-erbB-2 protein product expression and response to endocrine therapy in advanced breast cancer.
Br. J. Cancer
,
65
:
118
-121,  
1992
.
30
Elledge R. M., Green S., Ciocca D., Pugh R., Allred D. C., Clark G. M., Hill J., Ravdin P., O’Sullivan J., Martino S., Osborne C. K. HER-2 expression and response to tamoxifen in estrogen receptor-positive breast cancer: a Southwest Oncology Group Study.
Clin. Cancer Res.
,
4
:
7
-12,  
1998
.
31
Allred D. C., Clark G. M., Tandon A. K., Molina R., Tormey D. C., Osborne C. K., Gilchrist K. W., Mansour E. G., Abeloff M., Eudey L., McGuire W. L. HER-2/neu in node-negative breast cancer: prognostic significance of overexpression influenced by the presence of in situ carcinoma.
J. Clin. Oncol.
,
10
:
599
-605,  
1992
.
32
Stål O., Sullivan S., Wingren S., Skoog L., Rutqvist L. E., Carstensen J. M., Nordenskjöld B. c-erbB-2 expression and benefit from adjuvant chemotherapy and radiotherapy of breast cancer.
Eur. J. Cancer
,
31A
:
2185
-2190,  
1995
.
33
Muss H. B., Thor A. D., Berry D. A., Kute T., Liu E. T., Koerner F., Cirrincione C. T., Budman D. R., Wood W. C., Barcos M., Henderson I. C. c-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer.
N. Engl. J. Med.
,
330
:
1260
-1266,  
1994
.
34
Thor A. D., Berry D. A., Budman D. R., Muss H. B., Kute T., Henderson I. C., Barcos M., Cirrincione C., Edgerton S., Allred C., Norton L., Liu E. T. erbB-2, p53, and efficacy of adjuvant therapy in lymph node-positive breast cancer.
J. Natl. Cancer Inst.
,
90
:
1346
-1360,  
1998
.
35
Paik S., Bryant J., Park C., Fisher B., Tan-Chiu E., Hyams D., Fisher E. R., Lippman M. E., Wickerham D. L., Wolmark N. erbB-2 and response to doxorubicin in patients with axillary lymph node-positive, hormone receptor-negative breast cancer.
J. Natl. Cancer Inst.
,
90
:
1361
-1370,  
1998
.