Feasibility and Tolerability of Sequential Doxorubicin/Paclitaxel

Cancer Therapy: Clinical

Feasibility and Tolerability of Sequential Doxorubicin/Paclitaxel Followed by Cyclophosphamide, Methotrexate, andFluorouracilandItsEffectsonTumor Response as Preoperative Therapy Luca Gianni,1Jose¤Baselga,2 Wolfgang Eiermann,4 Vicente Guillem Porta,5 Vladimir Semiglazov,7 An‹ aLluch,6 Milvia Zambetti,1Dolores Sabadell,2 Gu«nther Raab,4 Antonio Llombart Cussac,5 Alla Bozhok,7 Angel Martinez-Agullo¤,6 Marco Greco,1Mikhail Byakhov,8 JuanJose' Lopez Lopez,3 Mauro Mansutti,9 Pinuccia Valagussa,1and Gianni Bonadonna1 for the European Cooperative Trial in Operable Breast Cancer Study Group

Abstract Purpose: The European CooperativeTrial in Operable breast cancer (ECTO) randomly tested whether efficacy of adjuvant doxorubicin followed by i.v. cyclophosphamide, methotrexate, and fluorouracil (CMF; doxorubicin ! CMF, arm A) could be improved by adding paclitaxel (doxorubicin/paclitaxel ! CMF) as adjuvant (arm B) or primary systemic therapy (PST, arm C). We report here feasibility, tolerability, locoregional antitumor activity, and breast conservation rate. Methods: A total of 1,355 women entered the study. Feasibility and safety were compared in arm A versus arms B plus C. Surgical findings were compared in arms A plus B versus arm C. Results: Grade 3 or 4 National Cancer Institute toxicities were low (<5%)inallarms.Neuropathy was more frequent in the paclitaxel-containing arms (grade 2, 20.5% versus 5.0%; grade 3, 1.3% versus 0.2%). At 31months of follow-up, asymptomatic drop of left ventricular ejection fraction was similar in all arms, whereas symptomatic cardiotoxicity was recorded in three patients (0.5%) in A and in three patients (0.3%) in B plus C. PST induced clinical complete plus partial remissionin 78%,withanin-breastpathologiccompleteresponse rate of 23%andanin-breastplus axillapathologiccompleteresponse rateof 20%.Inthemultivariateanalysis,onlyestrogenreceptor (ER)statuswas significantlyassociatedwithpathologiccompleteresponse(oddsratioforERneg- ative, 5.77; 95% confidence interval, 3.49-9.52; P < 0.0001). PTS induced a significant axillary downstaging (P < 0.001),andbreast sparing surgery was feasiblein 65% versus 34% (P <0.001). Conclusions: Doxorubicin/paclitaxel ! CMF is feasible, safe, and well tolerated. Given as PST, it is markedly active, allowing for breast-sparing surgery in a large fraction of patients.

