Characteristics associated with long-term progression-free survival following high-dose chemotherapy in metastatic breast cancer and influence of chemotherapy dose

A. Schneeweiss1,+, M. Hensel2, P. Sinn3, T. Khbeis1, R. Haas4, G. Bastert1 and A. D. Ho2

Departments of 1Gynecology and Obstetrics, 2Internal Medicine V and 3Pathology, University of Heidelberg, Heidelberg; 4Department of Hematology and Oncology, University of Düsseldorf, Düsseldorf, Germany

Received 27 September 2001; revised 14 December 2001; accepted 17 January 2002.


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background

The purpose of this study was to characterize long-term progression-free survivors (LTPFS) of metastatic breast cancer (MBC) following high-dose chemotherapy (HDCT) with autologous stem cell transplantation (ASCT) and to assess the influence of chemotherapy dose in order to identify patients who derive major benefit from this approach.

Patients and methods

We compared patient and tumor characteristics of 16 LTPFS with the characteristics of 118 MBC patients who received HDCT with ASCT at our institution between 1992 and 2000. To estimate the cumulative dose of chemotherapy received, the summation dose intensity product (SDIP) of the different chemotherapy regimens was calculated as recently described by Hryniuk et al. The SDIP of the induction regimens was added to that of the HDCT regimens to yield the total SDIP of the chemotherapy received. Multivariate analysis was performed to describe the influence of the total SDIP and other prognostic factors on progression-free survival (PFS).

Results

LTPFS were mostly <=50 years of age and had limited, chemotherapy-sensitive, hormone-responsive MBC. Due to an apparent dose–survival relationship, an increase by 10 units (U) in the SDIP increased the PFS time by 3 months. Independent predictors of an improved PFS were positive estrogen receptors (P = 0.001), positive combined hormone receptors (P = 0.020), and a complete remission/no evidence of disease status after HDCT (P <0.001). In patients who had a disease-free interval (DFI) >24 months after primary surgery, an SDIP of >55 U was independently associated with a longer PFS [hazard ratio (HR) = 2.73; 95% confidence interval 1.29–5.81; P = 0.009].

Conclusion

HDCT can achieve long-term PFS in young MBC patients with limited, hormone-responsive and chemotherapy-sensitive disease. After a DFI >24 months, a longer PFS is associated with a higher chemotherapy dose as measured by SDIP. These retrospective analyses suggest SDIP might be a tool for studying cumulative dose as a determinant of outcome of MBC chemotherapy. Thus far, however, we cannot clearly identify any subgroup of MBC patients in whom HDCT with ASCT is of particular benefit.

Key words: high-dose chemotherapy, long-term progression-free survival, metastatic breast cancer, summation dose intensity product


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Unlike early-stage breast cancer, metastatic breast cancer (MBC) is probably incurable in most cases [1]. Although some 60–70% of patients might initially respond to conventional therapy, most patients ultimately succumb to their disease [2]. However, typically 10–20% of patients achieve long-term progression-free survival (PFS) following stem cell-supported high-dose chemotherapy (HDCT) according to large numbers of phase II studies [3] and extensive data from the American and European bone marrow registries [4, 5]. This might be due to patient selection bias, as some patients could have achieved a similar survival following conventional-dose chemotherapy [2, 6]. On the other hand, there might be a certain subgroup of patients who derive particular benefit from very intensive or high-dose approaches. Most studies agree that the patients who are most likely to benefit from HDCT are younger patients (<45 to 50 years) in good clinical condition with hormone-sensitive, limited metastatic disease that responded well to conventional chemotherapy [710]. The question remains as to whether this subgroup would derive the same benefit from conventional-dose treatment as from HDCT. So far, the only randomized study in published form found no difference in PFS and overall survival (OS) between patients with chemotherapy-sensitive MBC randomized to HDCT or a standard-dose maintenance arm [11]. This study, however, has many limitations [12]. Therefore, a benefit from HDCT in a certain subgroup of patients might have been missed.

Hryniuk et al. [13] created a single scale, the summation dose intensity (SDI), with which the dose intensity of all chemotherapy regimens in breast cancer can be compared. In randomized trials that tested dose intensity in MBC in the conventional range, response rates and median survival correlated linearly with the SDIs of the scheduled treatments. To account not only for different dose intensities but also for different cumulative doses of treatment regimens, Hryniuk et al. [14] proposed using the SDI product (SDIP) as a tool to compare chemotherapy regimens.

In order to characterize further MBC patients who might derive benefit from HDCT, we compared patient and tumor characteristics of our long-term progression-free survivors (LTPFS) with the characteristics of all 118 MBC patients who received at least one stem cell-supported HDCT as front-line (110 patients) or second-line treatment at our institution between 1992 and 2000. We correlated the SDIP of the different chemotherapy regimens with the patient outcome, and performed univariate and multivariate analyses to describe the influence of SDIP and other prognostic factors on PFS.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients and systemic treatment
One hundred and eighteen patients with MBC were treated at the University of Heidelberg between 1992 and 2000 on sequential HDCT protocols with autologous stem cell transplantation (ASCT) as front-line treatment (110 patients) or after first relapse (eight patients). Inclusion criteria were: age between 18 and 60 years, Karnofsky performance score >=90%, normal hematological, cardiac, renal and hepatic function, and no relevant concomitant disease. The pretreatment patient and tumor characteristics, in detail, of all 118 patients, of those 46 patients who achieved a complete remission (CR) after chemotherapy or a status of no evidence of disease (NED) by surgical resection of metastases before chemotherapy, and of those 16 patients who remained progression-free for >3 years are given in Table 1. These characteristics include age, initial disease-free interval (DFI) after primary surgery, menopausal status at initial diagnosis, dominant metastatic site, number of metastatic sites, Possinger score, estrogen receptor (ER) status, combined hormone receptor status, HER2/neu expression, proportion of p53-positive tumor cells, and proliferation (S-phase fraction) and DNA content (DNA index) of tumor cells. The Possinger score is a prognostic index comprising hormone receptor status, initial DFI, and number and kind of metastatic sites, and was calculated as described by Possinger et al. [15]. HER2/neu and p53 were measured by immunohistochemical staining on primary tumor sections using the monoclonal antibodies 3B5 and DO7, respectively. DNA index and S-phase fraction were measured by DNA flow cytometry. All HDCT protocols were approved by the Joint Ethical Committee of the University of Heidelberg. Each patient gave her informed consent to be treated according to these protocols.


View this table:
[in this window]
[in a new window]
 
Table 1. Pretreatment characteristics of patients and tumors
 
A total of 112 patients received conventional induction chemotherapy followed by one (26 patients), two (45 patients), or three (41 patients) cycles of HDCT with ASCT. Six relapsed patients immediately received one (three patients) or two (three patients) cycles of HDCT without induction chemotherapy. Peripheral blood stem cells were reinfused 48 h after the end of each cycle of HDCT. The different chemotherapy regimens, including conventional induction chemotherapy and HDCT, and the number of patients who received these treatments are listed in Table 2. Irrespective of hormone receptor status, premenopausal patients received goserelin 3.6 mg s.c. once a month starting from the first induction chemotherapy for either 2 years (in cases of hormone receptor-negative tumors) or until the disease progressed. In cases of hormone receptor-positive tumor, postmenopausal patients received tamoxifen 20–30 mg/day orally (p.o.) or, if the metastases occurred during tamoxifen treatment, anastrozole 1 mg/day p.o., starting no later than 6 weeks after the last transplantation, until the disease progressed.


View this table:
[in this window]
[in a new window]
 
Table 2. Chemotherapy regimens administered, summation dose intensity product (SDIP) of different regimens, and number of patients who received those treatments
 
Calculation of the SDIP
The SDIP was calculated as described by Hryniuk et al. [13, 14]. Briefly, the unit dose intensity (UDI) for a single agent, which is the dose in mg/m2/week that produces a 30% CR plus partial remission (PR) rate as a single agent in first-line therapy for MBC, was fixed as defined by Hryniuk et al. [13]. UDIs of thiotepa and ifosfamide were fixed at 78 mg/m2/week and 1625 mg/m2/week, respectively (W. Hryniuk, personal communication) (Table 3). Then, separately for each induction and HDCT regimen, the planned dose intensities of each drug were divided by that drug’s UDI. The resulting decimal fractions were added to give the SDI. Multiplying the SDI by the intended treatment interval between two cycles measured in weeks and the number of cycles administered gives the SDIP. The treatment interval between two cycles of HDCT was fixed at 6 weeks, since that was the median time interval we observed within our trials [10]. Finally, for each patient we added the SDIP of induction chemotherapy (SDIPIND) and HDCT (SDIPHD) to get the total SDIP of the chemotherapy she received (SDIPTOT).


View this table:
[in this window]
[in a new window]
 
Table 3. Unit dose intensities of single agents in first-line therapy for metastatic breast cancer as described by Hryniuk et al. [13]
 
Statistical analysis
PFS was measured from the date of the first ASCT until the time of progression, death, or last contact. OS was measured from the date of the ASCT until death or last contact. PFS and OS curves were estimated using the Kaplan–Meier product limit method [16]. Further analyses were performed with PFS as a clinical outcome variable. The mean PFS ± standard deviation of different patient subgroups was plotted against the SDIPTOT of the chemotherapy those patients received. Patients of one subgroup received chemotherapy regimens with almost identical SDIPTOTs [within a range of 1.5 units (U)]. A simple regression analysis was performed to create a linear function that best described the relationship between SDIPTOT and PFS. Additionally, for patients in whom measurable lesions were not removed surgically, PFS was plotted against SDIPTOT and response to HDCT. Univariate analyses (log-rank tests [17]) and multivariate analyses (Cox regression analyses [18]) were performed to identify risk factors associated with PFS.

For all 118 patients, the following factors were examined in a univariate analysis: SDIPTOT (<=55 U versus >55 U), ER status (ER-positive/unknown versus negative), combined hormone receptor status [ER- or progesterone receptor (PgR)-positive/unknown versus ER- and PgR-negative], DFI after primary surgery (<=24 months versus >24 months), visceral disease (yes versus no), number of metastatic sites (<=2 versus >2), Possinger score (<=10 versus >10), status after HDCT (CR/NED versus no CR/NED), and number of cycles of HDCT (one or two versus three). Variables that were significant in the univariate analysis were included in a multivariate analysis as stated above. Additionally, we examined the influence of SDIPTOT (<=55 U versus >55 U) as a prognostic factor for PFS in various subgroups of patients, as listed in Table 4. If SDIPTOT (<=55 U versus >55 U) proved to be significant according to univariate analysis, we performed a multivariate analysis including all the aforementioned factors, i.e. ER status, combined hormone receptor status, DFI after primary surgery, visceral disease, number of metastatic sites, Possinger score, status after HDCT, and number of cycles of HDCT.


View this table:
[in this window]
[in a new window]
 
Table 4. Results of the univariate and multivariate analysis: total summation dose intensity product of the chemotherapy received (<=55 U versus >55 U) as a prognostic factor for progression-free survival in various subgroups of metastatic breast cancer patients treated with high-dose chemotherapy (HDCT)
 
All tests were performed using Systat software (version 7.0; Systat, Witzenhausen, Germany).


