Randomized trial of high-dose chemotherapy and hematopoietic progenitor-cell support in operable breast cancer with extensive lymph node involvement: final analysis with 7 years of follow-up

J. G. Schrama1,+, I. F. Faneyte2, J. H. Schornagel1, J. W. Baars1, J. L. Peterse2, M. J. van de Vijver2, O. Dalesio3, H. van Tinteren3, E. J. T. Rutgers4, D. J. Richel1,§ and S. Rodenhuis1

1Division of Medical Oncology, 2Pathology Department, 3Biometrics Department and 4Surgical Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands

Received 31 October 2001; revised 18 February 2002; accepted 8 March 2002.


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background

The aim of this study was to present an update of overall (OS) and disease-free survival (DFS) and to evaluate the correlation between outcome and pathological findings at surgery in a randomized trial of high-dose chemotherapy following neoadjuvant chemotherapy and surgery in high-risk breast cancer patients.

Patients and methods

Ninety-seven women <60 years of age with breast cancer and extensive axillary lymph node involvement received three courses of FE120C (5-fluorouracil 500 mg/m2, epirubicin 120 mg/m2, cyclophosphamide 500 mg/m2) followed by surgery. Eighty-one patients were randomized to receive either a fourth FE120C course alone or a fourth FE120C course followed by high-dose chemotherapy (cyclophosphamide 6 g/m2, thiotepa 480 mg/m2, carboplatin 1600 mg/m2). We performed a univariate analysis on possible prognostic factors and analyzed the sites of relapse.

Results

After a median follow-up of 6.9 years, 47 (48%) patients were alive, of whom 36 (38%) were without disease. Sixty patients relapsed after treatment. One patient died of myelodysplastic syndrome, without a relapse. In intention-to-treat analysis, the 5-year DFS rates were 47.5% in the conventional treatment arm and 49% in the high-dose arm, and the 5-year OS rates were 62.5% and 61%, respectively. In the univariate analysis, the clinical T-stage before chemotherapy and the number of tumor-positive axillary lymph nodes after induction chemotherapy (P = 0.027) were significant prognostic factors for OS. The same factors (both P = 0.06) plus the estrogen receptor (P = 0.08) were borderline significant factors for DFS.

Conclusions

After a median follow-up of 6.9 years there was no difference in OS or DFS rates between the two treatment groups. The number of tumor-positive axillary lymph nodes after induction chemotherapy and the clinical T-stage before chemotherapy were significant factors for OS.

Key words: breast cancer, hematopoietic progenitor-cell support, high-dose chemotherapy, randomized trial


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The value of high-dose chemotherapy as adjuvant treatment in high-risk primary breast cancer is still controversial. Uncontrolled data from several phase I and II studies indicated a survival improvement for high-dose chemotherapy with stem-cell support in the adjuvant treatment of breast cancer. As a result, high-dose therapy was initially considered to be a promising new strategy in breast cancer therapy. At the time of this writing, the results of six randomized adjuvant trials have been published [16], including a total of >2000 patients. One of these studies has proven to be unreliable [6].

Of the remaining five randomized studies, only the Dutch National study (preliminary analysis of the first 284 patients) [1] showed a modest survival benefit from high-dose chemotherapy. The American Intergroup study [2] was updated at the American Society of Clinical Oncology (ASCO) meeting in 2001 and did not show a survival advantage of high-dose therapy compared with intermediate-dose therapy. They found fewer relapses in the high-dose therapy group, but this was compensated by a high early toxicity mortality rate. The other three randomized studies have been published in peer-reviewed journals and showed no advantage of high-dose treatment [35]. One of these randomized trials was performed in our institute, and the results with a median follow-up of 4 years were published in 1998 [5]. In this report, we present the updated results of this trial with a median follow-up of >6.9 years and a lead follow-up of almost 10 years. The more mature data allow not only an update of survival, but also an evaluation of risk of second malignancies and possible long-term side effects of high-dose chemotherapy.

In this study, surgery was performed following three courses of chemotherapy, which gave the opportunity to analyze prognostic significance of the response to neoadjuvant chemotherapy. As well as response, we also analyzed other possible prognostic factors [estrogen (ER), progesterone receptor (PR) [7], tumor size, the axillary nodal status [8], HER-2/neu protein expression [9], proliferation markers] in order to identify patients that will benefit most from high-dose chemotherapy.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients
The patient selection for this study has been described previously [5]. Briefly, 97 women <60 years of age with breast cancer (operable according to classic Haagensen criteria [10] and with extensive axillary node metastases, confirmed by a tumor-positive infraclavicular lymph node biopsy [11, 12]) were included in the study. Women who underwent a tumorectomy for histological diagnosis were eligible if no axillary dissection had been carried out. Physical examination, chest radiography, liver ultrasound examination and a radionuclide bone scan had to be negative for distant metastases. A performance status of 0 or 1 (according to WHO criteria) and normal bone marrow, renal and hepatic functions were required for eligibility.

