help button home button The Oncologist
HOME HELP CONTACT US SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
First Published Online December 17, 2008
The Oncologist, Vol. 13, No. 12, 1235-1245, December 2008; doi:10.1634/theoncologist.2008-0073
© 2008 AlphaMed Press

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
theoncologist.2008-0073v1
13/12/1235    most recent
Right arrow eLetters: Submit a response to this article
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article link to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Reprints/Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Penault-Llorca, F.
Right arrow Articles by Durando, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Penault-Llorca, F.
Right arrow Articles by Durando, X.

Breast Cancer

Changes and Predictive and Prognostic Value of the Mitotic Index, Ki-67, Cyclin D1, and Cyclo-oxygenase-2 in 710 Operable Breast Cancer Patients Treated with Neoadjuvant Chemotherapy

Frédérique Penault-Llorcaac, Catherine Abriala, Inès Raoelfilsa, Philippe Cholletac, Anne Cayrea, Marie-Ange Mouret-Reyniera,b, Emilie Thivata, Florence Mishellanya,b, Pierre Gimberguesa,b, Xavier Durandoa,b

aCentre Jean Perrin and INSERM, Clermont Ferrand Cedex, France; bUniversité d'Auvergne, Faculté de Médecine, Clermont-Ferrand, France; cCentre d'Investigation Clinique, Clermont-Ferrand Cedex, France

Key Words. Breast cancer • Mitotic index • Ki-67 • Cyclin D1 • Cyclo-oxygenase-2

Correspondence: Catherine Abrial, Ph.D., Centre Jean Perrin, Bureau de Recherche Clinique, 58, rue Montalembert, BP 392, 63011 Clermont-Ferrand Cedex 1, France. Telephone: 33-4-73-27-80-05; Fax: 33-4-73-27-80-29; e-mail: Catherine.Abrial{at}cjp.fr

Received March 25, 2008; accepted for publication November 25, 2008; first published online in THE ONCOLOGIST Express on December 17, 2008.

Disclosure: The content of this article has been reviewed by independent peer reviewers to ensure that it is balanced, objective, and free from commercial bias. No financial relationships relevant to the content of this article have been disclosed by the authors, planners, independent peer reviewers, or staff managers.


   ABSTRACT
Top
 Abstract
Introduction
 Patients and Methods
 Results
 Discussion
 Conclusion
 Author Contributions
 References
 
The current study expands upon previous work using a database of 710 patients treated with neoadjuvant chemotherapy. First, we studied phenotypic characteristics of tumors before and after chemotherapy using the following factors: the mitotic index of the Scarff–Bloom–Richardson grade, Ki-67, cyclin D1, and cyclo-oxygenase-2. Second, the predictive value of these factors on response was assessed. Third, we measured the prognostic impact of these markers post-therapy in comparison with clinical and pathological responses according to the Chevallier and Sataloff classifications.

Patients were treated using different neoadjuvant chemotherapy combinations, mainly in successive prospective phase II trials. They received a median number of six cycles (range, 1–9). After neoadjuvant chemotherapy, patients underwent surgery and radiotherapy. In cases of important residual disease, some received additional courses of chemotherapy. In addition, menopausal patients with hormone receptor–positive tumors received tamoxifen for 5 years.

According to our analysis, we found significant variations before and after neoadjuvant chemotherapy only for cyclin D1 and the mitotic index. Concerning the predictive value of biomarkers for response, Ki-67 and the mitotic index were predictive on univariate analysis, both for objective clinical and pathological complete responses. Because these two factors were correlated, no multivariate analyses were conducted. We then assessed the prognostic impact of the biopathological factors. When the factors were measured before chemotherapy, all were prognostic. When evaluated after chemotherapy, the mitotic index, objective clinical response, and pathological complete response were prognostic. Because these factors were correlated, no multivariate model was done.

The main clinical fact is that there were significant correlations between clinical and pathological responses and variations in the biological factors studied.


   INTRODUCTION
Top
 Abstract
 Introduction
Patients and Methods
 Results
 Discussion
 Conclusion
 Author Contributions
 References
 
Neoadjuvant chemotherapy has been used to treat inflammatory and locally advanced breast carcinoma for the last three decades, first to improve local control and then, if possible, to improve patient survival. Recently this strategy was extended to the management of patients with operable disease, eligible for mastectomy, mainly in order to increase the rate of breast conservation [1]. This strategy most often produces a major objective regression of the tumor in order to avoid mastectomy (in about 50%–80% of cases) and to treat clinically undetectable micrometastatic disease. In addition, neoadjuvant chemotherapy permits the assessment of response of the primary tumor to a particular chemotherapy regimen and provides an early opportunity to change agents if the tumor appears clinically resistant. Thus, the assessment of predictive factors seems absolutely necessary. Moreover, pathological response assessment after chemotherapy (a prognostic factor) allows for the collection of interesting data on the chemosensitivity of residual disease and on patient survival. Predictive and prognostic factors have already been studied in the neoadjuvant setting, but the data are still scarce, and it would be very interesting to develop them.

