Yung-Chang Lin 1, Shin-Cheh Chen 2, Hsien-Kun Chang 1, Swei Hsueh 3, Chien-Sheng Tsai 4, Yung-Feng Lo 2, Tsann-Long Hwang 2 and Miin-Fu Chen 2

Transcription

1 Identifying Good Prognosis Group of Breast Cancer Patients with 1 3 Positive Axillary Nodes for Adjuvant Cyclophosphamide, Methotrexate and 5-Fluorouracil (CMF) Chemotherapy Yung-Chang Lin 1, Shin-Cheh Chen 2, Hsien-Kun Chang 1, Swei Hsueh 3, Chien-Sheng Tsai 4, Yung-Feng Lo 2, Tsann-Long Hwang 2 and Miin-Fu Chen 2 1 Division of Hematology/Oncology, Department of Internal Medicine, 2 Department of Surgery, 3 Department of Pathology and 4 Department of Radiation Oncology, Breast Cancer Study Group, Cancer Center, Chang Gung Memorial Hospital, Taipei, Taiwan Received December 15, 2004; accepted June 30, 2005 Jpn J Clin Oncol 2005;35(9) doi: /jjco/hyi143 Objective: We conducted a retrospective analysis of prognosis factors for survival in breast cancer patients with 1 3 axillary lymph node metastases and tried to identify a subset of patients with good prognosis suitable for cyclophosphamide, methotrexate and 5-fluorouracil (CMF) adjuvant chemotherapy. Methods: A cohort of 446 breast cancer patients received definite surgery and adjuvant chemotherapy with CMF at Chang Gung Memorial Hospital from 1990 to They were enrolled in the study. The median follow-up time was 69 months. Prognostic factors including age, tumor size, number of involved nodes, steroid receptor status, tumor ploidy, synthetic-phase fraction, histologic grade and administration of tamoxifen were analysed for disease-free survival (DFS) and overall survival (OS) by Cox regression model. Results: The estimated 5 year OS and DFS for all patients were 85.4 and 71.5%, respectively. Multivariate analysis revealed that tumor size, age and estrogen receptor (ER) status were independent prognostic factors for OS, and tumor size, age, ER status and number of involved nodes were independent prognostic factors for DFS. The 5 year OS rates of the low-risk group (age >40, tumor <3 cm and positive ER) and average-risk group (either age <40, tumor >3 cm or negative ER) were 98.8 and 82.4%, respectively (P = ). The 5 year DFS of the low-risk and high-risk group were 88.2 and 67.7%, respectively (P = ). Conclusion: Among breast cancer patients with 1 3 positive lymph nodes excellent survival rate was found in those who had favorable prognostic factors, including age >40, tumor size <3 cm and positive ER. Adjuvant chemotherapy with CMF regimen is optimal for these low-risk patients. Key words: breast cancer lymph node metastasis adjuvant chemotherapy prognostic factors INTRODUCTION Breast cancer is the second most common cancer, both in the West as well as in Taiwan (1,2). Adjuvant chemotherapy for breast cancer has been established for more than two decades following the initial work by the Milan Cancer Institute and the National Surgical Adjuvant Breast and Bowel Project (3,4). Despite the existence of extensive evidence from controlled clinical trials and meta-analysis to support the use of systemic chemotherapy for high-risk breast cancer patients, no standard regimen exists (5,6). In clinical practice, a breast cancer patient treated with adjuvant chemotherapy should balance the costs of For reprints and all correspondence: Shin-Cheh Chen, Department of surgery, Chang Gung Memorial Hospital, 199, Tun-Hwa North Road, Taipei 105, Taiwan. chensc@adm.cgmh.org.tw treatment in the form of short-term or long-term side effects, and weigh these against the benefits of treatment in reducing her risk of suffering recurrence of her cancer or death. Consequently, the fine-tuning of adjuvant chemotherapy based on prognostic factors, co-morbid illness and patient preference have become an important issue. The number of axillary lymph node metastases is the bestestablished prognostic factor for localized breast cancers (7). Accordingly, many adjuvant chemotherapy trials used the number of lymph nodes involved to separate patients into subgroups who were considered most likely to benefit from a given treatment regimen (5,6). In the past 10 years, we used a modified cyclophosphamide, methotrexate and 5-fluorouracil (CMF) regimen as adjuvant chemotherapy for patients with negative node status or with 1 3 positive axillary nodes. However, the use of methotrexate-based regimens still remains # 2005 Foundation for Promotion of Cancer Research

2 Jpn J Clin Oncol 2005;35(9) 515 controversial for patients with node-positive breast cancers, since a number of reports have considered CMF-like regimens to be suboptimal for such patients (8 13). We reported that a group of low-risk patients with 1 3 positive nodes have excellent survival rates after a CMF regimen based on a retrospective analysis of known prognostic factors for this patient population (14). However, our report was criticized at that time for the relatively short median follow-up period (47 months). In order to confirm our previous observation that certain subsets of patients with favorable prognostic factors have good survivals, we reanalysed the prognosis of patients with involvement of 1 3 nodes from the breast cancer data bank of our institution. PATIENTS AND METHODS From 1990 to 1998, a total of 2648 primary breast cancer patients who received breast cancer surgery were registered in the breast cancer database of Chang Gung Memorial Hospital (CGMH). This database is maintained and supervised by our Breast Cancer Study Group in the cancer center and was approved by the institution review board of CGMH. Patient s demographics, tumor characteristics, type of surgery, adjuvant chemotherapy, radiotherapy, hormonal therapy, disease status and survival outcomes were all prospectively collected. Based on the treatment guidelines established by our Breast Cancer Study Group for early breast cancer, patients were classified into four groups according to the number of axillary lymph node metastases (0, 1 3, 4 10, >10). The present analysis examines patients who had 1 3 axillary lymph nodes metastases. This subset of patients received primary breast surgery followed by adjuvant chemotherapy. The chemotherapy regimens consisted of cyclophosphamide 600 mg/m 2, methotrexate 40 mg/m 2 and 5-fluorouracil 600 mg/m 2 repeated every 3 weeks for nine cycles. Post-operative radiotherapy was performed only when the patient received partial mastectomy or the tumor size exceeded 5 cm after modified radical mastectomy. In patients whose tumors were positive for hormonal receptors, tamoxifen was given for 5 years. Medical information was obtained by periodic chart review and telephone interviews. The pathological features were reviewed by an experienced pathologist (S.H.). Mitotic index including DNA ploidy and synthetic-phase fraction (SPF) were evaluated by flow cytometry as described elsewhere (15). For this study, the last follow-up date was 30 June Disease-free survival (DFS) was calculated from the date of primary surgery to the date of documented recurrence. Patients who died without evidence of recurrence or were lost to follow-up were censored at the date of death or last visit. The overall survival (OS) was calculated from the date of primary surgery to the date of death. The dates of death were confirmed by the National Vital Statistical Data Base provided by the Taiwanese Department of Health. Prognostic factors including age (<40 versus >40), tumor size, steroid receptor status, tumor ploidy, SPF, Scarff Bloom-Richardson (SBR) grading in ductal carcinoma and administration of tamoxifen were analysed by a Cox regression model for DFS and OS. OS and DFS were estimated using the Kaplan Meier method, and the difference in survival rates was determined using the Log-Rank test. All statistics were calculated using SPSS 10.0 software (Chicago, IL, USA). RESULTS SURVIVAL OF ALL PATIENTS Among the 2648 primary breast cancer patients, 446 patients with 1 3 axillary lymph nodes metastases and who had received adjuvant chemotherapy with CMF were retrieved from the database. The median follow-up period at the time of analysis was 69 months. The 5 year OS and DFS were 85.4 and 71.5%, respectively. To date, 134 patients had disease recurrence and 95 patients expired. While six patients (1.3%) died of causes other than breast cancer, all were disease-free at their death. Table 1 summarizes the characteristics of this patient group. FACTORS FOR PREDICTING PROGNOSIS FOR PATIENTS RECEIVING ADJUVANT CMF Further subset analysis was performed and Tables 2 and 3 summarized the univariate analysis to identify prognostic factors affecting OS and DFS. In conclusion, age <40 years, tumor size >3 cm, more than one axillary lymph node involvement, negative estrogen receptor (ER) status and not using tamoxifen predicted a poor prognosis for DFS. The estimated 5 year DFS for age <40 years and >40 were 59.9 and 76.5%, respectively (P = 0.001); for tumor size <3 and >3 cm were 75.5 and 59.6%, respectively (P = 0.005); for one axillary nodes involvement and more than one were 78.5 and 65.6%, respectively (P = 0.004); for negative ER and positive ER were 68.7 and 80.4%, respectively (P = 0.006); and for tamoxifen user and non-user were 80.7 and 67.7%, respectively (P = 0.016). High SBR grade was marginally significant for poor prognosis (P = 0.05). While age >40 years, tumor size >3 cm and negative ER predicted poor prognosis for OS. The estimated 5 year OS for age <40 years and >40 were 79.6 and 87.9%, respectively (P = 0.002); for tumor size <3 and >3 cm were 89 and 74.1%, respectively (P = 0.002); for negative ER and positive ER were 79.5 and 94.9%, respectively (P = 0.001). Multivariate analysis revealed that age, tumor size, number of lymph node metastases (single versus multiple) and ER status were four independent prognostic factors for DFS, while only age, tumor size and ER status were independent prognostic factors for OS (Table 4). The relative risks of OS for tumor size >3 cm, age <40 and negative ER were 2.06, 1.97 and 2.56, respectively. IDENTIFICATION OF A SUBSET OF PATIENTS WITH FAVORABLE PROGNOSIS We stratified the 446 patients into two distinct groups according to age, primary tumor size and ER as prognostic

3 516 Prognosis of breast cancer patients Table 1. Patients demographics Characteristics Number Median age (range) 44 (24 77) Histologic type Ductal carcinoma 392 (87.9%) Lobular carcinoma 21 (4.7%) Medulary carcinoma 18 (4%) Tubular carcinoma 7 (1.6%) Mucinous carcinoma 3 (0.7%) Others 5 (1.1%) Tumor stage T1 155 (34.8%) T2 283 (63.4%) T3 8 (1.8%) Number of positive axillary node (45.3%) (32.1%) (22.6%) ER status Positive 145 (32.5%) Negative 243 (54.5%) Unknown 58 (13%) Surgery Modified radical mastectomy 386 (86.5%) Partial mastectomy 60 (13.5%) factors. Patients with age >40, positive ER status and tumor size <3 cm were categorized as a low-risk group (82 patients), while the remainder (including those with any of the three unfavorable prognostic factors) were deemed an averagerisk group (364 patients). The estimated 5 year OS rates for the low-risk and average-risk groups were 98.8 and 82.4%, respectively (P = 0.001; Fig. 1). The estimated 5 year DFS rates were 88.2 and 67.7%, respectively (P = ; Fig. 2). Table 2. Univariate analysis of 5 year overall survival Factors Events/patient number OS rate (%) P-value Age (years) <40 40/ >40 51/ Primary tumor size (mean) <3 cm 57/ >3 cm 33/ Number of positive axillary node 1 34/ or 3 57/ SBR grade Low 16/ High 15/ Unknown 60/ ER Positive 15/ Negative 60/ Unknown 16/ Ploidy Diploid 32/ Non-diploid 27/ Unknown 32/ SPF <4% 17/ >4% 50/ Unknown 24/ Tamoxifen Yes 19/ No 72/ Radiotherapy Yes 9/ No 82/ DISCUSSION This study confirms our previous conclusion that age, ER and tumor size are independent prognostic factors for OS and DFS in breast cancer patients with 1 3 positive axillary lymph nodes after adjuvant chemotherapy with a modified CMF regimen. The significance of the prognostic factors remains unchanged with the extent of follow-up period to almost 6 years. We also identified a subset of good prognosis patients who were above 40 years old, had a primary tumor size of no more than 3 cm and whose tumors were positive for ERs. The 5 year estimated DFS and OS for this group were 88.2 and 98.8%, respectively. In contrast, for patients with any unfavorable prognostic factor the estimated 5 year DFS and OS were 67.7 and 82.4%, respectively. The survival rates of the low-risk patients with 1 3 positive nodes were equal to highrisk node-negative patients reported (9). Here, we propose that patients with 1 3 lymph node metastases and favorable prognostic factors could be treated as patients of high-risk node-negative breast cancer as suggested by the 2001 St Gallen International Consensus Conference (16). For patients with 1 3 nodes in our high-risk group, their outcomes were poorer and anthracycline-based regimens with or without taxanes are recommended (17). The superiority of the anthracycline-based regimen over CMF was supported by the overview of the Early Breast Cancer Trials Cooperative Group in 1998 and by randomized controlled studies (8 13) while non-anthracycline based

4 Jpn J Clin Oncol 2005;35(9) 517 Table 3. Univariate analysis of disease-free survival Factors Events/patient number DFS (%) P-value Age (years) <40 54/ >40 80/ Primary tumor size (mean) <3 cm 88/ >3 cm 43/ Number of positive axillary node 1 47/ or 3 87/ SBR grade Low 30/ High 25/ Unknown 79/ ER Positive 29/ Negative 80/ Unknown 25/ Ploidy Diploid 41/ Non-diploid 45/ Unknown 48/ SPF <4% 26/ >4% 70/ Unknown 38/ Tamoxifen Yes 28/ No 106/ Radiotherapy Yes 13/ No 121/ Table 4. Multivariate analysis of 5 year OS and DFS Features OS DFS RR (95% CI) P-value RR (95% CI) P-value Size (>3 cm) 2.06 ( ) ( ) Age (<40 years) ( ) ( ) ER (negative) ( ) ( ) Number of nodes (>2) NS 1.62 ( ) RR (95% CI), relative risk (95% confidence interval) regimens such as CMF-type regimens remain useful particularly for patients with negative nodes as recommended by 2001 St Gallen International Consensus Conference (16) and The National Institute of Health 2000 Consensus Conference low risk (n=82) average risk (n=364) 0.1 Log Rank p= Years after operation No. at risk low risk# average risk* Cumulative survival rate low risk# : ER(+), size 3 and age >40 average risk* : any factor of ER (-), size >3cm or age 40 Figure 1. Overall survival of low-risk and average-risk patients with 1 3 positive axillary nodes received adjuvant CMF chemotherapy. Cumulative survival rate low risk (n=82) average risk (n=364) 0.1 Log rank p= Years after operation No. at risk low risk# average risk* low risk# : ER(+), size 3 and age >40 average risk* : any factor of ER ( ), size >3cm or age 40 Figure 2. Disease-free survival of low-risk and average-risk patients with 1 3 positive axillary nodes received adjuvant CMF chemotherapy. on Adjuvant Therapy for Breast Cancer (18). There were other advantages of a CMF-like regimen over an anthracyclinebased regimen in terms of secondary leukemia and heart disease (19 21). In present clinical practice, CMF-like regimens are commonly used in node-negative patients. Regarding patients with 1 3 positive nodes, the Milan Cancer Institute conducted a study in which patients were

5 518 Prognosis of breast cancer patients randomized to receive 12 cycles of CMF or 8 cycles of CMF followed by 4 cycles of doxorubicin (22). The CMF regimen was identical to that used in this study except that it included three additional treatment cycles. The addition of doxorubicin failed to improve relapse-free and OS compared to CMF alone after a median follow-up of 61 months. The overall 5 year relapse-free survival rate and OS rates were 74 and 89%, respectively, comparable to the survival rates in this study. The French Adjuvant Study Group conducted a randomized study to compare low-dose (50 mg/m 2 ) epirubicin with highdose (100 mg/m 2 ) eiprubicin in patients with more than three positive nodes or 1 3 nodes with SBR grade >2 or ER negativity. This study confirmed that the survival for patients with 1 3 lymph node involvement with low-dose epirubicin was equal to those with high-dose (23). In a subset analysis, the DFS and OS for patients with 1 3 nodes were 74 and 81%, respectively. Their results were modestly superior to our results for patients in our high-risk group. It could be attributed to the superiority of an epirubicin-based regimen over methotrexate. However, there were only 98 patients, which might be too small for validation. The Canada National Cancer Institute randomized premenopausal node-positive patients with classical CMF and CEF (cyclophosphamide, epirubicin, 5-fluorouracil). In patients with 1 3 positive nodes with CMF arm, the 5 year DFS and OS were 62 and 78%, respectively (12). The relatively younger population and the inclusion of primary T3 tumors might explain why their survival rate was inferior to other studies. The National Institute of Health 2000 Consensus Conference on Adjuvant Therapy for Breast Cancer reviewed the prognostic factors such as age, race, tumor size, histologic type, pathologic grading, hormonal receptor status and mitotic index (18). We compared these factors and found that histologic grading and mitotic index (DNA ploidy and SPF) had no additional prognostic value for survivals in our series. Patients with specific tumor type associated with a favorable prognosis, such as tubular carcinoma, were too few to validate its statistical significance. Steroid receptors and tumor size are well-established prognostic factors for breast cancer (5 7,16,18,24,25). In a study with a 10 year median followup period, Quiet et al. (26) found that tumor size <2 cm was a favorable prognostic factor even for patients with fewer than four positive nodes, the survival was comparable to patients with negative nodes. Another retrospective study on 512 patients found that tumor size of 3 cm, not 2 cm, was the cut-off value for predicting prognosis in patients with 1 3 lymph node metastases (27). The Milan study also postulated that relapse-free survival was longer for tumor size <2 cm and with positive estrogen or progesterone receptors for patients with 1 3 positive lymph nodes (22). Our observation is in accordance with these findings. Age as a factor has been increasingly studied recently for breast cancer. The National Cancer Data Base statistics indicate that women younger than 35 years had higher tumor grade and more advanced disease at presentation, and worse 5 year survival than do older premenopausal women (28). The Surveillance, Epidemiology and End Results Program of the National Cancer Institute also postulated that breast cancer is more lethal to women in their 20s and 30s than to any other age group (29). The International Breast Cancer Study Group (IBCSG) analysed outcome for 314 breast cancer patients with age <35 and found that relapse and death occurred earlier and more often in the younger group. Furthermore, younger patients with ER positive tumors had a significantly worse DFS rate than younger patients with ER negative tumor (30). In Taiwan, Cheng et al. (31) postulated that early-onset breast cancer (age <40) comprised 29.3% of a sample of 811 patients. Multivariate analysis of primary operable breast cancer revealed that early-onset breast cancer (age <40) was a poor prognostic factor. The clinical behavior of early-onset breast cancer was more aggressive than it was in the older age group, and the 5 year DFS rate for stage II disease among early-onset patients was only 66.5%. The analysis in the Milan study of patients with 1 3 positive nodes did not consider age to be a factor in prognosis. However, postmenopausal women survived longer than pre-menopausal women (22). Hormonal factors are believed to significantly influence the progression of breast cancer, particular in younger patients. The IBCSG thus suggested additional hormonal therapy to improve the survival of younger patients (30). The phenomenon discussed above may explain why younger estrogen positive patients (under 40 years old) have worse survival rates than older estrogen positive patients. The limitations of our study include the use of less than optimal CMF regimens (5), which may affect survival for patients in the high-risk group and exaggerate the differences in outcomes observed between low-risk and average-risk groups. The routine use of tamoxifen in ER positive patients may further increase the differences among groups, since tailored treatment with tamoxifen improved outcomes in premenopausal patients with positive ER (4). However, it is worth considering that the low-risk group of patients may represent a subset of patients whose disease follows a relatively indolent course. For these patients, the type of chemotherapy regimen may have little relationship with better survival rates. Given the cost of the benefits and the fine-tuning of adjuvant chemotherapy, this group of patients may benefit from less aggressive therapy, including a less aggressive chemotherapy regimen and ovarian ablation therapy for premenopausal women (6). Additionally, the retrospective nature of this study limits its clinical significance, but it may provide a rationale for subsequent trials applying less aggressive therapy to the low-risk group. Finally, this study did not review certain prognostic markers, such as erbb-2. A number of studies have suggested that erbb-2 over-expression predicts that CMF-type regimens and low-dose anthracycline will be less effective (32). In conclusion, retrospective analysis of breast cancer patients with 1 3 positive axillary nodes in a single institution revealed that the prognostic factors for worse OS were age <40, tumor size >3 cm and negative ER status. This study also identified a subset of patients with age >40, primary tumor size

6 Jpn J Clin Oncol 2005;35(9) 519 <3 cm and positive ER who had a favorable outcome after adjuvant CMF chemotherapy. References 1. Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics. CA Cancer J Clin 2001;51: Cancer Registry Annular Report (1998). Republic of China: Department of Health, The Executive Yuan 2002; Fisher B, Carbone P, Economou SG, Frelick R, Glass A, Lerner H, et al. L-phenylalanine mustard (L-PAM) in the management of primary breast cancer. A report of early findings. N Engl J Med 1975; 292: Bonadonna G, Brusamolino E, Valagussa P, Rossi A, Brugnatelli L, Brambilla C, et al. Combination chemotherapy as an adjuvant treatment in operable breast cancer. N Engl J Med 1976;294: Gianni L, Valagussa P, Zambetti M, Moliterni A, Capri G, Bonadonna G. Adjuvant and neoadjuvant treatment of breast cancer. Semin Oncol 2001;28: Hortobagyi GN. Adjuvant systemic therapy for early breast cancer: progress and controversies. Clin Cancer Res 2001;7: Saze RA, McGuire WL, Clark GM. Prognostic factors in breast cancer. Semin Surg Oncol 1989;5: Early Breast Cancer Trials Cooperative Group: Polychemotherapy for early breast cancer: an overview of the randomized trials. Lancet 1998;352: Hunchins L, Green S, Ravidin R, Lew D, Martino S, Abelogg M, et al. CMF versus CAF with and without Tamoxifen in high-risk node negative breast cancer patients and a natural history follow-up study in low-risk node negative patients: First Result of Intergroup Trials INT Proc Am Soc Clin Oncol 1998;17:1a (Abstract 2). 10. Piccart MF, Di Leo A, Beauduin M, Vindevoghel A, Michel J, Focan C, et al. Phase III trial comparing two dose levels of epirubicin combined with cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer. J Clin Oncol 2001;19: Fisher B, Anderson S, Tan-Chiu E, Wolmark N, Wickerham DL, Fisher ER, et al. Chemotherapy with or without tamoxifen for patients with ER-negative breast cancer and negative nodes: results from NSABP B23. J Clin Oncol 2001;19: Levine MN, Bramwell VH, Pritchard KI, Norris BD, Shepherd LE, Abu-Zahra H, et al. Randomized trial of intensive cyclophosphamide, epirubicin, and fluorouracil chemotherapy compared with cyclophosphamide, methotrexate, and fluorouracil in premenopausal women with node-positive breast cancer. J Clin Oncol 1998;16: Mouridsen HT, Andersen J, Anderson M, Domobernowsky P, Ejlertsen B, Rose C, et al. Adjuvant anthracycline in breast cancer: Improved outcome in premenopausal patients following substitution of methotrexate in the CMF combination with epirubicin. Proc Am Soc Clin Oncol 1999;18: 68a (Abstract 254). 14. Lin YC, Chen SC, Chang HK, Hsueh S, Hwang TL. Prognostic factors of breast cancer patients with 1 to 3 lymph node metastases after adjuvant chemotherapy. Euro J Cancer 2001;37(Suppl 6): 183 (abstract 667). 15. Chen SC, Chen MF, Hwang TL, Chao TC, Lo YF, Hsueh S, et al. Prediction of supraclavicular lymph node metastasis in breast carcinoma. Int J Radiat Oncol Biol Phys 2002;52: Goldhirsch A, Glick JH, Geller RD, Costes AS, Senn HJ. Meeting highlight: International Consensus Panels on the Treatment of Primary Breast Cancer Seventh International Conference on Adjuvant Chemotherapy of Primary Breast Cancer. J Clin Oncol 2001;19: Tan AR, Swain SM. Adjuvant chemotherapy for breast cancer: an update. Semin Oncol 2001;28: National Institute of Health Consensus Development Conference. J Natl Cancer Inst Monogr 2001;30: Chembers SK, Chopky RL, Chambers JT, Schwartz PE, Duffy TP. Development of leukemia after doxorubicin and cisplatin treatment for ovarian cancer. Cancer 1989;64: von Hoff DD, Rozencweig M, Piccart MJ. The cardiotoxicity of anticancer agents. Semin Oncol 1982;9: Zambetti M, Moliterni A, Materazzo C, Stefanelli M, Cipriani S, Valagussa P, et al. Long-term cardiac sequelae in operable breast cancer patients given adjuvant chemotherapy with or without doxorubicin and breast irradiation. J Clin Oncol 2001;19: Moliterni A, Bonadonna G, Valagussa P, Ferrari L, Zambetti M. Cyclophosphamide, methotrexate, and fluorouracil with and without doxorubicin in the adjuvant treatment of respectable breast cancer with one to three positive axillary nodes. J Clin Oncol 1991;9: French Adjuvant Study Group. Benefit of a high-dose eiprubicin regimen in adjuvant chemotherapy for node-positive breast cancer patients with poor prognosis factors: 5-year follow-up results of French Adjuvant study Group 05 randomized trial. J Clin Oncol 2001;19: Osborne CK, Yochmowitz MG, Knight WA 3rd, McGuire WL. The value of estrogen and progesterone receptors in the treatment of breast cancer. Cancer 1980;46: Fisher B, Slack NH, Bross IDJ. Cancer of the breast: size of neoplasm and prognosis. Cancer 1969;24: Quiet CA, Ferguson DJ, Weichselbaum RR, Hellman S. Natural history of node-positive breast cancer: the curability of small cancers with a limited number of positive nodes. J Clin Oncol 1996;14: Rakowsky E, Klein B, Kahan E, Derazne E, Lurie H. Prognostic factors in node-positive operable breast cancer patients receiving adjuvant chemotherapy. Breast Cancer Res Treat 1992;21: Winchester DP, Osteen RT, Menck HR. The National Cancer Data Base reports on breast carcinoma characteristics and outcome in relation to age. Cancer 1996;78: Swanson GM, Lin C-S. Survival patterns among younger women with breast cancer: the effects of age, race, stage, and treatment. J Natl Cancer Inst Monogr 1994;16: Aebi S, Gelber S, Castiglione-Gertsch M, Gelber RD, Collins J, Thurlimann B, et al. Is chemotherapy alone adequate for young women with estrogen-receptor-positive breast cancer? Lancet 2000; 355: Cheng SH, Tsou MH, Liu MC, Jian JJ, Cheng JC, Leu SY, et al. Unique features of breast cancer in Taiwan. Breast Cancer Res Treat 2000; 63: Gusterson BA, Gelber RD, Goldhirsch A, Price KN, Save-Soderborgh J, Anbazhagan R, et al. Prognostic importance of c-erb-2 expression in breast cancer: International (Ludwig) Breast Cancer Study Group. J Clin Oncol 1992;7: