F Montemurro 1, G Rondón 2, NT Ueno 2, M Munsell 3, JL Gajewski 2 and RE Champlin 2. Summary:

Transcription

1 (2002) 29, Nature Publishing Group All rights reserved /02 $ Breast cancer Factors affecting progression-free survival in hormone-dependent metastatic breast cancer patients receiving high-dose chemotherapy and hematopoietic progenitor cell transplantation: role of maintenance endocrine therapy F Montemurro 1, G Rondón 2, NT Ueno 2, M Munsell 3, JL Gajewski 2 and RE Champlin 2 1 Department of Oncology and Hematology, Institute for Cancer Research and Treatment, Candiolo, Torino, Italy; 2 Department of Blood and Marrow Transplantation, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; and 3 Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Summary: We retrospectively analyzed the effect of maintenance endocrine therapy (MET) after high-dose chemotherapy with hematopoietic progenitor cell transplant (HDCT) on the progression-free survival (PFS) of patients with hormone-dependent metastatic breast cancer (MBC). One hundred and nine consecutive patients with estrogen receptor (ER) and/or progesterone receptor (PgR)- positive MBC, who were progression free for at least 4 months after HDCT with cyclophosphamide, carmustine and thiotepa (CBT), were analyzed. Of these, 55 were non-randomly submitted to MET. After a median follow-up of 34.4 months (17.1 9), univariate analysis showed that MET was significantly associated with improved median PFS (31.1 vs 19.2 months, P = 22). Complete response to HDCT, pattern of metastatic spread, extent of the disease, single vs multiple metastatic sites, prior endocrine therapy for metastatic disease and prior exposure to any hormonal therapy (adjuvant and/or for the advanced disease) were also associated with PFS at univariate analysis. A multivariate Cox proportional hazard model was fitted to the data in order to correct the effect of MET for the other significant covariates. After correcting for these covariates, MET was still significant, predicting improved PFS (hazard ratio (HR) 80, 95% CI; 62 31). Administration of MET after optimal cytoreduction might result in increased efficacy of HDCT in hormonedependent metastatic breast cancer. (2002) 29, DOI: 1038/sj/bmt/ Keywords: breast neoplasm; metastatic; hematopoietic stem cell transplantation; maintenance endocrine therapy; hormonal therapy Correspondence: Dr F Montemurro, Dept of Oncology and Hematology, Institute for Cancer Research and Treatment, Strada Provinciale 142, Candiolo, Torino, Italy Received 27 August 2001; accepted 26 February 2002 Despite initial promising results, high-dose chemotherapy with hematopoietic progenitor cell transplant (HDCT) has not been clearly established. 1,2 Several randomized trials are still underway which may shed light on this controversial area of oncology. 3 Retrospective analyses of metastatic breast cancer patients, have found that the use of HDCT is consistently associated with higher objective response and complete response rate (CR), as a result of conversion into CR of a proportion of patients with partial or no response to standard-dose chemotherapy (SDC). 4 8 The achievement of a CR to either SDC or HDCT results in increased progression-free survival and a fraction of patients achieve long-term survival. 4,9,10 The use of endocrine therapy is well established in patients with hormone-dependent metastatic breast cancer. 11,12 The role of maintenance endocrine therapy (MET) in controlling the regrowth of hormone-dependent clones after maximum cytoreduction with SCD followed by HDCT has not been formally evaluated. Metastatic breast cancer patients, treated with cyclophosphamide, carmustine and thiotepa (CBT) HDCT in clinical trials at our institution, were allowed to receive MET, according to their treating physician, if they had achieved a CR, partial response (PR), or stable disease (SD) to the high-dose program. The purpose of the current retrospective analysis was to assess the impact of MET on progression-free survival after transplant in the subgroup of patients with positive estrogen and/or progesterone receptor metastatic breast cancer, after correcting for other covariates. Patients and methods Patients were selected from 232 consecutive women with MBC, who underwent HDCT consisting of cyclophosphamide, carmustine and thiotepa (CBT) on several different protocols at The University of Texas MD Anderson Cancer Center between August 1991 and December 1998.