The current study is a continuation of others we have published [2–4]. We have assessed several factors, such as hormone receptors, human epidermal growth factor receptor (HER)-2, Scarff–Bloom–Richardson (SBR) grade and modified (M)SBR grade, and nodal involvement. In this manuscript, we focus on other markers that are characteristic of tumor proliferation and aggressiveness, that is, the mitotic index of the SBR grade, Ki-67, cyclin D1, and cyclo-oxygenase (COX)-2. First, we studied these factors before and after chemotherapy. Second, we assessed their predictive value. Third, we measured the prognostic impact of these markers and the prognostic impact of a clinical response and pathological response according to the classification systems of Chevallier et al. [5] and Sataloff et al. [6].


   PATIENTS AND METHODS
Top
 Abstract
 Introduction
 Patients and Methods
Results
 Discussion
 Conclusion
 Author Contributions
 References
 
Patient Population
In the present study, 710 operable, stage II–III breast cancer patients received neoadjuvant chemotherapy. Patients with primary inflammatory carcinoma were excluded. The majority of these patients (500) were treated at Centre Jean Perrin (Clermont-Ferrand), and the others were treated at the Centre Hospitalier Universitaire (Tours), Centre Paul Papin (Angers), and Centre Hospitalier (Brive-la-Gaillarde).

The baseline workup included a complete history and clinical examination, bilateral mammography, and bilateral breast ultrasound. A diagnosis of carcinoma was established through fine-needle aspiration or core needle biopsy of the primary tumor and palpable lymph nodes. When invasive adenocarcinoma was demonstrated, the tumor was evaluated for Scarff–Bloom–Richardson (SBR) grade and hormone receptors were assessed by radioimmunology or immunohistochemistry (IHC). The laboratory assessment consisted of a CBC, blood chemistry analysis, and measurement of the tumor markers (carcinoembryonic antigen and cancer antigen 15.3). The absence of distant metastasis was confirmed by chest x-ray, bone scan, and liver ultrasound.

Treatment Modalities
Between 1982 and 2004, 710 patients were treated using different protocols of neoadjuvant chemotherapy—doxorubicin, vincristine, cyclophosphamide, and fluorouracil with or without methotrexate (AVCF/M); theprubicin, vinorelbine, cyclophosphamide, and fluorouracil (TNCF); vinorelbine, epirubicin, and methotrexate (NEM); vinorelbine, epirubicin, and paclitaxel (NET); docetaxel; fluorouracil, epirubicin, and cyclophosphamide (FEC)-50, FEC-75, FEC-100, fluorouracil, doxorubicin, and cyclophosphamide 50 (FAC 50); and docetaxel–TNCF (TAXOTERE-TNCF)—mostly in successive phase II trials published separately [7–11]. Table 1 gives the precise protocols for each regimen. Patients received a median number of six cycles of neoadjuvant chemotherapy (range, 1–9). Chemotherapy was administrated i.v. at 21- or 28-day intervals. After neoadjuvant chemotherapy, patients underwent appropriate surgery according to the size of their residual tumor. Nevertheless, we have to emphasize that 27% (n = 42) of patients who received AVCF/M did not undergo surgery and directly received radiotherapy. Radiation therapy was applied after the completion of surgery or chemotherapy if adjuvant chemotherapy was given. Postoperative irradiation treatment was delivered to the chest wall, internal mammary lymph nodes, and supraclavicular/axillary lymph nodes. In cases of important residual disease, that is, at least four involved nodes and depending on the residual size of the tumor in the breast, patients could have received additional courses of chemotherapy. Finally, patients with hormone receptor–positive tumors received tamoxifen for 5 years.


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

  		Table 1. Dosing for the different neoadjuvant chemotherapies used (n = 710 patients)

 
Assessment of Response
Clinical, mammographic, and ultrasound measurements were recorded before treatment, every two or three cycles during neoadjuvant chemotherapy, and at the end of cytotoxic treatment before surgery. Clinical, mammographic, and echographic responses were evaluated separately by a decrease in the tumor and node volumes (the product of the two greatest perpendicular dimensions) and were classified as follows: complete response, partial response (>50%), moderate response (25%–50%), no change (±25%) with no new lesions, and progression (>25%) or the appearance of new lesions.

Global clinical response was estimated using the mean of the response percentage obtained by the three methods of measurement in the majority of cases. In other cases, the global clinical response corresponded to the responses obtained by the two other methods of measurement if one of them was not available.

Pathological response was independently evaluated after surgical resection of the remaining tumor and nodes. Pathologic responses were classified as follows, according to the classification systems of Chevallier et al. [5] and Sataloff et al. [6].