The superior efficacy of the sequential approach with non– by Norton and Simon (1) and was clinically proven in a piv- cross-resistant drugs/regimens was postulated in the early 1980s otal study of the Istituto Nazionale Tumori with the regimen of doxorubicin and cyclophosphamide, methotrexate, and fluorouracil (doxorubicin ! CMF; refs. 2, 3). An independent study has confirmed that sequential epirubicin ! CMF is sig- Authors’ Affiliations: 1Istituto Nazionale Tumori, Milan, Italy; 2Hospital Vall d’Hebron and 3Hospital de San Pau, Barcelona, Spain; 4Frauenklinik vom Roten nificantly superior to CMF (4). In the mid-1990s, clinical Kreuz, Munich, Germany; 5Istituto Valenciano de Oncologia and 6Hospital Clinico studies indicated that paclitaxel had marked antitumor activity Universitario de Valencia, Valencia, Spain; 7N.N. Petrov Research Institute of in women with metastatic breast cancer while lacking complete 8 Oncology, St. Petersburg, Russia; Central Clinical Hospital named after N.A. clinical cross-resistance with anthracyclines (5, 6). Subse- Semashko, Moscow, Russia; and 9Azienda Ospedaliera Santa Maria della Misericordia, Udine, Italy quently, the combination of doxorubicin and paclitaxel Received 3/10/05; revised 9/15/05; accepted 9/27/05. resulted in a marked antitumor activity in women with Grant support: Bristol Meyers Squibb. metastatic breast cancer who had received minimal or no prior The costs of publication of this article were defrayed in part by the payment of page chemotherapy, although associated with higher than expected charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. cardiac toxicity (7, 8) that could later be in part attributed to Note: Presented in part at the 38th Annual Meeting of the American Society of pharmacokinetic and biochemical interactions between pacli- Clinical Oncology, May 18 to 21, 2002, Orlando, Florida. taxel and doxorubicin (9, 10). The cardiotoxic enhancement Requests for reprints: Luca Gianni, CTO Coordinating Center, Istituto Nazionale was not clinically apparent when doxorubicin was given for a Tumori,Via Venezian, 1, 20133 Milan, Italy. Phone: 39-02-2390-2789; Fax: 39-02- 2 2390-2678; E-mail: [email protected]. maximum total dose of 360 mg/m with paclitaxel (11). F 2005 American Association for Cancer Research. Based on the activity of the above regimens, in 1996, we doi:10.1158/1078-0432.CCR-05-0539 designed a trial [known as European Cooperative Trial in

www.aacrjournals.org 8715 Clin Cancer Res 2005;11(24) December 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Cancer Therapy: Clinical

Operable breast cancer (ECTO); Fig. 1] addressing the potential Between November 6, 1996 and May 29, 2002, a total of 1,355 role of introducing paclitaxel within a sequential regimen of patients were enrolled. Breast cancers were diagnosed by core biopsy non–cross-resistant chemotherapies. The ECTO study random- (82%) or fine-needle aspiration cytology (18%). Either method of ized patients with breast cancer > 2 cm to adjuvant doxorubicin diagnosis was evenly distributed among the three treatment arm. Approximately 50% of the tumor diagnosed by fine-needle aspiration ! CMF or doxorubicin/paclitaxel ! CMF (Fig. 1, arms A and B, had clinical involvement of axillary nodes, and after major surgical respectively). In the ECTO study, the planned delivery of 2 intervention, all tumors were classified as invasive carcinomas. maximum of four cycles of doxorubicin/paclitaxel (240 mg/m Assessment of hormonal receptor status and tumor grade was required. of total doxorubicin) before CMF seemed therefore consis- Treatment allocation was done centrally at the ECTO Operations tent with the goal of exploiting the potential benefits of Office using a minimization algorithm (19) to balance assignments doxorubicin/paclitaxel without exposing the patients to exces- according to primary tumor size (<4.0 versus >4.0 cm), tumor grade sive cardiac risk. (low versus intermediate versus high), and hormonal receptor status Based on kinetic considerations (12) and on animal experi- (estrogen and/or progesterone positive versus both negative). Geo- ments indicating that systemic therapy before surgery inhibited graphic area of the participating center was also taken into account the tumor growth stimulatory response triggered by surgery before randomization. Treatment. Patients in arm A first received surgery followed by (13, 14), the ECTO study is also addressing the effectiveness sequential single-agent doxorubicin and CMF; patients in arm B first of doxorubicin/paclitaxel ! CMF before surgery (Fig. 1, arm B underwent surgery followed by sequential doxorubicin/paclitaxel and versus C). The pivotal NSABP-B18 study showed that the CMF; patients in arm C received sequential doxorubicin/paclitaxel doxorubicin and cyclophosphamide combination afforded and CMF followed by surgery. equivalent efficacy when given either as primary or adjuvant Breast irradiation (50 Gy through two opposing tangential fields to systemic therapy (15, 16). The same investigation and other be started after completion of chemotherapy or within 4 weeks from studies also showed that eradication of the invasive tumor in surgery in arm C) was required after breast sparing surgery. In the the breast assessed at surgery (pathologic complete response) is presence of pT4 lesion after mastectomy, chest wall irradiation was an independent predictor of long term outcome (15–18). In recommended. 2 the ECTO trial, we reasoned that a prolonged administration of In arm A (Fig. 1), doxorubicin was given at the dose of 75 mg/m by a sequential non–cross-resistant chemotherapy containing i.v. bolus every 3 weeks. In arms B and C, doxorubicin was given as a 60 mg/m2 i.v. bolus followed by paclitaxel at 200 mg/m2 infused over paclitaxel could yield a large increment of the rate of pathologic 3 hours, and each course was repeated every 3 weeks. CMF consisted complete response that may increase the likelihood of in i.v. cyclophosphamide (600 mg/m2), methotrexate (40 mg/m2), and observing a difference of benefit. Analysis of the main efficacy fluorouracil (600 mg/m2) on days 1 and 8 every 4 weeks. No dose outcomes of the ECTO study awaits for the appropriate period reduction was allowed except following febrile neutropenia (as defined of observation. However, at median 31 months of follow-up, by absolute neutrophil count < 500/AL and fever > 38jC), grade >2 data on toxicity of the delivered regimens and on locoregional neuropathy due to paclitaxel, and grade >2 gastrointestinal toxicity on antitumor activity of primary systemic therapy, a secondary aim doxorubicin/paclitaxel regimen, when a 25% temporary reduction was of the ECTO trial, are available and reported here. required. In the presence of hematologic toxicity on the planned day of treatment, drug administration was delayed until marrow recovery. Before paclitaxel, premedication consisted of prednisone (25 mg orally Patients and Methods the evening before therapy), hydrocortisone (250 mg i.v.) plus chlorphenamine (10 mg i.v. or i.m.), and cimetidine (300 mg i.v.), Patient eligibility and study procedures. Women at participating all given 30 minutes before the taxane. Antiemetic therapy was ECTO centers who had primary operable breast tumor > 2.0 cm at administered according to the policy of the participating center. diagnosis (T2-T3,N0-N1,M0) were eligible for this study. Patients with At the end of the surgery-chemotherapy approach, all patients, locally advanced or bilateral breast carcinoma, prior anticancer regardless of age and receptor status, were originally offered tamoxifen treatment, inadequate bone marrow reserve, abnormal renal and liver (20 mg/d) for five consecutive years. The protocol was amended, and function tests, and history of relevant cardiac disease were not eligible after June 30, 2000, tamoxifen was offered only to women whose for the study. Also excluded were patients pregnant or lactating or with tumors were either estrogen or progesterone positive. active infection or with a history of other invasive malignancy. Primary tumor size determination and evaluation of primary systemic The study and the informed consent form had to be approved by therapy response. Clinical size of primary breast cancers was each local ethic committee. Patients were to undersign the written determined at diagnosis and before administration of each cycle of informed consent to enter the study. chemotherapy. Mammogram or breast ultrasound were planned at the end of each chemotherapy regimen. In the absence of clinical evidence of tumor in the breast and in the absence of any new lesions, treatment response was categorized as clinically complete. When the reduction in size of the breast tumor was >50% in the product of the two greatest perpendicular diameters, the response was judged to be partial; when the reduction was <50% but >25%, the response was categorized as minor. When there was an increase of >50% compared with a previous measurement or when a new lesion appeared, the patient was considered to have progressive disease. Patients whose tumor response did not meet any of the above definitions were considered nonresponders. Surgical breast and axillary node resection specimens were care- fully evaluated for pathologic tumor response according to the guide- lines provided by an ad hoc committee before starting the trial. Patients Fig. 1. ECTO study design.Women with breast tumor > 2 cm at diagnosis were randomized to either arm A, B, or C. A, doxorubicin; AT, doxorubicin and paclitaxel; who had no residual invasive cancer in the breast were considered to T,Taxol; CMF, cyclophosphamide, methotrexate, and fluorouracil; S, surgery. have had a pathologic complete response. Because this definition did

Clin Cancer Res 2005;11(24) December 15, 2005 8716 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Sequential Doxorubicin/Paclitaxel and CMF not take histologic nodal status into account, patients with noninvasive or no residual cancer could have had positive axillary lymph nodes. Ta b l e 1. Patient characteristics Side effects from chemotherapy. Side effects were assessed at each treatment cycle and graded according to the National Cancer Institute Arm A Arm B Arm C Common Toxicity Criteria (20). Electrocardiogram, physical examina- tion, and measurement of left ventricular ejection fraction (LVEF) were Patient status required at diagnosis, at the end of each chemotherapy regimen and Entered 453 451 451 every 6 months during at least the first 2 years of follow-up. Ineligible 2 5 1 Statistical methods. The primary objective of the ECTO study was Refused treatment 7 14 2 to assess the freedom from breast cancer progression or relapse as Evaluable for efficacy and 444 432 448 measured from the date of randomization. Based on the prior toxicity experience of the doxorubicin ! CMF regimen at the Milan Cancer Chemotherapy completed 89.2 89.1 90.6 Institute, it was planned to accrue 450 patients in each study arm, with as planned (%) an 80% power of the log-rank test to detect a 30% between-group Clinical tumor size, cm (%)* relative difference in the hazard of occurrence of end point events, at V4.0 80.4 80.3 80.4 the 5% significance level (two-sided test). The second objective of the study, which is the aim of this report, was >4.0 19.6 19.7 19.6 to compare the rate of breast-saving procedures and of pathologic nodal Clinical nodal status (%) status after primary chemotherapy (arm C) versus primary surgery N0 52.3 53.8 56.5 (arms A and B grouped together). Simple proportions were examined N1-N2 47.7 46.2 43.5 by use of the m2 test. A second aim was attempting to identify Hormonal receptor status (%)* pretreatment variables likely to predict clinical and pathologic response ER/PR+ 68.4 68.3 67.9 to primary chemotherapy with doxorubicin/paclitaxel ! CMF. Assess- ER/PR 30.7 30.8 31.2 ment was done by analyzing both simple proportions on univariate Not assessed 0.9 0.9 0.9 analyses and by performing logistic regression analyses using a Tumor grade (%)* backward procedure (21). All analyses are based on data received as L o w 11. 3 11. 6 11. 6 of July 31, 2003. Intermediate 55.8 54.6 55.1 High 30.4 31.2 31.1 Results Not assessed 2.5 2.6 2.2 Age (%) Patient population and tumor characteristics. A total of 1,355 <50 y 42.2 46.8 44.9 patients were enrolled into study (Table 1). Eight patients z50 y 57.8 53.2 55.1 (0.6%) failed to meet the eligibility criteria because of locally advanced disease (n = 2), distant metastases (n = 2), no invasive Abbreviation: PR, progesterone receptor. breast cancer on biopsy (n = 2), bilateral breast cancer (n = 1), *As reported at the time of randomization. and chronic hepatitis (n = 1). In addition, 23 women (1.7%) refused to be treated according to their allocated program. Main patient characteristics at randomization (Table 1) remained toxic events were similar whether or not paclitaxel was part of evenly distributed in the study arms. Approximately 20% of the chemotherapy regimen. As expected, reversible peripheral patients had tumors > 4.0 cm, and more than half were ages z50 neuropathy was significantly more frequent with the paclitaxel- years. Approximately one third of tumors were both estrogen containing regimen (Table 2). Signs and symptoms of receptor (ER) and progesterone receptor negative, and more peripheral neuropathy were documented after the second than half were categorized as intermediate grade. cycle of doxorubicin/paclitaxel, lasted in median 3 months, Therapy, compliance, and adverse events. The planned eight and completely recovered before the end of CMF. In only one cycles of therapy were completed in 89% of all patients woman did neuropathy last 8 months before complete irrespective of the allocated arm of treatment (Table 1). Of recovery. note, f97% of patients completed the first four courses of Serial measures of LVEF in 91% of all patients allowed either doxorubicin or doxorubicin/paclitaxel. Toxic events were for grading of cardiac effects for a median observation time reported as reason for treatment discontinuation in 2% of all of 31 months from randomization (Table 2). A total of 11.8% patients. Repeated episodes of nonfebrile neutropenia (0.8%) of patients developed grade 2 toxicity, that included either were the most frequent cause. One patient had a drop in LVEF LVEF values of <50% (6.2%) or a decrease of >20% from basal to 33% after the second cycle of CMF, and another woman values (5.6%). The incidence of either type of event was similar developed a myocardial infarction (see below for cardiac with doxorubicin ! CMF and doxorubicin/paclitaxel ! CMF events). Other causes of treatment discontinuation include (Table 2). Six patients (arm A, 3% or 0.5%; arms B plus C, 3% intervening nonneoplastic medical conditions (six patients) or 0.3%) developed grade 3 toxicity. All of these patients were and discontinuation of CMF after the third cycle because of alive after a minimum of 8 months to >5 years from occurrence poor patient tolerance (10 women). of the cardiac event. In view of the even distribution of main patient character- Comparison of surgical procedures and pathologic findings after istics and treatment delivery in the three arms of the study primary surgery or primary systemic therapy. A total of 296 (Table 1), further analysis of toxicity can be presented pooling women (34%) in the primary surgery arms underwent breast together observations and events in arms B and C, in which conservative surgery, with no differences between arm A and B. patients received the same regimen consisting of doxorubicin/ After primary chemotherapy, the rate of breast conservative paclitaxel ! CMF. Data in Table 2 show that almost all major surgery increased to 65% (284 of 438), showing a statistically