2 862 Patients received induction SDC with the intent to produce a maximum cytoreduction before starting HDCT. To be eligible for HDCT, patients had to be between 18 and 65 years old and have a Zubrod performance status 1, an estimated creatinine clearance 60 ml/min, SGOT, SGPT and bilirubin levels less than twice the upper limits of normal, adequate cardiac function (left ventricular ejection fraction 50%), and adequate pulmonary function (DLCO 50% of predicted). The extent of metastatic disease was assessed using the scoring system developed by Swenerton et al. 13 In brief, 12 anatomical sites: ipsilateral breast, contralateral breast, lymph nodes, skin/chest wall, lung, pleura, liver, mediastinum/abdomen intraperitoneal, peritoneum, bone, bone marrow, and central nervous system were examined. The extent of disease at each site was defined as follows: 0, no disease; 1, strong suspicion of involvement but insufficient laboratory or clinical information to define further; 2, minimal involvement; 5, moderate involvement; 10 extensive involvement. The total burden of metastatic disease was estimated as the sum of the scores for tumor extent at all known disease sites. The maximum extent of metastatic disease before induction chemotherapy and the extent of metastatic disease just prior to initiation of the HDCT were determined. Standard criteria for response were used. 14 Complete remission was defined as 100% disappearance of all known disease lasting for at least 1 month and normalization of all abnormal laboratory parameters. Complete remission in bone was defined as recalcification of all lytic osseous lesions and disappearance of all abnormal uptake areas previously noted on a bone scan, with complete absence of bone pain without analgesics. Partial remission was recorded as a 50% or greater reduction in the sum of the product of the two longest perpendicular diameters of all measurable lesions that persisted for at least 1 month. A PR in bone was considered an improvement or stabilization of radiographic assessment of disease with subjective improvement (a decrease in bone pain and lowering of analgesics by at least 50%) and improvement in performance status by one grade or more that lasted at least 3 weeks. Stable disease (SD) was defined as less than a 50% decrease or less than a 25% increase in the sum of the product of the two longest perpendicular diameters of all measurable lesions. Progressive disease (PD) was defined as a 25% or greater increase in the sum of the product of the two longer perpendicular diameters of one measurable lesion (even with regression of other lesions) or the appearance of new lesions. Early death (ED) was defined as death within 100 days after transplant, before any assessment of a response to the transplant could be made. The CBT HDCT regimen consisted of cyclophosphamide 2 g/m 2 i.v. on day 6, 5 and 4 (total dose 6 g/m 2 ), with Mesna 500 mg/m 2 before the first dose of cyclophosphamide and then Mesna 2 g/m 2 as a continuous i.v. infusion over 24 h for 3 days; thiotepa 240 mg/m 2 on days 6, 5 and 4 (total dose 720 mg/m 2 ); and carmustine i.v. 150 mg/m 2 on days 6, 5 and 4. Cryopreserved autologous peripheral blood progenitor cells or bone marrow were infused on day 0. Hematopoietic progenitor cells were collected differently in each protocol. One hundred and sixty-four patients received CVP regimen, as peripheral blood stem cell mobilization, in the high-dose program. This consisted of cyclophosphamide 1.5 g/m 2 i.v., etoposide 250 mg/m 2 i.v. and cisplatin 40 mg/m 2 on days 1 3, along with hematopoietic colony-stimulating factors (CSF). Only patients with estrogen receptor (ER) and/or progesterone receptor (PgR) positive metastatic breast cancer were selected for this analysis. The characteristics of patients receiving (MET) or not receiving (NoMET) post-transplant endocrine therapy were compared using the Chi-square test for categorical variables. Since age was normally distributed in the two groups, comparison was made by the Student s t-test. In order to avoid a bias potentially introduced by events, progression or death, occurring before the intended treatment could have been administered, a landmark for the survival analysis was established 4 months after the date of transplant. Overall survival (OS) was calculated from the date of transplant to the date of death from any cause. Progression-free survival (PFS) was calculated from the date of transplant to the date of disease progression or death from any cause. Kaplan Meier estimates of progression-free survival, were then found for each level of the various potential prognostic factors in a univariate fashion and compared by the logrank test. Those factors which were found to have statistically significant differences between strata (P value 5) were then analyzed together in a multivariate Cox proportional hazards regression model. Results are reported as estimated hazard ratio (HR) and confidence intervals for the HR. P values were calculated by means of the Wald statistic, and considered significant at the 5 level. Statistical analysis was performed using the SPSS-PC software (SPSS, Chicago, IL, USA). Results The selection process for this retrospective analysis is outlined in Figure 1. After high-dose chemotherapy and hematopoietic pro- 232 received HDCT 207 achieved CR, PR or SD 124 had ER and/or PgR+ MBC 65 received MET 52 had not received MET 10 excluded 5 excluded 55 MET 54 NoMET Figure 1 Flow chart summarizing the selection process for the retrospective analysis.

3 genitor cell transplantation, 103 patients were in CR (44%), 40 attained a further tumor response with respect to their status before transplant (PR, 17%), 64 patients (20 in SD, 38 in PR and six in PD before transplant) had no change in their disease status (SD, 28%), 14 patients progressed (PD, 6%), and 11 patients died from transplant-related complications (ED, 5%). Of the 207 patients who had attained an objective response or disease stabilization after transplant, 124 had ER and/or PgR receptor positive primary tumor. Of these patients, 65 received post-transplant maintenance endocrine therapy (MET). Nine patients in the MET group and one patient in the NoMET group were excluded from the analysis because they received posttransplant chemotherapy for residual disease. Another five patients, one in the MET group and four in the NoMET group, relapsed within 4 months from transplant and were excluded from the analysis (see Materials and methods). Table 1 summarizes the characteristics of the two groups of patients (MET and nomet). Although differences were not statistically significant, more patients in the NoMET group had been previously exposed to endocrine therapy (adjuvant and/or for the advanced disease, 59% vs 44%), and had visceral involvement (44% vs 31%). Moreover, NoMET patients tended to have more extensive disease, measured as maximum Swenerton score before induction chemotherapy. Most of the patients (87% and 85% in the MET and NoMET group, respectively) underwent HDCT after first-line conventional chemotherapy for metastatic disease. All the patients previously treated with adjuvant endocrine therapy received tamoxifen. Of the 30 patients who had received endocrine therapy for metastatic disease, nine (five in the MET and four in the NoMET groups) had also received a second-line hormonal compound upon progression to the first line used. Table 2 summarizes the endocrine treatments used as first-line, second-line and MET, respectively. At a median follow-up of 34.4 months (range ), the Kaplan Meier estimates of median OS and PFS for all the patients were 54.0 and 21.3 months, respectively. There was a trend for better OS in patients receiving MET (median 76.5 vs 41.7 months), but this was not statistically significant (P = 1) (Figure 2a). MET was significantly associated with prolonged PFS (median 31.1 vs 19.2 months, P = 22) (Figure 2b). The results of univariate analysis of factors affecting PFS are summarized in Table 3. Other covariates found to be significant at the univariate level were complete response at the end of the transplant program (CR), maximum extent of the disease at any time and extent of the disease at the time of HDCT as measured by the Swenerton score, single vs multiple site of metastatic involvement, dominant site of disease, prior exposure to endocrine therapy for metastatic disease and prior exposure to any endocrine therapy, whether in the adjuvant setting or for metastatic disease. A Cox proportional hazard model was fitted to these data in order to study the independent effect of post-transplant endocrine maintenance therapy on PFS, after correcting for the other significant covariates. Response to HDCT was dichotomized as CR vs PR and SD; dominant site of metastatic disease was dichotomized as bone vs others; extent of disease measured by the Table 1 Patient characteristics MET NoMET P values Number Mean age, years (range) 46 (33 61) (45 (30 62) 3 a DFI 2 years 26 (47) 26 (48) 2 years 29 (53) 28 (52) 3 Receptor status ER+ and PgR+ 39 (71) 32 (59) ER+ or PgR+ 16 (29) 22 (41) 0 Neo-adjuvant chemotherapy 38 (69) 35 (65) 4 Adjuvant endocrine 16 (29) 21 (39) 8 Endocrine for metastatic 13 (24) 17 (31) 6 disease Prior endocrine therapy b 24 (44) 32 (59) 0 Prior chemotherapy regimens for metastatic disease 1 48 (87) 46 (85) 2 7 (13) 8 (15) 6 Dominant site of disease Bone 27 (49) 17 (31) Soft tissue/nodes 11 (20) 13 (24) Liver 10 (18) 13 (24) Other viscera 7 (13) 11 (20) 4 Visceral involvement 17 (31) 24 (44) 4 Single metastatic site 29 (53) 23 (43) 8 Maximum extent of the disease c 5 21 (38) 14 (26) (38) 19 (35) (7) 13 (24) 20 9 (16) 8 (15) 0 Extent of the disease before HDCT d 5 36 (65) 33 (61) (24) 16 (30) (5) 5 (9) 20 3 (5) 0 7 CVP as part of the HD 35 (64) 32 (59) 4 program Final response to HDCT CR 26 (47) 24 (44) PR 11 (20) 9 (17) SD 18 (33) 21 (39) 8 Numbers in brackets are percentages. a Student s t-test for normally distributed data. All the other comparisons were made by the Chi-square test. b Prior exposure to any anti-endocrine therapy (adjuvant or for the advanced disease); P = 9. c Maximum extent of the disease before induction chemotherapy, as measured by the Swenerton score. d As measured by the Swenerton score. Because of rounding, sum of percentages is not always equal to 100. Swenerton score was dichotomized as 20 vs 20. The results of this analysis are summarized in Table 4. The following factors were found to be independently associated with improved PFS: CR to HDCT (HR 96, 95% CI; 72 10), and MET (HR 80, 95% CI; 62 31). The following factors were independently associated with worse PFS: maximum extent of metastatic disease 20 by the Swenerton score (HR 2.353, 95% CI; ) and prior exposure to endocrine therapy (any) (HR 1.651, 95% CI; ). Figure 3a and b shows the Kaplan Meier estimates of 863

4 864 Table 2 Summary of endocrine treatments Drug First line Second line Maintenance (No. of (No. of endocrine therapy patients) patients) (No. of patients) Tamoxifen Anastrozole Megestrol acetate Fluoxymesterone 1 Roloxifen 1 LH-RH inhibitors 2 Letrozole 1 a b Figure 2 Kaplan Meier estimates of survival (a) and progression-free survival (b) for patients receiving (solid line) and not receiving (dotted line) maintenance endocrine therapy. Log-rank test for survival, P = 1; log-rank test for PFS, P = 22. PFS according to MET in patients achieving (a) or not achieving (b) a CR to HDCT. MET was associated with better PFS in patients in CR after transplant (3 years PFS, 75.5% vs 38.4%), although this difference was of borderline statistical significance (log-rank test, P = 6) (Figure 3a). Similarly MET conferred a non-significant PFS benefit to patients failing to achieve a CR to HDCT (3-year PFS 21% vs 7.7%, log-rank test, P = 2) (Figure 2b). Discussion This retrospective analysis shows that the administration of MET resulted in a longer progression-free interval achieved by HDCT in hormone receptor-positive metastatic breast cancer patients. MET retained independent prognostic Table 3 analysis) Prognostic factors for progression-free survival (univariate Variable Median PFS Log rank (months) Age at transplant DFI years Receptor status ER and PgR ER or PgR Prior adjuvant endocrine therapy Yes No 26.6 Prior endocrine therapy for metastatic disease Yes No 26.6 Any previous exposure to endocrine therapy Yes No 32.8 Prior neo/adjuvant chemotherapy Yes No 2 Prior chemotherapy regimens for metastatic disease Dominant site of disease Bone 3 Soft tissue Liver 11.7 Other visceral 14.8 Maximum extent of the disease Extent of the disease before HDCT Number of sites CVP 62 Response to transplant CR 44.3 PR SD 9.8 Maintenance endocrine therapy 22 Yes Yes No No value in the multivariate model considering other important prognostic factors, like CR to HDCT, disease extent and pattern of metastatic disease. A few other retrospective analyses have had similar conclusions. Rowlings et al 5 analyzed data from the Autologous Blood and Marrow Transplant Registry regarding 1188 metastatic breast cancer patients treated with HDCT. This study showed a positive impact of post-transplant maintenance tamoxifen on PFS in

5 Table 4 Prognostic factors for progression-free survival (multivariate analysis) Variable Progression-free survival HR P values 95% Confidence intervals Maximum extent of the disease 20 Prior exposure to endocrine therapy (any) CR vs PR and SD Maintenance endocrine therapy 6 a b Figure 3 Kaplan Meier estimate of progression-free survival for patients receiving (solid line) and not receiving (dotted line) maintenance endocrine therapy according to their response to high-dose chemotherapy. For patients in CR (a), MET vs NoMET 3-year PFS was 75.5% vs 38.4% (log-rank test, P = 6). For patients achieving PR or SD (b) METvs NoMET 3-year PFS was 21% vs 7.7% (log-rank test 2). patients with hormone-dependent metastatic breast cancer. Another small study in metastatic breast cancer patients receiving conventional epirubicin therapy suggested a possible survival benefit from MET. 15 The only randomized trial assessing the role of MET in MBC patients has shown a modest improvement in PFS of MBC patients responding to six cycles of epirubicin and ifosfamide; 16 the administration of medroxyprogesterone acetate to 46 patients resulted in a PFS gain of 1.2 months compared to 44 patients in the control group. The small sample size, the inclusion of receptor-negative patients, and the use of an older generation endocrine compound limit the interpretation of this trial. In our study, 13 (23%) MET patients and 17 (31%) NoMET patients received endocrine therapy for metastatic disease and progressed before they were considered for HDCT. Their tumors could therefore be considered refractory to the hormonal compound administered. On univariate analysis, this factor had a detrimental effect on PFS (13.3 vs 26.6 months, P = 18), which disappeared in the multivariate model. Any prior exposure to endocrine therapy (adjuvant or for metastatic disease) retained an independent value in predicting worse PFS at the multivariate analysis (HR for progressive disease 1.651, P = 48), suggesting that selection of hormone refractory, aggressive clones might have taken place in extensively pretreated patients. The availability of newer hormonal compounds, with increased efficacy and improved tolerance, may provide more therapeutic options for MET in patients who have previously progressed during hormonal treatment We found a trend towards longer OS in patients receiving MET (Figure 1a), which may become significant with longer follow-up. We speculate that, since treatment upon progression after HDCT is expected to have only a marginal effect on the natural history of metastatic breast cancer, 23 delaying disease progression by maintenance therapy might eventually result in longer survival. Moreover, prolonged PFS would have a positive impact on quality of life in metastatic breast cancer patients, especially after the toxic burden of HDCT. The retrospective design and the small sample size do not allow us to draw any firm conclusion about the role of MET in metastatic breast cancer patients responding to conventional or high-dose chemotherapy, a randomized trial being the most appropriate method to address this issue. In fact, although no significant imbalance in the distribution of baseline prognostic factors was found between the two groups, more patients in the NoMET group had been previously exposed to endocrine therapy (adjuvant and/or for the advanced disease, 59% vs 44%), and had visceral involvement (44% vs 31%). Moreover, NoMET patients tended to have more extensive disease, measured as maximum Swenerton score before induction chemotherapy. One obvious consideration is that the choice of giving or not giving MET after HDCT might have been biased in favor of patients with better prognosis (non-visceral disease, limited disease extent, no prior exposure to endocrine therapy). In summary, MET was independently associated with better PFS in this selected group of patients. CR is a requisite for long-term survival, and MET confers an additional advantage to patients in CR (Figure 3a and b). Very extensive metastatic disease before HDCT (Swenerton score 20) predicts worse PFS. Patients pretreated with endocrine compounds had increased risk for relapse (more aggressive tumor behavior?). The detrimental effect of prior exposure to endocrine compounds, if confirmed in larger analyses or prospectively, is likely to have an impact on the optimal timing of administration of endocrine compounds in this group of 865

6 866 patients. In an ideal randomized trial, HDCT should be given upfront in endocrine-dependent metastatic breast cancer patients and, once maximum cytoreduction is achieved, endocrine therapy should follow. High-dose chemotherapy remains a promising approach to improve cytoreduction in patients with breast cancer. Ongoing randomized trials will hopefully clarify the efficacy of the general approach and identify subsets of patients in whom this approach is indicated. Our observation suggests that controlling tumor regrowth after maximally reduced tumor burden might result in increased efficacy of HDCT. The same rationale could be extended to other promising developmental strategies to eradicate drug resistant minimal residual disease, such as immunotherapy, or therapy with antiangiogenic factors. Acknowledgements This work was presented in part at the EBMT International Conference on High-Dose Chemotherapy in Breast and Ovarian Cancer, September 2000, Florence, Italy. FM was also supported by a grant from Institute for Cancer Research and Treatment, Ordine Mauriziano, Candiolo, Torino, Italy. References 1 Stadtmauer EA, O Neill A, Goldstein LJ et al. Conventionaldose chemotherapy compared with high-dose chemotherapy plus autologous hematopoietic stem-cell transplantation for metastatic breast cancer. Philadelphia Bone Marrow Transplant Group. New Engl J Med 2000; 342: Lotz JP, Cure H, Janvier M et al. Intensive chemotherapy and autograft of hematopoietic stem cells in the treatment of metastatic cancer: results of the national protocol Pegase 04. Hematol Cell Ther 1999; 41: Nieto Y, Champlin RE, Wingard JR et al. Status of high-dose chemotherapy for breast cancer: a review. Biol Blood Marrow Transplant 2000; 6: Rizzieri DA, Vredenburgh JJ, Jones R et al. Prognostic and predictive factors for patients with metastatic breast cancer undergoing aggressive induction therapy followed by highdose chemotherapy with autologous stem-cell support. J Clin Oncol 1999; 17: 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. JAMA 1999; 282: Ayash LJ, Wheeler C, Fairclough D et al. Prognostic factors for prolonged progression-free survival with high-dose chemotherapy with autologous stem-cell support for advanced breast cancer. J Clin Oncol 1995; 13: Williams SF, Gilewski T, Mick R et al. High-dose consolidation therapy with autologous stem-cell rescue in stage IV breast cancer: follow-up report. J Clin Oncol 1992; 10: Dunphy FR, Spitzer G, Fornoff JE et al. Factors predicting long-term survival for metastatic breast cancer patients treated with high-dose chemotherapy and bone marrow support. Cancer 1994; 73: Greenberg PA, 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: Tomiak E, Piccart M, Mignolet F et al. Characterisation of complete responders to combination chemotherapy for advanced breast cancer: a retrospective EORTC breast group study. Eur J Cancer 1996; 32A: Winer EP, Morrow M, Osborne KC, Harris JR. Malignant tumors of the breast. In: DeVita VT Jr, Hellman S, Rosenberg SA (eds). Cancer: Principles and Practice of Oncology. Lippincott Williams & Wilkins: Philadelphia, 2001, pp Hortobagyi GN. Treatment of breast cancer. New Engl J Med 1998; 339: Swenerton KD, Legha SS, Smith T et al. Prognostic factors in metastatic breast cancer treated with combination chemotherapy. Cancer Res 1979; 39: Miller AB, Hoogstraten B, Staquet M et al. Reporting results of cancer treatment. Cancer 1981; 47: Berruti A, Zola P, Buniva T et al. Prognostic factors in metastatic breast cancer patients obtaining objective response or disease stabilization after first-line chemotherapy with epirubicin. Evidence for a positive effect of maintenance hormonal therapy on overall survival. Anticancer Res 1997; 17: Kloke O, Klaassen U, Oberhoff C et al. Maintenance treatment with medroxyprogesterone acetate in patients with advanced breast cancer responding to chemotherapy: results of a randomized trial. Essen Breast Cancer Study Group. Breast Cancer Res Treat 1999; 55: Hortobagyi GN. Progress in the endocrine therapy of breast cancer. Cancer 1998; 83: Thürlimann B, Paridaens R, Serin D et al. Third-line hormonal treatment with exemestane in postmenopausal patients with advanced breast cancer progressing on aminoglutethimide: a phase II multicentre multinational study. Exemestane Study Group. Eur J Cancer 1997; 33: Dombernowsky P, Smith I, Falkson G et al. Letrozole, a new oral aromatase inhibitor for advanced breast cancer: doubleblind randomized trial showing a dose effect and improved efficacy and tolerability compared with megestrol acetate. J Clin Oncol 1998; 16: Goss PE, Strasser K. Aromatase inhibitors in the treatment and prevention of breast cancer. J Clin Oncol 2001; 19: Buzdar A, Douma J, Davidson N et al. Phase III, multicenter, double-blind, randomized study of letrozole, an aromatase inhibitor, for advanced breast cancer versus megestrol acetate. J Clin Oncol 2001; 19: Nabholtz JM, Buzdar A, Pollak M et al. Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: results of a North American multicenter randomized trial. J Clin Oncol 2000; 18: Bearman S, Vredenburgh J, Cagnoni P et al. High-dose therapy with autologous hematopoietic cell support as salvage treatment for patients with breast cancer who have relapsed after previous high-dose chemotherapy. Bone Marrow Transplant 1999; 24: