neu effect on adjuvant tamoxifen treatment

The Breast (2001) 10, 67–77 # 2000 Harcourt Publishers Ltd doi:10.1054/brst.2000.0225, available online at http://www.idealibrary.com on

ORIGINAL ARTICLE

Prognostic value of HER-2/neu and p53 expression in node-positive breast cancer. HER-2/neu effect on adjuvant tamoxifen treatment M. A. Climent,1 M. A. Seguı´ ,2 G. Peiro´,3 R. Molina,4 E. Lerma,3 B. Ojeda,3 J. J. Lo´pez-Lo´pez3 and C. Alonso3 1

Instituto Valenciano de Oncologı´a, 2Cosorci Hospitalari Parc Taulı´. Sabadell, 3Hospital de la Santa Creu i Sant Pau, 4Hospital Clinic, Barcelona, Spain S U M M A R Y . HER-2/neu and p53 expression, conventional clinical and pathologic prognostic factors, were evaluated in a retrospective series of 283 node-positive breast cancer patients. Overexpression was determined by immunohistochemistry in formalin-fixed paraffin-embedded tissue blocks. Twenty one percent were HER-2/neu positive and 40% p53 positive. HER-2/neu expression was related to axillary lymph node metastasis (P = 0.014), inflammatory infiltrates (P = 0.004), and the absence of oestrogen (ER) (P = 0.0026) and progesterone (P = 0.01) receptors (PR). p53 expression was related to lymph node involvement (P = 0.03), necrosis (P = 0.036), absence of ER (P = 0.028) and PR (P = 0.065). p53 was not associated with outcome. HER-2/neu was an unfavourable prognostic factor for disease-free (DFS) (P = 0.05) and overall survival (OS) (P = 0.02) in univariate analysis. Multivariate analysis showed that the number of involved axillary nodes (P 5 0.00001), age (P = 0.004), grade (P = 0.04), and PR (P = 0.04) were independent predictors for OS. ER-positive patients treated with adjuvant tamoxifen had shorter DFS and OS when they were HER-2/neu positive. # 2000 Harcourt Publishers Ltd

indicator of reduced disease-free survival (DFS) and of overall survival (OS) in lymph-node positive cancer. In lymph-node negative patients, the results are less clear.4–8 The 20-kilobase p53 gene is located on the short arm of chromosome 17(p13) and encodes a 53 kd nuclear phosphoprotein. Several lines of evidence indicate that p53 may function as a tumour suppressor gene.9,10 Mutations of p53 are one of the most commonly known genetic defects in human cancer and appear to be involved in the development and progression of many tumours. Mutant protein accumulates to high concentrations that can be detected by standard immunohistochemistry (IHC) techniques and IHC has been used to measure levels of mutant p53 protein in many types of cancer. So far, a number of reports have linked p53 overexpression to poor prognosis in breast cancer. Mutation of p53 has been associated with aggressive tumours and poor prognosis, especially in node-negative breast cancer.11–15 Overexpression of both genes has been correlated with treatment resistance. HER-2/neu expression has been

INTRODUCTION Accurate prognostic factors are critical for the counselling and care of breast cancer patients and for assisting with decisions regarding the type and intensity of adjuvant treatment. The identification of specific gene products that contribute to the process of carcinogenesis enables the study of biologically relevant prognostic factors. The HER-2/neu oncogene (also known as c-erbB-2) encodes a 185 kilodalton receptor-like transmembrane glycoprotein (p185) with 50% homology to epidermal growth factor receptor.1 The HER-2/neu oncoprotein is believed to function as a growth factor receptor and possesses an intracellular tyrosinekinase domain.2,3 Numerous reports have indicated that overexpression of the protein product of the HER-2/neu oncogene is an Address correspondence to: Dr M. A. Climent, Servicio de Oncologı´ a Me´dica, Instituto Valenciano de Oncologı´ a, C/Profesor Beltra´n Ba´guena 9 y 19, 46009 Valencia, Spain. Tel.: +34 963493411; Fax: +34 9611 14344 Published online: 12 December 2000

67

68

The Breast

found to be a strong prognostic factor in node-positive breast cancer patients or in node-negative patients who received adjuvant chemotherapy.16–19 Moreover, some authors suggest a relationship between oncogene expression and hormonal treatment resistance.20 At the same time, there is a trend for patients who are p53negative to receive greater benefit from chemotherapy.21 The present study was undertaken to help to clarify the prognostic significance of HER-2/neu and p53 overexpression, in addition to conventional prognostic factors, in a retrospective series of node-positive breast cancer patients treated in the same institution with long term clinical follow-up. We also evaluated its possible relationship to treatment response when relapse occurred.

PATIENTS AND METHODS This study was performed in a group of 283 patients with operable node-positive breast cancer who underwent modified radical mastectomy or breast conserving surgery in the Hospital de la Santa Creu i Sant Pau between January 1976 and December 1986. Patients included in the study met the following criteria: 1) nodepositive disease proven by histological examination of axillary lymph nodes; 2) all patients had to be treated and followed in the Hospital de la Santa Creu i Sant Pau; 3) follow-up was available at least 3 years from diagnosis. Clinical information including age and menopausal status at diagnosis, location of tumour, date of surgery, type of surgery, date and doses of radiation therapy, type of chemotherapy and hormone therapy agents, doses and number of cycles, local recurrence or metastasis, type of treatment at recurrence and response, and status at last follow-up was obtained from patients’ charts. Tumour size, number of lymph nodes examined, and number of positive lymph nodes were obtained from pathology reports. Histological type, tumour grade, presence of lymphatic and vascular invasion, necrosis, elastosis, degree of intraductal component, inflammatory features and nodal capsule involvement were reassessed by a trained pathologist. Tumour grade was determined following the semiquantitative method proposed by Scarff–Bloom–Richardson and modified by Elston and Ellis.22

