Multimodal treatment with primary single-agent epirubicin in operable breast cancer: 5-year experience of the Michelangelo Cooperative Group

G. Bonadonna1,+, M. Zambetti1, C. Bumma2, M. Donadio2, A. Bolognesi3, G. Robustelli della Cuna4, G. Ambrosini5, G. Lelli6, M. Mansutti7, P. Verderio1 and P. Valagussa1

1 Istituto Nazionale Tumori, Milano; 2 San Giovanni Antica Sede, Torino; 3 Istituto San Raffaele, Milano; 4 Fondazione Salvatore Maugeri, Pavia; 5 Ospedale S. Chiara, Trento; 6 Casa Sollievo della Sofferenza, San Giovanni Rotondo; 7 Ospedale S. Maria della Misericordia, Udine, Italy

Received 27 February 2001; revised 28 September 2001; accepted 23 October 2001


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

To assess the efficacy of primary single-agent epirubicin (120 mg/m2 every 3 weeks for three cycles) in reducing tumor burden in operable breast cancer >=2.5 cm in largest diameter at diagnosis and its effect on the rate of conservative surgery.

Patients and methods:

A total of 319 eligible patients, who were all candidates for mastectomy, were enrolled on to a multicenter prospective non-randomized study. Tumor response was assessed clinically and pathologically. Relapse-free and overall survival were assessed on major prognostic variables.

Results:

After primary epirubicin, complete disappearance of invasive neoplastic cells accounted for only 2.6% of patients, but 40% of patients had their primary tumor downstaged to <=2 cm in diameter. Conservative surgery was performed in 67% of patients. At a median follow-up of 5 years, both primary tumor size at diagnosis (P <0.01) and pathological nodal involvement (P <0.01) significantly influenced treatment outcome. After conserving surgery and breast irradiation, local recurrences accounted for ~4% of patients.

Conclusions:

This multicenter study confirms that three cycles of full-dose epirubicin can avoid mutilating surgery in a high proportion of patients with tumors not amenable to primary conservative surgery. This multimodal treatment can be safely administered outside of clinical trials in patients presenting with large tumors and with a desire to preserve their body integrity.

Key words: epirubicin, multicenter study, multimodal therapy, operable breast cancer, primary chemotherapy


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Primary chemotherapy for breast cancer was first introduced in 1973 for locally advanced mammary carcinoma and it now represents the treatment of choice for this stage of disease [1]. The aims of this approach were to achieve a higher rate of local control, by facilitating the delivery of locoregional therapy in tumors considered technically inoperable, as well as to prolong survival, by attempting to eradicate micrometastatic dissemination which is present in the vast majority of these patients.

In operable breast cancer, primary chemotherapy was attempted during the late 1980s [2, 3]. Although essentially based on the prior experiences achieved in locally advanced breast cancer, the rationale for the introduction of this treatment approach in early breast cancer rests largely on the clinical and conceptual advances derived from trials of conservative surgery in small cancers and of adjuvant chemotherapy in high-risk tumors. In fact, prospective randomized studies have clearly shown that in suitable tumors (typically <2–4 cm in their largest diameter) breast conserving therapy, i.e. limited surgery and breast irradiation, yields a cure rate similar to that achieved after mastectomy [4] and that local recurrences after conservative approaches are unrelated to the initial tumor size. Other randomized studies have continuously demonstrated that adjuvant systemic treatments have a favorable effect in improving long-term treatment outcome [5].

These considerations were relevant in the design of many of the randomized and non-randomized studies of primary chemotherapy. In fact, many of the initial studies were initiated in patients whose primary tumor diameter prevented a conservative approach [2, 3]. Investigators reasoned that the delivery of a few cycles of primary chemotherapy could achieve a downstaging of tumor size, such as to allow breast sparing surgery and control the risk of local recurrences. Overall, the rate of conservative surgery (lumpectomy or quadrantectomy) varied between 63% and 89%. Variations in this rate were essentially due to differences in patient selection, surgical techniques applied, surgeon training and beliefs, and above all, the criteria for defining the suitability of these approaches after primary chemotherapy. Also in this context, conservative surgery must comply with oncological and cosmetic requirements. Therefore, major contraindications include unfavorable tumor/breast ratios, multifocal disease and microcalcifications exceeding the area of a theoretically conservative resection. Other problems can be related to the presence of macro or microscopically involved surgical margins that may require a mastectomy to be removed.

In September 1993, the Michelangelo Cooperative Group for Multidisciplinary Treatment of Breast Cancer activated an observational non-randomized study of primary chemotherapy. The primary aims of the study were to assess tumor response and rate of conservative surgery in a multicenter environment by using a manageable, effective primary chemotherapy regimen that could be safely used outside of academic institutions. The study design was based on the previous experience of the Istituto Nazionale Tumori of Milan [6] and we report here the 5 year results of this cooperative study.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
From September 1993 to September 1996, 332 patients with primary operable mammary carcinoma, with tumors measuring >=2.5 cm at their largest diameter, were enrolled onto an observational prospective non-randomized study of primary chemotherapy.

All cancers were diagnosed by fine needle aspirates or core biopsy, while surgical biopsy was not allowed. At diagnosis, hormonal receptor status and tumor grading were evaluated according to the methods conventionally adopted at each single participating center.

Eligibility criteria included tumors classified as T2–3 (maximum diameter of 7 cm) N0–2 M0 clinical status according to the TNM system. Patients presenting with tumors measuring <2.5 cm or with skin infiltrating tumors, bilateral or metastatic breast cancer, multifocal disease or presence of microcalcifications exceeding the volume of a theoretical conservative surgery, as well as patients subjected to excisional surgical biopsy were not eligible for this study. History of other malignancies, cardiac diseases (arrhythmias, ischemic disease, previous congestive heart failure, uncontrolled arterial hypertension), pregnancy or active lactation and psychiatric disorders were also reasons for exclusion. The study protocol was approved by the Ethical Research Committee of the Istituto Nazionale Tumori of Milan and all patients had to provide written informed consent prior to enrollment.

Treatment plan
The treatment plan is shown in Figure 1. After registration at the Coordinating Center, the aim was for all patients to receive primary chemotherapy with single-agent epirubicin [120 mg/m2 intravenous (i.v.) bolus every 3 weeks] for three complete cycles. If the white blood count was <2500/mm3 and/or the platelet count was <100 000/mm3 on the commencement of chemotherapy, treatment was delayed for 1 week; drug dose reduction was considered only for prolonged severe hematological toxicity.



View larger version (29K):
[in this window]
[in a new window]
 
Figure 1. Treatment plan.

 
Following primary chemotherapy, patients were subjected to major surgery—the choice between modified radical mastectomy and quadrantectomy was largely dependent on the extent of primary tumor response (a suggested cut-off point for conservative surgery was a residual tumor diameter of <3.0 cm), but the final decision was left to local surgeons who also took into account the relation between tumor size at surgery and the volume of the breast. Skin incision was guided by the tattoo marks that were posed with China ink, marking the position and size of the primary tumor, before the start of primary chemotherapy. In the presence of positive surgical margins after conservative surgery, re-excision was highly recommended, but the protocol allowed local treatment policies. Full axillary nodal dissection was required in all patients.

After surgery, high risk patients (i.e. patients with histologically positive axillary nodes or negative axillary nodes associated with unfavorable prognostic factors defined as negative estrogen receptors or undifferentiated tumors) were subjected to postoperative adjuvant chemotherapy. Based on the previous experience from the Istituto Nazionale Tumori of Milan in women with >3 positive axillary nodes [7], we chose to deliver a regimen non-cross resistant with epirubicin. Six cycles of CMF (cyclophosphamide 600 mg/m2, methotrexate 40 mg/m2, fluorouracil 600 mg/m2 i.v. on days 1 and 8 every 4 weeks) were planned for patients with >=4 positive nodes, while the remaining patients with less extensive nodal involvement or with node-negative tumors and unfavorable prognostic indicators were randomly allocated to either three or six cycles of the same regimen. This program was amended in January 1995 to allow the delivery of postoperative high-dose chemotherapy in patients with >10 positive nodes. At the end of postoperative chemotherapy, tamoxifen (20 mg/day for 5 years) was also planned regardless of estrogen receptor and menopausal status. No systemic treatments were to be given to node-negative patients with favorable prognostic features (i.e. positive receptors and low or intermediate tumor grade).

Radiotherapy was always planned after conservative surgery and was delivered using cobalt or a 6-MeV linear accelerator. The target volume was to the whole breast, with wide margins, and the underlying chest wall and a total of 50 Gy plus a boost of 10 Gy were to be delivered in 6 weeks. No other regional radiotherapy was allowed. In patients subjected to postoperative chemotherapy, breast irradiation was delivered after the last dose of the chemotherapy regimen.

Examinations before and during chemotherapy treatment and follow-up
Tumor dimension was defined by physical examination and mammography in all patients, while echographic evaluation was performed only rarely. Baseline staging required chest X-ray, liver ultrasound, bone scan, blood cell count and biochemistry, and cardiac evaluation with ECG recording.

Complete blood counts with differential and platelets were required on the day of drug administration both for primary and postoperative chemotherapy, and biochemistry was planned every 3 months.

At the completion of primary chemotherapy, in addition to mammography, chest X-rays, cardiac evaluation and ECG had to be repeated.

At the end of postoperative CMF and before starting tamoxifen, all patients had a gynecological examination, which had to be repeated once a year during antiestrogen treatment.

After the completion of postoperative chemotherapy, as well as in all women who did not have adjuvant chemotherapy, physical examination, chest X-ray, blood counts and biochemistry were performed every 6 months for the first 5 years, while mammography (bilateral in women subjected to conservative surgery) was performed once a year. Other studies were repeated annually and included a bone scan, liver ultrasound and ECG. At the completion of the first 5 years of follow-up, all future examinations, including physical examination, were performed annually. When controversial clinical or radiological findings were present, examinations were performed more frequently than initially planned.

Assessment of response to primary chemotherapy and definition of treatment failure after surgery
Clinical examination with measurement of the primary tumor by palpation was performed before each cycle of primary chemotherapy. Complete remission was defined as complete tumor regression, while partial response was defined as a reduction of >50% in the product of the two largest tumor diameters. Minor response was defined as a decrease of between 25 and 50% in the product of the two largest tumor diameters. Any increase in tumor size was considered a disease progression: in this case, primary chemotherapy was discontinued and examinations to rule out the presence of distant metastases had to be performed. In the absence of distant tumor spread, major surgery was carried out soon after. In the presence of distant metastases, treatment was left at the discretion of the treating physician.

At the completion of primary chemotherapy and after clinical and radiological evaluation of tumor remission, extent of residual tumor was also pathologically measured on surgical specimens. Pathological guidelines required a complete assessment of the primary tumor, axillary nodes and surgical margins after conservative surgery. Tumor grading and hormonal receptors were also assessed. To measure the residual tumor extent, in the absence of definite macroscopic evidence of cancer, a minimum of 10 sections had to be thoroughly examined to rule out the presence of invasive neoplastic cells and to qualify for a pathological complete remission. Tumor remnants consisting of the non-invasive component alone were also reported separately. Finally, at least three sections were required for each examined lymph node.

In addition to disease progression while on primary chemotherapy, following surgery, the definition of primary treatment failure was the initial evidence of new disease manifestation(s), i.e. locoregional area(s), distant site(s), contralateral breast cancer or a combination of the above, as documented through clinical, radiological and, when feasible, histological means. Second primary cancers other than contralateral infiltrating breast carcinomas, as well as death due to causes other than disease progression, were not considered treatment failures.

Statistical analysis
This was an observational prospective study carried out over a 3-year period in 20 different centers and no attempts to calculate a sample size were carried out. The primary aims of the study were to assess tumor response and rate of conservative surgery after three cycles of epirubicin. A secondary aim of the study was to assess long-term results after a multimodal therapy. A special emphasis on this secondary aim was given by randomly allocating the duration of postoperative CMF (three versus six cycles) in high-risk patients with fewer than three positive axillary nodes on the 10-year relapse-free survival. Because the median follow-up at the time of this report is limited to 5 years, no data on this specific endpoint will be reported.

All analyses included all eligible patients and were performed according to the intention-to-treat principle.

The strength of association between objective response to primary chemotherapy and clinical tumor diameter (<5 cm versus >5 cm) as well as the existence of a general pattern of association between pathological nodal status (negative versus positive) at surgery were investigated by the Pearson chi-squared statistic.

Freedom from progression was considered as the time elapsed from starting primary chemotherapy to the first evidence of treatment failure for all eligible patients entered into the study, while relapse-free survival was considered as the time elapsed from surgery to the first evidence of recurrence only for those patients submitted to surgery and not failing on primary chemotherapy. All causes of death were taken into account when overall survival was estimated. The survival patterns were estimated by means of the product limit method [8]. For each considered variable the survival distributions were compared with the log-rank test [9]. The prognostic role of conventional variables (clinical tumor size, residual disease at surgery and pathological nodal involvement) was also investigated in univariate analyses using a Cox regression model [10]. In this model each of the regression coefficients is the logarithm of hazard ratio (HR), which is assumed constant over time. Under the null hypothesis (absence of prognostic role), HR is expected to be 1.00. The hypothesis of HR = 1.00 was tested by the Wald statistic.

When the risk of local relapses after conserving surgery was assessed, data were processed according to the approach for competing risks described by Marubini and Valsecchi [11].

Statistical analysis was performed using the SAS System (SAS Institute, Cary, NC) [12].


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patient characteristics
Among 332 registered patients, eight (2.4%) were considered not eligible because of extended surgical biopsy (two), concomitant second cancer (two), bilateral synchronous breast cancer (two), skin infiltrating cancer (one) and tumor diameter of >7 cm (one). Five additional patients (1.5%) withdrew their consent before starting primary chemotherapy. Finally, two patients could not be evaluated for tumor response and rate of conservative surgery. One patient after the first cycle of epirubicin presented abnormal values in the liver tests. The other patient suffered from a transient supraventricular arrhythmia within 48 h of the first administration of epirubicin. Primary chemotherapy was discontinued in both patients who underwent immediate surgery and they are presently alive. Therefore, they were evaluable only for freedom from progression and overall survival.

A total of 317 patients (95.5%) were fully eligible and assessable and their main characteristics are listed in Table 1. The median age was 49 years (range 25–70) and 58% of women were in premenopause. Tumor size measured <4.0 cm in 56% of patients and in 19% it exceeded a diameter of 5.0 cm. According to the TNM staging classification, approximately 80% of patients presented with stage II disease; the remaining women had stage III cancer. Fifty-three percent of patients presented with palpable axillary nodes. Hormonal receptors were evaluated in 53% of the patients; estrogen receptors were negative in 21% of the assays and progesterone receptors were recorded as negative in 24% of the tumors.


View this table:
[in this window]
[in a new window]
 
Table 1.  Main characteristics of eligible and evaluable patients
 
Primary chemotherapy: administered dose and toxicity
The median dose intensity of epirubicin was 0.99 (range 0.72–1.22). No cases of congestive heart failure were reported and no ECG abnormalities were recorded, except for one patient who developed non-specific ST-T trait alterations. As previously mentioned, one additional patient developed transient supraventricular arrhythmia 48 h from the first dose of epirubicin. Mild infections or fever were reported in seven patients (2%), but no septic episodes were observed. On the day of planned epirubicin administration grade 4 neutropenia (National Cancer Institute criteria) was documented in 7% of patients and thrombocytopenia was reported in 1.3%. Alopecia in 94% of patients and grade 3 mucositis in 3% of patients accounted for the main extramedullary toxicity. In one patient (0.3%) the liver test results were abnormal after the first cycle of epirubicin.

Clinical response after primary chemotherapy
Sixty-eight per cent of patients achieved a major tumor response, including 22 women (7%) with complete tumor disappearance on palpation. Tumor response <50% was documented in 16% of patients while 14% of patients had no tumor shrinkage. Four patients developed local disease progression while on primary chemotherapy and one patient developed ipsilateral supraclavicular involvement despite objective response of the primary tumor. Objective tumor response was more often observed in tumors measuring up to 5.0 cm (70%) compared to larger tumors (59%, P = 0.11).

Locoregional treatment
Conservative surgery could be performed in 67% of patients (including seven of eight women who were classified as complete responders) and it is important to mention that it became feasible also in 44% of patients (26 of 59) who presented at diagnosis with primary tumors that measured >5.0 cm in largest diameter (Table 2). The frequency of a conservative approach was higher in complete and partial responders (86% and 80%, respectively) than in minor or no responders (40%). Positive resection margins were reported in 34 specimens (16%), with no difference whether primary tumor at diagnosis measured <5.0 cm (16%) or >5.0 cm in diameter (15%). Surgical re-excision, always in breast-conserving surgery, was performed in 12 of 34 patients (35%), while in the remaining 22 women the treating surgeon selected not to perform a new surgical procedure.


View this table:
[in this window]
[in a new window]
 
Table 2.  Essential results after primary chemotherapy
 
As per protocol guidelines, 93% of patients subjected to conservative surgery received breast irradiation. Nine women received irradiation also on nodal regions not planned in the protocol, while breast irradiation was omitted in the remaining seven women: two patients developed disease relapse (local in one woman) while on CMF; one patient was subjected to modified radical mastectomy for a suspicious local recurrence not confirmed histologically; in three patients the radiation therapist responsible selected not to deliver breast irradiation because of the time that had elapsed since surgery; two women received irradiation outside of the treating center and no information was available.

Residual disease at pathology
Pathological complete remission of breast invasive disease was documented in eight patients (2.6%), including three in whom only the non-invasive component was documented (Table 2). A total of 124 additional women (40%) had residual disease measuring <=2.0 cm after histological examination. This finding correlated well with clinical tumor size after primary epirubicin in patients with a tumor diameter <4.0 cm at diagnosis (Table 2), but residual disease <2.0 cm was more frequently reported after histological examination in larger tumors, highlighting the fact that in these tumors there is a difficulty in differentiating between residual tumor and areas of non-specific thickening on palpation.

Nodal involvement was found in 178 of 312 patients (57%), including one of the eight women who achieved a pathological complete remission. As reported in Table 3, absence of nodal involvement was correlated with extent of residual disease at pathology (P = 0.001).


View this table:
[in this window]
[in a new window]
 
Table 3.  Nodal involvement according to residual disease at pathological examination in patients not failing on primary chemotherapy
 
Estrogen and progesterone receptors were assessed in 78% of specimens and were classified as negative in 47% and 59%, respectively. Undifferentiated tumors were recorded in 33% of specimens. At least one unfavorable prognostic factor was documented in 94 (70%) of 134 node-negative tumors.

Adjuvant treatment
According to the protocol plan, postoperative chemotherapy was to be delivered in 272 patients. Six cycles of CMF were given to 185 women, 82 patients received three CMF cycles and in 12 patients with >10 positive nodes high-dose chemotherapy was administered. Protocol violations were reported in 13 patients; three women with poor prognosis node-negative tumors were not given adjuvant chemotherapy while 10 patients received adjuvant CMF despite favorable prognostic indicators.

The median dose intensity of adjuvant CMF was 0.98 (0.17–1.09) for three cycles and 0.95 (0.34–1.13) for six cycles. At the time of planned cycles, grade 3–4 neutropenia was recorded in 1.3% of women and grade 3–4 stomatitis in 8% of patients. Grade 1 transient alterations of liver function tests were recorded in 21% of patients treated with CMF. No symptomatic episodes of cardiac failure were observed and only one woman developed non-specific alterations of ST-T trait. Six months after the end of adjuvant CMF one patient experienced a reactivation of hepatitis B virus and developed acute hepatic failure requiring liver transplantation.

Tamoxifen was delivered in 267 patients at the end of postoperative chemotherapy.

Results at 5 years
At a median follow-up of 5 years, 56 of the initial 319 patients had died and 87 showed disease progression, five while on primary chemotherapy and 82 after surgery. These results are graphically displayed in Figure 2: 5-year survival from the start of epirubicin treatment was 81% [95% confidence interval (CI ) 76% to 85%] and freedom from progression was 71% (95% CI 66% to 77%).



View larger version (10K):
[in this window]
[in a new window]
 
Figure 2. Five-year survival (––) and freedom from progression (- - ) for the entire series of 319 patients started on primary epirubicin.

 
The 5-year relapse-free and overall survival rates for the 312 patients not failing on primary chemotherapy are reported in Table 4. Overall, primary tumor size at diagnosis and extent of nodal involvement at surgery after primary chemotherapy were able to significantly influence relapse-free and total survival. Of note, patients with low-risk node-negative tumors after primary chemotherapy achieved the best treatment results (relapse-free survival, 90%; total survival, 97%), whereas in patients with node-negative high-risk tumors treatment outcome was not dissimilar (relapse-free survival, 73%; total survival, 76%) to that observed in women with limited nodal involvement (Table 4).


View this table:
[in this window]
[in a new window]
 
Table 4.  Five-year results after primary chemotherapy and surgery plus or minus postoperative systemic therapy (312 patients not failing on primary chemotherapy)
 
Undifferentiated tumors (relapse-free survival 64% versus 76%, P = 0.02; total survival 71% versus 86%, P < 0.01) and negative progesterone receptor status (relapse-free survival 67% versus 85%, P < 0.01; overall survival 75% versus 90%, P < 0.01) were also able to negatively influence 5-year results, although it is worth stressing that these variables were only available for approximately two-thirds of tumors.

Because of the limited number of patients classified as pathological complete responders and of the low number of events in this subset, the influence of residual tumor characteristics at pathology was not investigated by Cox regression model. As reported in Table 5, it is, however, worth mentioning that the 5-year relapse-free survival for the eight pathological complete responders was 88% (95% CI 65% to 100%). Only one patient with pathological complete response and negative axillary nodes has so far developed distant disease recurrence and she is presently alive with disease.


View this table:
[in this window]
[in a new window]
 
Table 5.  Five-year results according to pathological residual disease after primary chemotherapy
 
Among the 213 women who underwent conservative surgery after primary chemotherapy, eight presented local relapse as first event. The cumulative risk of this event was 3.9% (SE ±1.3%). As reported in Table 6, the risk of local relapse was not related to the increase in primary tumor size at diagnosis and to the residual disease at pathology. In contrast, patients in whom specimens of the resected tumors revealed infiltration of surgical margins and who were not subjected to re-excision were at a significantly higher risk of local spread compared to patients with negative or re-excised surgical margins. It is important to state that, as previously reported, the frequency of positive surgical margins was unrelated to the size of the primary tumor at diagnosis. Figure 3 shows that the risk of local relapse started approximately 6 months after quadrantectomy in tumors with positive surgical margins and after approximately 14 months in tumor with negative margins.


View this table:
[in this window]
[in a new window]
 
Table 6.  Cumulative risk of local relapse as first event after the primary epirubicin and breast-sparing approach
 


View larger version (9K):
[in this window]
[in a new window]
 
Figure 3. Cumulative risk of local (in-breast) recurrences following primary epirubicin and breast saving procedure: – – –, tumors with non excised positive surgical margins; ––, tumors with negative surgical margins.

 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The results presented here, achieved in a series of 319 eligible women prospectively enrolled on to an observational non-randomized study conducted in 20 different Italian hospitals, confirm the effectiveness and feasibility of primary chemotherapy. According to the local treatment policy in all participating centers, all of these women were considered candidates for modified radical mastectomy because the primary tumor size at diagnosis measured >=2.5 cm. In fact, in Italy, primary conservative surgery is restricted mainly to smaller tumors. Following single-agent epirubicin, 68% of patients achieved good tumor response and 67% of tumors (61% of those measuring >4.0 cm in largest diameter at diagnosis) were amenable to quadrantectomy plus full axillary dissection and breast irradiation. Overall, these findings are in agreement with similar experiences reported by several research groups and which are summarized in two review papers [2, 3].

This series of prospectively-treated patients followed for a median of 60 months allows us to draw some conclusions about the rate of breast-sparing surgery, the risk of subsequent local recurrences and the role of known prognostic factors, as documented in a multicenter environment. Pretreatment tumor size remains an important predictor of relapse-free and overall survival, but it is also the most important predictor of response to primary chemotherapy [6, 13]. In fact, in the present series the rate of objective remission in primary tumors that measured >5.0 cm in largest diameter at diagnosis was inferior to the rate observed in smaller tumors (59% versus 70%). As a consequence, the rate of conservative surgery was also inversely related to the primary tumor size at diagnosis, being 72% for tumors <5.0 cm and 44% in larger tumors. This latter findings, however, represents twice the rate of lumpectomies (22%) reported in the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-18 protocol in a similar series of patients [13]. The overall rate of breast-sparing procedure in the present multicenter study, however, was lower (67%) than that observed in a similar study in a single institution (83%) [6]. This could be accounted for, at least in part, by the reluctance of some surgeons at local centers to perform breast conserving surgery, after the mammary cancer had reduced following primary chemotherapy, for fear of not excising microscopic residual foci of the disease. The good rate of local control achieved in this patient subset could hopefully contribute to an increase in breast-sparing approaches in future similar experiences.

