| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
1 Department of Obstetrics and Gynecology, J. W. Goethe University, Theodor-Stern-Kai 7, 60590, Frankfurt am Main, Germany
2 German Breast Group, Neu-Isenburg, Germany
3 Praxis, Maximilianstrasse 50, 80538, München, Germany
4 St. Gertrauden Krankenhaus, Berlin, Germany
5 Department of Obstetrics and Gynecology, Otto-von-Guericke-University, Magdeburg, Germany
6 Department of Obstetrics and Gynecology, Klinikum Süd, Rostock, Germany
7 Department of Obstetrics and Gynecology, Christian-Albrecht-University, Kiel, Germany
8 Henriettenstiftung, Hannover, Germany
9 Klinikum Offenbach, Offenbach, Germany
10 HSK, Dr. Horst Schmidt Klinik, Wiesbaden, Germany
11 Wolfartklinik, Waldstrasse 7, 82166, Gräfelfing, Germany
Correspondence: Address correspondence and reprint requests to: Sibylle Loibl, MD; E-mail: loibl{at}em.uni-frankfurt.de
| ABSTRACT |
|---|
|
|
|---|
Methods: This study was an analysis of surgical data of a representative data subset of 607 patients enrolled in the GEPARDUO study. This prospective, multicenter, phase III study randomly assigned patients with operable breast cancer (
2 cm) to neoadjuvant 8-week dose-dense doxorubicin plus docetaxel or a 24-week schedule of doxorubicin plus cyclophosphamide followed by docetaxel (AC-DOC).
Results: Breast conservation was attempted in 493 (81.2%) patients, but 43 patients eventually required mastectomy, thus resulting in a breast-conserving surgery rate of 74.1%. Breast-conserving re-excision was performed in 61 patients (12.4%). Factors associated with a significantly higher breast-conserving surgery rate were a prechemotherapy tumor size
40 mm, nonlobular histological characteristics, treatment with AC-DOC, clinical response, postchemotherapy tumor size
20 mm, and treatment in a larger center (>10 enrolled patients). Nonlobular histological characteristics and intraoperative frozen-section analysis for margin evaluation were associated with significantly lower reoperation rates (P = .015).
Conclusions: Breast conservation after neoadjuvant chemotherapy is feasible in most patients with operable breast cancer. For surgical planning, tumor characteristics and response to neoadjuvant chemotherapy should be taken into account. Improved breast-imaging modalities are necessary to improve detection of residual disease after neoadjuvant chemotherapy, especially when breast cancer is of lobular invasive histology. Margin assessment by intraoperative frozen-section analysis is helpful to avoid reoperation. To achieve an optimal result, an interdisciplinary surgical approach is important.
Key Words: Neoadjuvant therapy Breast cancer Breast conservation Surgery Re-excision
| INTRODUCTION |
|---|
|
|
|---|
We recently reported the results of the GEPAR-DUO study, a neoadjuvant phase III study, which compared 8-week dose-dense doxorubicin plus docetaxel (ADOC) with a 24-week sequential schedule of doxorubicin plus cyclophosphamide followed by docetaxel (AC-DOC) in 913 patients with operable breast cancer and showed superiority of the AC-DOC regimen with respect to the pathologic response rate.19 The purpose of this study was to further analyze the details of the surgical procedures used in the GEPARDUO trial, focusing on the surgical techniques and the patient and tumor characteristics associated with successful breast-conserving surgery.
| PATIENTS AND METHODS |
|---|
|
|
|---|
Patients in the GEPARDUO trial were randomly assigned to receive either doxorubicin 50 mg/m2 plus docetaxel 75 mg/m2 each on day 1 every 14 days for four cycles with granulocyte colony-stimulating factor support (ADOC) or doxorubicin 60 mg/m2 plus cyclophosphamide 600 mg/m2 on day 1 every 21 days followed by docetaxel 100 mg/m2 every 21 days for four cycles (AC-DOC). Assessment of response was made by palpation and the most appropriate imaging method (ultrasonography, mammography, or magnetic resonance imaging). Surgery was performed within 14 to 28 days after administration of the last chemotherapy cycle. If the tumor was still too large for breast conservation, modified radical mastectomy was recommended. The patients were offered autologous or heterologous reconstructive surgery. If the tumor size allowed breast conservation, the following had to be considered:
All patients undergoing breast-conserving surgery received standard irradiation of the remaining breast after recovery from the operation. Application of the sentinel node biopsy technique was allowed in patients with tumors <2 cm and clinically uninvolved lymph nodes. A pCR was defined as no microscopic viable tumor cells (invasive or noninvasive) in the excised breast specimens and axillary lymph nodes.19
Descriptive data analysis was performed by using the SPSS software package, version 12.0 (SPSS, Inc., Chicago, IL). Statistical comparisons between groups were assessed by the
2 test. All comparisons were two tailed. A P value
.05 was considered statistically significant. For each evaluated variable, analysis was restricted to the cases with known values for the factor.