peroxidase complex (ABC) detection system. Briefly, 3 mm sections were cut from formalin-fixed, paraffinembedded tissue blocks, float-mounted on adhesivecoated glass slides, deparaffined, and rehydrated. Slides were then sequentially incubated, with intervening washes in Tris buffer (5 min), in 10% ovoalbumin/ 10% normal goat serum to block nonspecific protein binding (15 min), 3% hydrogen peroxide (H2O2/0.1% sodium azide to quench endogenous peroxidase activity (5 min), rabbit anti HER-2/neu polyclonal primary antibody at 10 mg/ml (2.5 h) [Lander Diagnostica] biotinylated goat anti rabbit secondary antibody (30 min), ABC (30 min), diaminobenzidine/H2O2 chromogen substrate (10 min), and 0.1% osmium tetroxide to enhance chromogen signal (1 min). Slides were then counterstained with methyl-green (2 min), rinsed in deionized water, dehydrated in graded alcohols, cleared in xylene, and coverslipped with a permanent mounting medium. Overexpression of p53 was determined by immunohistochemistry using a similar technique as for HER-2/ neu. Pab 1801 monoclonal antibody was utilized. Immunostained slides were examined by light microscopy, by two observers and in cases of discrepancy, a consensus view achieved. A formalin-fixed breast carcinoma was used as a positive control and included in every batch processed. Evidence of expression of HER-2/neu was defined as the presence of distinct staining of the surface membranes of tumour cells. Samples displaying clear membrane staining in more than 5% of malignant cells were considered as positive. Samples displaying clear nuclear staining in more than 5% of malignant cells were considered as positive for p53 expression. Statistical methods Rates of oncogene overexpression were compared using chi-square tests. The primary endpoints in this study were DFS and OS, and curves for these measurements of outcome were calculated using the method of Kaplan and Meier. The differences between curves were assessed with the log-rank test or Breslow test for censored survival data. Multivariate analysis was performed using the proportional hazard Cox analysis.

RESULTS

Immunohistochemistry

Patient characteristics

Overexpression of HER-2/neu was determined by immunohistochemistry using a standard avidin–biotin–

Patients’ age ranged from 22 to 84 years (mean 53.2 years). One hundred and eighteen (41%) patients were

Prognostic value of HER-2/neu and p53 expression premenopausal at diagnosis, 35 (12.4%) perimenopausal and 132 (46.6%) postmenopausal. Tumour size was smaller than 2 cm in 57 (20.1%) patients, from 2 to 5 cm in 192 (67.8%), larger than 5 cm in 31 (11%) and unknown in 3 (1.1%) patients. The number of axillary lymph nodes involved by metastatic carcinoma at diagnosis ranged from 1 to 42 (mean 5.5). Invasive ductal carcinoma (96.5%) was the most frequent histological type. Most patients (97.5%) underwent a mastectomy (radical or modified), whereas the remaining patients were treated by conservative surgery with axillary lymph node dissection. Patients were submitted to a variety of adjuvant therapies after surgery: 68 (24.1%) received adjuvant chemotherapy including mostly cyclophosphamide, methotrexate, and 5-fluorouracil for six cycles; 102 (36%) patients received adjuvant hormone therapy (mainly tamoxifen), and 81 (28.6%) received adjuvant chemotherapy and hormone therapy. No adjuvant treatment was given to 32 patients. Patients in the current study were not randomized, and the decision to add adjuvant therapy was related to a variety of factors, including menopausal status, hormone receptor content, and number of involved lymph nodes. Those receiving only adjuvant chemotherapy were pre and perimenopausal patients and had an absence of hormone receptors; hormonal therapy was added if they had positive hormone receptors. Those receiving only adjuvant hormone therapy were postmenopausal patients. Median follow-up time of the whole group ranged from 7 to 216 months (mean 94.3 months). In this time interval, distant metastasis developed in 48% of these patients and local recurrence in 12.7%. Pathological characteristics of the tumour are shown in Table 1. HER-2 oncoprotein expression Of the 248 cases for which the immunostaining was adequate, 197 (79%) had an absence of immunostaining, and 52 (21%) showed positive membrane immunoreaction. Table 2 shows the relationship of membranous staining for HER-2/neu and other recognized prognostic factors in breast carcinoma. It appears that HER-2/neu oncoprotein expression is related to several factors of poor prognosis, such as more involved axillary lymph nodes ( P = 0.014), a high degree of inflammatory infiltrate ( P = 0.0004), and an absence of estrogen ( P = 0.0026) and progesterone ( P = 0.01) receptors. Patients whose tumours were negative for HER-2/neu had a significantly better DFS and OS than those who had positive staining ( P = 0.05 for DFS and P = 0.02

Table 1

69

Pathologic characteristics of the patients

Characteristic

No. of cases

%

Tumour size (UICC–TNM classification) pT1 pT2 pT3 unknown

57 192 31 3

20.1 67.8 11 1.1

Histotype Ductal Lobulillar

273 10

96.5 3.5

Grading (SBR scheme) I II III

38 165 80

13.4 58.3 28.3

No. of positive axillary nodes 1–3 4–9 >10

153 82 48

54 29 17

Lymphatic vessel invasion present absent

133 150

47 53

Necrosis Absent þ þþ

151 96 36

53.4 34 36

Inflammatory infiltrate Absent þ þþ

137 114 32

48.4 40.3 11.3

ER Positive Negative Unknown

153 101 29

54 35.7 10.3

PR Positive Negative Unknown

56 197 29

20 69.7 10.3

for OS [Wilcoxon–Breslow]). Actuarial 5-year DFS was 56.8% for patients with negative immunostaining, as opposed to 37.6% for patients with positive staining. A comparable difference was seen on OS (78.2% versus 60.8%) (Figs 1 and 2). Response to chemotherapy or hormonal treatment for disseminated disease was determined for all patients who recurred. Thirty-five out of 80 (43.8%) patients without oncogene expression obtained an objective response (complete or partial) to any treatment in contrast to 6 out of 24 (25%) of the HER-2/neu positive patients (P = 0.09). Although differences in response rates were maintained for chemotherapy and hormonal therapy when analyzed separately, statistical significance was not reached. DFS and OS were also analyzed depending on the type of adjuvant treatment administered and oncogene

70

The Breast Table 2 Correlation of immunostaining for HER-2/neu and p53 with recognized prognostic factors % HER-2/neu overexpression

P

% p53 overexpression

P

Tumour size T1 T2 T3

17.6 22.4 10.3

0.28

50 36.5 41.4

0.3

Lymph node involvement 1–3 4–9 >10

1.3 26 28.3

0.04

35.4 23.7 45.7

0.09

Grade I II III

13.5 19 26

0.26

32.4 39.6 44.9

0.43

Inflammatory component Absent Moderate Intense

9.2 21.5 50

0.00001

30.1 43.3 68.9

0.0003

Lymphatic invasion Present Absent

21.1 19.4

Necrosis Absent Moderate Intense

16.8 24.4 24.2

0.3

37.4 36.7 60

0.036

Elastosis Absent Moderate Intense

42.9 18.4 20.4

0.3

28.6 43 38.4

0.82

Intraductal component Present Absent

24.5 17

0.14

36.2 42.1

0.21

Estrogen receptors Present Absent

14.1 29.9

0.0026

34 48

0.028

Progesterone receptors Present Absent

8.3 23.5

0.01

29.8 43.3

0.06

0.7

38.4 41.7

0.84

Fig. 1 Disease free survival depending on HER-2/neu overexpression (P = 0.05).

Fig. 2 Overall survival depending on HER-2/neu overexpression (P = 0.02).

expression. Although DFS and OS were shorter for HER-2/neu-positive patients who received adjuvant chemotherapy (mostly CMF), this did not reach statistical significance. Patients treated with adjuvant

tamoxifen (102 patients) had a statistically significant longer DFS and OS when HER-2/neu expression was negative compared to patients with oncogene expression. Median disease free survival for the HER-2/

Prognostic value of HER-2/neu and p53 expression neu-negative patients was 97 months compared with 49 months for HER-2/neu-positive patients (P = 0.03). Median overall survival for the positive oncogene expression group was 79 months compared with 157 months for those patients without HER-2/neu expression (P = 0.01) (Figs 3 and 4). This deleterous effect of HER-2/neu positivity remains statistically significant for patients with tumours positive for ER treated with adjuvant tamoxifen (most of them) compared with patients who have ER-negative tumours (Fig. 5).

71

The immunohistochemical expression of p53 was not significantly related to disease free survival or overall survival (Figs 6 and 7). Coexpression of p53 and HER-2/neu There was no statistically significant association between overexpression of p53 and HER-2/neu proteins. Overall, the frequencies of coexpression were as follows:

p53 protein overexpression Nuclear immunoreactivity for p53 was positive in sections from 100 patients (40% of the cases for which immunostaining was adequate); 150 (60%) cases had an absence of immunoreaction, and it was not adequate in 33 (11.7% of the samples). p53 expression was significantly related to number of involved lymph nodes (P = 0.03), the presence of marked necrosis (P = 0.036), on absence of oestrogen receptors (P = 0.028) and progesterone receptors (P = 0.065). Expression of p53 was related to histological grade although it did not reach statistical significance. Expression was higher in poorly differentiated (44.9%) than in intermediategrade (39.6%) and low-grade (32.4%) carcinomas (Table 2).

Fig. 3 Disease free survival in patients with hormonal adjuvant treatment (n = 102) depending on HER-2/neu overexpression (P = 0.04).

Fig. 4 Overall survival in patients with hormonal adjuvant treatment depending on HER-2/neu overexpression (P = 0.01).

Fig. 5 Disease free survival in ER-positive patients treated with adjuvant tamoxifen depending on HER-2/neu overexpression. (P = 0.04).

Fig. 6 Disease free survival depending on p53 overexpression (P = not significant).

Fig. 7 Overall survival depending on p53 overexpression (P = not significant).

72

The Breast

p53(þ)/HER(þ) 9.8%; p53(7)/HER(7) 49.6% p53(7)/ HER(þ) 30.1%; p53(þ)/HER(7) 10.5%. Univariate analysis for DFS and OS We found that HER-2/neu overexpression significantly predicted DFS and OS (P = 0.05 and P = 0.02, respectively). In contrast, p53 did not show any predictive value either for DFS or OS. In addition to HER-2/neu expression, other prognostic factors found to be significantly associated with DFS were the number of affected axillary nodes (P 5 0.00001), nodal capsule involvement (P 5 0.00001), presence of lymphatic invasion (P = 0.0001), tumour grade (P = 0.02) and tubular

differentiation (P = 0.01), menstrual status (P = 0.02) and PR content (P = 0.03). Moreover, the number of involved axillary nodes (P 5 0.00001), nodal capsule involvement (P = 0.00005), presence of lymphatic vessel invasion (P = 0.00009), tumour grade (P = 0.005) and tubular differentiation (P = 0.003), age (P = 0.007) and PR content (P = 0.04), were shown to be significantly associated with overall survival (Table 3). Multivariate analysis for DFS and OS Multivariate analysis of the effect of combining HER-2/ neu overexpression with other prognostic factors, which were found to be prognostically significant by univariate

Table 3 Univariate analysis of the prognostic value of the different factors for DFS and OS (P values for 10 year DFS and OS) Characteristic

Disease free survival

Overall survival

5 year %

10 year %

P

5 year %

10 year %

P

Age at diagnosis 5 40 years 4 40 years

45.9 62.7

36.5 49.7

0.06

65.3 77.3

36.6 60.2

0.007

Menopausal status Pre Peri Post

53.1 54.3 66.7

40.1 44.9 54.8

0.02

69.6 85.7 77.4

46.7 63.3 63

0.07

No. Positive lymph nodes 1–3 4–9 410

74.7 47.1 33.3

62.8 37.9 16.2

50.00001

87.3 65.4 54.2

70.3 44.2 30.5

50.00001

Nodal capsule involvement Present Absent

57.4 68.6

33.1 58.9

50.00001

63 84.6

46.5 62.9

0.0005

Grading I II III

78.9 59.3 50.9

68 43.6 45.2

0.02

94.4 74.7 67.2

80.6 53.1 52.5

0.005

Lymphatic vessel invasion Present Absent

50.1 68

36.3 57.5

0.0001

67.4 82.3

45.1 64.6

0.0009

Tubular differentiation 1þ2 3

75.7 53.2

59.8 42.5

0.01

87.1 70.6

72.7 49.4

0.003

ER 7 þ

54 60.7

45.9 45

0.6

67.9 80

52.1 58.2

0.2

PR 7 þ

55.4 71.1

45.3 60.5

0.03

72.7 83.6

51.3 67.5

0.04

HER-2/neu 7 þ

62.4 45.2

47.9 43.2

0.05

78.2 60.8

56.7 50.3

0.02

p53 7 þ

56.5 60.9

47.6 45.4

0.7

74.3 74.3

56.7 52.3

0.5

Prognostic value of HER-2/neu and p53 expression analysis, showed that the number of involved axillary lymph nodes was the strongest independent predictor for DFS and OS. Patients with more than 10 involved nodes were 2.56 times more likely to relapse and 1.86 times more likely to die than those with fewer than three nodes involved. PR content was the second most significant prognostic factor for DFS (P = 0.004); for patients with presence of progesterone receptors, the odds of recurrence were 0.5 times lower than those with PR negative carcinomas. The third most predictive independent prognostic factor for DFS was menopausal status (P = 0.005); for postmenopausal patients the chance of recurrence was 0.69 times lower than for pre and perimenopausal patients. The fourth most predictive independent prognosticator was the presence of lymphatic vessel invasion by tumour (P = 0.05); the odds of recurrence were 1.43 times higher for tumours with lymphatic vessel invasion. For OS, the number of involved axillary lymph nodes was the most significant independent variable to affect outcome (P = 0.00001). Age was the second most significant and independent prognostic factor (P = 0.004); patients older than 40 years had an odds of death 0.51 lower than younger patients. The third most predictive and independent variable was tumour grade (P = 0.04); the odds of death were 1.47 times higher in those patients with grade III (SBR) tumours than in those with grade I. The fourth most predictive independent factor was PR positivity (P = 0.04); the

73

odds of death of PR positive tumour patients was 0.57 lower than PR negative tumour patients. HER-2/neu and other prognostic variables failed to retain an independent and significant value for DFS and OS in multivariate analysis and they were excluded from the final model (Table 4).

DISCUSSION The prognostic usefulness of HER-2/neu amplification has been investigated by many techniques but, for clinical studies, immunochemistry seems to be more convenient than Northern, Southern and Western blot analyses, because immunohistochemistry is easier to perform and there is a high correlation between different techniques.23–26 Membrane HER-2/neu oncoprotein immunostaining was strongly associated with several other factors recognized as important prognostic indicators. As in our study, other investigators have reported an association between HER-2/neu amplification or overexpression and the number of involved lymph nodes,4,6,27–34 high inflammatory infiltrate35–37 and absence of oestrogen and progesterone receptors.4,6–8,27,28,30,32,34,36–42 Despite numerous conflicting data relating HER-2/ neu overexpression to prognosis, the consensus is that node-positive patients with overexpression have a poor prognosis,6,8,27,28,34,35,38,43–50 whereas this has

Table 4 Cox regression analysis of DFS and OS Characteristic

Disease free survival OR

95% CI

No. Positive lymph nodes 1–3 4–9 410

1 1.03 1.95

– 0.81–1.33 1.49–2.56

PR 7 þ

1 0.5

– 0.32–0.8

Menopausal status Pre-peri Post

1 0.69

– 0.65–0.94

Lymphatic vessel Invassion Absent Present

1 1.43

– 1.02–2.07

Age at diagnosis 540 years 440 years

– –

– –

Grading I II III

– – –

– – –

Overall survival P

OR

95% CI

1 1.2 1.84

– 0.42–1.56 1.39–5.45

1 0.57

– 0.34–0.98

– –

– –

– –

– –

1 0.51

– 0.38–0.8

1 1.36 1.47

– 0.97–1.98 1.03–2.13

50.00001

P 50.00001

0.004

0.04

0.005

–

0.05

–

–

0.004

–

0.04

74

The Breast

only been rarely reported in node-negative women.6,34,35,41,43,47–49,51–54 This difference between node-positive and node-negative patients raises the question of the interaction between adjuvant chemotherapy and HER-2/neu overexpression and its role as a predictive factor for response to treatment. Breast cancers are either oestrogen-dependent or oestrogen-independet. Despite the presence of oestrogen receptors, a proportion of patients will not respond to endocrine treatment. The growth of certain breast cancers may be regulated by growth factor-receptor interactions that are independent of oestrogen control. It has been demonstrated that introduction of a HER-2/ neu cDNA, converting non-overexpressing breast cancer cells to those which overexpress this receptor, results in the development of oestrogen-independent growth which is insensitive to both oestrogen and the antioestrogen, tamoxifen. Moreover, activation of the HER2 receptor in breast cancer cells by the peptide growth factor, heregulin, leads to direct and rapid phosphorylation of ER on tyrosine residues which is followed by the production of an oestrogen-induced protein, progesterone receptor. In addition, overexpression of HER-2/neu receptor in oestrogen-dependent tumour cells promotes ligand-independent down-regulation of ER and delayed autoregulatory suppression of ER transcripts.14,55 These data suggest a direct link between these two receptor pathways and one mechanism for the development of endocrine resistance in human breast cancers. Of clinical relevance, Wright et al.56 reported recently that coexpression of HER-2/neu reduced the response rate of ERpositive patients to first line hormone therapy of metastatic breast cancer from 48% to 20%. Leitzel et al.57 performed a study using serum from 300 patients with metastatic breast cancer randomized to receive a second-line endocrine treatment. All patients were ERþ, ER unknown or PRþ. The chance of a patient achieving clinical benefit from endocrine treatment in this study was half as great (20.7% vs. 40.9%) if the serum HER-2/ neu level was elevated and the survival was also shorter. Some other authors have recently reported similar results.58 Although statistical significance is not reached, our results show a higher response rate to both therapies (chemotherapy or hormonal) in patients without HER2/neu overexpression (43.8% vs. 25%). This difference was maintained in patients receiving hormonal treatment (44% vs. 25%). The role of HER-2/neu in hormone resistance in ERþ patients is also suspected from our OS and DFS results in adjuvant tamoxifen treated patients. The prognosis of ER-negative patients treated with adjuvant tamoxifen is not altered by HER2/neu overexpression. In contrast, DFS and OS of ERpositive patients with hormonal adjuvant treatment is

worse when they overexpress HER-2/neu oncogene (Fig. 5). Alterations in the p53 gene as a result of mutation, deletion, or other changes are associated with transformation. Under normal circumstances, the short half-life of p53 precludes detection of the protein by current immunohistochemical methods. A frequent feature of altered p53 expression appears to be the accumulation of p53 protein in tumour cells, which usually results in nuclear immunoreactivity to a variety of monoclonal and polyclonal antibodies now available. While immunohistochemical detection of p53 is indicative of perturbation in the gene and/or protein, the precise nature of the alteration cannot be determined by the immunohistologic methods presently available. There may be substantial differences in the molecular effects of p53 alterations included in any group of breast carcinoma patients reported to be p53-positive by immunohistochemistry. Comparison studies are further hampered by the diversity of reagents used by different investigators and the lack of standardized immunohistochemical criteria to classify a carcinoma as p53positive. With these limitations in mind, it is not surprising that the results of various reports differ with respect to the association between detection of p53, clinicopathologic features of breast carcinoma and prognosis. Our results are consistent with the findings of several investigators who have failed to detect a significant correlation of p53 expression with age at diagnosis or menopausal status.59–61 Although in this study p53 expression is more common as the number of axillary nodes affected increases, some studies have failed to find a significant relationship60,62–67 while other studies have confirmed this relationship.3,68–72 Two groups of investigators have found a positive association between the presence of inflammatory infiltrate and p53 expression.59,69 In both these studies, and in our own study, p53 expression is significantly more common in tumours with an abundant inflammatory infiltrate. Although not previously reported, a highly significant correlation was found in this study between p53 expression and the presence of abundant tumour necrosis. Although a significant relationship between p53 expression and tumour grade was not identified in this study, p53 expression is higher as tumoural grade increases, and our results confirm previous reports.59,60,62,64–67,72–76 As described in the present study, virtually all previous reports have found that p53 expression is more frequent in ER-negative patients.59–64,70–77 This association has been observed regardless of the method of detecting p53 and ER. p53 expression is also more frequent in PRnegative patients. Previously studies have not reported a

Prognostic value of HER-2/neu and p53 expression consistent relationship between p53 expression and breast carcinoma prognosis. The absence of a significant association in the present study supports the conclusion reached in many previous reports.3,59,67,79,80 On the other hand, some studies have found a relationship between p53 expression and prognosis but restricted to node-negative patients.60,61,73,77,81 Some other studies report a prognostic value of p53 expression but the median follow-up is short (around three years).71,72,74,76 Nevertheless, some studies report a significant prognostic relationship in node-positive breast carcinoma with long follow-up.75 The results obtained in this study indicate that the presence or absence of p53 expression alone cannot be considered a reliable prognostic indicator in nodepositive breast carcinoma patients, although there are significant correlations between the expression of p53 and other phenotypic characteristics of breast carcinomas that may have prognostic significance. In our analysis, the strong correlation between HER-2/neu expression and the degree of nodal involvement, the high inflammatory infiltrate, and oestrogen and progesterone receptor negativity suggests expression of this proto-oncogene product in a biologically more aggressive breast cancer. Although prognostic value was demonstrated in univariate analysis, HER-2/neu was not an independent prognostic factor in the multivariate analysis. Thus, we could not demonstrate that its assessment added significantly to the information provided by currently available, standard disease parameters. From our results, hormonal treatment resistance in ER-positive patients can be suspected for those patients with HER-2/neu expression. Although some previous studies have shown similar results using the presence of HER-2/neu expression as a factor in selecting effective therapy, prospective studies are necessary to corroborate this.

Acknowledgements The authors acknowledge support from Spanish grant F.I.S.S.

References 1. Coussens L, Yang-Feng T L, Liao Y C et al. Tyrosinekinase receptor with extensive homology to EGF receptor shares chromosomal location with the neu oncogene. Science 1985; 230: 1130 –1139. 2. Malkin D, Li F P, Strong L C et al. Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 1990; 250: 1233 –1238. 3. Martinazzi M, Crivelli F, Zampatti C, Martinazzi S. Relationship between p53 expression and other prognostic factors in human breast carcinoma. An immunohistochemical study. Am J Clin Pathol 1993; 100: 213 –217.

75

4. Slamon D J, Clarck G M, Wong S G, Levin W J, Ullrich A, McGuire W L. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987; 235: 177–182. 5. Slamon D J, Godolphin W, Jones L A et al. Studies of the HER-2/ neu proto-oncogene in human breast and ovarian cancer. Science 1989; 244: 707–712. 6. Tandon A K, Clark G M, Chamness G C, Ullrich A, McGuire W L. HER-2/neu oncogene protein and prognosis in breast cancer. J Clin Oncol 1989; 7: 1120–1128. 7. Wright C, Angus B, Nicholson S et al. Expression of c-erbB-2 oncoprotein: a prognostic indicator in human breast cancer. Cancer Res 1989; 49: 2087–2090. 8. Guterson B A, Gelber R D, Goldhirsch A et al. Prognostic importance of c-erbB-2 expression in breast cancer. J Clin Oncol 1992; 10: 1049–1056. 9. Levine A J, Momand J, Finlay C A. The p53 tumor suppressor gene. Nature 1991; 351: 453–456. 10. Lane P D. A death in the life of p53. Nature 1993; 362: 789–787. 11. Bardeesy N, Falkoff D, Petruzzi M J et al. Anaplastic Wilms tumour, a subtype displaying poor prognosis, harbour p53 gene mutations. Nat Genet 1994; 7: 91– 97. 12. Gaidano G, Ballerini P, Gong J Z et al. p53 mutations in human lymphoid malignancies: association with Burkitt lymphoma and chronic lymphocytic leukemia. Proc Natl Acad USA 1991; 88: 5413–5417. 13. Lowe S W, Ruley H E, Jacks T, Housman D E. p53-dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell 1993; 74: 954–967. 14. Elrouby S, Thomas A, Costin D et al. p53 gene mutation in B-cell chronic lymphocytic leukemia is associated with drug resistance and is independent of mdr1/mdr3 gene expression. Blood 1993; 82: 3452–3459. 15. Heimdal K, Lothe R A, Lystad S, Holm R, Fossa S D, Borresen A L. No germline TP53 mutations detected in familial and bilateral testicular cancer. Genes Chromosom Cancer 1993; 6: 92–97. 16. Costa S, Stamm H, Almendral A et al. Predictive value of EGF receptor in breast cancer. Lancet 1988; ii: 1258. 17. Nicholson S, Sainsbury J R C, Halcrow P, Chambers P, Farndon J R, Harris A L. Expression of epidermal growth factor receptors associated with lack of response to endocrine therapy in recurrent breast cancer. Lancet 1989; i: 182–184. 18. Osaki A, Taoi M, Yamada H, Kawami H, Kuroi K, Toge T. Prognostic significance of co-expression of c-erbB-2 oncoprotein and epidermal growth factor receptor in breast cancer patients. Am J Surg 1992; 164: 323–326. 19. Visscher D W, Zarbo R J, Jacobsen G et al. Multiparametric deoxyribonucleic acid and cell cycle analysis of breast carcinomas by flow cytometry: clinico-pathologic correlations. Lab Invest 1990; 62: 370–378. 20. Veronese S M, Gambacorta M. Detection of Ki67 proliferation rate in breast cancer. Correlation with clinical and pathologic features. Am J Clin Pathol 1991; 95: 30–34. 21. Waga S, Hannon G J, Beach D et al. The p21 inhibitor of cyclin-dependent kinases controls DNA replication by interaction with PCNA. Nature 1994; 369: 574 –578. 22. Chang F, Syrja¨nen S, Tervahauta A, Syrja¨nen K. Tumourigenesis associated with the p53 tumour suppressor gene. Br J Cancer 1993; 68: 653–661. 23. Dawkins H J, Robbins P D, Sarna M, Carrello S, Harvey J M, Sterrett G F. c-erb-2 amplification and overexpression in breast cancer: evaluation and comparison of Southern blot, slot blot, ELISA and immunohistochemistry. Pathology 1993; 25: 124–132. 24. Venter D J, Tuzi N L, Kumar S, Gullick W J. Overexpression of the c-erbB-2 oncoprotein in human breast carcinomas: immunohistological assessment correlates with gene amplification. Lancet 1987; July 11: 69–71. 25. Molina R, Ciocca D R, Tandon A K et al. Expression of HER-2/ neu oncoprotein in human breast cancer: a comparison of

76

26. 27.

28.

29.

30.

31.

32.

33.

34. 35. 36.

37.

38.

39. 40.

41. 42.

43.

The Breast immunohistochemical and western blot techniques. Anticancer Res 1992; (6B) 12: 1965–1971. McCann A, Dervan P A, Johnston P A, Gullick W J, Carney D N. C-erbB-2 oncoprotein expression in primary human tumors. Cancer 1990; 65: 88–92. Hartmann L C, Ingle J N, Wold L E et al. Prognostic value of c-erbB-2 overexpression in axillary lymph node positive breast cancer. Results from a randomized adjuvant treatment protocol. Cancer 1994; 74: 2956–2963. Teˆtu 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 1994; 73: 2359–2365. Berger M S, Locher G W, Saurer S et al. Correlation of c-erbB-2 gene amplification and protein expression in human breast carcinoma with nodal status and nuclear grading. Cancer Research 1988; 48: 1238–1243. Anbazhagan R, Gelber R D, Bettelheim R, Goldhisch A, Guterson B A. Association of c-erbB-2 expression and S-phase fraction in the prognosis of node positive breast cancer. Annals of Oncology 1991; 2: 47–53. Noguchi M, Koyasaki N, Ohta N et al. Internal mammary nodal status is a more reliable prognostic factor than DNA ploidy and c-erbB-2 expression in patients with breast cancer. Arch Surg 1993; 128: 242–246. Heatley M, Maxwell P, Whiteside C, Toner P G. C-erbB-2 oncogene product expression depends on tumour type and is related to oestrogen receptor and lymph node status in human breast carcinoma. Pathol Res Pract 1993; 189(3): 261–266. Gasparini G, Gullick W J, Bevilacqua P et al. Human breast cancer: prognostic significance of the c-erbB-2 oncoprotein compared with epidermal growth factor receptor, DNA ploidy, and conventional pathologic features. J Clin Oncol 1992; 10(5): 686–695. Borg A, Baldetorp B, Ferno M, Killander D, Olsson H, Sigurdsson H. ERBB2 amplification in breast cancer with high rate of proliferation. Oncogene 1991; 6(1): 137–143. Rilke F, Colnaghi M I, Cascinelli N et al. Prognostic significance of HER-2/neu expression in breast cancer and its relationship to other prognostic factors. Int J Cancer 1991; 49(1): 44–49. Lipponen H J, Aaltomaa S, Syrjanen S, Syrjanen K. c-erbB-2 oncogene related to p53 expression, cell proliferation and prognosis in breast cancer. Anticancer Res 1993; 13(4): 1147–1152. Descotes F, Pavy J J, Adessi G L. Human breast cancer: correlation study between HER-2/neu amplification and prognostic factors in an unselected population. Anticancer Res 1993; 13(1): 119–124. Schonborn I, Zschiesche W, Spitzer E et al. c-erbB-2 overexpression in primary breast cancer: independent prognostic factor in patients at high risk. Breast Cancer Res Treat 1994; 29(3): 287–295. Watanabe T, Fukutomi T, Tsuda H et al. Determination of c-erbB-2 protein in primary breast cancer tissue extract using an enzyme immunoasay. Jpn J Cancer Res 1993; 84(12): 1279–1286. Kallioniemi O P, Holli K, Visakorpi T, Koivula T, Helin H H, Isola J J. Association of c-erbB-2 protein over-expression with high rate of cell proliferation, increased risk of visceral metastasis and poor long-term survival in breast cancer. Int J Cancer 1991; 49(5): 650–655. Richner J, Gerber H A, Locher G W et al. c-erbB-2 protein expression in node negative breast cancer. Annals of Oncology 1990; 1: 263–268. Mourad W A, Erkman-Balis B, Livingston S et al. Argyrophilic nucleolar organyzer regions in breast carcinoma: correlation with DNA flow cytometry, histopathology, and lymph node status. Cancer 1992; 69: 1739–1744. Toikkanen S, Helin H, Isola J, Joensuu H. Prognostic significance of HER-2 oncoprotein expression in breast cancer: a 30-year follow-up. J Clin Oncol 1992; 10: 1044–1048.

44. Allred D C, Clark G M, Tandon A K et al. HER-2/neu in node-negative breast cancer: prognostic significance of overexpression influenced by the presence of in situ carcinoma. J Clin Oncol 1992; 10: 599–605. 45. Quenel N, Wafflart J, Bonichon F et al. The prognostic value of c-erbB-2 in primary breast carcinomas: a study on 942 cases. Breast Cancer Res Treat 1995; 35(3): 283–291. 46. Thor A D, Schwartz L H, Koener F C et al. Analysis of c-erbB-2 expression in breast carcinomas with clinical follow-up. Cancer Res 1989; 49: 7147–7152. 47. Tsuda H, Hirohashi S, Shimosato Y et al. Correlation between long-term survival in breast cancer patients and amplification of two putative oncogene-coamplification units: hst-1/int-2 and c-erbB-2-ear1. Cancer Res 1989; 49: 3102–3108. 48. Mittra I, Redkar A A, Badwe R A. Prognosis of breast cancer: evidence for interaction between c-erbB-2 overexpression and number of involved axillary lymph nodes. J Surg Oncol 1995; 60(2): 106–111. 49. Clarck G M, McGuire W L. Follow-up study of HER-2/neu amplification in primary breast cancer. Cancer Res 1991; 51(3): 944–948. 50. Elwood J M, Godolphin W. Oestrogen receptors in breast tumours. Associations with age, menopausal status and epidemiological and clinical features in 735 patients. Br J Cancer 1980; 42: 635. 51. Bianchi S, Paglierani M, Zampi G et al. Prognostic significance of c-erbB-2 expression in node negative breast cancer. Br J Cancer 1993; 67(3): 625–629. 52. Ottestad L, Andersen T I, Nesland J M et al. Amplification of c-erbB-2, int-2 and c-myc genes in node-negative breast carcinomas. Relationship to prognosis. Acta Oncol 1993; 32: 289–294. 53. Rosen P P, Lesser M L, Arroyo C D, Cranor M, Borgen P, Norton L. Immunohistochemical detection of HER-2/neu in patients with axillary lymph node negative breast carcinoma. A study of epidemiologic risk factors, histologic features, and prognosis. Cancer 1995; 75(6): 1320–1326. 54. Spyratos F, Maudelonde T, Brouillet J P et al. Cathepsin D: an independent prognostic factor for metastasis of breast cancer. Lancet 1989; 8672: 1115–1118. 55. Pietras R J, Arboleda J, Reese D M et al. HER-2 tyrosine kinase pathway targets estrogen receptor and promotes hormoneindependent growth in human breast cancer cells. Oncogene 1995; 10(12): 2435–2446. 56. Wright C, Cairns J, Cantwell B J et al. Responses to mitoxantrone in advanced breast cancer: correlation with expression of c-erbB-2 protein and glutathione S-transferases. Br J Cancer 1992; 65: 271–274. 57. Leitzel K, Teramoto Y, Konrad K et al. Elevated serum c-erbB-2 antigen levels and decreased response to hormone therapy of breast cancer. J Clin Oncol 1995; 13: 1129–1135. 58. Chin K V, Ueda K, Pastan I et al. Modulation of activity of the promoter of human MDR 1 gene by Ras and p53. Science 1992; 255: 459–462. 59. Rosen P P, Lesser M L, Arroyo C D, Cranor M, Bargen P, Norton L. p53 in node-negative breast carcinoma: an immunohistochemical study of epidemiologic risk factors, histologic features, and prognosis. J Clin Oncol 1995; 13: 821–830. 60. Isola J, Visakorpi T, Holli K, Kallionemi O P. Association of overexpression of tumour suppressor protein p53 with rapid cell proliferation and poor prognosis in node-negative breast cancer patients. J Natl Cancer Inst 1992; 84: 1109–1114. 61. Silvestrini R, Benini E, Dainone M G et al. P53 as an independent prognostic marker in lymph node-negative breast cancer patients. J Natl Cancer Inst 1993; 85: 965–970. 62. Thor A D, Moore D H, Edgerton S M et al. Accumulation of p53 tumor suppressor gene protein: an independent marker of prognosis in breast cancers. J Natl Cancer Inst 1992; 84: 854–855.

Prognostic value of HER-2/neu and p53 expression 63. Caleffi M, Teague M W, Jensen R A, Vnencak-Jones C L, Dupont W D, Parl F F. p53 gene mutations and steroid receptor status in breast cancer. Cancer 1994; 73: 2147–2156. 64. MacGrogan G, Bonichon F, de Mascarel I et al. Prognostic value of p53 in breast invasive ductal carcinoma: an immunohistochemical study on 942 cases. Breast Cancer Res Treat 1995; 36(1): 71–81. 65. Poller D N, Hutchings C E, Galca M et al. p53 protein expression in human breast carcinoma: relationship to expression of epidermal growth factor receptor. Br J Cancer 1992; 66: 583–588. 66. Barbareschi M, Leonardi F, Maari F A et al. p53 and c-erbB-2 protein expression in breast carcinomas. Am J Clin Pathol 1992; 98: 408–418. 67. Walker R A, Dearing S J, Lane D P et al. Expression p53 protein in infiltrating and in situ breast carcinomas. J Pathol 1991; 165: 203–211. 68. Davidoff A M, Kerns B-J M, Iglehart J D et al. Maintenance of p53 alterations throughout breast cancer progression. Cancer Res 1991; 51: 2605–2610. 69. Davidoff A M, Kerns B J M, Pence J C et al. p53 alterations in all stages of breast cancer. J Surg Oncol 1991; 48: 260–267. 70. Stenmark-Askmalm M, Stal O, Sullivan S et al. Cellular accumulation of p53 protein: as independent progonstic factor in stage II breast cancer. Eur J Cancer 1994; 30A: 175–180. 71. Andersen T I, Holm R, Nesland J M, Heimdal K R, Ottestad L, Borresen A L. Prognostic significance of TP53 alterations in breast carcinoma. Br J Cancer 1993; 68: 540–548. 72. Visscher D W, Sarkar F H, Wykes S, Kothari K, Macosta J, Crissman J. Clinicopathologic significance of p53 immunostaining in adenocarcinoma of the breast. Arch Pathol Lab Med 1993: 117: 973–976.

77

73. Stenmark-Askmalm M, Stal O, Olsen K, Nordenskjold B. p53 as a prognostic factor in stage I breast cancer. South-East Sweden Breast Cancer Group. Br J Cancer 1995; 72(3): 715–719. 74. Friedrichs K, Gluba S, Eidtmann H, Jonat W. Overexpression of p53 and prognosis in breast cancer. Cancer 1993; 72: 3641–3647. 75. Ostrowski J L, Sawan A, Henry L et al. p53 expression in human breast cancer related to survival and prognostic factors: an immunohistochemical study. J Pathol 1991; 164: 75–81. 76. Lipponen P, Aaltomaa S, Syrjanen S, Syrjanen K. p53 protein expression in breast cancer as related to histopathological characteristics and prognosis. Int J Cancer 1993; 55: 51–56. 77. Allred D C, Clarck G M, Elledge R et al. Association of p53 protein expression with tumour cell proliferation rate and clinical outcome in node-negative breast cancer. J Natl Cancer Inst 1993; 85: 200–206. 78. Davidoff A M, Herndon J E, Glover N S et al. Relation between p53 overexpression and established prognostic factors in breast cancer. Surgery 1991; 110: 259–264. 79. Hansal E, Gitsch G, Kohlberger P, Dadak C, Miechwiecka N, Breitenecker G. Immunohistochemical detection of mutant p53-suppressor gene product in patients with breast cancer: influence on metastasis-free survival. Anticancer Res 1992; 12: 2325–2329. 80. Van Agthoven T, Van Agthoven T L A, Portengen H et al. Ectopic expression of epidermal growth factor receptors induces hormone independence in ZR-75-1 human breast cancer cells. Cancer Res 1992; 52: 5082–5088. 81. Lewis S, Locker A, Todd J H et al. Expression of epidermal growth factor receptor in breast carcinoma. J Clin Pathol 1990; 43: 385–389.