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Characteristics of LTPFS
Of the 118 MBC patients who received at least one HDCT with ASCT as front-line (110 patients) or second-line treatment, 46 patients (39%) achieved a CR. In 15 of these 46 patients the measurable lesions were surgically removed before chemotherapy. These patients were classified as showing NED. Forty-seven patients (40%) achieved a PR and in 25 patients (21%), disease was stable or progressed during chemotherapy. The PFS and OS curves with a median follow-up time of 48 months (range 10–103 months) according to the maximum response achieved are shown in Figures 1 and 2, respectively. One patient died of septicemia with multiple organ failure while in PR. All other patients died of MBC. Table 1 compares the characteristics of all patients, of the patients who achieved a CR/NED, and of the patients who have lived progression-free for >3 years. Besides younger age (<=50 years), the characteristics correlating with limited disease (dominant soft tissue metastases, one metastatic site, and Possinger score <=10) occurred more commonly in the CR/NED population than in the total population and were even more frequently present among LTPFS. For example, although only 54% of the total population had one metastatic site, 76% of patients with CR/NED and 81% of patients with long-term PFS were in this category. More LTPFS had positive ERs or a positive combined hormone receptor status compared with the total population and the CR/NED group. No significant difference was found on the basis of DFI, menopausal status, HER2/neu-expression, p53 positivity, S-phase fraction and DNA index.



View larger version (21K):
[in this window]
[in a new window]
 
Figure 1. Progression-free survival curves of 118 patients with metastatic breast cancer who received at least one cycle of peripheral blood stem cell-supported high-dose chemotherapy with a median follow-up of 4 years according to response to therapy. CR/NED, complete remission/no evidence of disease; NC, no change; PR, partial remission; PD, progressive disease.

 


View larger version (21K):
[in this window]
[in a new window]
 
Figure 2. Overall survival curves of 118 patients with metastatic breast cancer who received at least one cycle of peripheral blood stem cell-supported high-dose chemotherapy with a median follow-up of 4 years according to the response to therapy. CR/NED, complete remission/no evidence of disease; NC, no change; PR, partial remission; PD, progressive disease.

 
SDIP and survival
The different induction and HDCT regimens, their calculated SDIPIND, SDIPHD and SDIPTOT, and the number of patients who received those regimens are listed in Table 2. The SDIPTOT of the different regimens extended from 25.44 U up to 81.18 U. The median SDIPTOT was 64.32 U. When we correlated the SDIPTOT with the PFS time and performed a simple linear regression analysis, we found an apparent dose–survival relationship. An increase of 10 U in the SDIPTOT increased the PFS time by 3.0 months on average (Figure 3). All patients in whom metastatic lesions were not surgically removed before chemotherapy with a PFS time of 40 months or longer had achieved a CR after HDCT and had received a chemotherapy regimen with an SDIPTOT >55 U (Figure 4).



View larger version (23K):
[in this window]
[in a new window]
 
Figure 3. Progression-free survival (mean ± standard deviation) of different metastatic breast cancer patient subgroups who received at least one cycle of high-dose chemotherapy compared with the total summation dose intensity product of the chemotherapy those patients received (SDIPTOT). Each circle represents a subgroup of all 118 patients who received chemotherapy with an almost identical SDIPTOT. Circle size is proportional to the number of patients. p, P value of univariate analysis; r, regression coefficient of regression analysis; solid line, dose–survival line calculated by regression analysis.

 


View larger version (29K):
[in this window]
[in a new window]
 
Figure 4. Progression-free survival versus the total summation dose intensity product in 103 metastatic breast cancer patients according to their response to high-dose chemotherapy (HDCT). The 15 patients in whom metastases were surgically removed before chemotherapy were excluded from this analysis. Each symbol represents one patient. CR, complete remission; PR, partial remission; NC, no change; PD, progressive disease.

 
Univariate- and multivariate analyses
According to univariate analysis, positive/unknown ERs (P = 0.006), a positive/unknown combined hormone receptor status (P = 0.020), no more than two metastatic sites (P = 0.035), Possinger score <=10 (P <0.001), and a CR/NED after HDCT (P <0.001) were associated with a better PFS. We observed a non-significant trend for a longer PFS after chemotherapy with an SDIPTOT >55 U than after chemotherapy with an SDIPTOT <=55 U (P = 0.058). The number of cycles of HDCT (1 or 2 versus 3) had no influence on survival (P = 0.786). According to multivariate analysis, positive/unknown ERs [hazard ratio (HR) = 2.08 (95% confidence interval (CI) 1.33–3.21); P = 0.001], a positive/unknown combined hormone receptor status [HR = 1.72 (95% CI 1.18–2.50); P = 0.020] and a CR/NED after HDCT [HR = 3.58 (95% CI 2.11–6.04); P <0.001] remained independent predictors of longer PFS (Table 5).


View this table:
[in this window]
[in a new window]
 
Table 5. Results of the univariate and multivariate analysis: prognostic factors for progression-free survival in 118 metastatic breast cancer patients treated with high-dose chemotherapy (HDCT)
 
In the following subgroups of patients, an SDIPTOT >55 U was associated with longer PFS according to univariate analysis: patients with ER-positive/unknown tumors (P = 0.012), patients with ER- or PgR-positive/unknown tumors (P = 0.018), patients with a DFI >24 months (P = 0.012), patients with no more than two metastatic sites (P = 0.033), patients with a Possinger score <10 (P = 0.049), and patients who received one or two cycles of HDCT (P = 0.018). In patients with a DFI >24 months, an SDIPTOT >55 U remained an independent prognostic factor of later relapse in multivariate analysis [HR = 2.73 (95% CI 1.29–5.81); P = 0.009] (Table 4). As a consequence, patients with a DFI >24 months who received a regimen showing an SDIPTOT >55 U were estimated to have a PFS of 53% at 2 years, whereas in all other patients who received a less intensive chemotherapy, disease progressed within 24 months (Figure 5).



View larger version (19K):
[in this window]
[in a new window]
 
Figure 5. Progression-free survival curves of 51 patients with metastatic breast cancer who had an initial disease-free interval of >24 months according to the total summation dose intensity product of their chemotherapy (SDIPTOT). p, P value of multivariate analysis.

 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
With conventional treatment <5% of patients with MBC remain in remission for >5 years [2]. Following stem cell-supported HDCT, however, typically 10–20% achieve long-term PFS [35]. This might be due to patient selection bias. Rahman et al. [6] assessed the impact of selection process and found similar response and survival rates among potential HDCT candidates who received an anthracycline-containing standard-dose chemotherapy. On the other hand, there might be a certain subgroup of patients who derive particular benefit from very intensive or high-dose approaches with or without stem cell support.

Our data confirm that ~20% of MBC patients remain progression-free for >3 years following HDCT. The prerequisite was the achievement of a CR/NED status, as in all other patients disease progressed within 38 months. Achievement of CR/NED and no progression for longer than 3 years was more likely in younger patients (<=50 years) with a lower tumor burden, i.e. with soft tissue metastases at one site only and a Possinger score <=10. The response to HDCT was not related to the ER or combined hormone receptor status. LTPFS, however, were more often receptor-positive. This reflects, at least in part, the efficacy of antihormonal treatment in hormone-responsive MBC [19].

These results are in line with the findings of Rizzieri et al. [9] who analyzed 425 MBC patients. They found an increased long-term disease-free survival in ER-positive patients with smaller, non-visceral metastases and a longer DFI in those who received immediate or delayed HDCT. In a smaller study of HDCT in 20 MBC patients, all three LTPFS had predominantly nodal disease, and two of them achieved a CR after induction chemotherapy [20]. With respect to these characteristics, LTPFS after HDCT does not differ from LTPFS after conventional chemotherapy [2, 21–25]. However, it is noteworthy that in our analysis the proportion of patients with overexpression of HER2/neu, which is associated with poorer outcome after conventional chemotherapy [26], is nearly the same among LTPFS as among all patients. Because of the small sample size one should be cautious in interpreting this observation; nevertheless, it could be hypothesized that HDCT might be able to overcome, at least in part, the unfavorable prognostic impact of HER2/neu overexpression. There was no appreciable impact of p53 positivity, S-phase fraction or DNA index, but too few data are available to draw valid conclusions regarding these factors.

We found no relationship between the PFS and the SDI, which accounts only for dose intensity and activity of each chemotherapy drug (data not shown). However, when we accounted for all three relevant factors, namely dose intensity, activity and cumulative dose of each drug, by using the SDIP, we observed an apparent dose–survival relationship between the PFS of our patients and the SDIPTOT of the chemotherapy they received. The PFS increased by an average of 3 months for each increase of 10 U SDIPTOT. Besides the patients with an NED status as a result of surgical removal of metastases before chemotherapy, all patients with a PFS of >=40 months had achieved a CR after HDCT and had received chemotherapy with an SDIPTOT of >55 U. To the best of our knowledge the relationship between SDIP and PFS in patients with MBC has not been evaluated by others either in the conventional or in the high-dose setting. Reports on most phase II and all phase III trials of HDCT in MBC lack exact information on induction chemotherapy. Therefore, SDIPTOT could not be calculated from the available data [11, 27–29]. Hryniuk et al. [13] found a linear correlation between SDI as a single scale and survival of MBC patients in randomized trials that tested the influence of dose intensity in the conventional range. An increment of 1 SDI unit increased the median survival by 3.75 months. To account not only for the dose intensity and activity of different drugs but also for their total dose, the investigators proposed multiplying the SDI by the treatment duration to get an extended score, the SDIP, which was used in this analysis [14].

Positive ERs, positive combined hormone receptors, and CR/NED after HDCT remained independent prognostic factors for PFS in multivariate analysis, which confirmed our preliminary analysis on the first 76 patients [10] and agrees with the findings of others [2, 8]. The number of metastatic sites was no longer an independent prognostic factor, because it correlates with the probability of the achievement of a CR/NED, which has proven to be the strongest predictor of long-term PFS (P <0.001). Considering all 118 patients, only in univariate analysis did we find a non-significant trend towards better PFS for patients who received chemotherapy with an SDIPTOT of >55 U (P = 0.058). In patients with a DFI >24 months, however, an SDIPTOT >55 U was found to be an independent predictor of later relapse after adjustment for all other prognostic factors [HR = 2.73 (95% CI 1.29–5.81); P = 0.009]. It could be hypothesized that, in those patients, the higher-dosed chemotherapy was more capable of eradicating minimal residual disease, which persists in less aggressively treated patients, thereby leading to earlier relapse and ultimately death. On the other hand, this survival difference could result from a selection bias, as patients in better general condition with no disease progression during chemotherapy—who intrinsically have a better prognosis—tolerated and received more cycles of chemotherapy. However, thus far two or more cycles of stem cell-supported HDCT have not improved survival of MBC patients as compared with one cycle of HDCT [10, 30, 31]. Prospective, randomized trials comparing different chemotherapy regimens as defined by different SDIPs are required to confirm our findings. Provided that an optimal HDCT regimen is chosen, SDIP should be increased by prolonging the induction chemotherapy rather than by adding multiple cycles of HDCT.

So far only one phase III study comparing stem cell-supported HDCT versus a conventional maintenance chemotherapy as consolidation treatment in patients with MBC has been published as a full paper [11]. This trial showed no difference in PFS and OS after 3 years of follow-up. It has, however, a number of limitations: small number of randomized patients (n = 199, only n = 45 complete responders); substantial number of treatment refusals and arm crossover (n = 15); almost twice the cumulative cyclophosphamide dose in the conventionally treated than in the high-dose group; suboptimal high-dose regimen (only 7% conversion from PR to CR); and possible tumor regrowth due to delay in the HDCT of up to 3 months. Three other randomized trials published in abstract form only found an improved PFS for patients treated with early HDCT [2729]. So far, there has been no improvement in OS, but the follow-up time of 2–3 years in two trials is still too short [28, 29]. Therefore, it is conceivable that a benefit of HDCT in a certain subgroup of patients could have been missed [3, 12]. However, it has to be stressed that thus far with our current knowledge we cannot clearly identify any subgroup of patients with MBC in whom HDCT with stem cell support is of benefit.

In summary, our data suggest that HDCT may be a possible tool for eradicating minimal residual disease, at least in patients 50 years of age or younger with hormone-responsive and chemotherapy-sensitive, limited MBC and a long DFI (>24 months). However, the authors are aware that retrospective analyses such as the present one are only useful for generating hypotheses. Prospective testing of chemotherapy doses as measured by SDIP is required to confirm our findings. Furthermore, HDCT with ASCT will only, if ever, solve the problem of breast cancer for a small group of patients. Therefore, it should be considered as one component of a multistep approach together with new strategies in order to improve long-term survival or even to cure those patients.


    Acknowledgements
 
We thank William Hryniuk for his helpful comments on earlier drafts of the manuscript.


    Footnotes
 
+ Correspondence to: Dr A. Schneeweiss, University of Heidelberg, Department of Gynecology and Obstetrics, Vossstrasse 9, D-69115 Heidelberg, Germany. Tel: +49-6221-567856; Fax: +49-6221-567920; E-mail: andreas_schneeweiss@med.uni-heidelberg.de Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Henderson IC, Harris JR, Kinne DW et al. Cancer of the Breast. In DeVita VT, Hellman S, Rosenberg SA (eds): Cancer—Principles and Practice of Oncology, 2nd edition. Philadelphia, PA: Lippincott 1989; 1197–1268.

2. Greenberg PAC, Hortobagyi GN, Smith TL et al. Long-term follow-up of patients with complete remission following combination chemotherapy for metastatic breast cancer. J Clin Oncol 1996; 14: 2197–2205.[Abstract]

3. Peters WP, Dansey RD, Klein JL et al. High-dose chemotherapy and peripheral blood progenitor cell transplantation in the treatment of breast cancer. Oncologist 2000; 5: 1–13.[Abstract/Free Full Text]

4. Antman KH, Rowlings PA, Vaughan WP et al. High-dose chemotherapy with autologous hematopoietic stem-cell support for breast cancer in North America. J Clin Oncol 1997; 15: 1870–1879.[Abstract]

5. Rosti G, Ferrante P, Maranglolo M. European trends and the EBMT database. In Dicke KA, Keating A (eds): Autologous Blood and Marrow Transplantation. Proceedings of the Ninth International Symposium, Arlington, Texas. Charlottesville, VA: Carden Jennings Publishing 1999.

6. 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: 3171–3177.[Abstract]

7. deVries EGE, Rodenhuis S, Schouten HC et al. Phase II study of intensive chemotherapy with autologous bone marrow transplantation in patients in complete remission of disseminated breast cancer. Breast Cancer Res Treat 1996; 39: 307–313.[ISI][Medline]

8. Rowlings PA, Williams SF, Antman KH et al. Factors correlated with progression-free survival after high-dose chemotherapy and hematopoietic stem cell transplantation for metastatic breast cancer. J Am Med Assoc 1999; 282: 1335–1343.[Abstract/Free Full Text]

9. Rizzieri DA, Vredenburgh JJ, Jones R et al. Prognostic and predictive factors for patients with metastatic breast cancer undergoing aggressive induction therapy followed by high-dose chemotherapy with autologous stem cell support. J Clin Oncol 1999; 17: 3064–3074.[Abstract/Free Full Text]

10. Schneeweiss A, Hensel M, Goerner R et al. Comparison of double and triple high-dose chemotherapy with autologous blood stem cell transplantation in patients with metastatic breast cancer. Stem Cells 2001; 19: 151–160.[Abstract/Free Full Text]

11. 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; 325: 1371–1372.[ISI][Medline]

12. Rodenhuis S. The status of high-dose chemotherapy in breast cancer. Oncologist 2000; 5: 369–375.[Abstract/Free Full Text]

13. Hryniuk W, Frei E III, Wright FA. A single scale for comparing dose-intensity of all chemotherapy regimens in breast cancer: summation dose intensity. J Clin Oncol 1998; 16: 3137–3147.[Abstract]

14. Hryniuk WM, Peters WP, Ragaz J. Determinants of outcome in adjuvant chemotherapy of breast cancer: dose intensity vs total dose vs dose size. Breast Cancer Res Treat 2000; 64: 66 (Abstr).

15. Possinger K, Williams W. Palliative Therapieführung zur Hemmung der Tumorprogression bei Patientinnen mit metastasiertem Mammakarzinom. Internist 1993; 34: 340–350.[ISI][Medline]

16. Kaplan EL, Meier P. Non-parametric estimation from incomplete observation. J Am Stat Assoc 1958; 47: 457–481.

17. Peto R, Peto J. Asymptotically efficient rank invariant test procedures. J R Stat Soc (A) 1972; 135: 185–210.[ISI]

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

19. Buzdar AU. Endocrine therapy in the treatment of metastatic breast cancer. Semin Oncol 2001; 28: 291–304.[ISI][Medline]

20. Baker WJ, Vukelja SJ, Burrell LM et al. High-dose cyclophosphamide, etoposide and carboplatin with autologous bone marrow support for metastatic breast cancer: long-term results. Bone Marrow Transplant 1998; 21: 775–778.[ISI][Medline]

21. Decker DA, Ahmann DL, Bisel HF et al. Complete responders to chemotherapy in metastatic breast cancer. J Am Med Assoc 1979; 242: 2075–2079.[Abstract]

22. Swenerton KD, Legha SS, Smith T et al. Prognostic factors in metastatic breast cancer treated with combination chemotherapy. Cancer Res 1979; 39: 1552–1562.[Abstract]

23. Nash CH, Jones SE, Moon TE et al. Prediction of outcome in metastatic breast cancer treated with adriamycin combination chemotherapy. Cancer 1980; 46: 2380–2388.[ISI][Medline]

24. Paterson AHG, Zuck VP, Szafran O et al. Influence and significance of certain prognostic factors on survival in breast cancer. Eur J Cancer Clin Oncol 1982; 18: 937–943.[ISI][Medline]

25. Pedrazzini A, Cavalli F, Brunner KW et al. Complete remission following endocrine or combined cytotoxic and hormonal treatment in advanced breast cancer. Oncology 1987; 44: 51–59.[ISI][Medline]

26. Ross JS, Fletcher JA. The HER-2/neu oncogene in breast cancer: prognostic factor, predictive factor, and target for therapy. Stem Cells 1998; 16: 413–428.[Abstract/Free Full Text]

27. Peters WP, Jones RB, Vredenburgh J et al. A large prospective, randomized trial of high-dose combination alkylating agent (CPB) with autologous cellular support (ABMS) as consolidation for patients with metastatic breast cancer achieving complete remission after intensive doxorubicin-based induction therapy (AFM). Proc Am Soc Clin Oncol 1996; 15: 121a (Abstr 149).

28. Lotz JP, Cure H, Janvier M et al. High dose chemotherapy with hematopoietic stem cell transplantation for metastatic breast cancer: results of the French protocol PEGASE 04. Proc Am Soc Clin Oncol 1999; 18: 43a (Abstr 161).

29. Crump M, Gluck S, Stewart D et al. A randomized trial of high-dose chemotherapy (HDC) with autologous peripheral blood stem cell support (ASCT) compared to standard therapy in women with metastatic breast cancer: a National Cancer Institute of Canada (NCIC) Clinical Trials Group study. Proc Am Soc Clin Oncol 2001; 20: 21a (Abstr 82).

30. Hu WW, Negrin RS, Stockerl-Goldstein K et al. Four-cycle high-dose therapy with hematopoietic support for metastatic breast cancer: No improvement in outcomes compared with single-course high-dose therapy. Biol Blood Marrow Transplant 2000; 6: 58–69.[ISI][Medline]

31. Bashey A, Corringham S, Garrett J et al. A phase II study of two cycles of high-dose chemotherapy with autologous stem cell support in patients with metastatic breast cancer who meet eligibility criteria for a single cycle. Bone Marrow Transplant 2000; 25: 519–524.[ISI][Medline]