| RESULTS |
|---|
|
|
|---|
Patient and tumor characteristics at the time of breast cancer diagnosis are summarized in Table 1
. Among the 607 patients studied in the current analysis, 306 had randomly been assigned to ADOC and 301 to AC-DOC. The median clinical tumor size was 40 mm (range, 15160 mm) before and 10 mm (range, 085 mm) after neoadjuvant chemotherapy. The clinical response rates (complete and partial response) assessed by palpation and imaging were 77.4% and 69.5%, respectively. The pCR rate in this subset of patients was 8.5% for ADOC and 15% for AC-DOC (P < .001), as compared with 7.0% and 14.3%, respectively, in the main analysis.
|
|
Placement of a titan clip to mark the tumor before commencement of chemotherapy was used in 7% of the patients. Preoperative localization of the tumor either by ultrasound- or radiographic-guided wire placement or ultrasound-guided dye application was used in 194 patients (32.2%) and was more frequently applied in case of a clinical complete response (43.9%). Two patients received preoperative localization by magnetic resonance imagingguided wire placement. Conventional axillary node dissection was performed in most patients (n = 604; 99.3%), whereas only 3 patients were treated by sentinel node biopsy. One patient did not receive axillary dissection, for unknown reasons. The mean number of removed axillary lymph nodes was 15 (range, 154). In 49 patients (8.1%), fewer than 10 lymph nodes were removed.
The breast-conserving surgery rate was significantly higher among patients who had received AC-DOC compared with patients treated with ADOC (78.1% vs. 70.3%; P < .028). Patients with a clinical complete response after neoadjuvant chemotherapy were more likely to receive breast-conserving surgery than patients with a partial response or with stable or progressive disease. The type of surgery correlated strongly with clinical tumor size after chemotherapy (P < .0001). Patients treated by tumorectomy had a mean tumor diameter of 10.9 mm after chemotherapy, as compared with 22.3 mm in the mastectomy group (Fig. 1
). Ductal invasive breast cancer histology was associated with a higher breast-conservation rate than lobular invasive histology (78.6% vs. 61.0%; P < .0001).
|
10 enrolled patients) were more likely to receive breast-conserving surgery (77.8%) than 138 patients treated in one of the 40 smaller centers (<10 enrolled patients) (62.3%; P < .0001). The median clinical tumor size after chemotherapy did not differ (13 vs. 14 mm), and treatment regimens were balanced (n = 236 for ADOC vs. n = 233 for AC-DOC) in centers recruiting more or less than 10 patients. All preoperative characteristics analyzed for a possible association with breast-conserving surgery rate are summarized in Table 3
|
|
|
| DISCUSSION |
|---|
|
|
|---|
More than 70% of the patients in our study were treated by breast conservation. Free margins could mostly be achieved with the first surgical procedure, but 21.1% of the patients required re-excision either by breast-conserving surgery (12.4%) or mastectomy (8.7%). Preoperative planning and intraoperative macroscopic evaluation of the tumor-free margins seem more difficult after tumor downstaging by pre-operative systemic therapy, and this leads to additional surgical procedures. Re-excision rates after breast-conserving surgery after neoadjuvant chemotherapy have rarely been reported.4 Sadetzki et al.18 studied 100 patients with stage II and III breast cancer who were designated for lumpectomy after neoadjuvant chemotherapy. Nineteen percent of the patients proceeded to mastectomy, and 10% required re-excision lumpectomy. The higher rate of secondary mastectomies may be explained by the inclusion of patients with larger tumors, with a mean tumor size of 50 mm before chemotherapy compared with 40 mm in our study, as well as the use of a nontaxane-containing regimen. The frequency of residual cancer in re-excised tissue for close or positive margins was 42% and is comparable to the frequency observed after primary breast-conserving surgery (without prior systemic treatment).20,21
Conversion of patients who were initially candidates for mastectomy into candidates for breast conservation can be considered the most appealing benefit of neoadjuvant chemotherapy in patients with operable breast cancer.4,22 Optimizing neoadjuvant treatment has been shown to improve response rates and often translates into increased breast-conservation rates.23,24 Unfortunately, we were unable to retrospectively determine the percentage of patients eligible for breast-conserving surgery up front. However, the fact that the breast-conservation rate was improved with the more effective AC-DOC treatment in the GEPARDUO study indicates that surgeons took advantage of the observed tumor shrinkage by resecting the breast tumor according to its new margins. The frequent use of a lumpectomy as a surgical technique to achieve breast conservation in our analysis supports this hypothesis.
When performing a lumpectomy, the surgeon is guided by the (residual) breast mass and therefore most likely operates according to its new margins. Some authors have reported higher re-excision rates after lumpectomy as opposed to quadrantectomy in patients treated by primary surgery.25,26 Although there was a trend toward more frequent re-excisions in patients receiving lumpectomy in this analysis, this did not reach statistical significance. Intraoperative frozen-section analysis of margins was associated with a significantly lower re-excision rate in our study (13.3% vs. 27.0%) and might therefore be a useful instrument to minimize reoperation after attempted breast conservation. Data on the use of intraoperative margin assessment after neoadjuvant chemotherapy have not been reported to date, but a comparable reduction in number of re-excisions has been demonstrated for intraoperative margin assessment in patients undergoing primary breast-conserving surgery.27,28
Most of the factors we found to be associated with a higher breast-conservation rate are related to the initial tumor stage and the response to chemotherapy. Naturally, neoadjuvant chemotherapy is less likely to shrink a large tumor enough to enable breast conservation than it is a small tumor. Larger tumors have also been demonstrated to respond less well to neoadjuvant chemotherapy than smaller tumors.22 However, patients with tumors showing a clinical response to neoadjuvant therapy after the first two cycles have a higher chance to be treated by breast-conserving surgery (86.3%) than patients who did not respond after the first two cycles (66%).29 Special attention needs to be paid to patients with lobular invasive breast cancer histology. Only 56% of the patients with lobular invasive breast cancer were able to undergo breast conservation in our study, and the re-excision rate was significantly higher in these patients. Lobular invasive breast cancer has lower clinical and pathologic response rates after neoadjuvant chemotherapy, and this can be an explanation for the lower breast-conservation rate.3032 Moreover, the extent of disease is known to be difficult to judge in tumors of lobular histology.
It is interesting to note that the likelihood of a patients undergoing breast conservation also depended on treatment center size. Patients treated in large centers, which enrolled
10 patients in the GEPARDUO study, were more likely to receive breast-conserving therapy than patients treated in smaller centers, which enrolled <10 patients. The number of enrolled study subjects can be expected to correlate with hospital volume and surgeon experience in our study. This observation has already been reported for nonstudy patients in an epidemiological study in Germany.33 Similar observations have been made by Hiotis et al.,34 who demonstrated that not only the extent of locoregional disease but also hospital volume, surgeon volume, and surgeon specialization, affect the likelihood of a patients receiving breast-conserving surgery. Other factors such as socioeconomic status and patient choice have been demonstrated to influence breast-conservation rates as well, but this study focused on retrospective collection of surgical and pathologic reports only.34,35
One difficulty of breast-conserving surgery after neoadjuvant chemotherapy lies in the limited ability to predict which patients will have a pCR. The main analysis of the GEPARDUO study revealed that neither palpation nor any of the imaging techniques used to assess tumor response were associated with a meaningful positive predictive value for pCR of the breast tumor.19 Nevertheless, imaging techniques can be helpful in determining the location and extent of residual disease and facilitate surgery.18,36 In this study, one third of the patients received ultrasound-or mammography-guided preoperative localization of the tumor site. Although breast conservation could be achieved in most of these patients, the number of re-excisions was still 22.3%, thus suggesting that more accurate methods of detecting residual tumor after neoadjuvant chemotherapy are necessary. Magnetic resonance imaging, which was applied to evaluate the tumor response in 7.6% of the study population only, might be superior to other imaging techniques in identifying residual lesions, but this has not yet been fully established.37,38 To overcome the difficulty of later localization of the tumor site in patients with a clinical complete response to neoadjuvant chemotherapy, different techniques of marking the tumor bed either before initiation of chemotherapy or after the first cycle, according to response, have been advocated: (1) invasive procedures, such as placement of metallic markers or platinum embolization coils or marking the position and size of the tumor on the breast skin with a tattoo, and (2) noninvasive techniques, such as preparing a transparent grid or sketch and photographing the breast and tumor location.15,35,39,40 Only 7% of the patients in our analysis received placement of a titan clip before chemotherapy began, but information on the number and kind of noninvasive tumor markings by drawing or photography was not available and could therefore not be analyzed. In all protocols of consecutive neoadjuvant chemotherapy studies conducted by the German Breast Group, marking of the primary tumor site is recommended, according to Kaufmann et al.15
Recently, Chen et al.41 presented the M.D. Anderson prognostic index as an instrument for selecting patients for breast conservation after neo-adjuvant chemotherapy on the basis of their risk of developing local failure. Factors predictive of increased local failure rates in their study (e.g., multi-focal disease and lymphovascular space invasion) were similar to the postoperative variables associated with lower breast-conservation rates in our study. Unfortunately, most factors associated with increased local failure in Chen and associates study can be determined only after surgery and are therefore not available for initial surgical procedure planning.
Our study has several limitations. First, we could not measure the cosmetic benefit many patients may have experienced, because tumor shrinkage after neoadjuvant chemotherapy allowed resection of a smaller volume of tissue. Second, this analysis was focused on the practical aspects of surgery after neoadjuvant chemotherapy, and, therefore, data on local failure were not presented. However, analysis of disease-free (intrabreast, local, regional, and distant) and overall survival is in preparation.
In conclusion, the results of this study demonstrate that breast conservation after neoadjuvant chemotherapy is feasible in most patients with operable breast cancer. For surgical planning, it is important to take into account the initial tumor stage and clinical response to neoadjuvant chemotherapy. Improved breast-imaging modalities are necessary to improve the accuracy of detection of residual disease after neoadjuvant chemotherapy, especially when breast cancer is of lobular invasive histology. To facilitate surgery, we recommend marking the tumor site before commencement of chemotherapy and intraoperative frozen-section analysis for margin assessment to avoid reoperation. Surgery after neo-adjuvant chemotherapy for breast cancer is challenging and should therefore be performed by an experienced breast surgeon.
| ACKNOWLEDGMENTS |
|---|
Received for publication November 22, 2005. Accepted for publication February 13, 2006.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. von Minckwitz, S. Kummel, P. Vogel, C. Hanusch, H. Eidtmann, J. Hilfrich, B. Gerber, J. Huober, S. D. Costa, C. Jackisch, et al. Neoadjuvant Vinorelbine-Capecitabine Versus Docetaxel-Doxorubicin-Cyclophosphamide in Early Nonresponsive Breast Cancer: Phase III Randomized GeparTrio Trial J Natl Cancer Inst, April 16, 2008; 100(8): 542 - 551. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. von Minckwitz, S. Kummel, P. Vogel, C. Hanusch, H. Eidtmann, J. Hilfrich, B. Gerber, J. Huober, S. D. Costa, C. Jackisch, et al. Intensified Neoadjuvant Chemotherapy in Early-Responding Breast Cancer: Phase III Randomized GeparTrio Study J Natl Cancer Inst, April 16, 2008; 100(8): 552 - 562. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Kaufmann, G. von Minckwitz, H. D. Bear, A. Buzdar, P. McGale, H. Bonnefoi, M. Colleoni, C. Denkert, W. Eiermann, R. Jackesz, et al. Recommendations from an international expert panel on the use of neoadjuvant (primary) systemic treatment of operable breast cancer: new perspectives 2006 Ann. Onc., December 1, 2007; 18(12): 1927 - 1934. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |