ANTICANCER RESEARCH 24: 2061-2068 (2004)

Tumor Histology and Stage but Not p53, Her2-neu or Cathepsin-D Expression are Independent Prognostic Factors in Breast Cancer Patients

D. P. KORKOLIS1, E. TSOLI2,3, D. FOUSKAKIS2, J. YIOTIS4, G. J. KOULLIAS1, D. GIANNOPOULOS5, E. PAPALAMBROS6, N. I. NIKITEAS7, C. A. SPILIOPOULOU3, E. PATSOURIS8, P. ASIMACOPOULOS9,10 and V.G. GORGOULIS2

1Department of Surgery and 4Department of Biochemistry, "St Savvas" Hospital, Athens; 2Molecular Carcinogenesis Group, Department of Histology-Embryology, 3Department of Forensic Medicine and Toxicology, 7Second Department of Propaedeutic Surgery, 8Department of Pathology and 9Department of Surgery, School of Medicine, University of Athens, Athens; 5Department of Pathology, "Evangelismos" Hospital, Athens; 6Department of Surgery, "Laiko" Hospital, School of Medicine, University of Athens, Athens, Greece; 10Baylor College of Medicine, Houston, Texas, U.S.A.

Abstract. Background: Several factors are currently employed for both negative nodal status and Her2-neu overexpression tended to prognosis assessment and treatment determination in breast display an elevated risk of death. Conclusion: Our results support cancer. An array of molecular parameters, such as p53, Her2-neu the prognostic power of tumor histology and stage and emphasize (c-erbB 2) and Cathepsin-D, are also examined to improve clinical the need for further studies on the prognostic impact of p53, Her2- patient management. We have conducted a statistically powerful neu and Cathepsin-D in breast cancer. Additionally, our analysis study of the prognostic value of conventional factors and of the indicates that deregulation of Her2-neu might characterize a investigational factors p53, Her2-neu and Cathepsin-D in patients subgroup of node-negative patients with poor prognosis who could with invasive breast carcinoma, in order to compare their benefit from an aggressive adjuvant therapy. significance. Our analysis was extended to determine the associations of p53 and Her2-neu with risk of death and relapse Primary breast cancer (PBC) is the most common among patients with and without lymph node metastases. malignancy in the female population worldwide with more Materials and Methods: In a set of 125 primary breast tumors, p53 than a million new cases being diagnosed every year (1). and Her2-neu expression were immunohistochemically evaluated. Extensive studies have shown that tumor clinicopathological Cathepsin-D, estrogen and progesterone receptor concentrations parameters are well established breast cancer prognostic were determined in cytosols by a standard immunoradiometric factors and useful for treatment decision (2). assay. Results: Over a mean of 62 months, 49 patients (39%) had Carcinogenesis represents a complex process that a relapse and 29 patients (23%) died. Overexpression of p53, involves multiple changes in the controlling pathways of cell Her2-neu and Cathepsin-D was observed in 31%, 46% and 88% proliferation, apoptosis, invasiveness and metastatic spread of cases, respectively. Overall survival was associated with histology (3). For this reason, much effort has been placed on the (hazard ratio 0.04, 95% confidence interval: 0.01, 0.49 for lobular identification of novel molecular prognostic factors that tumors) and stage (hazard ratio 5.94, 95% confidence interval: alone or in combination with clinicopathological factors may 1.30, 27.15 for stage III samples). Disease-free survival was also improve the prediction of clinical outcome and determine related to histology (hazard ratio 0.23, 95% confidence interval: the appropriate therapeutic approach. 0.08, 0.73 for lobular tumors) and stage (hazard ratio 4.27, 95% The oncosuppressor protein p53 is a stress response confidence interval: 1.36, 13.36 for stage III tumors). Patients with protein that mediates growth suppression through cell cycle arrest or induction of apoptosis in response to DNA damage (4). Inactivation of the p53 protein in breast tumors is caused mainly by chromosomal deletion and gene Correspondence to: Vassilis G. Gorgoulis, Antaiou 53 Str., mutations, although epigenetic mechanisms also exist, Lamprini, Ano Patissia, Athens, Greece, GR-11146. Tel: ++3-01- 6535894, Fax: ++3-01-6535894, e-mail: [email protected] including mislocalization, increased degradation and inactivation through binding to amplified mdm2 (4). In Key Words: Breast cancer, p53, Her2-neu, Cathepsin D, survival. breast cancer, overexpression of the growth factor receptor

0250-7005/2004 $2.00+.40 2061 ANTICANCER RESEARCH 24: 2061-2068 (2004)

Her2-neu is another pathway leading to increased cell antibodies (Abs) were used: mouse monoclonal antibody against proliferation and evasion from apoptosis by mechanisms p53 (Clone DO-7, code COM108, YLEM) and rabbit polyclonal independent of p53 action (5). Her2-neu overexpression antibody against Her2-neu (code A 0485, DAKO, Denmark). b) Method. Immunohistochemistry was performed according to the mainly results from gene amplification and represents a indirect streptavidin-biotin-peroxidase method (34). Antibodies were critical event in the process of breast carcinogenesis and used at 1:250 and 1:200 dilutions for p53 and Her2-neu, respectively. chemoresistance (5). On the other hand, invasiveness of c) Evaluation. Fifteen high power fields (x 400) were examined in breast cancer cells is linked to lysosomal digestion of ECM each slide under light microscopy. Samples were considered p53- fragments, plasma proteins and cellular debris that enters positive when more than 10% of tumor cells exhibited nuclear cells by phagocytosis (6). Cathepsin-D is an important expression (21, 24). According to the HerceptTest TM scoring enzyme that participates in this process (6). system, samples with more than 10% of cells exhibiting staining of the entire membrane were considered Her2-neu-positive (35). Slide Published evidence of the prognostic impact of p53, Her2- examination was performed by two independent observers (VG, neu and Cathepsin-D expression on PBC has been quite DG) and inter-observer variability was minimal (p<0.01). confusing due to different evaluation methodologies, cut-off levels, statistical tests and particularly controversial findings (7- Assessment of the levels of Cathepsin-D, estrogen and progesterone 31). Interestingly, general recommendations on the assessment receptors. Primary tumor samples were collected and stored at -70ÆC. of these molecular parameters have been given to increase the Cytosols were prepared from the frozen tissue by homogenization quality of data collection and subsequently enable application in 10 mmol/1 Tris-HCl buffer, pH 7.4, containing 1.5 mmol EDTA, 0.5 mmol/1 monothioglycerol and 10 mmol sodium molybdate and of these data in clinical patient management (2, 32). subsequent centrifugation at 10,000xg, for 30 min at 4ÆC. Cathepsin-D In view of these controversies, we performed a concentrations were measured in the prepared cytosols by a standard statistically powerful analysis of the prognostic significance immunoradiometric assay (IRMA, ELSA Cath-D kits, CiS Bio- of the conventional clinicopathological markers, p53, Her2- International, Gif-sur-Yvette, France), following the manufacturer's neu and Cathepsin-D expression in a set of females with instructions (36). The cut-off value for CD positivity/negativity was primary breast cancer. Emphasis was given to multivariate 60 pmol/mg protein, as previously described (26). analysis after adjustment for age, tumor diameter, histology, Estrogen and progesterone receptor (ER, PgR) levels were determined in the same cytosols by the Ligand Binding Assay nodal status, grade, stage, hormone receptor status, p53, procedure using the Dextran-Coated Charcoal technique; the cut- Her2-neu and Cathepsin-D expression, type of surgery and off value for both ER and PgR was 10fmol/mg protein according therapeutic approach to specify the best model for survival to previous work (25). prediction. The possible interactions between nodal status, p53 and Her2-neu on death and relapse were also examined Statistical evaluation. All analyses were carried out using STATA in an attempt to discriminate patients with different risks software, version 8 (37). We used Cox proportional hazard models who might need a particular therapeutic approach. to assess the influence of clinicopathological and molecular parameters on the incidences of death and relapse. Examination of Materials and Methods the interaction between nodal status, p53 and Her2-neu, as well as between p53 and Her2-neu on death and relapse, was based on likelihood ratio tests comparing models with and without the Patients. This study included 125 consecutive women suffering from relevant explanatory variables. Such analysis could not be invasive breast carcinoma, grade I to III who were assessed and treated performed for Cathepsin-D because of the small number of failures at the Hellenic Anticancer Institute, "Saint Savvas" Hospital in Athens, and relapses in patients with negative Cathepsin-D expression. Greece, during the period 1993-1998. Forty-nine of these patients (39%) underwent modified radical mastectomy whereas the remaining 76 (61%) followed a breast-conserving treatment with limited excision. Results After surgery, all patients were given adjuvant treatment such as radiotherapy and/or systemic therapy (hormono- and/or Tumor clinicopathological characteristics. The age of the chemotherapy). The mean postsurgical follow-up time of patients was patients ranged between 24 and 89 years, whereas more than 62 months (range 52-72 months). Among them, 49 patients (39%) had half of them were older than 55 years (56%). Tumor size was a relapse and 29 patients (23%) died of the disease. In all cases, a complete recording of histopathological data was 1 cm for 37 specimens (30%), 2 cm for 69 specimens (55%) performed. Clinical staging was determined according to the and 3-4 cm for 19 cases (15%). Regarding histology, 105 Postsurgical International Union Against Cancer Tumor-Node- tumors (84%) were ductal infiltrative and 20 lobular infiltrative Metastasis (pTNM) classification and histological grading was (16%). Axillary nodal involvement was observed in 73 patients estimated as described by Bloom and Richardson (33). Patients (58%). Histological grading showed that 89 specimens (71%) with bilateral PBC, Paget’s breast disease, widespread disease in were grade I and II, whereas, 36 (29%) were grade III. time of diagnosis, carcinoma in situ and those who received only According to stage evaluation, 18 tumors (14%) were stage I, supportive treatment, were excluded from the analysis. 81 (65%) stage II and 26 (21%) stage III. One hundred and Immunohistochemical assessment of p53 and Her2-neu three tumors (82%) were positive for estrogen receptors and a) Antibodies. For immunohistochemical analysis, the following 110 (88%) were positive for progesterone receptors.

2062 Korkolis et al: Breast Cancer Histology and Stage as Prognostic Factors

Forty-nine patients (39%) underwent modified radical TableI. Distribution of patients who died or had a relapse plus the full mastectomy and 76 (61%) breast-conserving surgery. data set. Regarding therapy, 106 patients (85%) were irradiated, 92 Characteristics Death Relapse Full data set (74%) received hormonotherapy and 43 (34%) were treated (n=29) (n=49) (n=125) with chemotherapy. The clinicopathological features of the patients are presented in Table I. Age 1 < 45 years 5 (17%) 9 (18%) 24 (19%) p53, Her 2-neu and Cathepsin D expression. Overexpression 2 45-55 years 7 (24%) 14 (29%) 31 (25%) 3 >55 years 17 (59%) 26 (53%) 70 (56%) of p53 and Her2-neu was found in 39 (31%) and 57 (46%) of the tumors respectively. High Cathepsin-D levels were Tumor Diameter observed in 110 specimens (88%). Tumor p53, Her 2-neu 1 cm 5 (17%) 13 (26%) 37 (30%) 2 cm 13 (45%) 26 (53%) 69 (55%) and Cathepsin-D status are shown in Table I. 3-4 cm 11 (38%) 10 (20%) 19 (15%)

Prognostic factors influencing overall and disease-free survival. Histology Ductal infiltrative 28 (97%) 45 (92%) 105 (84%) In a univariate analysis, overall survival was associated with Lobular infiltrative 1 (3%) 4 (8%) 20 (16%) tumor diameter (hazard ratio 7.53, 95% confidence interval 3.26, 17.36 for tumors with 3 cm diameter), grade (hazard Nodal Status ratio 2.16, 95% confidence interval 1.02, 4.56 for grade III N0 11 (38%) 20 (41%) 52 (42%) N1 18 (62%) 29 (59%) 73 (58%) carcinomas), stage (hazard ratio 7.06, 95% confidence interval 3.19, 15.64 for stage III patients), estrogen receptor Grade status (hazard ratio 2.82, 95% confidence interval 1.24, 6.42 I&II 17 (59%) 30 (61%) 89 (71%) III 12 (41%) 19 (39%) 36 (29%) for tumors with negative estrogen receptor status), type of surgery (hazard ratio 4.32, 95% confidence interval 1.96, Stage 9.61 for radical mastectomy) and chemotherapy treatment I 2 (7%) 6 (12%) 18 (14%) (hazard ratio 2.47, 95% confidence interval 1.17, 5.19 for II 11 (38%) 26 (53%) 81 (65%) III 16 (55%) 17 (35%) 26 (21%) patients treated with chemotherapy). Histology and stage were independent prognostic factors in a multivariate Estrogen receptor status survival analysis (hazard ratio 0.04, 95% confidence interval Negative 8 (28%) 13 (27%) 22 (18%) Positive 21 (72%) 36 (73%) 103 (82%) 0.01, 0.49 for lobular tumors and hazard ratio 5.94, 95% confidence interval 1.30, 27.15 for stage III patients), Progesterone receptor status whereas, estrogen receptor status was of borderline Negative 5 (17%) 7 (14%) 15 (12%) significance (hazard ratio 3.98, 95% confidence interval Positive 24 (83%) 42 (86%) 110 (88%) 1.00, 15.94 for negative estrogen receptor tumors). Type of surgery Disease-free survival was related to tumor diameter Conservative surgery 10 (35%) 26 (53%) 76 (61%) (hazard ratio 3.37, 95% confidence interval 1.60, 7.10 for Radical mastectomy 19 (65%) 23 (47%) 49 (39%) tumors with 3 cm diameter), grade (hazard ratio 2.03, 95% Radiotherapy confidence interval 1.14, 3.62 for grade III carcinomas), stage No 3 (10%) 7 (14%) 19 (15%) Yes 26 (90%) 42 (86%) 106 (85%) (hazard ratio 3.42, 95% confidence interval 1.84, 6.33 for stage III patients), estrogen receptor status (hazard ratio 2.78, Hormonotherapy 95% confidence interval 1.47, 5.26 for tumors with negative No 8 (28%) 16 (33%) 33 (26%) Yes 21 (72%) 33 (67%) 92 (74%) estrogen receptor status), type of surgery (hazard ratio 1.92, 95% confidence interval 1.09, 3.37 for radical mastectomy), Chemotherapy chemotherapy treatment (hazard ratio 1.86, 95% confidence No 14 (48%) 27 (55%) 82 (66%) Yes 15 (52%) 22 (45%) 43 (34%) interval 1.06, 3.27 for patients treated with chemotherapy) and p53 staining (hazard ratio 2.02, 95% confidence interval p53 1.15, 3.55 for tumors with abnormal p53 expression) in a Negative 16 (55%) 27 (55%) 86 (69%) Positive 13 (45%) 22 (45%) 39 (31%) univariate analysis. When all factors examined were included in the analysis, only histology and stage displayed prognostic Her2-neu value (hazard ratio 0.23, 95% confidence interval 0.08, 0.73 Negative 12 (41%) 24 (49%) 68 (54%) Positive 17 (69%) 25 (51%) 57 (46%) for lobular tumors and hazard ratio 4.27, 95% confidence interval 1.36, 13.36 for stage III patients). The hazard ratios Cathepsin-D of the prognostic factors for overall and disease-free survival Negative 3 (10%) 4 (8%) 15 (12%) are summarized in Table II. Positive 26(90%) 45 (92%) 110(88%) ANTICANCER RESEARCH 24: 2061-2068 (2004)

TableII. Non-adjusted and fully adjusted hazard ratios for patients who died and patients who had a relapse.

Variable Univariate analysis Multivariate analysis Died Relapse Died Relapse Hazard 95% CI Hazard 95% CI Hazard 95% CI Hazard 95% CI ratio ratio ratio ratio

Age < 45 years 0.89 0.33, 2.41 1.03 0.48, 2.20 0.91 0.24, 3.54 1.26 0.50, 3.20 45-55 years 0.83 0.33, 2.11 1.40 0.73, 2.69 0.56 0.15, 2.11 1.36 0.64, 2.92 > 55 years 1.00 1.00 1.00 1.00 Tumor Diameter 1 cm 0.83 0.29, 2.37 0.96 0.49, 1.86 3.24 0.72, 14.76 2.17 0.76, 6.23 2 cm 1.00 1.00 1.00 1.00 3-4 cm 7.53 3.26, 17.36 3.37 1.60, 7.10 1.99 0.52, 7.59 1.16 0.39, 3.46 Histology Ductal infil. 1.00 1.00 1.00 1.00 Lobular infil. 0.19 0.02, 1.38 0.42 0.15, 1.17 0.04 0.01, 0.49 0.23 0.08, 0.73 Nodal Status N0 0.69 0.32, 1.49 0.90 0.51, 1.60 0.84 0.26, 2.73 1.65 0.71, 3.84 N1 1.00 1.00 1.00 1.00 Grade I&II 1.00 1.00 1.00 1.00 III 2.16 1.02, 4.56 2.03 1.14, 3.62 2.32 0.75, 7.17 1.57 0.69, 3.57 Stage I 0.84 0.18, 3.85 1.03 0.42, 2.50 1.21 0.14, 10.53 0.60 0.14, 2.58 II 1.00 1.00 1.00 1.00 III 7.06 3.19, 15.64 3.42 1.84, 6.33 5.94 1.30, 27.15 4.27 1.36, 13.36 Estrogen receptor status Negative 2.82 1.24, 6.42 2.78 1.47, 5.26 3.98 1.00, 15.94 2.68 0.94, 7.62 Positive 1.00 1.00 1.00 1.00 Progesterone receptor status Negative 1.68 0.58, 4.88 1.71 0.76, 3.83 2.03 0.45, 9.18 1.43 0.52, 3.93 Positive 1.00 1.00 1.00 1.00

Type of surgery Conservative 1.00 1.00 1.00 1.00 Radical 4.32 1.96, 9.61 1.92 1.09, 3.37 3.04 0.89, 10.40 1.77 0.75, 4.17 Radiotherapy No 0.46 0.11, 1.94 1.01 0.45, 2.26 0.59 0.10, 3.38 1.12 0.44, 2.84 Yes 1.00 1.00 1.00 1.00

Hormonotherapy No 1.13 0.50, 2.58 1.52 0.83, 2.77 0.40 0.10, 1.70 0.67 0.25, 1.76 Yes 1.00 1.00 1.00 1.00 Chemotherapy No 1.00 1.00 1.00 1.00 Yes 2.47 1.17, 5.19 1.86 1.06, 3.27 2.02 0.66, 6.16 0.98 0.41, 2.34 p53 Negative 1.00 1.00 1.00 1.00 Positive 2.04 0.97, 4.29 2.02 1.15, 3.55 1.21 0.48, 3.01 1.77 0.91, 3.43 Her2-neu Negative 1.00 1.00 1.00 1.00 Positive 1.67 0.79, 3.54 1.24 0.71, 2.18 1.97 0.83, 4.67 1.14 0.62, 2.09 Cathepsin-D Negative 1.08 0.33,3.60 0.77 0.28, 2.15 2.29 0.51, 10.27 1.14 0.36, 3.60 Positive 1.00 1.00 1.00 1.00

2064 Korkolis et al: Breast Cancer Histology and Stage as Prognostic Factors

Table III. Examination of the association between nodal status and risk of death among patients with different p53 or Her 2-neu expression and examination of the association of Her 2-neu expression with risk of death among patients with different p53 status.

Nodal Status p53 Negative Positive No of cases Hazard ratio 95% CI No of cases Hazard ratio 95%CI

Negative 8 0.71 0.19, 2.61 3 0.80 0.16, 3.90 Positive 8 0.80 0.26, 2.45 10 1.00

Likelihood test for interaction: LR Chi2 (df 1) = 0.01 P-value=0.91

Nodal Status Her 2-neu Negative Positive No of cases Hazard ratio 95% CI No of cases Hazard ratio 95%CI

Negative 3 0.25 0.04, 1.50 8 1.71 0.45, 6.57 Positive 9 0.98 0.33, 2.89 9 1.00

Likelihood test for interaction: LR Chi2 (df 1) = 3.70 P-value=0.05

Her 2-neu p53 Negative Positive No of cases Hazard ratio 95% CI No of cases Hazard ratio 95%CI

Negative 8 0.41 0.14, 1.19 4 0.49 0.10, 2.26 Positive 8 0.78 0.24, 2.58 9 1.00

Likelihood test for interaction: LR Chi2 (df 1) = 0.01 P-value=0.95

Table IV. Examination of the association of nodal status with risk of relapse across patients with different p53 or Her 2-neu expression and examination of the association of Her 2-neu expression with risk of relapse across patients with different p53 status.

Nodal Status p53 Negative Positive No of cases Hazard ratio 95% CI No of cases Hazard ratio 95%CI

Negative 12 0.94 0.35, 2.54 8 2.21 0.68, 7.14 Positive 15 0.67 0.29, 1.56 14 1.00

Likelihood test for interaction: LR Chi2 (df 1) = 0.47 P-value=0.49

Nodal Status Her 2-neu Negative Positive No of cases Hazard ratio 95% CI No of cases Hazard ratio 95%CI

Negative 9 1.45 0.51, 4.13 11 2.64 0.89, 7.87 Positive 15 1.27 0.55, 2.93 14 1.00

Likelihood test for interaction: LR Chi2 (df 1) = 1.64 P-value=0.20

Her 2-neu p53 Negative Positive No of cases Hazard ratio 95% CI No of cases Hazard ratio 95%CI

Negative 16 0.55 0.25, 1.20 8 0.94 0.34, 2.57 Positive 11 0.62 0.25, 1.55 14 1.00

Likelihood test for interaction: LR Chi2 (df 4) = 0.01 P-value=0.93

2065 ANTICANCER RESEARCH 24: 2061-2068 (2004)

Association of nodal status with risk of death or relapse among 31% of the specimens. Similar percentages have been obtained patients with different p53 or Her2-neu expression. No evidence by others with the use of the same cut-off levels (21, 24). of statistical interaction was found between nodal status and However, a variety in p53 immunodetection has also been p53 with respect to their association with death (pinteraction = observed and it has been supported that differences in the 0.91, Table III) or relapse (pinteraction = 0.49, Table IV). antibody used, tissue fixation, sensitivity of the Evidence of interaction term was observed between nodal immunohistochemical technique or even in the evaluation of status and Her2-neu on death (pinteraction = 0.05, Table III). the results may account for this variation (13). However, the The hazard ratio of death was higher in patients combining antibody DO7, which was used in the present study, has been negative lymph nodes and Her2-neu overexpression. The considered very effective in immunolocalising p53 (13). lowest risk was seen in patients with no lymph node In our work, patients with positive p53 staining had an metastases and normal Her2-neu status (Table III). Such an increased risk of death in a univariate analysis, which was interaction between nodal status and Her2-neu was not abrogated when all parameters were examined. No differences found in the case of relapse (pinteraction = 0.20, Table IV). with regard to survival were found among patients with The above findings should be interpreted with caution in different nodal status and p53 expression. These findings view of the small number of events in each cell. support the concept that the p53 pathway is more important than the protein alone (38) and that some p53 mutations may Association of Her2-neu expression with risk of death or relapse cause downstream effects with minor impact on survival. In among patients with different p53 status. No evidence of statistical this respect, examination of components of the p53 pathway interaction was found between Her2-neu and p53 status with may represent a promising alternative approach to study the respect to their association with death (pinteraction = 0.95, Table impact of p53 on survival (38). The absence of association III) and relapse (pinteraction = 0.93 for relapse, Table IV). between p53 immunostaining and survival has been reported irrespective of nodal status (16) and even in large studies with Discussion extensive follow-up periods conducted in node-negative patients (24, 27). At this point it should be stressed that, in the The present study is a comprehensive survival analysis that latter studies, the p53 prognostic value was overrated by tumor was undertaken in order to clarify the prognostic power of clinicopathological features in accordance with our findings p53, Her2-neu and Cathepsin-D expression, in a panel of (24, 27). On the other hand, p53 expression has been female breast cancer patients. The potential additive effect suggested as an independent prognostic factor in breast cancer of nodal status, p53 and Her2-neu on death and/or relapse (8, 9) and as a marker of early relapse in node-positive was also investigated in a search for high-risk individuals. patients with ductal carcinomas (17). Given the fact that Multivariate regression analysis revealed that tumor different methodologies and evaluation standards have been histology and stage were independent prognostic parameters applied, further large-scale and statistically robust studies with for overall and disease-free survival, whereas the status of uniform p53 assessment in combination with examination of estrogen receptors was of borderline significance for overall downstream effectors or even functional tests for p53 activities survival. According to our findings, ductal infiltrative tumors may clarify the issue of the prognostic power of p53. of advanced stage were associated with high incidence of Abnormal Her2-neu protein expression usually reflects death and local relapse. The presence of estrogen receptors, amplification of the corresponding gene in approximately one- which is currently estimated on a routine base in breast third of breast tumors (5, 44) and was detected by IHC in 46% cancer patients to predict response to anti-estrogen therapy of our samples. Survival analysis revealed the absence of a (hormonotherapy), conferred a better prognosis in the case relationship between patient outcome and Her2-neu of disease-free survival. Our findings are well in line with expression in the total number of patients. When the the consensus statement of the American College of combination of nodal status and Her2-neu expression was Pathologists (2) that characterizes tumor clinicopathological examined, a trend was found for node-negative patients parameters as invaluable factors in prognosis assessment overexpressing Her2-neu to be related to poor prognosis, and patient care design. whereas node-negative patients with normal Her2-neu status Mutations of the p53 gene mainly lead to prolonged half- displayed the best prognosis, in agreement with the findings of life and accumulation of the encoded protein, thus making Fernandez et al. (16). Our data further support the concept immunohistochemistry (IHC) a generally accepted method to that Her2-neu alteration occurs very early during malignant detect p53 gene status (13). Although, it has been reported that transformation (16, 39) and point out that a group of patients, p53 alterations may be better detected by sequencing rather traditionally treated in a conservative fashion, might need a than IHC, the sequencing technique misses p53 protein defects more aggressive approach. caused by epigenetic changes (4). In our cohort, overexpression The majority of studies attribute a strong prognostic of the p53 protein was immunohistochemically observed in significance to Her2-neu status in breast cancer (31).

2066 Korkolis et al: Breast Cancer Histology and Stage as Prognostic Factors

However, the use of non-standardized molecular methods or debate (11, 23) that addresses immunohistochemistry as the IHC have generated a variety of data, currently not applicable most appropriate method of Cathepsin-D assessment and in clinical patient management (2). More specifically, in the questions the validity of results obtained by other case of IHC-based studies, differences in fixation, antigen methodologies. retrieval, used antibodies and arbitrary cut-off levels may In the search for patient subgroups with different prognosis, contribute to the lack of uniformity of the collected results. In the survival effect of concomitant p53 and Her 2-neu this context, it should be mentioned that direct detection of deregulation was examined. Considering that both the c-erbB2 gene status by the methods of fluorescence and corresponding genes are located on chromosome 17, it has chromogenic in situ hybridization (FISH and CISH) has been been suggested that their deregulation might be related to or repeatedly associated with poor survival (31). Concordance even caused by the same event, allowing malignant growth by analyses between IHC and FISH or CISH have suggested two separate mechanisms; loss of cellular proliferation control percentages of invasive stained cells equal or higher than 50% and self sufficiency in growth signals (43). The present analysis to be the cut-off value for Her2-neu overexpression (40, 41). revealed that either overall survival or disease free-survival Such values are significantly higher compared to the was not altered among patients with different p53 and Her2- HerceptTest TM positive score (35), which was adopted in the neu status. Therefore, it seems that p53 and Her2-neu present work. Furthermore, since both protein overexpression simultaneous deregulation does not confer an additive and amplification of the c-erbB 2 gene represent surrogate advantage, possibly because alterations from other pathways markers of protein activation, their evaluation is not a direct are crucial for cells to acquire the fully malignant phenotype. measurement of Her2-neu activity. Immunohistochemical However, several investigators support that combined studies with a novel antibody against the phosphorylated deregulation of p53 and Her2-neu is related to poor survival (activated) Her 2-neu protein demonstrate that the receptor in breast cancer (9, 14, 15, 19, 20, 29, 30). Particularly, in the is activated only in a small number of patients overexpressing study of Nakopoulou et al. (14), overall survival was not Her 2-neu by classical IHC detection (28). These studies related to the expression of one of these molecules but to their consider that either the activated receptor or Her2-neu co-expression. With regard to nodal status, Beenken et al. (29) overexpression in at least 80% of tumor cells is associated with reported that p53 and Her2-neu co-expression was a poor survival (28). The aforementioned discrepancies show significant prognostic marker of overall survival only in node- that the prognostic strength of Her2-neu remains elusive. In positive patients, is contrast to the findings of Albanell et al. addition, development of Her2-neu-blocking antibodies that (15) in node-negative patients. Disease-free survival has been may induce regression of Her2-neu overexpressing tumors associated to p53 and Her2-neu co-expression irrespective of (42) recommends the discrimination of patients with altered nodal status (29, 30). However, it is important to note that Her2-neu protein expression that would benefit from this type some of these studies have been conducted with small patient of therapy. numbers (15, 19, 20) and different evaluation standards and Overexpression of Cathepsin-D was observed in 88% of the statistical methods have been used (9, 14, 15, 19, 20, 29, 30). tumors by a standard immunoradiometric assay. This In conclusion, the present findings indicate that tumor percentage is higher than the corresponding one detected by histology and stage possess a strong prognostic value in our the same technique in a large study of Foekens et al. (25), group of breast cancer patients. Node-negative patients with possibly due to sample composition and tumor heterogeneity. Her2-neu overexpression tend to show poor prognosis and No association was found between Cathepsin-D status and may require a more aggressive treatment. The prognostic patient prognosis, in accordance with studies based on impact of p53, Her2-neu and Cathepsin-D in breast cancer immunohistochemical detection of Cathepsin-D (7, 12). Others remains unclear and needs further investigation. have also failed to show Cathepsin-D prognostic significance in the specific sub-group of node- negative patients (10, 18). In References contrast to our findings, this lysosomal protease estimated by an immunoradiometric assay has been considered as an 1 Ferlay J, Bray F, Pisane P and Parkin DM: Cancer Incidence, Mortality and Prevalence Worldwide. Globocan 2000 (CD-ROM). independent prognostic factor irrespective of nodal IARC Press 2001. involvement (25), most probably because the considerable 2 Fitzgibbons PL, Page DL, Weaver D et al: Prognostic factors in sample size of the latter analysis allowed statistical detection breast cancer. College of American pathologists consensus of such a correlation. Furthermore, Scorilas et al. (26) support statement 1999. Arch Pathol Lab Med 124: 966-978, 2000. that high Cathepsin-D concentrations predict early relapse, 3 Hanahan D and Weinberg RA: Hallmarks of cancer. Cell 100: 57- whereas Ardavanis et al. (22) suggest that Cathepsin-D is a 70, 2000. 4 Oesterreich S and Fuqua SAW: Tumor suppressor genes in breast prognostic factor for loco-regional recurrence in node-negative cancer. Endocrine-Related Cancer 6: 405-419, 1999. patients. However, the prognostic significance of stromal or 5 Harari D and Yarden Y: Molecular mechanisms underlying tumor cell Cathepsin-D concentrations represents a matter of ErbB2/HER2 action in breast cancer. Oncogene 19: 6102-6114, 2000.

2067 ANTICANCER RESEARCH 24: 2061-2068 (2004)

6 Thorton H and Baum M: Prognostic value of Cathepsin-D in breast 26 Scorilas A, Trangas T, Yotis J et al: Determination of c-myc cancer. Br J Cancer 81(8): 1426-1428, 1999. amplification and overexpression in breast cancer patients: evaluation 7 Kandalaft PL, Chang KL, Ahn CW et al.: Prognostic significance of of its prognostic value against c-erbB-2, Cathepsin-D and immunohistochemical analysis of Cathepsin D in low stage breast clinicopathological characteristics using univariate and multivariate cancer. Cancer 71: 2756-2763, 1993. analysis. Br J Cancer 81(8): 1385-1391, 1999. 8 Friedrichs K, Gluba S, Eidtmann H and Jonat W: Overexpression of 27 Reed W, Hannisdal E, Boehler PJ et al: The prognostic value of p53 and p53 and prognosis in breast cancer. Cancer 72: 3641-3647, 1993. c-erbB-2 immunostaining is overrated for patients with lymph node 9 Marks JR, Humphrey PA, Wu K et al: Overexpression of p53 and negative breast carcinoma. A multivariate analysis of prognostic factors HER-2/neu proteins as prognostic markers in early stage breast in 613 patients with a follow up of 14-30 years. Cancer 88: 804-813, 2000. cancer. Annals Surg 219(4): 332-341, 1994. 28 Thor AD, Liu S, Edgerton S et al: Activation (tyrosine phosphorylation) 10 Ravdin PM, Tandon AK, Allred DC et al: Cathepsin D by Western of ErbB-2 (HER-2/neu): a study of incidence and correlation with blotting and immunohistochemistry: failure to confirm correlations outcome in breast cancer. J Clin Oncol 18(18): 3230-3239, 2000. with node negative breast cancer. J Clin Oncol 12: 467-474, 1994. 29 Beenken SW, Grizzle WE, Crowe R et al: Molecular biomarkers for 11 O’Donoghue AE, Poller DN, Bell JA et al: Cathepsin D in primary breast cancer prognosis: Coexpression of c-erbB-2 and p53. Annals breast carcinoma: adverse prognosis is associated with expression of Surg 233(5): 630-638, 2001. Cathepsin D in stromal cells. Breast Cancer Res Treat 33: 137-145, 1995. 30 Tsutsui S, Ohno S, Murakami S et al: Prognostic significance of the 12 Aaltonen M, Lipponen P, Kosma V-M et al: Prognostic value of coexpression of p53 protein and c-erbB2 in breast cancer. Am J Surg Cathepsin-D expression in female breast cancer. Anticancer Res 15: 185(2): 165-167, 2003. 1033-1038, 1995. 31 Ross JS, Fletcher JA, Linette GP et al: The HER-2/neu gene and 13 Horne GM, Anderson JJ, Tiniakos DG et al: p53 protein as a protein in breast cancer 2003: Biomarker and target of therapy. prognostic indicator in breast carcinoma: a comparison of four Oncol 8(4): 307-325, 2003. antibodies for immunohistochemistry. Br J Cancer 73: 29-35, 1996. 32 Bast RC, Ravdin P, Hayes DF et al: 2000 update of recommendations 14 Nakopoulou LL, Alexiadou A, Theodoropoulos GE et al: Prognostic for the use of tumor markers in breast and colorectal cancer: clinical significance of the co-expression of p53 and c-erbB-2 proteins in practice guidelines of the American Society of Clinical Oncology. J breast cancer. J Pathol 179: 31-38, 1996. Clin Oncol 19(6): 1865-1878, 2001. 15 Albanell J, Bellmunt J, Molina R et al: Node-negative breast cancers 33 Bloom HJG and Richardson WW: Histological grading and with p53 (-)/HER2-neu(-) status may identify women with very good prognosis in breast cancer. Br J Cancer 11: 359-377, 1957. prognosis. Anticancer Res 16: 1027-1032, 1996. 34 Tsoli E, Gorgoulis VG, Zacharatos P et al: Low levels of p27 in 16 Fernandez AMJ, Farina GJ and Aragoncillo BP: Immunohistochemical association with deregulated p53-pRb protein status enhance tumor expression of p53 and c-erbB-2 in breast carcinoma: relation with proliferation and chromosomal instability in non-small cell lung epidemiologic factors, histologic features and prognosis. Gen Diagn carcinomas. Mol Med 7(6): 418-429, 2001. Pathol 142(5-6): 289-96, 1997. 35 HercpepTest package insert. Glostrup Denmark: Dako Corporation. 17 Eissa S, Khalifa A, El-Gharib A et al: Multivariate analysis of DNA 36 EORTC Breast Cancer Cooperative Group. Revision of the ploidy, p53, c-erbB-2 proteins, EGFR and steroid hormone receptors standards for the assessment of hormone receptors in human breast for short term prognosis in breast cancer. Anticancer Res 17: 3091- cancer. Eur J Cancer 16: 1513-1515, 1980. 3098, 1997. 37 Stata Corporation, Stata Statistical software, release 8.0 College 18 Charpin C, Garcia S, Bouvier C et al: Cathepsin D detected by station, Texas: Stata Corporation, 2003. automated and quantitative immunohistochemistry in breast 38 Steele RJC, Thompson AM, Hall PA and Lane DP: The p53 tumor carcinomas: correlation with overall and disease free survival. J Clin suppressor gene. Br J Surg 85: 1460-1467, 1998. Pathol 50: 586-590, 1997. 39 Rennstam K, Baldetorp B, Kytola S et al: Chromosomal imbalances 19 Barbati A, Cosmi EV, Sidoni A et al: Value of c-erbB-2 and p53 and gene amplification precede aneuploidization in the genetic oncoprotein co-overexpression in human breast cancer. Anticancer evolution of breast cancer. Cancer Res 61: 1214-1219, 2001. Res 17: 401-406, 1997. 40 Vincent-Salomon A, MacGrogan G, Couturier J et al: Calibration of 20 Bebenek M, Bar JK, Harlozinska A and Sedlaczek P: Prospective immunohistochemistry for assessment of HER2 in breast cancer: studies of p53 and C-erbB-2 expression in relation to clinicopathological results of the French Multicentre GEFPICS study. Histopathology parameters of human ductal breast cancer in the second stage of clinical 42: 337-347, 2003. advancement. Anticancer Res 18: 619-624, 1998. 41 Tanner D, Gancberg D, Di Leo A et al: Chromogenic in situ 21 Gasparini G, Toi M, Verderio P et al: Prognostic significance of p53, hybridization. A practical alternative for fluorescence in situ angiogenesis, and other conventional features in operable breast hybridization to detect HER-2/neu oncogene amplification in archival cancer: Subanalysis in node-positive and node-negative patients. Int breast cancer samples. Am J Pathol 157(5): 1467-1472, 2000. J Oncol 12: 1117-1125, 1998. 42 Huston JS and George AJ: Engineered antibodies take center stage. 22 Ardavanis A, Scorilas A, Loukeri A et al: Cathepsin D may help in Hum Antibodies 10: 127-142, 2001. discriminating node-negative breast cancer patients at risk for local- 43 Barbareshi M, Leonardi E, Mauri F et al: p53 and c-erbB-2 protein regional recurrence. Anticancer Res 18: 2885-2890, 1998. expression in breast carcinomas. An immunohistochemical study 23 Losch A, Tempfer C, Kohlberger P et al: Prognostic value of including correlations with receptor status, proliferation markers and Cathepsin D expression and association with histomorphological clinical stage in human breast cancer. Am J Clin Pathol 98: 408-418, 1992. subtypes in breast cancer. Br J Cancer 78(2): 205-209, 1998. 44 Schulze G. Her-2/neu gene product in serum-an oncoprotein in the 24 Rudolph P, Olsson H, Bonatz G et al: Correlation between p53, c- diagnosis and therapy of breast carcinoma. Anticancer Res 23: 1007- erbB-2 and Topoisomerase II · expression, DNA ploidy, hormonal 10, 2003. receptor status and proliferation in 356 node-negative breast carcinomas: prognostic implications. J Pathol 187: 207-216, 1999. Received November 6, 2003 25 Foekens JA, Look MP, Bolt-de Vries J et al: Cathepsin-D in primary breast cancer:prognostic evaluation involving 2810 patients. Br J Revised March 12, 2004 Cancer 79(2): 300-307, 1999. Accepted April 19, 2004

2068