Skip Navigation


Annals of Oncology Advance Access originally published online on March 19, 2007
Annals of Oncology 2007 18(6):1004-1014; doi:10.1093/annonc/mdm074
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
18/6/1004    most recent
mdm074v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Treré, D
Right arrow Articles by Derenzini, M
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Treré, D
Right arrow Articles by Derenzini, M
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2007 European Society for Medical Oncology

breast cancer

Prognostic relevance of a novel semiquantitative classification of Bcl2 immunohistochemical expression in human infiltrating ductal carcinomas of the breast

D Treré1,*, L Montanaro1, C Ceccarelli2, S Barbieri1, G Cavrini3, D Santini2, M Taffurelli4 and M Derenzini1

1 Department of Experimental Pathology, Unit of Clinical Pathology
2 Department of Surgical Pathology
3 Department of Statistics
4 Department of Surgery, University of Bologna, Italy

* Correspondence to: Dr D. Treré, Alma Mater Studiorum, Università di Bologna, Dipartimento di Patologia Sperimentale, Via San Giacomo 14, 40126 Bologna, Italy. Tel: +39-051-302874; Fax: +39-051-306861; E-mail: davide.trere{at}unibo.it


   Abstract
Top
 Abstract
introduction
 patients and methods
 results
 discussion
 Acknowledgements
 References
 
Background: Bcl2 is an important prognostic parameter in human breast cancer. However, the evaluation of Bcl2 expression by immunohistochemistry is carried out using arbitrary scoring criteria. In the present study, we evaluated the clinical relevance of a novel, semiquantitative classification of the Bcl2 immunostaining based on both the distribution and the intensity of the staining reaction.

Patients and methods: The proposed classification was first validated in 69 breast cancer specimens by comparing the Bcl2 immunostaining with the Bcl2 messenger RNA (mRNA) levels evaluated by real-time RT-PCR. Since a highly significant association was found between protein and mRNA for Bcl2, the immunohistochemical scoring system was applied to 442 patients with infiltrating ductal carcinomas of the breast with long-term follow-up (median observation time 106 months).

Results: In the entire series, the Bcl2 variable was an independent predictor of clinical outcome, and its prognostic independence was maintained when lymph node-negative and -positive patients were considered separately. In this regard, of particular interest was the observation of a subgroup of node-negative breast cancer patients with a negative Bcl2 immunostaining, who had a very high probability of relapse or death (respectively about five and seven times greater than patients with a positive Bcl2 immunostaining). Moreover, the Bcl2 variable retained prognostic significance also in subgroups of patients treated with either adjuvant endocrine therapy or chemotherapy.

Conclusions: Our results demonstrated that in breast cancer, Bcl2 protein expression parallels its mRNA level, and it has a highly significant and independent prognostic relevance.

Key words: adjuvant therapy, breast cancer, Bcl2 expression, immunohistochemistry, prognosis, real-time RT-PCR


   introduction
Top
 Abstract
 introduction
patients and methods
 results
 discussion
 Acknowledgements
 References
 
Bcl2 is expressed by a wide range of normal tissues and tumors and belongs to a family of proteins which regulate the programmed cell death or apoptosis [1]. Since Bcl2 protein is a prosurvival factor which blocks apoptosis, its overexpression in human tumors would be expected to be associated with a more aggressive phenotype as well as to confer resistance to those forms of treatments inducing cell death. This is actually true for some human tumor types, including acute myeloid leukemia [2], aggressive non-Hodgkin's lymphoma [3], prostate carcinomas [4], neuroblastoma [5], and testicular carcinomas [6], where high Bcl2 expression was found to be associated with a worse clinical outcome and radio- [7] and chemoresistance [5, 8, 9]. However, in other human tumors including non-small-cell lung carcinoma [10], esophageal squamous cell carcinoma [11], cervical carcinoma [12], melanoma [13], and breast carcinoma [14–17], Bcl2 expression was positively associated with favorable prognostic features and better survival. In particular, in breast cancer it has been definitively shown that high Bcl2 levels are related to low tumor grade, positive estrogen receptor (ER)/progesterone receptor (PGR) status [16, 18], low proliferation rate [19, 20], favorable prognosis [19, 21–23], and good response to adjuvant therapies [20, 22, 24, 25].

In clinical–pathological practice, Bcl2 analysis is usually carried out by immunohistochemistry, using specific monoclonal antibodies (mAbs). Nevertheless, quantitative immunohistochemistry is far from being a reliable standardized procedure. Immunohistochemical Bcl2 quantification mainly relies on the evaluation of the staining intensity, which is subject to variables linked to the staining reaction itself, observer's subjectivity, and arbitrary scoring criteria. This technical limitation may be considered responsible for the differing data reported in literature with regard to the frequency of Bcl2 expression and the clinical utility of the Bcl2 parameter [13, 18, 26].

In order to obtain a reliable immunohistochemical scoring of Bcl2 expression, in the present study we have scored 69 breast cancer samples into three immunohistochemical classes on the basis of the distribution and intensity of the staining reaction. This scoring system was validated by comparing the Bcl2 protein expression with the Bcl2 messenger RNA (mRNA) levels evaluated by RT-PCR on frozen specimens from the corresponding tumor samples. Since a highly significant association was found between protein and mRNA for Bcl2, the Bcl2 immunostaining classification was applied to a larger series of infiltrating ductal carcinomas of the breast (n = 442) in order to test its interobserver reproducibility and clinical relevance.


   patients and methods
Top
 Abstract
 introduction
 patients and methods
results
 discussion
 Acknowledgements
 References
 
patients
Two different cohorts of patients were considered for the two different purposes of our study.

For the definition of the immunohistochemical scoring system, on the basis solely of frozen tissue availability, a first cohort was selected from a series of consecutive patients who underwent surgical resection for primary infiltrating carcinoma of the breast at the Surgical Department of the University of Bologna from 1994 to 1995. This first group was made up of 72 cases aged from 28 to 87 years (median age 62 years).

For the evaluation of the clinical relevance of Bcl2 expression—as assessed according to our scoring criteria—a second cohort was used including infiltrating ductal carcinomas only, due to the different prognosis which characterizes the other breast cancer histotypes. This second group was made up of 442 consecutive patients aged from 25 to 89 years (median age 61 years) who underwent surgical resection for primary infiltrating ductal carcinoma of the breast at the Surgical Department of the University of Bologna from 1991 to 1995.

In both series, tumors were histologically diagnosed according to the World Health Organization criteria. Histological grading (G) was carried out according to Elston and Ellis [27]. Tumors were also typed by nuclear grading as follows: mild (NG1), moderate (NG2), and severe (NG3) nuclear atypia. Tumor size was evaluated in freshly obtained tissue, before formalin fixation, and coded according to the International Union Against Cancer pT recommendations. Axillary node status was assessed by pathological staging after axillary node dissection. The number of axillary lymph nodes examined ranged from 8 to 63 [mean number ± standard deviation (SD) 26.4 ± 9.2]. Axillary lymph node metastases were reported as absent (N0) or present (N+). ER and PGR status, Ki67 antigen expression, HER2 status, and Bcl2 protein expression were assessed on histological sections by standard immunohistochemistry, as reported below. All immunohistochemical analyses were carried out at the time of diagnosis. Patients were then regularly followed up every 6 months for a median observation time of 106 months (range 4–139 months).

The present study was approved by the Senior Staff Committee: the board which, at the time of patient enrollment, regulated noninterventional studies and was comparable to an Institutional Review Board.

immunohistochemical assessment
Four-micrometer thin serial sections were cut, collected on 3-ethoxy-aminoethyl-silane-treated slides, and allowed to dry overnight at 37°C. Sections were then processed for immunohistochemistry according to stretavidin–biotin–peroxidase complex protocol combined with an antigen retrieval pretreatment at 98°C in citrate buffer solution (pH 6.0) for 40 min, and highlighted using a peroxidase/diaminobenzidine (DAB) enzymic reaction. Sections were finally counterstained with hematoxylin, dehydrated, and mounted with a specific resin (BioMount, Bioptica, Milan, Italy). The following mAbs were applied overnight at room temperature at the predetermined optimal concentrations: anti-Ki67 (clone MIB-1), anti-ER (clone 1D5), anti-PGR (clone 1A6), anti-HER2 internal domain (clone CB11), and anti-Bcl2 (clone 100), all from BioGenex Laboratories (San Ramon, CA). Negative controls were stained along with test sections; control sections received the same treatment, except for application of the primary antibody.

Nuclear immunostaining of ER, PGR, and Ki67 was assessed by image cytometry using the Cytometrica program (C & V, Bologna, Italy), as previously detailed [28]; it was expressed as the percentage of labeled nuclear area over the total neoplastic nuclear area in the section (labeling index area: LIa). For each case, at least 2000 cells were evaluated. HER2 membrane immunostaining pattern and intensity were assessed by direct microscope evaluation following the four classes scoring system (0, +1, +2, +3) according to the published recommendations [29]. Cases were then classified as negative (classes 0 and +1) and positive (classes +2 and +3).

Bcl2 immunostaining was blindly assessed by direct microscopic analysis with no knowledge of either the clinical outcome or other clinicopathological data. The staining intensity was scored into three classes by a senior pathologist (MD) according to the criteria detailed in the ‘results’ section. To test the interobserver reproducibility of the scoring system, two other investigators (DT and LM) evaluated the same slides without knowledge of the former classification. The classification of the senior pathologist was used in the statistical analysis.

quantification of Bcl2 mRNA expression by real-time RT-PCR
Total RNA was extracted from the frozen samples using Trizol reagent (Invitrogen, Darmstadt, Germany). For each sample, 10 µg of total RNA was reverse transcribed using a High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA), following the manufacturer's instructions.

The complementary DNA (cDNA) was subjected to real-time PCR analysis in a Gene Amp 7000 Sequence Detection System (Applied Biosystems) using the TaqMan approach. Cycling conditions were the following: 50°C for 2 min, 95°C for 10 min, 40 cycles at 95°C for 15 sec, and 60°C for 1 min. For each sample, three replicates were analyzed. The Bcl2-specific set of primers and fluorogenic probe was purchased from Applied Biosystems. The relative amounts of Bcl2 mRNA expression were calculated using the expression of human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an endogenous control (TaqMan gene expression assays, Applied Biosystems). The final results were determined as follows: N target = 2–({Delta}Ct sample–{Delta}Ct calibrator), where {Delta}Ct values of the sample and calibrator were determined by subtracting the Ct value of the endogenous control gene from the Ct value of Bcl2 gene. In each single determination, the value obtained using a cDNA preparation from the LOVO cell line was used as calibrator. The LOVO cell line was cultured in standard conditions in our laboratory.

In order to quantify leukocyte infiltration for each tumor sample, the mRNA level of CD45 (common leukocyte antigen) was assessed by real-time RT-PCR. For this purpose, we used primers and probe with the following sequences (Pennington et al. 2001): forward 5'-GAAGTGCTGCAATGTGTCATT; reverse 5'-CTTGACATGCATACTATTATCTGATGTCA; probe 5'-FAM-ACAACTAAAAGTGCTCCTCCAAGCCAGGTCT-MGB. The relative amount of CD45 mRNA expression was calculated using the expression of human GAPDH as an endogenous control; the final results, expressed as N-fold differences in target gene expression relating to both the endogenous control gene expression and the calibrator, were determined using the 2–{Delta}{Delta}Ct method, as previously reported. The calibrator used to quantify CD45 expression was obtained from a mixture of LOVO cells (60%) and lymphocytes (40%).

statistical analysis
Differences and associations among categorical variables were analyzed using the Mann–Whitney, Kruskal–Wallis, or chi-square tests, as appropriate. Interobserver agreement for Bcl2 immunostaining classification was assessed by the kappa statistic. Disease-free survival (DFS) and overall survival (OS) analyses were carried out by applying univariate and stepwise forward multivariate Cox proportional hazards regression models [30]. The P values were considered statistically significant if <0.05. In multivariate Cox analyses, when the P value of a variable was between 0.051 and 0.060, we considered the likelihood ratio test between the models including and excluding the variable, and if this test was statistically significant the variable was maintained in the model. All statistical analyses were carried out using the STATA/SE 9.0 software (Stata Corporation, College Station, TX).


   results
Top
 Abstract
 introduction
 patients and methods
 results
discussion
 Acknowledgements
 References
 
identification of three classes of Bcl2 protein expression determined by immunohistochemical analysis
Bcl2 immunostaining was carried out in the first cohort of 72 patients who had been selected solely on the basis of frozen tissue sample availability. This group included 59 ductal carcinomas (81.9%), nine lobular carcinomas (12.5%), two pure mucoid carcinomas (2.8%), and two medullary carcinomas (2.8%). In positive cells, Bcl2 protein was expressed in the cytoplasm. In the samples examined, both the staining intensity and distribution were highly heterogeneous, ranging from the total absence of immunostaining to various intermediate staining patterns, up to a strong and homogenous immunostaining. In most of the studies reported in literature, Bcl2 immunostaining had been assessed by counting the number of labeled cells out of the total number of cancer cells within a variable number of microscopic fields, and defining different groups according to variable and arbitrary cut-off points [15, 17, 19–24]. In our opinion, this quantitative method is not suitable for Bcl2 evaluation, since it does not take into account the staining intensity, which may be highly variable in cases with the same number of positive cells. Figure 1 shows five ductal carcinomas, which are characterized by various Bcl2 protein expressions. Figure 1A and B show the two extreme situations observed in our samples, ranging from the total absence of immunostaining (Figure 1A) to an intense cytoplasmic and plasmatic staining of all cancer cells (Figure 1B). In between, variable staining patterns were observed, including cases with a weak/moderate homogeneous cytoplasmic staining (Figure 1C) or with a heterogeneous staining distribution within the tumor area (Figure 1D and E). Considering the high variability of the immunostaining patterns observed, here we have a tested semiquantitative classification based on both the distribution and the intensity of the staining reaction. Cases with no immunostaining in any cancer cell were scored as class 0; cases showing an intense immunostaining in all cancer cells were scored in class 2; all the remaining cases, including those with a weak/moderate immunostaining homogeneously distributed throughout the cancer cell population, and those showing variable intratumor immunostaining intensities (also comprising cases with isolated very intensively stained clusters of cells), were scored in a unique class named class 1. According to this classification, 13 cases (18.1%) were scored in class 0, 17 cases (23.6%) in class 1, and 42 cases (58.3%) in class 2.


Figure 1
View larger version (82K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Different Bcl2 immunostaining patterns in infiltrating ducal carcinomas of the breast. (A) Negative immunostaining of cancer cells (note that stromal cells—used as positive internal control—stain positive). (B) Intense immunostaining of all cancer cells. (C) Homogeneously weak staining of cancer cells. (D) and (E) Marked intratumor heterogeneity for bcl2 immunostaining. The case reported in (A) was scored in class 0; the case reported in (B) was scored in class 2; the cases reported in (C), (D), and (E) were score in class 1.

 
relationship between Bcl2 mRNA expression and Bcl2 immunostaining
In the same group, Bcl2 mRNA expression was assessed by real-time RT-PCR. The expression level of Bcl2 mRNA was standardized to the expression level of mRNA of the GAPDH housekeeping gene. Bcl2 mRNA expression measured in the LOVO human cancer cell line was used as calibrator. Since neoplastic lesions are often infiltrated by lymphocytes expressing high levels of Bcl2, and this variable may constitute a possible misleading source of Bcl2 mRNA, in each tumor sample leukocyte infiltration was also evaluated by quantifying the levels of CD45 (common leukocyte antigen) mRNA by real-time RT-PCR. In these analyses, the calibrator was obtained from a mixture of LOVO (60%) and lymphocytes (40%). Three cases (including one ductal carcinoma, one lobular carcinoma, and one medullary carcinoma) displayed a CD45 mRNA expression greater than the calibrator, thus indicating diffuse lymphocyte infiltration, and were excluded from the study. Therefore, Bcl2 mRNA expression was considered in 69 patients only. In this series, Bcl2 mRNA values, expressed in arbitrary units, ranged from 0.24 to 324.03, with a mean (±SD) value of 51.11 (±65.86) and a median value of 26.54. Figure 2 reports the Bcl2 mRNA level distribution within the three Bcl2 immunostaining classes. The mean (±SD) mRNA values were 1.17 (±1.14) in the 11 patients scored in class 0, 11.44 (±11.96) in the 17 patients scored in class 1, and 80.96 (±71.06) in the 41 patients scored in class 2. The difference in the Bcl2 mRNA values among the three immunohistochemical classes was highly significant (Kruskal–Wallis test P < 0.001; Mann–Whitney test class 0 versus class 1, P < 0.001; class 0 versus class 2, P < 0.001; class 1 versus class 2, P < 0.001).


Figure 2
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Box plots showing Bcl2 mRNA values within the three Bcl2 immunostaining classes. The median Bcl2 value is depicted by horizontal bars, the interquartile range by boxes and the minimum and maximum values by vertical bars.

 
relationship between Bcl2 protein expression and histopathological and clinical data
Our three class-based immunohistochemical scoring system was applied to 442 ductal breast carcinomas. In order to determine the interobserver reproducibility of the proposed classification, the Bcl2 immunostaining reaction was first assessed by a senior pathologist (MD). Fifty-two cases (12.7%) were scored in class 0, 184 (41.6%) in class 1, and the remaining 202 (45.7%) in class 2. All slides were then independently and blindly scored by two additional investigators (DT and LM), who were totally unaware of all case data. Interscorer agreement was assessed by the kappa statistic. Kappa ({kappa}) is a measure of agreement: a value of 1.00 indicates perfect agreement and 0 indicates a level of agreement by chance only. Values >0.6 are generally taken to denote acceptable concordance between scorers. The {kappa} values obtained in the present study—by comparing the scores obtained by the three independent investigators—were 0.87, 0.83, and 0.82, respectively, thus indicating a good interobserver reproducibility of the proposed classification.

Table 1 reports the associations between Bcl2 immunostaining and the histopathological characteristics of the patients enrolled in the study. The Bcl2 parameter showed the closest associations with ER status, nuclear grade, proliferate activity, and HER2 status, while a weaker association was observed with tumor size and lymph node status.


View this table:
[in this window]
[in a new window]

  		Table 1. Association between bcl2 immunostaining and histopathological characteristics in 442 ductal carcinomas

 
All patients were prospectively followed up for a median observation period of 106 months (range 4–139 months). Among them, 292 (66.1%) underwent mastectomy and 150 (33.9%) underwent conservative breast surgery followed by radiation therapy. One hundred and eleven patients received adjuvant systemic chemotherapy (cyclophosphamide, methotrexate, and 5-fluorouracil). One hundred and sixty-nine patients who did not receive systemic chemotherapy received adjuvant endocrine therapy alone (tamoxifen, 20 mg daily). The 11-year cumulative DFS and OS estimates for the whole series of patients were 67.65% and 72.47%, respectively. For survival analysis, all variables were categorized. The Ki67 variable was dichotomized using the cut-off point of 20% which, in a previous study, provided the best sensitivity and specificity for distinguishing high- from low-risk patients according to the receiving operator characteristic (ROC) curve analysis [30]. Patients were divided by age into two groups: a first group included 108 patients (24.4%) aged ≤50 years, while the second group included 334 patients (75.6%) of an age >50 years. Since histological grade and nuclear grade were closely associated with each other (Pearson chi-square = 189.439; P < 0.001), in survival analyses we considered only the nuclear grade, which expressed the highest prognostic value. Due to the low number of pT3 cases in our series (n = 11; 2.5%), pT3 and pT4 tumors were considered together.

In the univariate DFS and OS analyses carried out on the entire series of patients, all variables tested achieved statistical significance, with the exception of patients’ age (Table 2). When variables were included in a multivariate model, Bcl2 expression, N status, and Ki67 LIa emerged as independent predictors for both DFS and OS, while tumor size was significant in the OS analysis only (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2. Cox proportional hazards analysis of prognostic factors for DFS and OS applied to the whole series of cases

 
In a second analysis, cases were split according to the most informative prognostic parameters reported in literature (i.e. N status), and lymph node-negative and -positive patients were examined separately. In the group of node-negative patients, Bcl2 expression and tumor size were capable of significantly predicting both DFS and OR in multivariate analysis (Table 3), whereas in the group of patients with lymph node involvement, Bcl2 expression and Ki67 LIa showed an independent prognostic relevance (Table 4).


View this table:
[in this window]
[in a new window]

  		Table 3. Cox proportional hazards analysis of prognostic factors for DFS and OS applied to lymph node-negative patients

 

View this table:
[in this window]
[in a new window]

  		Table 4. Cox proportional hazards analysis of prognostic factors for DFS and OS applied to lymph node-positive patients

 
An additional survival analysis was carried out in subgroups of patients treated with different adjuvant protocols. Two groups were considered separately: a first group consisted of 169 patients treated with endocrine therapy only and the second group consisted of 111 patients who received chemotherapy either alone or associated with endocrine therapy. Multivariate analysis carried out in patients receiving endocrine therapy indicated Bcl2 expression, N status, and Ki67 LIa as significant predictors for both DFS and OS, while patient age was significant only in DFS analysis and tumor size was significant only in OS analysis (Table 5). In the group of patients who received chemotherapy as adjuvant treatment, Bcl2 immunostaining and N status and HER2 status emerged as independent predictors of clinical outcome in both DFS and OS multivariate analyses (Table 6).


View this table:
[in this window]
[in a new window]

  		Table 5. Cox proportional hazards analysis of prognostic factors for DFS and OS applied to patients treated by endocrine therapy only

 

View this table:
[in this window]
[in a new window]

  		Table 6. Cox proportional hazards analysis of prognostic factors for DFS and OS applied to patients treated by chemotherapy

 

   discussion
Top
 Abstract
 introduction
 patients and methods
 results
 discussion
Acknowledgements
 References
 
In the present study, we have applied a simple, semiquantitative classification of Bcl2 immunostaining to a group of 442 infiltrating ductal carcinomas of the breast. The classification, which takes into account both the distribution and the intensity of the staining reaction, consists of the definition of three well-cut groups of tumors characterized by: (i) the absence of immunostaining (class 0); (ii) the presence of an intense homogenous immunostaining (class 2); and (iii) the presence of intermediate staining intensities and/or the coexistence of variable distribution patterns within the same tumor area (class 1). Before starting the study, we validated this classification in 69 breast cancer specimens by comparing the Bcl2 immunostaining with the Bcl2 mRNA levels evaluated by real-time RT-PCR on frozen samples from the corresponding tumors. Our results, showing a highly significant association between Bcl2 protein and -mRNA values, demonstrated that the proposed Bcl2 scoring system permits a reliable assessment of Bcl2 expression in breast tumors.

The immunohistochemical expression of Bcl2 protein was assessed in patients who were prospectively followed up for a long period, in order to define its clinical relevance. Consistent with the data found in literature, a high Bcl2 expression resulted to be associated with prognostic factors which express a well-differentiated phenotype (low histological/nuclear grade, positive ER/PRG status, and low proliferation rate), while a weaker correlation was found between Bcl2 levels and parameters indicating tumor progression (tumor size and nodal status). In the entire group, patients expressing high Bcl2 levels had significantly improved DFS and OS. Moreover, when compared with the other well-established variables that are commonly considered for the prognostic definition of breast cancer patients, Bcl2 expression emerges as an independent predictor of both DFS and OS and maintained its independence even when patients were considered separately according to their lymph node status. Our results appear particularly interesting in node-negative patients since, within this subgroup, biomolecular markers capable of accurately predicting survival are still lacking [31]. Five previous investigations have assessed the prognostic impact of Bcl2 protein expression on node-negative breast cancer patients: three of them [15, 16, 32] failed to demonstrate any prognostic value of the Bcl2 parameter, while the other two showed a moderate prognostic relevance in univariate analysis only [19, 33]. All these investigations have considered two groups of patents only, characterized by absent-to-weak versus moderate-to-strong immunostaining. In the present study, we have applied a three class-based scoring system which showed a greater prognostic impact. In particular, our classification allows the identification of a subset of node-negative patients characterized by a negative Bcl2 immunostaining, who have a very high probability of relapse or death (respectively about five and seven times greater than patients with a positive Bcl2 immunostaining), and who could probably benefit from additional treatments.

The prognostic impact of the Bcl2 expression was also separately evaluated within subsets of patients who received different adjuvant treatments. In the 169 cases who received endocrine therapy only, Bcl2 immunostaining showed an independent prognostic capacity together with cell proliferation rate and N status, and proved to be a more accurate predictor of survival than ER status, in accordance with previous investigations [15, 24, 25]. Also within the group of 111 patients treated with chemotherapy, Bcl2 expression disclosed a significant prognostic relevance, and in the multivariate analysis it emerged as the only independent predictor of both DFS and OS, together with the lymph node status. These results are in line with previous studies obtained in breast cancers [17, 20, 22]; they are, however, totally in contrast with the results obtained both in other tumor types [8, 9, 34] and in in vitro models [35–37], including human breast cancer cell lines [38, 39]. In this regard, Yang et al. [40] evaluated the in vitro chemosensitivity of 117 breast carcinoma specimens to five chemotherapeutic drugs (mitomicin D, 5-fluorouracil, adriamicin, cisplatinum, and cyclophosphamide), and found that the susceptibility of Bcl2-negative tumors to the tested drugs was significantly higher than that of Bcl2-positive tumors. In another clinical study, however, the same research group demonstrated that in breast cancer patients treated with chemotherapy, Bcl2 overexpression was a significant favorable predictive factor for survival [20]. Similar results had been reported by Gasperini et al. [22] and Le et al. [17], who found that a high Bcl2 expression was associated with a better outcome in patients treated with chemotherapy. Our results are consistent with these latter observations and confirm that, in contrast to in vitro data on drug resistance of breast cancer cells, breast cancer patients treated with chemotherapy have a better clinical outcome when expressing high Bcl2 levels.

In conclusion, the present results indicate that the proposed three class-based classification of Bcl2 immunostaining, validated by its close association with Bcl2 mRNA expression, makes it possible to obtain a reliable prognostic indication of clinical outcome for patients with infiltrating ductal carcinomas of the breast. The Bcl2 parameter was shown to be particularly predictive in node-negative cancers.


   Acknowledgements
Top
 Abstract
 introduction
 patients and methods
 results
 discussion
 Acknowledgements
References
 
This work was supported by grants from Pallotti's Legacy for Cancer Research and Ministero dell'Università e della Ricerca, Finanziamenti per la Ricerca Fondamentale Orientata.

Received for publication November 15, 2006. Revision received January 31, 2007. Accepted for publication February 1, 2007.


   References
Top
 Abstract
 introduction
 patients and methods
 results
 discussion
 Acknowledgements
 References
 
1. Adams JM, Cory S. The Bcl-2 protein family: arbiters of cell survival. Science (1998) 28:1322–1326.

2. Karakas T, Maurer U, Weidmann E, et al. High expression of bcl-2 mRNA as a determinant of poor prognosis in acute myeloid leukemia. Ann Oncol (1998) 9:159–165.[Abstract/Free Full Text]

3. Hermine O, Haioun C, Lepage E, et al. Prognostic significance of bcl-2 protein expression in aggressive non-Hodgkin's lymphoma. Groupe d'Etude des Lymphomes de l'Adulte (GELA). Blood (1996) 87:265–272.[Abstract/Free Full Text]

4. Quinn DI, Henshall SM, Sutherland RL. Molecular markers of prostate cancer outcome. Eur J Cancer (2005) 41:858–887.[CrossRef][ISI][Medline]

5. Castle VP, Heidelberger KP, Bromberg J, et al. Expression of the apoptosis-suppressing protein bcl-2, in neuroblastoma is associated with unfavorable histology and N-myc amplification. Am J Pathol (1993) 143:1543–1550.[Abstract]

6. Eid H, Gulyas M, Geczi L, et al. Expression of bcl-2 in testicular carcinoma: correlation with tumor progression and MDR1/Pgp. Cancer (1998) 83:331–336.[CrossRef][ISI][Medline]

7. Rosser CJ, Reyes AO, Vakar-Lopez F, et al. Bcl-2 is significantly overexpressed in localized radio-recurrent prostate carcinoma, compared with localized radio-naive prostate carcinoma. Int J Radiat Oncol Biol Phys (2003) 56:1–6.[ISI][Medline]

8. Campos L, Rouault JP, Sabido O, et al. High expression of bcl-2 protein in acute myeloid leukemia cells is associated with poor response to chemotherapy. Blood (1993) 81:3091–3096.[Abstract/Free Full Text]

9. Lotem J, Sachs L. Regulation by bcl-2, c-myc, and p53 of susceptibility to induction of apoptosis by heat shock and cancer chemotherapy compounds in differentiation-competent and -defective myeloid leukemic cells. Cell Growth Differ (1993) 4:41–47.[Abstract]

10. Pezzella F, Turley H, Kuzu I, et al. bcl-2 protein in non-small-cell lung carcinoma. N Engl J Med (1993) 329:690–694.[Abstract/Free Full Text]

11. Ohbu M, Saegusa M, Kobayashi N, et al. Expression of bcl-2 protein in esophageal squamous cell carcinomas and its association with lymph node metastasis. Cancer (1997) 79:1287–1293.[CrossRef][ISI][Medline]

12. Dimitrakakis C, Kymionis G, Diakomanolis E, et al. The possible role of p53 and bcl-2 expression in cervical carcinomas and their premalignant lesions. Gynecol Oncol (2000) 77:129–136.[CrossRef][ISI][Medline]

13. Divito KA, Berger AJ, Camp RL, et al. Automated quantitative analysis of tissue microarrays reveals an association between high Bcl-2 expression and improved outcome in melanoma. Cancer Res (2004) 64:8773–8777.[Abstract/Free Full Text]

14. Charpin C, Garcia S, Bonnier P, et al. bcl-2 automated and quantitative immunocytochemical assays in breast carcinomas: correlation with 10-year follow-up. J Clin Oncol (1998) 16:2025–2031.[Abstract]

15. Hellemans P, van Dam PA, Weyler J, et al. Prognostic value of bcl-2 expression in invasive breast cancer. Br J Cancer (1995) 72:354–360.[ISI][Medline]

16. Joensuu H, Pylkkanen L, Toikkanen S. Bcl-2 protein expression and long-term survival in breast cancer. Am J Pathol (1994) 145:1191–1198.[Abstract]

17. Le MG, Mathieu MC, Douc-Rasy S, et al. c-myc, p53 and bcl-2, apoptosis-related genes in infiltrating breast carcinomas: evidence of a link between bcl-2 protein over-expression and a lower risk of metastasis and death in operable patients. Int J Cancer (1999) 84:562–67.[CrossRef][ISI][Medline]

18. Charpin C, Garcia S, Bouvier C, et al. Automated and quantitative immunocytochemical assays of CD44v6 in breast carcinomas. Hum Pathol (1997) 28:289–296.[CrossRef][ISI][Medline]

19. Silvestrini R, Veneroni S, Daidone MG, et al. The Bcl-2 protein: a prognostic indicator strongly related to p53 protein in lymph node-negative breast cancer patients. J Natl Cancer Inst (1994) 86:499–504.[Abstract/Free Full Text]

20. Yang QF, Sakurai T, Yoshimura G, et al. Expression of Bcl-2 but not Bax or p53 correlates with in vitro resistance to a series of anticancer drugs in breast carcinoma. Breast Cancer Res Treat (2000) 61:211–216.[CrossRef][ISI][Medline]

21. Berardo MD, Elledge RM, de Moor C, et al. bcl-2 and apoptosis in lymph node positive breast carcinoma. Cancer (1998) 82:1296–1302.[CrossRef][ISI][Medline]

22. Gasparini G, Barbareschi M, Doglioni C, et al. Expression of bcl-2 protein predicts efficacy of adjuvant treatments in operable node-positive breast cancer. Clin Cancer Res (1995) 1:189–198.[Abstract]

23. McCallum M, Baker C, Gillespie K, et al. A prognostic index for operable, node-negative breast cancer. Br J Cancer (2004) 90:1933–1941.[CrossRef][ISI][Medline]

24. Elledge RM, Green S, Howes L, et al. bcl-2, p53, and response to tamoxifen in estrogen receptor-positive metastatic breast cancer: a Southwest Oncology Group study. J Clin Oncol (1997) 15:1916–1922.[Abstract/Free Full Text]

25. Gee JM, Robertson JF, Ellis IO, et al. Immunocytochemical localization of BCL-2 protein in human breast cancers and its relationship to a series of prognostic markers and response to endocrine therapy. Int J Cancer (1994) 59:619–628.[ISI][Medline]

26. Schiller AB, Clark WS, Cotsonis G, et al. Image cytometric bcl-2:bax and bcl-2:bcl-x ratios in invasive breast carcinoma: correlation with prognosis. Cytometry (2002) 50:203–209.[CrossRef][ISI][Medline]

27. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology (1991) 19:403–410.[ISI][Medline]

28. Faccioli S, Chieco P, Gramantieri L, et al. Cytometric measurement of cell proliferation in echo-guided biopsies from focal lesions of the liver. Mod Pathol (1996) 9:120–125.[ISI][Medline]

29. Ellis IO, Bartlett J, Dowsett M, et al. Updated recommendations for HER2 testing in the UK. J Clin Pathol (2004) 57:233–237.[Abstract/Free Full Text]

30. Cox DR. Regression models and life-tables. J R Stat Soc B (1972) 34:87–220.

31. Treré D, Ceccarelli C, Migaldi M, et al. Cell proliferation in breast cancer: a major determinant of clinical outcome in node-positive, but not in node-negative patients. Appl Immunohistochem Mol Morphol (2006) 14:314–323.[CrossRef][ISI][Medline]

32. Lipponen P, Pietilainen T, Kosma VM, et al. Apoptosis suppressing protein bcl-2 is expressed in well-differentiated breast carcinomas with favourable prognosis. J Pathol (1995) 177:49–55.[CrossRef][ISI][Medline]

33. van Slooten HJ, van de Vijver MJ, van de Velde CJ, van Dierendonck JH. Loss of Bcl-2 in invasive breast cancer is associated with high rates of cell death, but also with increased proliferative activity. Br J Cancer (1998) 77:789–796.[ISI][Medline]

34. Pepper C, Bentley P, Hoy T. Regulation of clinical chemoresistance by bcl-2 and bax oncoproteins in B-cell chronic lymphocytic leukaemia. Br J Haematol (1996) 95:513–517.[CrossRef][ISI][Medline]

35. Miyashita T, Reed JC. Bcl-2 oncoprotein blocks chemotherapy-induced apoptosis in a human leukemia cell line. Blood (1993) 81:151–157.[Abstract/Free Full Text]

36. Reed JC, Kitada S, Takayama S, Miyashita T. Regulation of chemoresistance by the bcl-2 oncoprotein in non-Hodgkin's lymphoma and lymphocytic leukemia cell lines. Ann Oncol (1994) 1:61–65.

37. Sartorius UA, Krammer PH. Upregulation of Bcl-2 is involved in the mediation of chemotherapy resistance in human small cell lung cancer cell lines. Int J Cancer (2002) 97:584–592.[CrossRef][ISI][Medline]

38. Davis JM, Navolanic PM, Weinstein-Oppenheimer CR, et al. Raf-1 and Bcl-2 induce distinct and common pathways that contribute to breast cancer drug resistance. Clin Cancer Res (2003) 9:1161–1170.[Abstract/Free Full Text]

39. Teixeira C, Reed JC, Pratt MA. Estrogen promotes chemotherapeutic drug resistance by a mechanism involving Bcl-2 proto-oncogene expression in human breast cancer cells. Cancer Res (1995) 55:3902–3907.[Abstract/Free Full Text]

40. Yang Q, Sakurai T, Yoshimura G, et al. Prognostic value of Bcl-2 in invasive breast cancer receiving chemotherapy and endocrine therapy. Oncol Rep (2003) 10:121–125.[ISI][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
18/6/1004    most recent
mdm074v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Treré, D
Right arrow Articles by Derenzini, M
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Treré, D
Right arrow Articles by Derenzini, M
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?