Written informed consent was obtained from all patients before enrolment in the study, according to institutional guidelines. Patients again gave their consent before up-front chemotherapy, before surgery and before randomization. The protocol was approved by The Netherlands Cancer Institute’s protocol review committee and the Committee for Medical Ethics of The Netherlands Cancer Institute.

Study design
The study was designed as a single-institution phase II study. Earlier results concerning efficacy and feasibility have been published previously [5, 13]. The treatment sequence started with three up-front courses of FE120C (5-fluorouracil 500 mg/m2, epirubicin 120 mg/m2, cyclophosphamide 500 mg/m2) (Figure 1). After the third course, all patients underwent surgery. The type of surgery (mastectomy or breast-conserving therapy and axillary lymph node dissection) had been decided before the start of chemotherapy. After surgery, patients with a response to chemotherapy or stable disease were randomized to either conventional chemotherapy (a fourth course of FE120C, followed by radiotherapy and tamoxifen) or high-dose chemotherapy [the same treatment but with additional high-dose chemotherapy and peripheral-blood progenitor cell (PBPC) support]. In the high-dose arm, the fourth FE120C course was used for mobilization of stem cells. For this purpose granulocyte colony stimulating factor (filgrastim 300 µg, daily subcutaneous dose) was started 24 h after the chemotherapy. PBPCs were harvested by leukocytaphereses, starting when the white blood cells exceeded 3 x 109/l and continued until at least 3 x 106 CD34-positive cells per kilogram of body weight had been collected. High-dose chemotherapy started preferentially 21 days after the fourth FE120C course. The high-dose regimen consisted of cyclophosphamide 6 g/m2, thiotepa 480 mg/m2 and carboplatin 1600 mg/m2, divided over 4 days [14]. The PBPC reinfusion was performed 2 days after the end of the chemotherapy. The prophylactic use of anti-emetics and antibiotics has been described elsewhere [5]. Anti-emetics were used prophylactically and then as needed. All patients received prophylactic antibiotics, including oral ciprofloxacin (500 mg bd) and oral amphotericin B (500 mg qds), starting 4 days before high-dose treatment. Penicillin (1 x 106 U qds) was administered 2 days before reinfusion and amphotericin B (0.25 mg/kg daily) on the day of reinfusion. All antibiotics were discontinued when the neutrophil cell count exceeded 0.5 x 109/l. Patients who were positive for antibodies to herpes simplex virus were given oral aciclovir 400 mg twice daily. Irradiated platelet transfusions were administered to maintain platelet count above 10 x 109/l and irradiated leukocyte-free red blood cells were given if hemoglobin was <5.5 mmol/l. The chemotherapy was followed by radiation therapy (to the chest wall, parasternal, axilla and supraclavicular and infraclavicular fields) and by tamoxifen (40 mg/day for 2 years).



View larger version (19K):
[in this window]
[in a new window]
 
Figure 1. Study design.

 
Assessment of response
Clinical complete response (cCR) was defined as disappearance of all palpable tumors in the breast and in the axilla, as assessed by a surgeon and a medical oncologist. A clinical partial remission (cPR) was defined as a decrease of at least 50% of the sum of all perpendicular diameters of palpable lesions in breast and axilla. Progressive disease (PD) was defined as an increase of 25% of all palpable tumor or the appearance of new tumor lesions. Stable disease (SD) was a decrease of <50% or an increase <25%. All patients with a cCR, cPR or SD were offered randomization. To begin with, randomization was only offered to patients with SD if they showed minimal clinical response, a subjective response or pathological evidence of response. Because of lack of reproducibility of the evaluations, randomization was later offered to all patients with SD.

Pathological examination
All breast specimens were handled according to institutional protocols. Tumor size was measured in fresh specimen, and the largest diameter was used to assess pathological tumor stage (pT). pT was checked microscopically by complete cross sections of the gross tumor. At least 10 lymph nodes were harvested from the axillary specimens; all lymph nodes were completely embedded and evaluated by hematoxylin–eosin (HE) and immunohistochemical staining using cam 5.2. Immunohistochemistry was performed in case of residual carcinoma to assess ERs and PRs, HER2/neu, p53 overexpression, Ki-67 and BCL-2. Pathological response was evaluated and scored as complete in case of absence of invasive tumor in the breast and lymph nodes. Histological classification was carried out according to the WHO criteria. Grading was carried out according to criteria described by Elston and Ellis [15]. The results of the proliferation markers (Ki-67 and BCL-2) and the pathological response will be published elsewhere.

Immunohistochemistry
Formalin-fixed, paraffin-embedded tissue samples were stained with antibodies against HER-2/neu (3B5; 1:10 000), p53 (D07; 1:8000; mouse IgG2b; DAKO, Copenhagen, Denmark), ER (1D5; 1:150; mouse IgG1; DAKO), Ki-67 (MIB-1; 1:1000; DAKO) and BCL-2 (clone 124; 1:100 mouse IgG1; DAKO). The methods of the immunohistochemical staining will be published elsewhere. Briefly, 5-µm sections were deparaffinized in xylol and alcohol, and endogenous peroxidase was quenched in methanol–peroxide (3%; 20'). Slides were pre-treated (antigen retrieval) in citrate buffer, blocked with normal goat serum and then incubated with the primary antibody overnight at 4°C. Binding of the monoclonal antibody was detected with biotin-labeled goat anti-mouse IgG (1:200; 30' DAKO) and horseradish peroxidase avidin–biotin complex (1:100; 30' DAKO). Bound peroxidase was developed with 3,3'-diaminobenzidine tetrachloride (Sigma, St. Louis, MO, USA) and 0.02% H2O2 in phosphate-buffered saline (PBS). All reagents were diluted in a 1% bovine albumin solution in PBS. Anti-mouse IgG solution was admixed with 10% normal human serum to prevent non-specific binding. Replacement of the primary antibody with 1% bovine albumin solution in PBS served as negative control.

Immunohistochemical results were mostly scored semi-quantitatively on a six-point scale for percentage of positively staining tumor cells (0; 1, <10%; 2, 10% to 25%; 3, 25% to 50%; 4, 50% to 75%; 5, 75% to 100%). HER-2/neu was scored according to the system that has recently come into use for clinical testing (0; 1+ >10% cells weakly positive; 2+ moderate homogenous staining; 3+ strong homogenous staining).

Statistical analysis
This report is an update of our study published in 1998 [5]. Treatment assignment was done by the minimization technique with stratification for cCR (yes or no) and postmenopausal status (yes or no). The primary endpoints were overall survival (OS) and relapse-free survival.

The treatment comparisons were done by intention-to-treat analysis. Survival curves were constructed by the Kaplan–Meier method and compared by the log-rank test. We used Cox’s model to analyze the factors predictive of progression or death and to adjust the comparison of treatment for the patients’ baseline characteristics.

In the univariate analysis, possible prognostic factors for relapse and overall disease-free survival (DFS) from the start of treatment were examined. For each variable, Kaplan–Meier curves and the graph of the log (–log survival function) versus log (time) were plotted and checked for proportionality of hazards. For these plots, each variable was screened univariately by means of Cox’s proportional hazards model. Discrete variables with more than two categories were analyzed by means of the categories or indicator variables. A forward stepwise procedure was planned with those variables found to be significant in the screening. No variable passed this step.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Between April 1991 and December 1995, 97 patients were enrolled in the study (Figure 1). Eight patients were not evaluable for clinical response because the primary tumor had been completely removed for diagnostic reasons. These patients remain in the analysis except for the analysis of tumor response. In six other patients the primary breast tumor had been excised, but they had palpable axillary lymph nodes to evaluate response. One patient went off-study after the first FE120C course, due to patient refusal. Eighty patients underwent a modified radical mastectomy and 16 patients had breast-conserving surgery. After surgery, 81 patients were randomized. Fifteen patients were not randomized, 11 because of patient’s refusal to undergo high-dose therapy if randomized to that arm, and four other patients because of unresponsiveness to FE120C. Thirty-five of the 41 patients scheduled for high-dose chemotherapy actually received the treatment. Five patients refused high-dose chemotherapy after initial consent. Another patient did not receive high-dose chemotherapy because mobilization of PBPCs did not succeed in this patient. All these patients were treated with a fourth FE120C only.

Feasibility, toxicity and data on PBPC mobilization have been described in earlier reports [5, 16].

Overall and progression-free survival
At the time of this analysis, 47 (48%) of the 97 patients were alive, of whom 11 (11%) had disease and 36 (37%) were without disease (Table 1). The 5-year DFS and OS were 45% and 59%, respectively.


View this table:
[in this window]
[in a new window]
 
Table 1. Survival status
 
There was no significant difference in survival rates between the two treatment groups after a median follow-up of 6.7 years from randomization (with a lead follow-up of 9.6 years). The 5-year DFS were 47.5% and 49% in the conventional and high-dose arms, respectively. The 5-year OS rates were 62.5% and 61% in the conventional arm and the high-dose arm, respectively (Figures 2 and 3). The two groups were comparable with respect to clinical and pathological factors before and after up-front chemotherapy (Tables 2 and 3).



View larger version (19K):
[in this window]
[in a new window]
 
Figure 2. Disease-free survival by treatment (n = 81). Median follow-up 6.7 years after randomization. CONV, conventional treatment; HD, high-dose treatment.

 


View larger version (18K):
[in this window]
[in a new window]
 
Figure 3. Overall survival by treatment (n = 81). Median follow-up 6.7 years after treatment. CONV, conventional treatment; HD, high-dose treatment.

 

View this table:
[in this window]
[in a new window]
 
Table 2. Patient characteristics
 

View this table:
[in this window]
[in a new window]
 
Table 3. Histological and immunohistochemical tumor characteristics
 
Sites of relapse
In theory, high-dose chemotherapy could fail to eradicate microscopic disease in pharmacological sanctuaries, such as the central nervous system. We evaluated the sites of relapse in the two treatment groups. Sixty (62%) of the 97 patients relapsed: 22 patients in the high-dose group, 27 in the conventional dose group and 11 of the 16 non-randomized patients. One patient in the high-dose group died of myelodysplasic syndrome (MDS) without a relapse. The sites of the relapses are shown in Table 4. Locoregional relapse was seen in 16 patients as the primary site of relapse, nine patients in the high-dose treatment arm and seven in the FE120C arm. Some patients had more than one site of relapse. The number of patients with a relapse was not significantly different between the two treatment groups. Liver and bone metastases occurred more frequently in the FE120C group, but the difference was not significant.


View this table:
[in this window]
[in a new window]
 
Table 4. Sites of relapse
 
Prognostic factors
Many (possible) prognostic factors have been described in patients receiving adjuvant therapy for breast cancer [79, 1721]. As surgery was carried out after three courses of chemotherapy, it was possible to evaluate both the clinical and the pathological response as predictive factors. In our previous report only clinical T-stage approached significance [5]. In the present analysis, we additionally evaluated the number of tumor-positive lymph nodes in the axilla after up-front chemotherapy as a separate prognostic factor. In the univariate analysis (Table 5), clinical T-stage before chemotherapy (P = 0.016) and number of lymph nodes with residual tumor after chemotherapy (P = 0.027) were significant prognostic factors for OS. No factor reached significance for DFS, but clinical T-stage (P = 0.06), number of axillary lymph nodes (P = 0.06) and a positive ER status (P = 0.08) were nearly significant for DFS. Pathological investigation of the number of tumor-positive lymph nodes has been performed after neoadjuvant chemotherapy; therefore, the number of tumor-positive lymph nodes is related to pathological response. None of the other possible prognostic factors (age, surgery, hormone-receptors, HER-2/neu, p53 and pathological response) was significantly related to outcome. Because only two established prognostic factors were significant, further multivariate analysis was not performed.


View this table:
[in this window]
[in a new window]
 
Table 5. Univariate analysis: summery of potentially prognostic factors
 
Subgroup of patients with a DFS >5 years
The group of patients with a DFS >5 years is of special interest, because it might be that this group has one or more specific characteristics that predispose long-term survival. We performed a separate analysis of the patient group with a DFS >=5 years (Table 6).


View this table:
[in this window]
[in a new window]
 
Table 6. Characteristics of patients with a DFS <5 or >=5 years
 
Forty-four (45%) of the 97 patients had a DFS >=5 years. Not surprisingly, the group of patients with a DFS <5 years contained more non-randomized patients, because patients who were unresponsive to FE120C did not proceed to randomization. There was no difference in treatment between the patients with a DFS more or less than 5 years (high-dose and conventional therapy ±50% in both patient groups). Comparison between characteristics of the patients with a DFS >5 years and the remainder of the group showed no clear differences. There was also no difference in relapse sites between these two groups: most primary relapses occurred in bone. ER positivity was not significantly different between the two groups (P = 0.26). Thus, patients who remained free of disease for 5 years or more after the start of treatment had no distinguishing characteristics.

Long-term side effects and second tumors
The median follow-up of 6.9 years after registration with a lead follow-up of 9.8 years is still too short for a complete evaluation of risk of second tumors (particularly solid tumors) and long-term side effects (Table 7). Radiation pneumonitis is a relatively early side effect. Five patients (four in the high-dose group and one in the conventional therapy arm) developed a radiation pneumonitis between 3 and 12 weeks after radiation therapy; all were reversible after treatment with prednisone. Cardiac problems were reported with expected frequency: one patient had a myocardial infarction. Van Dam et al. [22] have evaluated cognitive toxicity in these patients, and found that cognitive function impairment was more frequent after high-dose treatment for breast cancer.


View this table:
[in this window]
[in a new window]
 
Table 7. Long-term effects and second tumors/malignanciesa
 
At the time of this analysis, three solid (second) neoplasms have been diagnosed: one basal cell carcinoma and one tubular adenoma of the colon (both in the high-dose therapy arm), and one tubulovillous adenoma of the sigmoid in the conventional therapy arm. Two other patients in the high-dose treatment group developed a MDS. One of these patients already had (retrospectively) myelodysplastic features in the bone marrow before transplantation. The other patient presented with MDS 8 years after high-dose chemotherapy. One patient developed antibodies against platelets and neutrophil granulocytes. She responded favorably to prednisone.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The primary aim of this study was to develop a practical approach that could be used in a multi-center study of high-dose chemotherapy in operable breast cancer with extensive lymph node involvement. All patients had a tumor-positive infraclavicular lymph node biopsy [11, 12], which is usually associated with extensive lymph node involvement and has a poor prognosis [23, 24]. We have previously published the results concerning efficacy and feasibility of this study [5, 13].

After a median follow-up of 6.9 years and a lead follow-up of almost 10 years in 97 patients with breast cancer and extensive lymphadenopathy, there was no statistically significant relapse-free survival advantage for high-dose chemotherapy over conventional therapy in this study. Since the study was small and did not have the power to detect a difference of >30% in (relapse-free) survival between the treatment groups, a smaller but clinically important advantage for high-dose therapy may have been missed. In fact, 27 of 40 patients randomized to the conventional treatment died or relapsed, while there were only 23 (22 deaths from or relapse of disease, and one death from myelodysplasia) of 41 patients in the high-dose group who experienced such an ‘event’. At the time of writing, 32% of the conventionally treated patients remain alive and free of disease compared with 44% of the patients who received high-dose therapy as a consolidation. This corresponds to a reduction in hazard rate of almost 20%. Although statistically not significant, this is consistent with preliminary data of the large Dutch National study, which were reported at the ASCO meeting in 2000 [1]. This study showed a reduction of odds of relapse of 20% in 885 patients at a median follow-up of 35 months after randomization. Thus, the present study cannot be regarded as ‘negative’, because it is too small to reject an equivalence hypothesis.

A second possible reason for the absence of an effect of high-dose therapy could be the fact that chemotherapy was interrupted for an appreciable amount of time to allow surgery. Substantial regrowth of tumor cells might take place between the third and the fourth FE120C courses, thereby losing some of the beneficial effect of the induction chemotherapy. It is of interest that a similar interval between conventional dose induction chemotherapy and high-dose therapy was present in the Philadelphia study [25], which showed no advantage for high-dose therapy. Mathematical models of tumor growth suggest that this could eliminate most of the benefit of the induction therapy [26, 27].

The high-dose chemotherapy regimen used in this study and in the Dutch National study [1] is related but not identical to the STAMP-V or CTCb regimen, which was employed in the Philadelphia study [25] and in the Scandinavian randomized adjuvant trial [3]. The main differences are that the carboplatin dose in the Dutch regimen is twice as high (1600 mg/m2), and the three alkylating agents are administered as short-term infusions rather than as 96-h continuous infusions. The latter difference could be important, since we have shown recently that even low concentrations of thiotepa can inhibit the activation route of cyclophosphamide [28]. If thiotepa is continuously present in the circulation, a sustained reduction of 4-hydroxycyclophosphamide formation may be present, accounting for less active metabolites of cyclophosphamide, less toxicity and possibly a lower efficacy.

Although our study failed to show an advantage for high-dose therapy, the OS is appreciably better than that of a similar series of patients reported from our institute [29]. This could result from factors such as patient selection [30, 31] and stage migration. Moreover, these patients were treated with the CMF (cyclophosphamide, methotrexate and 5-fluoruracil) regimen instead of FE120C. It is likely, however, that some real improvements in adjuvant therapy, such as the prolonged administration of tamoxifen after chemotherapy and radiation therapy of the chest wall, have contributed to the better OS.

An important disadvantage of high-dose therapy or other forms of intensive chemotherapy in the adjuvant treatment of breast cancer is long-term side effects, including the risk of therapy-induced second malignancies. After a lead follow-up of 10 years we identified five neoplasms: two MDS and three solid tumors. Both cases of MDS developed in the high-dose therapy arm, but one of these patients retrospectively had signs of a myelodysplastic syndrome in her bone marrow before high-dose therapy. This incidence is comparable to the results of Sobecks et al. [32], who reported an incidence of 0.3%, and Laughlin et al. [33], who reported a 4-years probability of developing MDS/acute leukemia of 1.6% after high-dose therapy. In the study by Bergh et al. [3], no cases of MDS/acute leukemia were seen in the high-dose therapy arm, but in the 251 patients treated with ‘tailored’ FEC (dose adjusted to tolerance), six cases of acute myelogenous leukaemia and three cases of MDS were diagnosed. This high incidence was probably caused by the high cumulative dose of epirubicin in the conventional arm.

Since MDS and leukemia usually develop within 6 years after treatment [33, 34], we believe that the incidence will not appreciably rise with further follow-up in the present study.

Only three solid tumors have been diagnosed in our patient group: one basal cell carcinoma (high-dose treatment) and two colorectal tubular/tubulovillous adenomas (one in the high-dose arm and one in the conventional treatment arm). These malignancies are also common in the normal population. We found only one contralateral second breast tumor in the conventional dose arm. This is remarkable, because the incidence of second contralateral breast tumors is 7–8 per 1000 women-years of follow-up [3537]. In this group of 97 patients with 670 women-years of follow-up, one would expect ~4–5 cases of contralateral breast cancers. So, it could be that the adjuvant hormonal and chemotherapeutic treatment has a protective effect for contralateral breast cancer. At this time it is difficult to estimate the number of second malignancies after high-dose therapy in breast cancer, since there is insufficient follow-up in all of the available studies.

A potential complication of high-dose therapy in breast cancer is radiation pneumonitis. We observed this complication in 5% of our patients: four patients in the high-dose group and one in the conventional arm, all responding well to prednisone without developing clinically significant long-term pulmonary toxicity. This is an important difference with the cisplatin–BCNU–cyclophosphamide regimen used in the American Intergroup study [2]. In this trial, pulmonary toxicity was an important cause of therapy-related death.

We recently reported that neuropsychological disorders may be present after high-dose chemotherapy, although their clinical importance is as yet unclear [22]. At this point, it is not known whether or not abnormalities at neuropsychological testing are also present after one of the other high-dose regimens used in breast cancer. This would be interesting to determine, while theoretically it is possible that the higher dose carboplatin in our regimen has influence on neuropsychological functioning.

If high-dose chemotherapy proves to benefit patients with breast cancer, it is quite likely that this benefit will be limited to a (possibly small) subgroup of patients. It would be important to identify or confirm prognostic factors that characterize this patient group. In the present study, pathological specimens were taken both before and after induction chemotherapy, allowing the study of a range of potential prognostic factors in relation to treatment outcome. Pathological characteristics of primary tumors have been reported to have predictive value [1821] and similar results have been obtained in small series of patients receiving intensive chemotherapy [79]. In the present study, only clinical T-stage and the number of tumor-positive axillary lymph nodes were prognostic factors for OS. Clinical T-stage, the number of tumor-positive axillary lymph nodes and positivity for the ER were of borderline prognostic value for DFS. Unfortunately, it was not possible to identify a patient group that specifically benefited from high-dose therapy. Clearly, the number of patients in our study may have been too small to investigate this in a meaningful way, and similar analyses will be crucial when the large American and European studies of high-dose therapy and the adjuvant chemotherapy of breast cancer have reached sufficient follow-up time.

Although this study did not show a benefit of high-dose therapy in patients with high-risk breast cancer, the Dutch National study [1] has reported preliminary results that could be consistent with a modest benefit for high-dose therapy. An update of the American Intergroup study [2] presented at ASCO 2001 showed fewer relapses in the high-dose therapy group compared with the intermediate-dose group. This did not result in a survival advantage, probably because of the high early toxicity rate. The maturation of these and other ongoing large studies should be awaited before the validity of the high-dose concept in breast cancer is abandoned or adopted.


    Acknowledgements
 
This study was supported in part by a grant from the Schumacher-Kramer Foundation, Amsterdam, The Netherlands.


    Footnotes
 
+ Correspondence to: Dr J. Schrama, Division of Medical Oncology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands. Tel: +31-20-5122870; Fax: +31-20-5122572; E-mail: j_schrama@nki.nl Back

§ Present address: Division of Medical Oncology, Academical Medical Centre, Amsterdam, The Netherlands. Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Rodenhuis S, Bontenbal M, Beex LVAM et al. Randomized phase III study of high-dose chemotherapy with cyclophosphamide thiotepa and carboplatin in operable breast cancer with 4 or more axillary lymph nodes. Proc Am Soc Clin Oncol 2000; 19: 1a (Abstr 286).

2. Peters WP, Posner G, Vredenburgh J et al. Updated results of a prospective, randomized comparison of two doses of combination alkylating agents (AA) as consolidation after CAF in high-risk primary breast cancer involving ten or more axillary lymph nodes (LN): CALBG 9082. SWOG9114. NCIC MA-13. Proc Am Soc Clin Oncol 2001; 20: 21a (Abstr 81).

3. Bergh J, Wiklund T, Erikstein B et al. Tailored fluorouracil, epirubicin, and cyclophosphamide compared with marrow-supported high-dose chemotherapy as adjuvant treatment for high-risk breast cancer: a randomized trial. Scandinavian Breast Cancer Group 9401 study. Lancet 2000; 356: 1384–1391.[ISI][Medline]

4. Hortobagyi GN, Buzdar AU, Theriault RL et al. Randomized trial of high-dose chemotherapy and blood cell autografts for high-risk primary breast carcinoma. J Natl Cancer Inst 2000; 92: 225–233.[Abstract/Free Full Text]

5. Rodenhuis S, Richel D, Van der Wall E et al. Randomised trial of high-dose chemotherapy and haemopoietic progenitor-cell support in operable breast cancer with extensive axillary lymph node involvement. Lancet 1998; 352: 515–521.[ISI][Medline]

6. Bezwoda WR. Randomised, controlled trial of high dose chemotherapy versus standard dose chemotherapy for high-risk, surgically treated, primary breast cancer. Proc Am Soc Clin Oncol 1999; 18: 2a (Abstr 4).

7. Somlo G, Doroshow JH, Forman SJ et al. High-dose chemotherapy and stem-cell rescue in the treatment of high-risk breast cancer: prognostic indicators of progression-free and overall survival. J Clin Oncol 1997; 15: 2882–2892.[Abstract]

8. Nieto Y, Cagnoni PJ, Shpall EJ et al. A predictive model for relapse after high-dose chemotherapy with peripheral progenitor cell support for high-risk primary breast cancer. Clin Cancer Res 1999; 5: 3425–3431.[Abstract/Free Full Text]

9. Nieto Y, Cagnoni PJ, Nawaz S et al. Evaluation of the predictive value of HER-2/neu overexpression and p53 mutations in high-risk primary breast cancer patients treated with high-dose chemotherapy. J Clin Oncol 2000; 18: 2070–2080.[Abstract/Free Full Text]

10. Haagensen CD. Diseases of the Breast, 2nd edition. Philadelphia, PA: Saunders 1976.

11. Van Dongen JA. Subclavicular biopsy as a guideline for the treatment of breast carcinoma. World J Surg 1977; 1: 306–308.[Medline]

12. Van Tienhoven G, Borger JH, Passchier DH et al. The prognostic significance of the axillary apex biopsy in clinically operable breast cancer. Eur J Cancer 1995; 31A: 1965–1968.

13. Van der Wall E, Schaake-Koning CCE, Van Zandwijk N et al. The toxicity of radiotherapy following high-dose chemotherapy with peripheral stem cell support in high risk breast cancer: a preliminary analysis. Eur J Cancer 1996; 32A: 1490–1497.

14. Rodenhuis S, Baars J, Schornagel JH et al. Feasibility and toxicity study of a high-dose regimen incorporating carboplatin, cyclophosphamide and thiotepa. Ann Oncol 1992; 3: 855–860.[Abstract]

15. 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 1997; 19: 403–410.

16. Van der Wall E, Nooijen WJ, Baars JW et al. High-dose carboplatin, thiotepa and cyclophosphamide (CTC) with peripheral blood stem cell support in the adjuvant therapy of high-risk breast cancer. Br J Cancer 1995; 71: 857–862.[ISI][Medline]

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

18. Kuerer HM, Newman LA, Buzdar AU et al. Residual metastatic axillary lymph nodes following neoadjuvant chemotherapy predicts disease free survival in patients with locally advanced breast cancer. Am J Surg 1998; 176: 502–508.[ISI][Medline]

19. Collecchi P, Baldini E, Gianessi 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.[ISI][Medline]

20. Honkoop AH, Diest PJ, de Jong JS et al. Prognostic role of clinical, pathological and biological characteristics in patients with locally advanced breast cancer. Br J Cancer 1998; 77: 621–626.[ISI][Medline]

21. Hamilton A, Piccart M. The contribution of molecular markers to the prediction of response in the treatment of breast cancer: a review of the literature on Her-2, P53 and Bcl-2. Ann Oncol 2000; 11: 647–663.[Abstract]

22. Van Dam FS, Schagen SB, Muller MJ et al. Impairment of cognitive function in women receiving adjuvant treatment for high-risk breast cancer: high-dose versus standard-dose chemotherapy. J Natl Cancer Inst 1998; 90: 210–218.[Abstract/Free Full Text]

23. Danforth DN Jr, Findlay PA, McDonald HD et al. Complete axillary lymph node dissection for stage I–II carcinoma of the breast. J Clin Oncol 1986; 4: 655–662.[Abstract]

24. Du Toit RS, Locker AP, Ellis IO et al. Evaluation of the prognostic value of triple node biopsy in early breast cancer. Br J Surg 1990; 77: 163–167.[ISI][Medline]

25. Stadtmauer EA, O’Neill A, Goldstein LJ et al. Conventional-dose chemotherapy compared with high-dose chemotherapy plus autologous hematopoietic stem-cell transplantation for metastatic breast cancer. N Engl J Med 2000; 15: 1069–1075.

26. Skipper HE, Perry S. Kinetics of normal and leukemic leukocyte populations and relevance to chemotherapy. Cancer Res 1970; 30: 1883–1897.[ISI][Medline]

27. Surbone A, Norton L. Kinetic concepts in the treatment of breast cancer. Ann N Y Acad Sci 1993; 698: 48–62.[ISI][Medline]

28. Huitema ADR, Kerbusch T, Tibben MM et al. Reduction of cyclophosphamide-bioactivation by thioTEPA: critical sequence dependency in high-dose regimens. Cancer Chemother Pharmacol 2000; 46: 119–127.[ISI][Medline]

29. Schaake-Koning C, van der Linden EH, Hart G, Engelsman E. Adjuvant chemo- and hormonal therapy in locally advanced breast cancer: a randomized clinical study. Int J Radiat Oncol Biol Phys 1985; 11: 1759–1763.[ISI][Medline]

30. Rahman ZU, Frye DK, Buzdar AU et al. Impact of selection process on response rate and long-term survival of potential high-dose chemotherapy candidates treated with standard-dose doxorubicin containing chemotherapy in patients with metastatic breast cancer. J Clin Oncol 1997; 15: 207–215.[Abstract]

31. Garcia-Carbonero R, Hodalgo M, Paz-Ares L et al. Patient selection in high-dose chemotherapy trials: relevance in high-risk breast cancer. J Clin Oncol 1996; 15: 3178–3184.[Abstract]

32. Sobecks RM, Le Beau MM, Anastasi J, Williams SF. Myelodysplasia and acute leukemia following high-dose chemotherapy and autologous bone marrow or peripheral blood stem cell transplantation. Bone Marrow Transplant 1999; 23: 1161–1165.[ISI][Medline]

33. Laughlin MJ, McGaughey DS, Crews JR et al. Secondary myelodysplasia and acute leukemia in breast cancer after autologous bone marrow transplant. J Clin Oncol 1998; 16: 1008–1012.[Abstract]

34. Chaplain G, Milan C, Sgro C et al. Increased risk of acute leukemia after adjuvant chemotherapy for breast cancer: a population-based study. J Clin Oncol 2000; 18: 2836–2842.[Abstract/Free Full Text]

35. Kaas R, Hart AA, Besnard AP et al. Impact of mammographic interval on stage and survival after the diagnosis of contralateral breast cancer. Br J Surg 2001; 88: 123–127.[ISI][Medline]

36. Chaudary MA, Millis RR, Hoskins EOL et al. Bilateral primary breast cancer: a prospective study of disease incidence. Br J Surg 1984; 71: 711–714.[ISI][Medline]

37. Storm HH, Jensen OM. Risk of contralateral breast cancer in Denmark 1943–1980. Br J Cancer 1986; 54: 483–492.[ISI][Medline]