In our experience, the risk of local relapses observed after conservative surgery was 3.9% (± 1.3%) at 5 years from surgery. The risk of local relapses was higher in large tumors at diagnosis, but not significantly different from that of smaller tumors (Table 6). On the contrary, it was unrelated to the residual tumor size at surgery after primary chemotherapy. It should be noted that none of the seven patients classified as pathological complete responders and subjected to conservative surgery developed local relapse, thus confirming the true absence of invasive cancer. The overall risk of local relapses confirm that, provided conservative surgery is technically adequate, the rate of local recurrence after primary chemotherapy is not measurably different from that of small tumors that are suitable for breast-sparing surgery at the time of diagnosis [14]. In contrast, tumors in which surgically positive margins at resection were not subjected to re-excision were at a significantly higher risk of developing local recurrence (13.9% ± 7.4) compared to tumors with surgically negative margins (2.7% ± 1.2). The multicenter protocol recommended, but did not require, a surgical re-excision of positive margins after a conservative procedure and this recommendation was followed by the vast majority of participating centers. The present findings, however, highlight the need also for surgical re-excision to enable long-term good local control after primary chemotherapy.

As far as follow-up results are concerned, the study design, i.e. prospective non-randomized, prevents any sound conclusion on the true ability of primary chemotherapy plus or minus postoperative adjuvant therapy to control distant metastases. Nevertheless, considering that this case series included only women with primary tumors measuring >=2.5 cm and that after primary chemotherapy only 14% of patients presented with node-negative low-risk tumors, the proposed treatment plan was able to achieve therapeutic results which were at least superimposable with those observed after conventional postoperative chemotherapy in a large series of patients treated at the Istituto Nazionale Tumori [15]. In addition, they confirm that, as has also been reported in other similar studies [1618], the classical pathological variables, i.e. extent of nodal involvement at surgery and primary tumor size at diagnosis, retain their relevance even after primary chemotherapy.

As reported in other studies, and also supported by the present limited series of patients who achieved pathological complete remission of invasive cancer (eight women), residual tumor extent, as confirmed by thorough microscopic examinations, can be considered a marker of treatment outcome [6, 17, 19]. This low rate of pathological complete remissions can plausibly be explained by the thorough pathological examination suggested and conducted on surgical specimens. In addition, in contrast to the NSABP [13] and Royal Marsden [16] studies where patients with T1 lesions were also eligible, our study included patients whose tumors only measured >=2.5 cm at diagnosis and 44% of them presented with cancer >4.0 cm at diagnosis.

In conclusion, our results confirm that, overall, the multimodal approach devised for this study and including three cycles of primary epirubicin is at least as effective as classical postoperative systemic approaches [15] and can also achieve good local and distant tumor control outside the strict environment of an academic institution. This is reassuring in that, when properly performed, this treatment approach can safely offer the advantage of breast-sparing modalities in a large proportion of women with large tumors, also an added advantage over classical adjuvant therapies outside the context of clinical trials.

Although findings from randomized studies indicate that primary chemotherapy results in an overall outcome equivalent to classical postoperative chemotherapy [1618], it is important to recognize that additional and more appropriate studies can now be performed without fear of putting patients at a disadvantage. For this reason, and despite the reassuring results achieved with a simple, safe and effective regimen like the one devised in this multicenter study, we encourage patients and doctors alike to continue participation in newer well designed controlled studies aimed at assessing whether the value of primary chemotherapy remains solely to increase the rate of breast-conserving approaches or, as suggested by experimental and biologic data [2022], it can also improve the cure rate of operable breast cancer.


    Acknowledgements
 
This work was partially supported by a grant from the Associazione Italiana Ricerche sul Cancro, Milano, and by Pharmacia & Upjohn, Milano. The authors wish to thank Mrs Susanna Nave for her cooperation in data management.


View this table:
[in this window]
[in a new window]
 
Appendix A.  Michelangelo Cooperative Study of Primary Chemotherapy in Operative Breast Cancer: participating centers
 

    Footnotes
 
+ Correspondence to: Dr G. Bonadonna, Fondazione Michelangelo Istituto Nazionale Tumori, Via Venezian 1, 20133 Milano, Italy. Tel: +39-02-23902352; Fax: +39-02-23902678; E-mail: bonadonna@istitutotumori.mi.it {dagger} Michelangelo Cooperative Group participating centres listed in Appendix A. Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Hortobagyi GN, Singletary SE, McNeese MD. Treatment of locally advanced and inflammatory breast cancer. In Harris JR, Lippman ME, Morrow M, Hellmann S (eds): Disease of the Breast. Philadelphia, PA: Lippincott-Raven 1996; 585–599.

2. Bonadonna G, Valagussa P. Primary chemotherapy in operable breast cancer. Semin Oncol 1996; 23: 464–474.[ISI][Medline]

3. Hortobagyi GN, Buzdar AU. Preoperative (neoadjuvant) chemotherapy for operable breast cancer. In Bonadonna G, Hortobagyi GN, Gianni AM (eds): Textbook of Breast Cancer: A Clinical Guide to Therapy. London: Martin Dunitz 2000; 119–137.

4. Effects of radiotherapy and surgery in early breast cancer. An overview of the randomized trials. Early Breast Cancer Trialists’ Collaborative Group. N Engl J Med 1995; 333: 1444–1455.[Abstract/Free Full Text]

5. Polychemotherapy for early breast cancer: an overview of the randomised trials. Early Breast Cancer Trialists’ Collaborative Group. Lancet 1998; 352: 930–942.[ISI][Medline]

6. 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]

7. Bonadonna G, Zambetti M, Valagussa P. Sequential or alternating doxorubicin and CMF regimens in breast cancer with more than three positive nodes. Ten-year results. JAMA 1995; 273: 542–547.[Abstract]

8. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

9. Peto R, Peto J. Asymptotically efficient rank invariant test procedures. J R Stat Soc [A] 1972; 135: 185–207.[ISI]

10. Cox DR. Regression models and life tables. J R Stat Soc [B] 1972; 34: 187–220.[ISI]

11. Marubini E, Valsecchi MG. Analysing survival data from clinical trials and observational studies. Chichester: Wiley 1995.

12. SAS Institute. SAS/STAT User’s Guide, version 6, 4th edition. Cary, NC: SAS Institute 1989.

13. Fisher B, Brown A, Mamounas E et al. Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 1997; 15: 2483–2493.[Abstract]

14. Veronesi U, Saccozzi R, Del Vecchio M et al. Comparing radical mastectomy with quadrantectomy, axillary dissection, and radiotherapy in patients with small cancers of the breast. N Engl J Med 1981; 305: 6–11.[Abstract]

15. Bonadonna G. Current and future trends in the multidisciplinary approach for high-risk breast cancer. The experience of the Milan Cancer Institute. Eur J Cancer 1996; 32A: 209–214.

16. Makris A, Powles TJ, Ashley SE et al. A reduction in the requirements for mastectomy in a randomized trial of neoadjuvant chemoendocrine therapy in primary breast cancer. Ann Oncol 1998; 9: 1179–1184.[Abstract]

17. Fisher B, Bryant J, Wolmark N et al. Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 1998; 16: 2672–2685.[Abstract]

18. Mauriac L, Macgrogan G, Avril A et al. Neoadjuvant chemotherapy for operable breast carcinoma larger than 3 cm: a unicentre randomized trial with a 124-month median follow-up. Institut Bergonie Bordeaux Groupe Sein (IBBGS). Ann Oncol 1999; 10: 47–52.[Abstract]

19. Kuerer HM, Newman LA, Smith TL et al. Clinical course of breast cancer patients with complete pathologic primary tumor and axillary lymph node response to doxorubicin-based neoadjuvant chemotherapy. J Clin Oncol 1999; 17: 460–469.[Abstract/Free Full Text]

20. Simpson-Herren L, Sanford AH, Holmquist JP. Effects of surgery on the cell kinetics of residual tumor. Cancer Treat Rep 1976; 60: 1749–1760.[ISI][Medline]

21. Gunduz N, Fisher B, Saffer EA. Effect of surgical removal on the growth and kinetics of residual tumor. Cancer Res 1979; 39: 3861–3865.[Abstract]

22. Fisher B, Gunduz N, Saffer EA. Influence of the interval between primary tumor removal and chemotherapy on kinetics and growth of metastases. Cancer Res 1983; 43: 1488–1492.[Abstract]