Optimizing Adjuvant Breast Cancer Chemotherapy: Rationale for the MA.21 Study

ABSTRACT: Recently initiated is a phase III randomized trial (MA.21 trial)of adjuvant chemotherapy for node-positive and high-risk node-negative,premenopausal and postmenopausal (£ 60 years) women with breast cancer who haveno distant metastases. Conducted by the National Cancer Institute of Canada-ClinicalTrials Group, the trial will compare two standard therapies, CEF(cyclophosphamide [Cytoxan, Neosar], epirubicin [Ellence], fluorouracil) andAC®T (doxorubicin [Adriamycin], cyclophosphamide, followed by paclitaxel[Taxol]), and includes a third arm consisting of a dose-dense, dose-intenseEC®T regimen (epirubicin, cyclophosphamide, followed by paclitaxel). Theseregimens were chosen for study based on results of previous clinical assessmentsof adjuvant therapies, which, taken together, suggest that CEF, FEC 100 (where100 represents the dose in mg/m² of epirubicin in FEC [fluorouracil,epirubicin, cyclophosphamide]), CAF (cyclophosphamide, doxorubicin,fluorouracil), and AC®T may all be superior to standard AC or CMF(cyclophosphamide, methotrexate, fluorouracil) regimens. This article reviewstrial results that support the testing of the regimens chosen for the MA.21trial. The intent of the MA.21 study is to advance our ability to provideoptimal adjuvant therapy for patients with breast cancer. [ONCOLOGY 15(Suppl 7):7-13,2001]

Introduction

Cooperative groups and researchgroups on national and multinational levels have initiated new and innovative randomized clinical trialsdesigned to advance adjuvant chemotherapy for breast cancer. Following theirsuccess in the metastatic setting, new classes of agents such as the taxanes andtrastuzumab (Herceptin) have become available to study in the adjuvant setting.This has resulted in a large number of ongoing studies of combination orsequential therapies, particularly focused on the anthracyclines and thetaxanes, and the incorporation of trastuzumab for patients whose tumorsoverexpress HER2/neu.

In Canada, the National Cancer Institute of Canada—ClinicalTrials Group (NCIC-CTG) will soon initiate a phase III randomized clinicaltrial, MA.21 (the group’s 21st breast study). MA.21 is based on results ofprevious clinical trials, some conducted through the NCIC-CTG and others throughother cooperative group mechanisms, and builds on results of the meta-analysesof the Early Breast Cancer Trialists Collaborative Group (EBCTCG) as well.[1]The MA.21 study compares two standard therapies: CEF (cyclophosphamide [Cytoxan,Neosar], epirubicin [Ellence], fluorouracil[5-FU]) and AC®T (doxorubicin [Adriamycin] plus cyclophosphamide followed bypaclitaxel [Taxol]). In addition to comparing these two standard therapies, thestudy tests whether it is necessary to add a taxane to an optimal anthracyclineregimen for adjuvant therapy, ie, epirubicin and cyclophosphamide (EC). Thestudy therefore includes a third arm consisting of a dose-dense, dose-intenseregimen of EC®T (epirubicin and cyclophosphamide followed by paclitaxel).

Background and Rationale

Reduction in Risk of Recurrence and Mortality

In its most recent publications, the EBCTCG summarized theproportional reductions in the risk of recurrence and death associated withovarian ablation,[1] tamoxifen (Nolvadex),[2] and chemotherapy.[3] Ovarianablation in premenopausal women, ie, less than 50 years old, results in a 24%reduction in the proportional risk of deathat 10 years compared to no treatment. Tamoxifen treatment reduces the riskcompared to no treatment, by 20% to 22% in both premenopausal and postmenopausalwomen, and CMF (cyclophosphamide, methotrexate, 5-FU) chemotherapy results in a14% reduction in risk for all ages. These findings translate into an approximate10% absolute reduction in mortality at 10 years for node-positive patients and a5% absolute reduction in mortality for node-negative patients.

The most recent meta-analysis[3] demonstrated thatanthracycline-based adjuvant regimens are modestly more effective than non-anthracycline-containingregimens, with an 11% proportional reduction in mortality risk, or anapproximate 3% absolute reduction in mortality risk. Furthermore, thecombination of tamoxifen and chemotherapy is superior to either modality alonein hormone-receptor-positive patients, while tamoxifen is of no benefit inreceptor-negative patients with respect to the primary breast cancer.

Recommended Regimens

Various guidelines for adjuvant therapy have been developed,including the 2000 National Comprehensive Cancer Network (NCCN) guidelines.[4-6]These guidelines recommend adjuvant chemotherapy for hormone-receptor-negativepatients regardless of nodal status. For patients with hormone-receptor-positivetumors, chemotherapy ± tamoxifen is recommended for high-risk node-negative ornode-positive patients. Those whose tumors measure 1 to 2 cm and who are node-negative may be offered tamoxifen ± chemotherapy. For patients with tumors lessthan 1 cm and no poor prognostic factors, tamoxifen or no therapy can berecommended.

Potential regimens for node-negative patients are: CMF; FAC(5-FU, doxorubicin [Adriamycin], cyclophosphamide)/CAF (cyclophosphamide,doxorubicin, 5-FU); or AC. In the node-positive setting,anthracycline-containing regimens are preferred, although CMF remains anacceptable treatment option. FAC/CAF or CEF may be recommended. AC followed bypaclitaxel is another option, as is doxorubicin followed by CMF, which has beenshown to be superior to doxorubicin alternating with CMF in women with four ormore positive lymph nodes.[7]

Doxorubicin-based regimens have been compared to CMF-basedregimens in several studies. The National Surgical Adjuvant Breast and BowelProject (NSABP) B-15[8] and B-23[9] studies have demonstrated the equivalence offour cycles of AC vs six cycles of CMF. While in one study by the SoutheastCancer Study Group, an FAC regimen was shown to be equivalent to CMF,[10] a morerecent intergroup study involving node-negative patients showed superiority of aCAF regimen over CMF.[11]

A Cancer and Leukemia Group B (CALGB)/Intergroup studydemonstrated that AC followed by paclitaxel[12] was superior to AC alone. A morerecent study at M. D. Anderson Cancer Center did not show that paclitaxel × 4followed by FAC × 4 was superior to FAC × 8, although the results are stillearly.[13]

Epirubicin-containing regimens have also been compared to CMF.The NCIC-CTG study of CEF vs CMF[14] showed superiority for the CEF regimen. TheInternational Cancer Collaborative Group showed that FEC 50 was equivalent toCMF,[15] where 50 represents the epirubicin dose (mg/m²) in the FEC regimen. TheFrench Adjuvant Study Group, however, showed that FEC 100 is superior to FEC50.[16] Hence, direct or indirect comparisons would suggest that CEF, FEC 100,CAF, and AC®paclitaxel may all be superior to standard AC or CMF regimens.

Choosing the Anthracycline

The anthracycline chosen by the various collaborative groups hasbeen based on both drug availability and associated toxicities. Epirubicin ismetabolized not only into 4-aglycone and epirubicinol—similar to what occursin doxorubicin metabolism—but also is glucuronidated as epirubicin glucuronideand epirubicinol glucuronide. This results in a shorter half-life and fewerhematologic, nonhematologic, and cardiac toxicities than those associated withdoxorubicin; however, no decrease in efficacy is seen.[17] The probability ofdeveloping congestive heart failure with epirubicin treatment is less than 5% ata total dose of 950 mg/m², while a comparable level of risk is reached at atotal doxorubicin dose of 500 mg/m².[18,19]

In 1998, Cancer Care Ontario (Canada) published a practiceguideline[20] that compared doxorubicin and epirubicin in women with metastaticbreast cancer. The guideline reviewed 13 randomized controlled trials, 11published reports and 2 abstracts. The guideline concluded that epirubicin wasas effective as doxorubicin in the metastatic setting but caused less toxicity.

Three of these randomized controlled trials compared FEC 50 toFAC 50, where 50 represents the dose (mg/m²) of the anthracycline in therespective regimen.[21-23] Response rates were not significantly differentwithin any of the trials and median survival was not affected by whetherdoxorubicin or epirubicin was used. However, the FEC regimens resulted in lesscardiac toxicity as determined by ECG alterations or percent of patients withcongestive heart failure. Furthermore, less grade 3/4 leukopenia, anemia, andnausea/vomiting occurred with the epirubicin-based regimens.

The guideline also examined randomized trials comparingescalating doses of epirubicin in metastatic breast cancer.[24-26] These studiesdemonstrated an increase in response rate with increasing doses of epirubicin atleast to 100 mg/m². The increased response rates did not, however, translateinto increased overall survival in the metastatic setting. Increased responserates with escalating doses of doxorubicin have also been demonstrated, butthese studies were limited by hand-foot syndrome and cardiac toxicity.

Results of Clinical Trial Comparing CEF vs CMF

In 1987, a pilot study was conducted by the Ontario ClinicalOncology Group and formed the basis for the NCIC-CTG clinical trial of CEF vsCMF (MA.5).[27] The pilot regimen was developed based on the concept that, atthat time, anthracyclines were the most active chemotherapeutic agents foradvanced breast cancer and the expectation that dose-intensive chemotherapywould improve outcome. The study was initiated prior to the routine use of 5-HT₃antagonists as antiemetics and before recognition of the advantages ofepirubicin over doxorubicin with respect to both emesis and lack of congestiveheart failure.

At the optimal dose level chosen in the pilot study, the rate offebrile neutropenia in women less than 50 years old was 17%. Several patientswere accrued at these dosages; the subsequent use of concurrent antibioticprophylaxis resulted in a reduction in hospital admissions for febrileneutropenia to 5% of patients.

The Drug Regimens

In the MA.5 trial, the CEF regimen consisted of oralcyclophosphamide at 75 mg/m² daily on days 1 through 14, and intravenous (IV)epirubicin at60 mg/m² and IV fluorouracil at 500 mg/m², both given on days 1 and 8(Figure 1). Each cycle was repeated monthly for 6 months.

Based on the pilot study results with the use of antibiotics toreduce febrile neutropenia, in the MA.5 protocol, all patients assigned to theCEF regimen also received prophylactic antibiotics throughout the trial withtrimethoprim-sulfamethoxazole (Co-Trimoxazole, TMP-SMZ), norfloxacin (Noroxin),or ciprofloxacin (Cipro).

Standard CMF with oral cyclophosphamide was administered in theother study arm. Patients were stratified with respect to type of surgeryperformed (total or partial mastectomy), hormone receptor status, and number ofpositive lymph nodes.

Differences in Survival andAdverse Events

A total of 710 premenopausal node-positive patients wereregistered in the study—351 in the CEF arm and 359 in the CMF arm. In the twoarms, respectively, 36% and 39% of patients had T1 tumors; 55% and 49% had T2tumors. Among all patients, 61% had one to three positive nodes and 59% wereestrogen-receptor positive.

CEF resulted in a 29% reduction in risk of recurrence and a 19%reduction in risk of mortality compared with CMF. At 5 years, overall survivalwas 77% in the CEF arm and 70% in the CMF arm, a statistically significantdifference (Figure 2).

During treatment, grade 3/4 adverse events included 8.5% ofpatients with fever or infection in the CEF group and 1.1% in the CMF group. Asexpected, alopecia was significantly more frequent in the CEF arm. Rates ofstomatitis were 12.3% for CEF and 1.9% for CMF, and vomiting occurred in 11.4%of CEF and 4.1% of CMF patients. Of note, no 5-HT3 antagonists orcolony-stimulating factors were administered.

Quality of life was studied using standard European Organizationfor the Research and Treatment of Cancer (EORTC) questionnaires. Except for thehigher frequency of alopecia with CEF in cycle 1, no significant differences inquality of life indices were noted between the two treatments.

Late adverse events included congestive heart failure and acuteleukemias. In the CEF arm, four patients (1.1%) developed congestive heartfailure while one (0.3%) in the CMF arm did so. Five cases of acute leukemiaoccurred on the CEF arm—four acute myelocytic leukemias (AML), and one acutelymphoblastic leukemia (ALL). One patient in the CMF group developed AML. Casesof leukemia were reported between 18 and 40 months after study enrollment. Nocases of leukemia have been reported between years 4 and 6. Of note, mortalitysecondary to leukemia is included in the overall survival statistics; thus,overall survival benefits with respect to breast cancer outweigh this risk.

CEF Becomes Standard Treatment in Canada

Based on results of the MA.5 trial, CEF has become standardtherapy in Canada for node-positive, premenopausal women with breast cancer. Infact, in clinical practice, high-risk, node-negative, and young postmenopausalwomen, ie, less than 60 years old, are also treated with CEF.

The AC®T vs AC Trial

The second "standard" treatment chosen for the MA.21study is based on results from the CALGB 9344 study,[12] which assessed AC doseintensity with and without the addition of paclitaxel. A total of 3,121premenopausal and postmenopausal node-positive women were randomly assigned toreceive one of three doses of doxorubicin (60, 75, or 90 mg/m²) pluscyclophosphamide at 600 mg/m².After four cycles of AC, patients received either no further therapy or fourcycles of paclitaxel at 175 mg/m² given over 3 hours. Patients with positivehormone-receptor status were offered 5 years of tamoxifen therapy (Figure3).

Addition of Paclitaxel Improves Survival Rates

The initial results of the CALGB 9344 study were presented atthe American Society of Clinical Oncology meeting in 1998.[12] At that time,analysis at 18 months showed no difference in disease-free or overall survivalrates related to doxorubicin dose. The addition of paclitaxel, however, improvedboth disease-free and overall survival. Rates of disease-free survival were 86%for AC and 90% for AC®T, while overall survival rates were 95% and 97%,respectively. The differences between treatments for both of these results werestatistically significant.

Three-year disease-free survival results have been updated,[28]and are 76.6% for patients who received paclitaxel vs 73% for those who did not—a difference of 3.6%. However, according to subgroup analysis with respect tohormone-receptor status, the benefit of therapy was limited to patients whosetumors were estrogen-and progesterone-receptor negative. Among these individuals, the difference in3-year survival between those who did and did not receive paclitaxel was 10.5%(67.3% vs 56.8%, respectively).

Disease-free survival rate for hormone-receptor-positivepatients was approximately 82% regardless of whether they received paclitaxel.However, this was a retrospective subgroup analysis, and results may change overtime as events occur in the more favorable receptor-positive population.

AC Followed by PaclitaxelIs Standard Therapy in US

Based on the results of this study, AC followed by paclitaxelhas become standard therapy for node-positive patients in the United States. AC®T is the second standard arm in the MA.21 study.

Only a head-to-head comparison of CEF and AC®T will determinewhether these regimens are comparable or one is superior to the other. Anotherquestion is whether the addition of paclitaxel to AC improves outcome based onthe taxane itself or because more cycles of chemotherapy are administered.

The EC vs CEF Trial

The third arm of the MA.21 study is based on the comparison ofCEF to a dose-intense, dose-dense regimen of epirubicin and cyclophosphamide(EC) (Figure 4). This regimen was studied in the neoadjuvant setting in patientswith locally advanced or inflammatory breast cancer. The study was undertaken bythe EORTC, NCIC-CTG, and the Swiss Group for Clinical Cancer Research (SAKK)(NCIC MA.10).[29]

In this trial, CEF was given as in the MA.5 trial. The ECregimen was assessed in a pilot study by theEORTC, and consisted of epirubicin at 120 mg/m² IV and cyclophosphamide at 830mg/m² IV day 1 every 2 weeks for six cycles. The granulocyte colony-stimulatingfactor (G-CSF [Neupogen]) filgrastim at 5 mg/kg was administered continuously ondays 2 through 13.

Following chemotherapy, local management could include surgery,radiotherapy, or a combination of both, in any order. Tamoxifen was alsoprescribed when appropriate. Prophylactic antibiotics were used in the CEF arm.

Eligible patients had locally advanced or inflammatory breastcancer. Any age was acceptable as long as performance status was adequate(Eastern Cooperative Oncology Group) [ECOG] 0 or 1). Patients with severecardiac disease were not eligible.

A total of 437 patients were enrolled in the study—217 in theCEF group and 220 in the EC plus G-CSF group. The median age was 49 years;approximately half the patients were premenopausal and half postmenopausal.Approximately 46% to 48% of patients had inflammatory breast cancer, and therest had locally advanced disease. Receptor status was available in a smallpercentage of patients.

Toxicities and Responses Vary

Hematologic toxicity (grade 3/4) consisted of febrileneutropenia in 14% of CEF patients and 8.7% of EC patients; anemia occurred in15.9% and 50.9%; and thrombocytopenia occurred in 32.2% and 21.1% of patients,respectively. With respect to cardiac toxicity, fewer than 1% of CEF patientsand 1% (two) EC plus G-CSF patients experienced congestive heart failure.Myelodysplastic syndrome was seen in one patient taking EC plus G-CSF and nonein the CEF arm. No cases of leukemia were documented. Approximately 30% ofpatients in both arms were hospitalized for toxicities. Two toxic deaths wereseen in the CEF arm.

The overall response rate to CEF was 58%, 29% complete responseand 29% partial response. In the EC plus G-CSF arm, the overall response ratewas 57%, 19% complete and 38% partial response. (For patients with inflammatorybreast cancer, only complete response, stable disease, or progressive diseasewas recorded). Pathologic complete responses were seen in patients in both arms—14%with CEF and 10% with EC plus G-CSF.

EC Plus G-CSF Considered Safe and Effective Alternative

Based on this study, it was concluded that EC plus G-CSF was asafe and effective alternative to CEF. Prior to publication of final results ofthis study, a cost-effectiveness analysis of the regimens will be undertaken andfindings included with the final report.

The New Trial

The results of MA.10, showing that EC plus G-CSF is as effectiveas CEF in locally advanced breast cancer, provide justification for using EC,instead of CEF, in the third arm of the study. Routine erythropoietin treatmentwill be added for all patients receiving EC plus G-CSF due to the high rate ofanemia seen in the MA.10 study. Because EC is a dose-dense regimen that iscompleted in six cycles (or 12 weeks), it was thought that addinga taxane after EC would be feasible and would allow treatment to be completedwith 24 weeks (eg, EC®T), similar to AC followed by paclitaxel or to CEF.

The MA.21 protocol will be open to node-positive and high-risknode-negative women up to 60 years of age who are premenopausal andpostmenopausal. They may have undergone total or partial mastectomy and have nodistant metastases. The three arms of the study will be: standard CEF for sixcycles; AC × 4 followed by paclitaxel × 4; and EC plus G-CSF pluserythropoietin for six cycles followed by paclitaxel for four cycles (Figure5).There is currently no plan to add trastuzumab to any of the arms for patientswhose tumors overexpress HER2.

Optimal Adjuvant RegimenIs Still To Be Determined

In summary, the optimal adjuvant chemotherapy regimen forpatients with breast cancer is still to be defined. The type of chemotherapychosen for low-risk patients may be more flexible than for high-risk patients,for whom anthracycline-based regimens such as CEF or FAC should be consideredfor a duration of six cycles.

The final role of taxanes with respect to optimalanthracycline-based regimens remains to be defined. The use of atrastuzumab-containing regimen would be recommended only within the context ofother ongoing clinical trials. The MA.21 trial will open shortly, with a plannedaccrual of 1,500 patients. Although the results will not be available for someyears, it is hoped that this study will advance our ability to identify the bestchemotherapy regimen for adjuvant treatment of patients with breast cancer.

Addendum

Following the presentation of this paper at the M.D. AndersonInvestigators’ Workshop, new data have become available. The NationalInstitutes of Health Consensus Conference was held in November 2000 whereupdated results from the CALGB 9344 trial were presented at 50 months offollow-up, demonstrating a loss of statistical significance in overall survivalfor the addition of paclitaxel to AC. The NSABP B-28 study comparing AC to ACfollowed by paclitaxel (225 mg/m²) was also presented, and no advantage has yetbeen seen at 36 months of follow-up for the addition of paclitaxel to AC. Inboth studies, CALGB 9344 and NSABP B-28, the majority of patients receivedtamoxifen as well as chemotherapy, and in these patients no benefits were seenfor the addition of paclitaxel. However, in those patients who did not receivetamoxifen, similar positive trends for the addition of paclitaxel are seen inboth studies.

Of note, the MA.21 trial opened to accrual of patients inDecember 2000. MA.21 becomes an even more important trial in light of theexperience to date with CALGB 9344 and NSABP B-28.

References:

1. Early Breast Cancer Trialists’ Collaborative Group: Ovarianablation in early breast cancer: Overview of the randomised trials. Lancet348:1189-1195, 1996.

2. Early Breast Cancer Trialists’ Collaborative Group:Tamoxifen for early breast cancer: An overview of the randomised trials. Lancet351:1451-1467, 1998.

3. Early Breast Cancer Trialists’ Collaborative Group:Polychemotherapy for early breast cancer: An overview of the randomised trials.Lancet 352:930-942, 1998.

4. National Comprehensive Cancer Network Guidelines, 2000. http://www.nccn.org

5. Goldhirsh A, Glick JH, Gelber RD, et al: Commentary MeetingHighlights: International consensus panel on the treatment of primary breastcancer. J Natl Cancer Inst 90:1601-1608, 1998.

6. The Steering Committee on Clinical Practice Guidelines forthe Care and Treatment of Breast Cancer: Adjuvant systemic therapy for womenwith node-negative breast cancer. Can Med Assoc J 158 (suppl 3):S43-S51, 1998;and Adjuvant systemic therapy for women with node-positive breast cancer. CanMed Assoc J 158 (suppl 3):S52-S64, 1998.

7. Bonadonna G, Zambetti M, Valagussa P: Sequential oralternating doxorubicin and CMF regimens in breast cancer with more than threepositive nodes. Ten-year results. JAMA 273 (7):542-547, 1995.

8. Fisher B, Brown AM, Dimitrov NV, et al: Two months ofdoxorubicin-cyclophosphamide with and without interval reinduction therapycompared with 6 months of cyclophosphamide, methotrexate, and fluorouracil inpositive-node breast cancer patients with tamoxifen-nonresponsive tumors:Results from the National Surgical Adjuvant Breast and Bowel Project B-15. JClin Oncol 8:1483-1946, 1990.

9. Fisher B, Anderson S, Wolmark N, et al: Chemotherapy with orwithout tamoxifen for patients with ER negative breast cancer and negativenodes: Results from NSABP B23 (abstr 277). Proc Am Soc Clin Oncol 19:72a, 2000.

10. Carpenter JT, Velez-Garcia E, Aron BS, et al: Five-yearresults of a randomized comparison of cyclophosphamide, doxorubicin, andfluorouracil versus cyclophosphamide, methotrexate, and fluorouracil fornode-positive breast cancer. Proc Am Soc Clin Oncol 13:68, 1994.

11. Hutchins L, Green S, Ravdin P, et al: CMF versus CAF withand without tamoxifen in high risk node-negative breast cancer patients and anatural history follow-up in low risk node-negative patients: First results ofintergroup trial INT 0102 (abstr 1). Proc Am Soc Clin Oncol 17:1A, 1998.

12. Henderson IC, Berry D, Demetri G, et al: for CALGB, ECOG,SWOG, and NCCTG: Improved disease-free (DFS) and overall survival (OS) from theaddition of sequential paclitaxel (T) but not from the escalation of doxorubicin(A) dose level in the adjuvant chemotherapy of patients (pts) with node-positiveprimary breast cancer (abstr 390A). Proc Am Soc Clin Oncol 17:101a, 1998.

13. Thomas E, Buzdar A, Theriault R, et al: Role of paclitaxelin adjuvant therapy of operable breast cancer: Preliminary results ofprospective randomized clinical trial (abstract 285). Proc Am Soc Clin Oncol19:74a, 2000.

14. Levine MN, Bramwell VH, Pritchard KI, et al: for theNational Cancer Institute of Canada Clinical Trials Group: Randomized trial ofintensive cyclophosphamide, epirubicin, and fluorouracil chemotherapy comparedwith cyclophosphamide, methotrexate, and fluorouracil in premenopausal womenwith node-positive breast cancer. J Clin Oncol 16:2651-2658, 1998.

15. Coombes RC, Bliss JM, Wils J, et al: Adjuvantcyclophosphamide, methotrexate, and fluorouracil versus fluorouracil,epirubicin, and cyclophosphamide chemotherapy in premenopausal women withaxillary node-positive operable breast cancer: Results of a randomized trial. JClin Oncol 14:35-45, 1996.

16. Bonneterre J, Roché H, Bremond A, et al: for the FrenchAdjuvant Study Group (FASG): Results of a randomized trial of adjuvantchemotherapy with FEC 50 vs FEC 100 in high risk node-positive breast cancerpatients (abstract 473). Proc Am Soc Clin Oncol 17:124a, 1998. (Updated based onOncology Drugs Advisory Committee [OCAD] presentation, June 7, 1999).

17. Launchburg AP, Habboubi N: Epirubicin and doxorubicin: Acomparison of their characteristics, therapeutic activity and toxicity. CancerTreat Rev 19:197-228, 1993.

18. Praga C, Trave F, Petroccione A: Anthracycline-inducedcardiotoxicity and its relevance in cancer treatment, in Nimmo WS, Tucker GT(eds): Clinical Measurement in Drug Evaluation, pp 131-142. London, Wolfe, 1991.

19. Von Hoff D-D, Layard MW, Basa P, et al: Risk factors fordoxorubicin-induced congestive heart failure. Ann Intern Med 91:710-717, 1979.

20. Findlay BP, Walker-Dilks C, for the Provincial Breast CancerDisease Site Group and the Provincial Systemic Treatment Disease Site Group:Epirubicin, alone or in combination chemotherapy, for metastatic breast cancer.Cancer Prev Control 2:140-146, 1998.

21. French Epirubicin Study Group: A prospective randomizedphase III trial comparing combination chemotherapy with cyclophosphamide,fluorouracil, and either doxorubicin or epirubicin. J Clin Oncol 6:679-688,1988.

22. Italian Multicentre Breast Study With Epirubicin: Phase IIIrandomized study of fluorouracil, epirubicin, and cyclophosphamide vsfluorouracil, doxorubicin, and cyclophosphamide in advanced breast cancer: AnItalian multicentre trial. J Clin Oncol 6:976-982, 1998.

23. Lopez M, Papaldo P, Di Lauro L, et al: 5-fluorouracil,Adriamycin, cyclophosphamide (FAC) vs 5-fluorouracil, epirubicin,cyclophosphamide (FEC) in metastatic breast cancer. Oncology 46:1-5, 1989.

24. Focan C, Andrien JM, Closon M Th, et al: Dose-responserelationship of epirubicin-based first-line chemotherapy for advanced breastcancer: A prospective randomized trial. J Clin Oncol 11:1253-1263, 1993.

25. Bastholt L, Dalmark M, Gjedde SB, et al: Dose-responserelationship of epirubicin in the treatment of postmenopausal patients withmetastatic breast cancer: A randomized study of epirubicin at four differentdose levels performed by the Danish Breast Cancer Cooperative Group. J ClinOncol 14:1146-1155, 1996.

26. Focan C, Andrien JM, Closon M Th, et al: Dose-responserelationship of epirubicin-based first-line chemotherapy for advanced breastcancer: A prospective randomized trial. J Clin Oncol 11:1253-1263, 1993.

27. Levine MN, Bramwell V, Pritchard KI, et al: A pilot study ofintensive cyclophosphamide, epirubicin and fluorouracil in patients withaxillary node-positive or locally advanced breast cancer. Eur J Cancer 29:37-43,1993.

28. Food and Drug Administration Oncology Drugs AdvisoryCommittee presentation, April 1999.

29. Therasse P, Mauriac L, Welnicka M, et al: Neo-adjuvant doseintensive chemotherapy in locally advanced breast cancer (LABC): AnEORTC-NCIC-SAKK randomized phase III study comparing FEC (5FU, epirubicin,cyclophosphamide) vs high dose intensity EC + G-CSF (Filgrastim) (abstract 472).Proc Am Soc Clin Oncol 17:124a, 1998.