The Chevallier et al. Classification

Class 1: disappearance of all tumor on either macroscopic or microscopic assessment.
Class 2: presence of in situ carcinoma in the breast, no invasive tumor, and no tumor found in the lymph nodes.
Class 3: presence of invasive carcinoma with stromal alteration, such as sclerosis or fibrosis.
Class 4: no or few modifications in the tumor appearance.

The Sataloff et al. Classification
Breast

T-A: total or nearly total therapeutic effect.
Sataloff T-A tumor corresponds to minimal residual disease, that is, scattered cells accounting for <5% of the tumor surface assuming that a thorough sampling of the supposed tumor site has been done (up to 15 samples). The reproducibility is excellent with these rules; tumors can be either focal or widespread (there is no difference).
T-B: therapeutic effect subjectively >50%.
T-C: therapeutic effect <50%, but evident effect.
T–D: no therapeutic effect.

Nodes

N-A: therapeutic effect, but no metastasis.
N-B: no metastasis, no therapeutic effect.
N-C: therapeutic effect, but metastasis.
N-D: metastasis, no therapeutic effect.

Mitotic Counts
To obtain the mitotic count, we counted the number of mitotic figures in 10 high-power fields using a 40x objective and 10x ocular lens in areas of the tumor with the highest cellularity and number of mitotic figures. We divided the number of mitoses into three scores as used in the SBR grade [12]. All the mitotic counts were performed with the same microscope to ensure homogeneity.

IHC Studies
The Ki-67, cyclin D1, and COX-2 status were determined using IHC on 3-µm paraffin sections, conducted on whole slides before neoadjuvant chemotherapy on microbiopsies and after neoadjuvant chemotherapy on residual disease. After deparaffinization, antigen retrieval was carried out for 3 minutes in citrate buffer (pH, 7.3) in a pressure cooker for Ki-67, cyclin D1, and COX-2. Immunostaining was performed with a Nexes automated immunostainer (Ventana, Illkirch, France). The incubation time was 16 minutes and the visualization was done with the AEC detection kit (Ventana). The clones used and dilution and cutoff values for positivity are detailed in Table 2. Because many cutoff values have been assessed in the literature for biological parameters and because various results have been obtained, we have chosen to give the percentage of positive cells and then different cutoffs were applied.


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

  		Table 2. Antibodies used for immunohistochemistry

 
Statistical Analysis
The {chi}2 test was used for a descriptive analysis of the population and to assess correlations between tumor characteristics and qualitative data. The Kruskal-Wallis H-test was used to compare categorical and quantitative data. Matched-pairs tests were used to study variations in the biological markers before and after neoadjuvant chemotherapy. Response rates are presented with their 95% confidence intervals (CIs). Results were last updated in February 2008. The disease-free survival (DFS) duration was defined as the time elapsed between the date of first diagnosis and the date of first relapse, regardless of the site of relapse. The overall survival (OS) duration was the time between the date of initial diagnosis and the date of the last status report, whether the patient was alive or dead and regardless of the cause of death. Survival curves were designed using the Kaplan–Meier method [13] and were compared using the log-rank test. We used the 20-year survival rate and Rothman 95% CI to document survival. A p-value < .05 was considered significant. In the statistical hypothesis tested next, the p-value is a measure of how much evidence exists against the null hypothesis. The null hypothesis represents no change or no effect. So, the p-value is the threshold below which we consider that the observed difference in a comparison is statistically significant; that is, with a low risk for error, the null hypothesis is rejected.


   RESULTS
Top
 Abstract
 Introduction
 Patients and Methods
 Results
Discussion
 Conclusion
 Author Contributions
 References
 
Patient Population
Table 3Go lists the main characteristics of the patients. The median age was 49.5 years (range, 26–81), and 324 of the 710 patients were menopausal. The median largest diameter of the primary tumor was 4 cm (range, 1–13). In all, 417 patients had clinical lymph-node involvement. The pathological diagnosis was 554 (78%) invasive ductal carcinoma, 90 (12.6%) invasive lobular carcinoma, and 182 SBR grade III tumor. Hormone receptors were positive in 51.1% of the tumors.


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

  		Table 3. Patient characteristics, responses, and adjuvant treatments

 


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

  		Table 3. (Continued)

 
Clinical and Pathological Responses
Because of technical reasons or incomplete treatments, 10 patients were not evaluable for global clinical response. The response was consequently assessed in 700 patients. The objective response rate was 70.28% (95% CI, 66.9%–73.7%), of which 16.14% were complete responses.

Fifty-four patients were not included in the pathological assessment after chemotherapy. Most of them experienced an objective clinical response, but some did not want to undergo surgery, and after a certain period, complete responses were directly treated using radiotherapy. Of the 656 patients who underwent surgery, 94 (14.3%; 95% CI, 11.6%–17.0%) achieved a pathological complete response (pCR) in the breast and nodes according to the Chevallier et al. [5] classification, and 135 patients (25.8%; 95% CI, 22.0%–29.6%) achieved a pCR according to the Sataloff et al. [6] classification.

Surgery and Adjuvant Treatments
After neoadjuvant chemotherapy, 66% of the patients underwent conservative surgery; of the 521 patients with an axillary dissection, 265 had nodal involvement. Axillary dissection was not possible for the other patients (n = 135). In fact, 20 years ago, axillary dissections were not done systematically when patients were in complete clinical response after treatment, in order to avoid lymphedema. The median diameter of the residual tumor was 1.5 cm (range, 0–10). In all, 95% of patients received radiotherapy, 17% received adjuvant chemotherapy, and 52% received hormonal therapy with tamoxifen for 5 years.

DFS and OS
On February 29, 2008, the median follow-up period was 133 months (range, 48–312 months). The follow-up corresponds to the period between the date of diagnosis and February 29, 2008. The median follow-up was calculated from this period using software. The obtained value of 133 months means that, for 50% of the patients in this database, we have a follow-up ≥133 months, and for the other 50% of the patients, the follow-up is <133 months. The 20-year OS and DFS rates were 55.2% (95% CI, 51.5%–58.9%) and 48.4% (95% CI, 44.7%–52.1%), respectively. At 20 years, 236 recurrence events had been reported, with six patients remaining at risk, and 164 deaths had been reported with eight patients remaining at risk.

Immunohistological Data
Table 4 lists the tumor markers we assessed. These data were measured on biopsies before neoadjuvant chemotherapy. We also studied these parameters in the surgical resection after neoadjuvant chemotherapy. For each factor, there are missing data. Because our database is retrospective and long term, for several patients there is no longer a paraffin-embedded block on which to conduct the IHC analyses. For other patients, we did not obtain usable results because of insufficient residual tumor. Nevertheless, patients with missing marker data are not statistically different from the others, so this will not induce a bias in the analyses.


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

  		Table 4. Tumor markers assessed before and after neoadjuvant chemotherapy

 
Table 5 lists variations in the tumor markers before and after treatment. No significant difference in expression was observed for Ki-67 (p = .47) and COX-2 (p = .31). Conversely, we found a significant difference for cyclin D1 expression and the mitotic index. Indeed, there was a significantly higher percentage of cells that stained positive for cyclin D1 after treatment (p = .016), and there was significantly more downgrading of the mitotic index (score 2/3 to score 1, 96 patients) than upgrading (score 1 to score 2/3, 38 patients; p = 5.4 x 10–7). For patients with an objective response, we observed a significantly smaller pathological residue (p = 10–7) and a lower SBR grade and mitotic index (p = 5.3 x 10–4 and p = 1.6 x 10–5, respectively). For patients who had progressed, there was more frequently a significantly greater number of tumors that became cyclin D1 positive after treatment than the reverse (p = .046).


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

  		Table 5. Immunohistochemical measures of biomarker expression evaluated before and after neoadjuvant chemotherapy

 
Predictive Value of Biomarkers on Response to Neoadjuvant Chemotherapy
Only Ki-67 and the mitotic index were predictive of clinical response and pCR to neoadjuvant chemotherapy. A positive Ki-67 status (cutoff, 1%) was associated with an objective clinical response (p = 1.3 x 10–2) and with a pCR (2 x 10–2 and 8.6 x 103, respectively, for the Chevallier and Sataloff classifications). Patients who had tumors with a mitotic index of 3 had a significantly higher clinical response rate (p = 1.3 x 10–3) and a higher pCR rate (p = 3 x 10–6 and p = 3.7 x 10–7, respectively, for the Chevallier and Sataloff classifications). As these two factors were highly correlated, no multivariate analysis was conducted.

In Table 6, we report the percentage of positive and negative values with respect to response or nonresponse to chemotherapy for each marker.


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

  		Table 6. Predictive value of biopathological parameters measured before neoadjuvant chemotherapy

 
Prognostic Value of Biomarkers on DFS and OS
The biomarkers analyzed for the prognostic study were assessed before and after neoadjuvant chemotherapy. When we considered the biomarkers before neoadjuvant chemotherapy, none were prognostic. When we considered the biomarkers after neoadjuvant chemotherapy, the mitotic index, clinical response, and pCR according to the Chevallier and Sataloff classifications were prognostic (Table 7). The prognosis was better for mitotic index 1 tumors than for mitotic index 2 or 3 tumors (p = 1.2 x 10–6 for OS and 2.6 x 10–5 for DFS). Patients with an objective clinical response had a longer OS time than other patients (p = 2.7 x 10–2). pCR predicted longer OS and DFS times—p = 4.2 x 10–3 for OS and p = 5.6 x 10–3 for DFS according to the Chevallier classification; p = 2.1 x 10–2 for OS and p = 1.9 x 10–2 for DFS according to the Sataloff classification. Because the factors assessed in the univariate analysis were correlated with each other, we did not conduct a multivariate analysis.


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

  		Table 7. Prognostic value of biopathological parameters measured after neoadjuvant chemotherapy

 

   DISCUSSION
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
Conclusion
 Author Contributions
 References
 
Biopathological Marker Variations
Since the 1960s, it has been known that cytotoxics modify the phenotype and cellular characteristics of breast cancers [14]. After chemotherapy, tumor cells increase in size and vacuoles appear. Apoptosis mechanisms are activated. One of the interests in neoadjuvant chemotherapy is to allow for the in vivo analysis of phenotypic variations in tumor cells induced by chemotherapy. Many biomarkers have been assessed before and after neoadjuvant chemotherapy in order to observe potential changes in their expression.

According to our results, and contrary to Penault-Llorca et al. [15], in this larger series, no significant difference in expression was observed for Ki-67 (p = .47). Moreover, no significant difference was found for COX-2 (p = .31). In the literature, we did not find any study on COX-2 variations.

Conversely, significant variations have been observed for cyclin D1 and for the mitotic index. Indeed, there was a significant increase in the percentage cells that stained positive for cyclin D1 after treatment (p = .016), and there was significantly more downgrading of the mitotic index (score 2/3 to score 1, 96 patients) than upgrading (score 1 to score 2/3, 38 patients; p = 5.4 x 10–7). No study assessing cyclin D1 variations was found in the literature. Our results concerning the mitotic index agree with the work of Amat et al. [16] on variations in the SBR grade before and after neoadjuvant chemotherapy; indeed, they reported that there were more SBR grade I tumors after chemotherapy than before.

Because our results are retrospective, it would be interesting to develop prospective studies to assess variation in these markers. It would be equally interesting to evaluate whether or not variation in these factors is prognostic.

Predictive Factors

Univariate Analysis

Mitotic Index. The mitotic index is one of the three components of the SBR grade. It is one of the older parameters studied and a more simple way to assess cellular proliferation.

Concerning its predictive value, our results show that a high mitotic index (score of 3) is predictive of an objective clinical response and pCR. These results are similar to the work of Vincent-Salomon et al. [17], who showed in a smaller series of patients that a high mitotic index is predictive of a pCR to neoadjuvant chemotherapy (50% pCR rate for high mitotic index versus 7% pCR rate for low mitotic index). Our results are also comparable with those of Amat et al. [16], who reported that SBR grade III tumors responded better to neoadjuvant treatment than SBR grade I tumors.

Ki-67. The proportion of proliferative cells can be assessed by measuring the percentage of cells with positive immunostaining for the Ki-67 antigen. This antigen is present in cells engaged in the cell cycle.

A positive Ki-67 status was associated with an objective clinical response and with a pCR, according to both the Chevallier and Sataloff classifications. These results are comparable with those obtained by Collecchi et al. [18], Chang et al. [19], Vincent-Salomon et al. [17], and Burcombe et al. [20], who concluded that proliferative activity before treatment is significantly linked to tumor response.

The results obtained for the mitotic index and for Ki-67 are very informative for physicians. Indeed, it is now recognized that a proliferative tumor responds better to neoadjuvant treatment. For example, a retrospective study published by Méklati et al. [21] on metastatic pancreatic tumors showed that the etoposide plus cisplatin combination was a very effective treatment when the tumor was proliferating. Nevertheless, it is necessary to develop prospective studies with different chemotherapies in order to know which of them is most effective in proliferating tumors.

Cyclin D1. The oncogene CCND1 is located on chromosome 11. It codes for cyclin D1, a protein implicated in cell cycle control and, in particular, the G1–S transition.

In our database, no significant correlation was found between cyclin D1 expression and response to treatment. Moreover, in the literature, we did not find any study assessing the predictive value of cyclin D1. More investigations should be performed to define more accurately the predictive role of this protein. It could, perhaps, be interesting to define a score combining the percentage of cells expressing cyclin D1 and the intensity of expression. In a prospective study, such a score could be validated.

COX-2. COX-2 is an enzyme that catalyses the transformation of arachidonic acid to prostaglandin H2. In normal tissue, COX-2 is practically not assessable, but tumor cells, and particularly breast tumor cells, are responsible for an increase in the level of this protein [22]. When tumor progression occurs, prostaglandins can induce many mechanisms, such as cellular proliferation, apoptosis, immune system modulation, and angiogenesis.

In our dataset, we did not find a significant correlation between COX-2 expression and response to treatment. In addition, just as for cyclin D1, no study has been reported in the literature. Consequently, it would be very interesting to conduct retrospective studies to build a score measuring the expression of COX-2, to validate it with retrospective trials, and then to assess the predictive impact of this biomarker.

Prognostic Factors

Univariate Analysis
The biomarkers analyzed for the prognostic study were assessed before and after neoadjuvant chemotherapy. When we considered the biomarkers before neoadjuvant chemotherapy, none were prognostic. When we considered the biomarkers after neoadjuvant chemotherapy, the mitotic index, clinical response, and pCR according to the Chevallier and Sataloff classifications were prognostic. In fact, results of our earlier work have already led us to conclude that only factors assessed after neoadjuvant chemotherapy are prognostic [2, 16].

Mitotic Index. Concerning the prognostic value of the mitotic index, according to our results, the prognosis was better for patients with low mitotic index (score 1) tumors. These results are comparable with those of Simpson et al. [23] and Laroye and Minkin [24] in the adjuvant setting. In the neoadjuvant setting, Penault-Llorca et al. [15] also concluded that the mitotic count was a prognostic factor.

Ki-67. According to our results, the expression of Ki-67 was not prognostic at a cutoff of 1%. Our conclusions were similar using a cutoff of 10% or 20%. Nevertheless, many authors have reported that Ki-67 nonexpression (cutoff <10% or 20%, according to the authors) is significantly correlated with longer OS and DFS times (with a cutoff of 10% [25] or 20% [15]). It would be very important to validate these results in prospective studies.

Cyclin D1. We did not find a significant correlation between the expression of cyclin D1 (with a cutoff of 1%) and survival, even using a cutoff of 10% or 20%. Our results are comparable with those of Michalides et al. [26], Barbareschi et al. [27], and Umekita et al. [28]. Conversely, Gillett et al. [29] concluded that overexpression of cyclin D1 is predictive of a good prognosis.

Nevertheless, we have to emphasize that these authors proposed different scores to measure cyclin D1 expression. It would be interesting to define a universal score combining the intensity of IHC and the percentage of cells expressing cyclin D1 in retrospective studies, and then validate it in prospective studies.

COX-2. According to our results, the expression of COX-2 is not prognostic with a cutoff of 1%. We obtained the same results using a cutoff of 10% or 20%. Conversely, Costa et al. [22], Denkert et al. [30], Ristimäki et al. [31], and Sivula et al. [32] concluded that overexpression of COX-2 is significantly correlated with shorter OS and DFS times. This could be explained by an increase in prostaglandin production, which could be responsible for tumor progression (cellular proliferation, immune system modulation, and angiogenesis).

As for cyclin D1, the scores proposed for COX-2 in the literature are variable, so it would be interesting to propose a universal score that combines the percentage of cells expressing COX-2 and the intensity of expression. Further prospective studies could validate this score.

As mentioned above, many cutoff values have been assessed in the literature and various results have been obtained. Thus, we chose to fix the cutoff at 1%. But, because we don't know if this value is informative, we chose to conduct other analyses using 10% or 20% as the cutoff in order to have more results to compare with those in the literature.

Response. Patients with an objective response had a significantly longer OS time than other patients. As expected from previously published studies [2, 33–38], the OS and DFS times were longer for patients with a pCR according the Chevallier or Sataloff classifications.


   CONCLUSION
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
 Conclusion
Author Contributions
 References
 
To conclude, according to our results, we observed variations before and after neoadjuvant chemotherapy in the expression of cyclin D1 and of the mitotic index. Concerning Ki-67, we did not find the same results as previously published, and no variation was observed for COX-2. Because few or no studies are published on variations in cyclin D1, Ki-67, and COX-2, it would be interesting to further investigate these biomarkers in order to confirm or refute our results.

Concerning the predictive impact of the biomarkers studied, we found that only Ki-67 and the mitotic index were predictive. Because we did not find any correlation among cyclin D1, COX-2, and response to treatment, and because we found no data in the literature, it is necessary to study these biomarkers in prospective studies.

Finally, concerning the prognostic value, as previously published, when the biomarkers were measured before neoadjuvant treatment, they were not prognostic. However, when the factors were assessed after neoadjuvant treatment, significant correlations were found: the mitotic index, objective clinical response, and pCR were prognostic.

There were correlations among the different domains of evaluation: clinical, pathological, and biological. From our results, and according to the literature, it appears that we can separate patients into two groups: (a) patients with an objective clinical response, with a decrease in the SBR grade mitotic index, with little pathological residual disease, and with less aggressive biological markers and (b) patients with stable disease or who have progressed clinically, with an increase in the SBR grade mitotic index, with a large amount of pathological residual disease, and with more aggressive biological markers.


   AUTHOR CONTRIBUTIONS
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
 Conclusion
 Author Contributions
References
 
Conception/design: Frédérique Penault-Llorca, Philippe Chollet

Provision of study materials: Xavier Durando, Marie-Ange Mouret-Reynier, Pierre Gimbergues

Collection/assembly of data: Inès Raoelfils, Florence Mishellany

Data analysis: Catherine Abrial

Manuscript writing: Catherine Abrial, Emilie Thivat, Florence Mishellany

Final approval of manuscript: Frédérique Penault-Llorca, Philippe Chollet

Other: Anne Cayre, immunohistochemistry


   ACKNOWLEDGMENT
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
 Conclusion
 Author Contributions
 References
 
The authors thank Dr. Olivier Tacca for help in reading the manuscript.


   REFERENCES
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
 Conclusion
 Author Contributions
 References
 

  1. Buchholz TA, Hunt KK, Whitman GJ et al. Neoadjuvant chemotherapy for breast carcinoma: Multidisciplinary considerations of benefits and risks. Cancer 2003;98:1150–1160.[CrossRef][Medline]
  2. Abrial C, Penault-Llorca F, Delva R et al. High prognostic significance of residual disease after neoadjuvant chemotherapy: A retrospective study in 710 patients with operable breast cancer. Breast Cancer Res Treat 2005;94:255–263.[CrossRef][Medline]
  3. Penault-Llorca F, Abrial C, Mouret-Reynier MA et al. Achieving higher pathological complete response rates in HER-2-positive patients with induction chemotherapy without trastuzumab in operable breast cancer. The Oncologist 2007;12:390–396.[Abstract/Free Full Text]
  4. Tacca O, Penault-Llorca F, Abrial C et al. Changes in and prognostic value of hormone receptor status in a series of operable breast cancer patients treated with neoadjuvant chemotherapy. The Oncologist 2007;12:636–643.[Abstract/Free Full Text]
  5. Chevallier B, Roche H, Olivier JP et al. Inflammatory breast cancer. Pilot study of intensive induction chemotherapy (FEC-HD) results in a high histologic response rate. Am J Clin Oncol 1993;16:223–228.[Medline]
  6. Sataloff DM, Mason BA, Prestipino AJ et al. Pathologic response to induction chemotherapy in locally advanced carcinoma of the breast: A determinant of outcome. J Am Coll Surg 1995;180:297–306.[Medline]
  7. Chollet P, Charrier S, Brain E et al. Clinical and pathological response to primary chemotherapy in inoperable breast cancer. Eur J Cancer 1997;36:862–866.
  8. Van Praagh I, Curé H, Leduc B et al. Efficacy of a primary regimen combining vinorelbine, epirubicin and methotrexate (VEM) as neoadjuvant treatment in 89 patients with operable breast cancer. The Oncologist 2002;7:418–423.[Abstract/Free Full Text]
  9. Amat S, Bougnoux P, Penault-Llorca F et al. Neoadjuvant docetaxel for operable breast cancer induces a high pathological response and breast-conservation rate. Br J Cancer 2003;88:1339–1345.[CrossRef][Medline]
  10. Abrial C, Mouret-Reynier MA, Amat S et al. Tumor parameters, clinical and pathological responses, medical management, and survival through time on 710 operable breast cancers. Med Oncol 2005;22:233–240.[CrossRef][Medline]
  11. Amat S, Mouret-Reynier MA, Penault-Llorca F et al. Sequential addition of an anthracycline-based regimen to docetaxel as neoadjuvant chemotherapy in patients with operable breast cancer. Clin Breast Cancer 2006;7:262–269.[Medline]
  12. 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.[Medline]
  13. Kaplan EL, Meier P. Non parametric estimation from incomplete observation. J Am Stat Assoc 1957;53:457–471.[CrossRef]
  14. Waller U. [Giant nuclei after myleran therapy and splenic irradiation in chronic myeloid leukemia.]. Pathol Microbiol (Basel) 1960;23:283–290; In German.
  15. Penault-Llorca F, Cayre A, Bouchet Mishellany F et al. Induction chemotherapy for breast carcinoma: Predictive markers and relation with outcome. Int J Oncol 2003;22:1319–1325.[Medline]
  16. Amat S, Penault-Llorca F, Cure H et al. Scarff-Bloom-Richardson (SBR) grading: A pleiotropic marker of chemosensitivity in invasive ductal breast carcinomas treated by neoadjuvant chemotherapy. Int J Oncol 2002;20:791–796.[Medline]
  17. Vincent-Salomon A, Rousseau A, Jouve M et al. Breast Cancer Study Group. Proliferation markers predictive of the pathological response and disease outcome of patients with breast carcinomas treated by anthracycline-based preoperative chemotherapy. Eur J Cancer 2004;40:1502–1508.[CrossRef][Medline]
  18. Collecchi P, Baldini E, Giannessi P et al. Primary chemotherapy in locally advanced breast cancer (LABC): Effects on tumour proliferative activity, bcl-2 expression and the relationship between tumour regression and biological markers. Eur J Cancer 1998;34:1701–1704.[CrossRef][Medline]
  19. Chang J, Ormerod M, Powles TJ et al. Apoptosis and proliferation as predictors of chemotherapy response in patients with breast carcinoma. Cancer 2000;89:2145–2152.
  20. Burcombe RJ, Makris A, Richman PI et al. Evaluation of ER, PgR, HER-2 and Ki-67 as predictors of response to neoadjuvant anthracycline chemotherapy for operable breast cancer. Br J Cancer 2005;92:147–155.[CrossRef][Medline]
