The relationship between c-erbB-2 expression, S-phase fraction and prognosis in breast cancer.

The relationship between c-erbB-2 gene expression (assessed immunohistochemically), S-phase fraction (SPF) and prognosis has been analysed in 172 women with primary breast cancer. c-erbB-2 staining was independent of age, tumour size, number of nodes involved, tumour grade and DNA ploidy, but was more common in oestrogen receptor (ER) negative tumours (P = 0.02) and progesterone receptor (PgR) negative tumours (P = 0.03). A weak correlation between c-erbB-2 staining and SPF was observed (r = 0.18). Amongst women with node negative disease, SPF was significantly related to relapse free survival (RFS, P = 0.04) while c-erbB-2 staining was not (P = 0.2). In contrast, both SPF (P = 0.002) and c-erbB-2 staining (P = 0.016) provided significant prognostic information on RFS for women with node positive disease. Multivariate analysis showed that c-erbB-2 staining and SPF gave independent information on RFS for women with node positive disease.     

c-erbB-2 oncoprotein overexpression identifies a subgroup of estrogen receptor positive (ER+) breast cancer patients with poor prognosis

Purpose:To investigate the predictive value of c-erbB-2 oncoprotein expression as compared with established histopathological and cytometric indicators of disease evolution in breast carcinoma. Patients and methods:A short-term retrospective study was conducted on a series of 306 breast cancer patients. Classic prognostic factors included tumour size, nodal involvement, histological grading, and hormone receptor status. Flow cytometric DNA ploidy and S-phase fraction (SPF) were also assessed. A Cox proportional hazards regression model was used for multivariate statistical analysis. Results:c-erbB-2 overexpression was present in 43 out of 295 (14.6%) tumours, and showed a statistically significant correlation with high histological grade, DNA aneuploidy, high SPF and lack of estrogen receptors (ER). Univariate analysis revealed its association with worse disease-free survival (DFS) and overall survival (OS). The combined evaluation of c-erbB-2 with ploidy and SPF defines a variable (P + S + c) that showed a significant correlation with disease outcome. By multivariate analysis, only nodal status (P < 0.001) and P + S + c subgrouping (group 2: P = 0.002; group 3: P = 0.001) in relation to DFS, and nodal status (P = 0.001) and DNA ploidy (P = 0.006) in relation to OS, retained independent prognostic significance. Subset analyses showed that cytometric parameters, P + S + c subgrouping and hormone receptors were significantly correlated with disease outcome in node-positive patients, whereas in node-negative subgroup no prognostic indicators were found. c-erbB-2 overexpression exhibited a trend in node-positive breast cancer (DFS: P = 0.068; OS: P = 0.086), and significant correlation with poor clinical evolution in ER positive patients (DFS: P = 0.015; OS: P = 0.004), mostly receiving tamoxifen. Conclusions:c-erbB-2 is an independent prognostic indicator of DFS when evaluated in conjunction with ploidy and SPF. It also seems to predict response to tamoxifen therapy, by identifying a subgroup of ER positive (ER+) breast cancer patients with poor prognosis.     

Prognostic significance of expression of c-ERBB-2 oncoprotein in breast cancer patients]

The c-erbB-2 oncogene, is a member of tyrosine kinase oncogene family and expected to play a role in the regulation of cell growth. In the present study, prognostic significance of the expression of c-erbB-2 oncoprotein was investigated by immunocytochemical assay with 130 primary breast cancer patients. The positive expression of c-erbB-2 oncoprotein was found in 53 out of 130 patients (40.8%). There was no significant correlation between expression of c-erbB-2 oncoprotein and conventional prognostic factors, estrogen receptor (ER), axillary lymph node metastasis and epidermal growth factor receptor (EGFR). Relapse free survival rate of the patients with positive c-erbB-2 expression was significantly worse than those with negative at 49 month after operation. Similar result was found in overall survival rate but the difference was not significant. Furthermore, when patients were stratified by regional nodal status, in the patients with lymph node metastasis, the prognosis of the patients with positive c-erbB-2 expression was significantly worse than those with negative, while no significant difference was found in the patients without lymph node metastasis. These results suggest that c-erbB-2 oncoprotein may act as a prognostic factor independently, predicting the prognosis of breast cancer patients.     

Prognostic significance of c-erbB-2 and estrogen receptor status in human breast cancer.

Using the 21N polyclonal antibody, we immunohistochemically stained 314 primary breast carcinomas to identify those tumors overexpressing the c-erbB-2 oncoprotein and to ascertain the prognostic significance of this expression on disease-free and overall survival. Positive membrane staining was present in 52 (17%) of these carcinomas of which 7 (13%) were ductal carcinomas in situ. There was no significant relationship between c-erbB-2 positivity and (a) age at diagnosis, (b) menopausal status, (c) tumor size, (d) lymph node status, (e) estrogen receptor status, or (f) whether or not the patient had disseminated disease outside the axillary fields. However, c-erbB-2-positive tumors were significantly associated with poorer grade (P = 0.02). Patients who were positive for this oncoprotein had a shorter disease-free survival (P = 0.002) and reduced overall survival (P = 0.0001). Overexpression of this oncoprotein was predictive of a worse prognosis in lymph node-positive disease (P = 0.003) and in patients presenting with grade II tumors (P = 0.001). Stratifying the patients on the basis of estrogen receptor status suggested that c-erbB-2+/estrogen receptor-negative status was predictive of a poorer prognosis when compared with the other subgroups (P less than 0.001). Primary and recurrent tumor tissues were available from 42 of the 314 patients. Identical patterns of c-erbB-2 expression occurred in 95% of cases, arguing against a direct role for c-erbB-2 expression in the process of tumor dissemination. The high incidence of staining in ductal carcinomas in situ suggests that expression of this oncoprotein is an early event in tumorigenesis. Finally, multivariate analysis indicated that the c-erbB-2 oncoprotein was an independent prognostic indicator for overall survival in breast carcinoma patients.     

Role of epidermal growth factor receptor expression in primary breast cancer: results of a biochemical study and an immunocytochemical study

Summary We have conducted two series of studies, a biochemical study and an immunocytochemical study, to investigate the role of epidermal growth factor receptor (EGFR) expression in primary breast cancer patients. In the biochemical study, a consecutive 115 patients were included and EGFR was measured by a competitive binding assay with multipoint Scatchard analysis. In the immunocytochemical study comprising 126 patients, EGFR status was determined by immunostaining with anti-EGFR antibody EGFR1. Several agreements were found from these two studies. EGFR status was inversely correlated with estrogen receptor (ER) status. No significant correlation was found between EGFR status and tumor size, nodal metastases, or the expression of c-erbB-2 protein. Ki-67 immunoreactivity, a cellular proliferation marker, was enhanced in EGFR positive tumors over EGFR negative tumors, suggesting a linkage of EGFR expression to cellular proliferative activity. Post-operative follow up showed that relapse-free survival for EGFR positive patients was significantly worse than that for EGFR negative patients, particularly in node-positive patients. Multivariate analysis demonstrated a significance of EGFR status as an independent prognostic indicator in primary breast cancer. The group expressing EGFR and c-erbB-2 protein indicated a particularly high risk for relapse.     

Prognostic significance of HER-2 oncoprotein expression in breast cancer: a 30-year follow-up.

PURPOSE: To study retrospectively the long-term prognostic significance of HER-2 oncoprotein expression in breast cancer (BC). PATIENTS AND METHODS: Two hundred nine consecutive female patients with invasive operable BC from a defined urban population were observed for a median of 30 years. Tissue expression of HER-2 oncoprotein was demonstrated by using an immunoperoxidase procedure. RESULTS: Fifty-five (26%) patients had cancer and a positive HER-2 oncoprotein stain reaction. They had significantly worse 10- and 25-year survival rates than those patients who had a negative stain reaction in their cancer (31% v 48% and 31% v 39%, respectively; P = .004). HER-2 expression was also associated with a poorer survival among patients who had axillary nodal metastases (P = .003; n = 104), but not among those patients who did not have metastases. HER-2 expression was related to the ductal histologic type, poor histologic grade, and high mitotic count, but not to tumor size, axillary nodal status, DNA ploidy, or S-phase fraction (SPF). In a multivariate analysis among patients with nodal metastases, HER-2 expression was an independent prognostic factor (P = .04) that predicted poor survival. However, if the entire series was entered onto the analysis, it did not emerge as an independent factor. CONCLUSIONS: HER-2 oncoprotein expression has long-term prognostic significance for predicting poor survival in BC, and it has an independent prognostic value among patients who presented with axillary nodal metastases. This result is contradictory to the argument that HER-2 expression is only a marker for drug resistance because the patients were not given adjuvant drug therapy.     

Comparison of EGFR, c-erbB-2 product and ras p21 immunohistochemistry as prognostic markers in primary breast cancer.

In a prospective study of 75 breast carcinomas, monoclonal antibodies EGFR1, Anti-serum 21N (As21N) and RAP-5 were used to assess immunohistochemically expression of Epidermal Growth Factor Receptor (EGFR), c-erbB-2 oncoprotein and ras protein p21. A careful comparison was made of their relative prognostic significance. Positive staining was seen with EGFR1 in 12/71 cancers (17%) and with As21N in 16/75 cancers (21%). Positive staining with RAP-5 occurred in all cancers and benign breast tissue, but varied in intensity. EGFR expression correlated with the number of involved lymph nodes, histological grade, estrogen and progesterone receptor (ER, PR) levels and the Melbourne Prognostic Index. C-erbB-2 and ras expression both correlated with ER levels and EGFR, but not with the Prognostic Index. Based on an immunohistochemical technique, EGFR expression emerges as the parameter with strongest prognostic associations.     

The prognostic significance of DNA flow cytometry in breast cancer: results from 881 patients treated in a single centre.

In this single-centre study of 881 patients, S-phase fraction (SPF) was shown to be a significant prognostic marker in terms of overall survival (OS), relapse-free survival (RFS) and survival after relapse (SAR). Further, SPF had independent prognostic significance when considering a range of other clinicopathological variables, namely tumour grade and stage, nodal status, patient age, tumour size, menstrual status and treatment details. For OS and RFS, SPF was the second strongest predictor of the clinical course of the disease after nodal status, and for SAR it was the strongest prognostic marker. SPF correlated positively with histological grade but was the stronger predictor of survival. The distribution of SPF values was markedly different for the two ploidy classes of tumour, with DNA aneuploid tumours having a significantly higher average SPF. However, SPF retained its independent prognostic ability when DNA diploid and aneuploid tumours were analysed separately, DNA ploidy itself also proved to be an independent prognostic marker but the survival difference between the two ploidy classes was much less than that seen for different levels of SPF. Tumours with several DNA aneuploid populations (multiploid tumours) tended to have a worse prognosis than other aneuploid tumours but this trend did not reach statistical significance. In this and other studies from this centre, SPF has proved to be a robust predictor of clinical outcome in carcinoma of the breast.     

c-erbB-2 oncoprotein assay in ovarian carcinoma and its clinical correlation with prognostic factors.

Overexpression of the c-erbB-2 oncoprotein has been detected in human adenocarcinoma of the breast, cervix and salivary gland, in all of which an association between the overexpression of the c-erbB-2 and a poor prognosis of the disease has been reported. However, the prognostic role of c-erbB-2 oncoprotein in ovarian carcinoma remains controversial. We measured c-erbB-2 oncoprotein with an enzyme-linked immunosorbent assay (ELISA). Patients with invasive ovarian cancer were found to have significantly higher median c-erbB-2 oncoprotein expression than patients with either benign ovarian cyst (P = 0.002) or control groups (P = 0.001). Overexpression of c-erbB-2 oncoprotein was found in seven (21.9%) of 32 epithelial ovarian cancers. Our results suggest that quantitative analysis of c-erbB-2 oncoprotein may be used to define the prognostic significance of ovarian carcinoma.     

Predictive value of c-erbB-2 and cathepsin-D for Greek breast cancer patients using univariate and multivariate analysis.

The value of various prognostic factors in breast cancer patients has been determined in a number of studies. One hundred thirty-eight Greek women were followed up over a 5-year period after surgery for breast cancer. Amplification and overexpression of c-erbB-2 was found in 22.4% and 29.7% of the respective cases, and the concentration of total cytosolic Cathepsin-D (CD) in 46.4% of them was high (> or = 60 pmol/mg protein). The examined biological variables were compared with standard clinicopathological prognostic factors for the disease and related to early relapse (ER; before 3 years), relapse-free survival (RFS; median, 5 years), and overall survival (OS; median, 5 years). It was found that high CD levels significantly shorten ER of both node-negative and node-positive patients (P < 0.0001 and P = 0.002, respectively) and have prognostic value for RFS and OS of node-negative patients (P = 0.0012 and P = 0.0288, respectively), but lose their value as relapse predictors for node-positive patients for periods longer than 3 years. Overexpression of c-erbB-2 was found to be predictive for OS of node-positive and -negative patients (P = 0.0048 and P = 0.0285, respectively), but its predictive power was weak for ER (P = 0.0456) and RFS (P = 0.0455) of node-negative patients and disappeared for node-positive patients. c-erbB-2 amplification offers minimal assistance to the prediction. In conclusion, high CD concentration is indicative of ER of patients, and c-erbB-2 overexpression correlates with OS of patients.     

Prognostic value of p53 expression in early-stage breast-carcinoma compared with tumor angiogenesis, epidermal growth-factor receptor, C-erbb-2, cathepsin-d, DNA-ploidy, parameters of cell-kinetics and conventional features

p53 expression detected by immunocytochemistry is emerging as a novel potentially useful prognostic indicator in breast carcinoma. However, additional research is warranted because a consensus has not yet been achieved on: i) methodology and quality control issues; ii) its association with other new biological prognostic indicators; iii) its prognostic value in multivariate analysis including conventional and new pathobiological features and; iv) its clinical usefulness either as a prognostic and predictive factor. This study was undertaken in a series of 165 early-stage breast cancer patients (median follow-up of 5 years) to compare the prognostic role of p53 expression with that of several other markers that have been found to be of value, using a multivariate statistical analysis. These factors are: tumour angiogenesis, epidermal growth factor receptor (EGFR), c-erbB-2 expression, cathepsin D, growth fraction by Ki-67 antibody, DNA ploidy and S-phase fraction. The main results observed were: i) 47 of 165 (28.5%) carcinomas had pAb 1801 staining and were considered as p53-positive; ii) p53 expression was weakly associated with S-phase fraction by flow cytometry (OR=1.86; p=0.085); iii) p53 expression was significantly associated with recurrence (p53 negative [-] versus weak positive [+] tumours: p=0.07 and odds ratio of 2.21; p53 negative [-] versus high positive [++] tumours: p=0.01 and odds ratio of 2.86) and death (p53-versus +: p=0.53 and odds ratio of 1.35; p53- versus ++: p=0.05 and odds ratio of 2.53); iv) the determination of p53 is able to identify a subset of high risk patients in c-erbB-2 negative tumours, this group being generally considered at good prognosis; v) In multivariate analysis on relapse-free survival including all the above markers only tumour angiogenesis, cathepsin D, EGFR and S-phase fraction and nodal status retained significance, and for overall survival only tumour angiogenesis was significant and independent. This new information on p53 expression could be useful to the clinician for a more rationale approach in defining prognosis of breast cancer patients. The prognostic value of p53 depends on which other markers are additionally analyzed and previous studies have not always assayed tumour angiogenesis, which is the most important factor in this series. p53 still need to be assessed as a potential predictor of response to chemo or radiotherapy, because of its role in monitoring DNA damage.     

Immunohistochemical and other features of breast carcinomas presenting clinically compared with those detected by cancer screening.

Features of 111 mammary carcinomas derived from breast cancer screening were compared with those of 69 carcinomas presenting 'clinically'. Screen detected cancers were smaller, had less likelihood of nodal metastases, included a higher proportion of in situ tumours and if invasive, tended to be of lower grade. Using immunohistochemical methods, the expression of c-erbB-2 oncoprotein, epidermal growth factor receptor (EGFR) and cathepsin D were compared in the two groups. A similar proportion of screened and unscreened tumours expressed c-erbB-2 oncoprotein and EGFR but expression of the oestrogen regulated protein cathepsin D was significantly more frequent in the screened group (P less than 0.05). Although a relatively small series, the results suggest a biological difference between 'screened' and 'clinical' tumours.     

Prognostic significance of immunohistochemical c-erbB-2 proto-oncogene expression and nuclear DNA content in human breast cancer.

Immunoreactivity of the c-erbB-2 proto-oncogene product and nuclear DNA content were assessed in specimens from 211 breast cancer patients with a mean follow-up of 16 years (range 13-19 years). A routine immunoperoxidase technique was used and cytometrical DNA assessments were performed on cytodiagnostically identified tumour nuclei, using image analysis. C-erbB-2 cell membrane staining was observed in 29% of the cases and was found to be related to tumour size (P = 0.02), histopathological grade (P = 0.02) and nuclear DNA content (P < 0.01). In univariate analysis immunohistochemical c-erbB-2 expression was of prognostic significance among node-positive patients (P = 0.02), but not among women with node-negative disease. This prognostic ability was reduced by multivariate analysis and was no longer significant. In contrast, nuclear DNA content was significantly related to distant recurrence-free survival even in multivariate analysis after adjustment for nodal status and tumour size (P < 0.01). In conclusion, the findings of the present study indicate that c-erbB-2 expression is of limited prognostic value in a subgroup of patients, whereas nuclear DNA content seems to provide significant prognostic information even in node-negative patients.     

Node-negative breast cancer: prognostic subgroups defined by tumor size and flow cytometry

Adjuvant systemic therapy for women with node-negative breast cancer is most easily justified for those patients at highest risk of relapse. We have examined the impact of tumor size, histologic grade, estrogen receptor (ER) status, tumor ploidy, and S-phase fraction (SPF) on relapse-free survival (RFS) for 169 patients with node-negative breast cancer in order to identify groups of patients at high and low risk of relapse. Patients with small tumors (less than or equal to 1.0 cm) had a significantly better RFS than those with larger tumors (P = .005), with 96% remaining relapse-free at 5 years. Patients with tumors less than or equal to 1.0 cm were thus excluded from analysis when attempting to define a group with a poor prognosis. Within the group of patients with tumors greater than 1.0 cm, tumor ploidy (P = .63), ER status (P = .3), or progesterone receptor (PgR) status (P = .24) did not predict for RFS. Patients with grade 1 or 2 infiltrating ductal tumors had a significantly better prognosis than those with grade 3 tumors (P = .04). The prognostic factor that gave the widest separation between subgroups, however, was SPF. Patients whose tumors were greater than 1.0 cm with an SPF less than or equal to 10% had a 5-year RFS of 78% compared with a 5-year RFS of 52% for those with an SPF greater than 10% (P = .006). We have combined tumor size and SPF to identify three prognostic groups: (1) tumor less than or equal to 1.0 cm, 5-year RFS 96%; (2) tumor greater than 1.0 cm plus SPF less than or equal to 10%, 5-year RFS 78%; 3) tumor greater than 1.0 cm plus SPF greater than 10%, 5-year RFS 52%. These prognostic groupings may help identify patients most suitable for adjuvant therapy.     

Value of epidermal growth factor receptor status compared with growth fraction and other factors for prognosis in early breast cancer.

The epidermal growth factor receptor (EGFR) is a transmembrane glycoprotein whose expression is important in the regulation of breast cancer cell growth. The relationship between EGFR status (determined by an immunocytochemical assay) and various prognostic factors was investigated in 164 primary breast cancers. Overall 56% of tumours were EGFR-positive and the expression of EGFR was unrelated to axillary node status, tumour size and histological grade; and it was poorly associated with the tumour proliferative activity measured by Ki-67 immuno-cytochemistry. The relapse-free survival (RFS) probability at 3-years was significantly worse for patients with EGFR positive tumours (P = 0.003) and for those whose Ki-67 score was > 7.5% (P = 0.0027), as well as in patients with axillary node involvement (P = 0.01) and with poorly differentiated tumours (P = 0.04). Immunocytochemical determination of EGFR and cell kinetics gave superimposable prognostic information for predicting RFS with odds ratios of 3.51, when evaluated singly. In our series of patients EGFR, Ki-67 and node status retain their prognostic value concerning RFS in multivariate analysis. The 3-year probability of overall survival (OS) was significantly better in node-negative patients (P = 0.04) and was similar in EGFR-positive and negative patients. In conclusion, EGFR status appears to be a significant and independent indicator of recurrence in human breast cancer and the concomitant measurement of the tumour proliferative activity seems to improve the selection of patients with different risks of recurrence.     

Tumour DNA ploidy as an independent prognostic factor in breast cancer

We determined nuclear DNA content from 308 archival paraffin-embedded malignant breast tumours and evaluated the survival of the patients by univariate and multivariate statistical analyses. The overall 8-year survival rate of stage I-III breast cancer patients was 74.3% in DNA-diploid and 51.2% in DNA-aneuploid tumours (P less than 0.0001). DNA ploidy had prognostic significance in both node-negative and node-positive breast cancer, primarily in cases with steroid receptor-positive tumours. In a Cox multivariate analysis DNA ploidy (P = 0.001), primary tumour size (P = 0.0007), nodal status (P = 0.04) and the content of progesterone receptors (P = 0.0008) emerged as significant independent prognostic factors, whereas oestrogen receptor status, age and menopausal status of the patients had no significant independent prognostic value. If the histological grade of ductal carcinomas was also included in the Cox model, both grade and DNA ploidy had independent prognostic effect. In conclusion, our results indicate that the analysis of DNA ploidy is a useful adjunct in the assessment of prognosis for breast cancer patients.     

c-erbB-2 protein overexpression in breast cancer is a risk factor in patients with involved and uninvolved lymph nodes.

The c-erbB-2 gene is overexpressed in about 20% of human breast cancers. Four hundred and eighty-three cases previously examined by immunohistochemical staining for c-erbB-2 expression were analysed to assess the risk associated with the elevated protein expression. Oncoprotein expression was correlated with increasing tumour grade but not with oestrogen receptor status, nodal involvement, tumour size or age. There was an increased risk of relapse and death associated with c-erbB-2 expression irrespective of nodal involvement. This marker thus appears to be a significant prognostic factor in the early as well as the late stages of breast cancer.     

Prognostic impact of cathepsin D and c-erbB-2 oncoprotein in a subgroup of node-negative breast cancer patients with low histological grade tumors

Some node-negative breast cancer patients, with initially good prognosis, relapse from their cancer and are poorly identified. In the present study, based on prospective data of 197 tumors, we measured cathepsin D (cath D, n=197), pS2 protein (n=125), c-erbB-2 oncoprotein (n=100) and epidermal growth factor receptor (EGF-R, n=99) to better define the risk of relapse of node-negative patients in comparison with that defined by the clinical and histological factors. The median follow-up in surviving patients was 75 months. Univariate analysis indicated that patients with histological grade III tumors (the Scarff, Bloom and Richardson classification) had a much poorer prognosis than those with histological grade I or II tumors (P=0.0027 for relapse-free survival and P=0.0156 for overall survival). When the population of node-negative patients was divided by tertiles, high cath D levels showed a significant association with an early relapse (P=0.0316). Using cut-off values, patients with high cath D (> or =25 pmol/mg protein) or c-erbB-2 oncoprotein (> or =4 Human Neu Unit/microg protein) levels, had a significant worse relapse-free survival (P=0.0147 and 0.0417, respectively). No prognostic information was supported by pS2 protein or EGF-R measurements. In multivariate analysis, histological grade, cath D and c-erbB-2 oncoprotein remained independent predictors of recurrence (P=0.005, 0.0361 and 0.0321, respectively). By combining low levels of cath D and c-erbB-2 oncoprotein in histological grade I or II tumors, we identified a subgroup of patients with a 100% relapse-free survival probability at 6 years of follow-up. Moreover, the subgroup of patients with histological grade I or II tumors and high values of both cath D and c-erbB-2 oncoprotein showed a prognosis as poor as the subgroup defined by histological grade III alone, respectively 66% and 70% relapse-free survival at 6 years of follow-up. In conclusion, the combination of conventional prognostic factor (histological grade) and biochemical factors (cath D and c-erbB-2 oncoprotein) enabled us to identify, in this preliminary study, a subgroup of patients having an increased risk of relapse in a group (node-negative patients with low histological grade tumors) considered as good prognosis.     

Expression and co-expression of the members of the epidermal growth factor receptor (EGFR) family in invasive breast carcinoma

The epidermal growth factor receptor (EGFR) family plays an important role in breast carcinogenesis. Much interest has been focused recently on its members because of their potential role as prognostic indicators in breast cancer and their involvement in cancer therapy. We have evaluated more than 1500 cases of invasive breast carcinoma immunohistochemically using tissue microarray technology to examine the expression of EGFR family receptor proteins. We have found that 20.1 and 31.8% of cases were positive for EGFR and c-erbB-2, respectively, and 45 and 45.1% of tumours overexpressed for c-erbB-3 and c-erbB-4, respectively. The expression of either EGFR or c-erbB-2 was associated with other bad prognostic features and with poor outcome. Neither c-erbB-3 nor c-erbB-4 had any association with survival. c-erbB-2 had an independent prognostic effect on overall and disease-free survival (DFS) in all cases, as well as in the subset of breast carcinoma patients with nodal metastases. Several hetero- and homodimeric combinations have been reported between the EGFR members. Those dimers can evoke diverse signal transduction pathways with variable cellular responses. We stratified cases according to their co-expression of receptors into distinct groups with different receptor-positive combinations. Patients whose tumours co-expressed c-erbB-2 and c-erbB-3, as well as those whose tumours co-expressed EGFR, c-erbB-2 and c-erbB-4 showed an unfavourable outcome compared with other groups, while combined c-erbB-3 and c-erbB-4 expression was associated with a better outcome. In cases showing expression of one family member only (homodimers), we found a significant association between c-erbB-4 homodimer-expressing tumours and better DFS. In contrast, patients with c-erbB-2 homodimer-expressing tumours had a significant poorer DFS compared with other cases. These data imply that the combined profile expression patterns of the four receptor family members together provide more accurate information on the tumour behaviour than studying the expression of each receptor individually.