  21. Méklati el-HM, Lévy P, O'Toole D et al. Granular cell tumor of the pancreas. Pancreas 2005;31:296–298.[Medline]
  22. Costa C, Soares R, Reis-Filho JS et al. Cyclo-oxygenase 2 expression is associated with angiogenesis and lymph node metastasis in human breast cancer. J Clin Pathol 2002;55:429–434.[Abstract/Free Full Text]
  23. Simpson JF, Gray R, Dressler LG et al. Prognostic value of histologic grade and proliferative activity in axillary node-positive breast cancer: Results from the Eastern Cooperative Oncology Group Companion Study, EST 4189. J Clin Oncol 2000;18:2059–2069.[Abstract/Free Full Text]
  24. Laroye GJ, Minkin S. The impact of mitotic index on predicting outcome in breast carcinoma: A comparison of different counting methods in patients with different lymph node status. Mod Pathol 1991;4:456–460.
  25. Rozan S, Vincent-Salomon A, Zafrani B et al. No significant predictive value of c-erbB-2 or p53 expression regarding sensitivity to primary chemotherapy or radiotherapy in breast cancer. Int J Cancer 1998;79:27–33.[CrossRef][Medline]
  26. Michalides R, Hageman P, Van Tinteren H et al. A clinicopathological study on overexpression of cyclin D1 and of p53 in a series of 248 patients with operable breast cancer. Br J Cancer 1996;73:728–734.[Medline]
  27. Barbareschi M, Pelosio P, Caffo O et al. Cyclin-D1-gene amplification and expression in breast carcinoma: Relation with clinicopathologic characteristics and with retinoblastoma gene product, p53 and p21WAF1 immunohistochemical expression. Int J Cancer 1997;74:171–174.[CrossRef][Medline]
  28. Umekita Y, Ohi Y, Sagara Y et al. Overexpression of cyclinD1 predicts for poor prognosis in estrogen receptor-negative breast cancer patients. Int J Cancer 2002;98:415–418.[CrossRef][Medline]
  29. Gillett C, Smith P, Gregory W et al. Cyclin D1 and prognosis in human breast cancer. Int J Cancer 1996;69:92–99.[CrossRef][Medline]
  30. Denkert C, Winzer KJ, Mller BM et al. Elevated expression of cyclooxygenase-2 is a negative prognostic factor for disease free survival and overall survival in patients with breast carcinoma. Cancer 2003;97:2978–2987.[CrossRef][Medline]
  31. Ristimäki A, Sivula A, Lundin J et al. Prognostic significance of elevated cyclooxygenase-2 expression in breast cancer. Cancer Res 2002;62:632–635.[Abstract/Free Full Text]
  32. Sivula A, Talvensaari-Mattila A, Lundin J et al. Association of cyclooxygenase-2 and matrix metalloproteinase-2 expression in human breast cancer. Breast Cancer Res Treat 2005;89:215–220.[CrossRef][Medline]
  33. Brain E, Garrino C, Misset JL et al. Long-term prognostic and predictive factors in 107 stage II/III breast cancer patients treated with anthracycline-based neoadjuvant chemotherapy. Br J Cancer 1997;75:1360–1367.[Medline]
  34. Bonadonna G, Valagussa P, Brambilla C et al. Primary chemotherapy in operable breast cancer: Eight-year experience at the Milan Cancer Institute. J Clin Oncol 1998;16:93–100.[Abstract/Free Full Text]
  35. Eltahir A, Heys SD, Hutcheon AW et al. Treatment of large and locally advanced breast cancers using neoadjuvant chemotherapy. Am J Surg 1998;175:127–132.[CrossRef][Medline]
  36. Fisher B, Costantino JP, Wickerham DL et al. Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371–1388.[Abstract/Free Full Text]
  37. Kuerer HM, Newman LA, Buzdar AU et al. Residual metastatic axillary lymph nodes following neoadjuvant chemotherapy predict disease-free survival in patients with locally advanced breast cancer. Am J Surg 1998;176:502–509.[CrossRef][Medline]
  38. Chollet P, Amat S, Cure H et al. Prognostic significance of a complete pathological response after induction chemotherapy in operable breast cancer. Br J Cancer 2002;86:1041–1046.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
theoncologist.2008-0073v1
13/12/1235    most recent
Right arrow eLetters: Submit a response to this article
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article link to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Reprints/Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Penault-Llorca, F.
Right arrow Articles by Durando, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Penault-Llorca, F.
Right arrow Articles by Durando, X.

HOME HELP CONTACT US SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
THE ONCOLOGIST STEM CELLS CME ALPHAMED PRESS JOURNALS