www.aacrjournals.org 8717 Clin Cancer Res 2005;11(24) December 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Cancer Therapy: Clinical

Ta b l e 2 . Main toxicities

Events Arm A Arms B + C A (%) CMF (%) AT (%) CMF (%) Neutropenia (<500 ANC/AL) 1.3 1.0 0 1.7 Febrile neutropenia* 2.5 3.0 4.3 3.4 c Infection G2-G3 0.2 0.5 1.1 0.6 Allergic reaction

G2 0.7 0.7 0.5 0.2

G4 00 0.10 Neuropathy

G2 5.0 2.5 20.5 7.0

G3 0.2 0 1.3 0.1 Stomatitis

G3 3.3 1.6 3.4 2.3

G4 0 0.2 0.5 0.5 Cardiac events CTC 0 18.1 17.8 CTC1 67.8 70.8 CTC 2 13.6 11.1 LVEF <50% 8.0 5.7 >20% decrease from basal values 5.6 5.4 CTC 3 0.5 0.3

Abbreviations: A, doxorubicin; AT, doxorubicin/paclitaxel; ANC, absolute neutrophil count; CTC, CommonToxicity Criteria. *ANC < 500/ALandfever> 38jC. cUnrelated to febrile neutropenia or surgical reasons.

significant reduction (P < 0.001) in the requirement for Clinical response after primary systemic therapy. Major mastectomy. Figure 2 shows the effects of primary systemic shrinkage of the primary tumor was measured in 78% therapy on locoregional disease. Breast-sparing surgical proce- of the patients (clinically complete response, 49%; partial dures were significantly more frequent after primary systemic response, 29%) after primary doxorubicin/paclitaxel ! CMF. therapy in all categories of initial tumor size. This is especially An analysis of the degree of response after doxorubicin/ marked for tumors > 4 cm at diagnosis (Fig. 2A). The low paclitaxel and after CMF was feasible in 373 patients and rate of breast conservative surgery in patients presenting with is reported in Table 3. Overall, 100 patients (27%) achieved tumors measuring up to 3.0 cm in arms A and B can in part be a clinically complete response after the first four cycles explained by the fact that many centers in Europe still perform of doxorubicin/paclitaxel. Of note, 151 of 273 patients mastectomy in women with tumors > 2.5 cm, especially if (55%) having a residual clinically palpable tumor mass axillary nodes are clinically palpable. after doxorubicin/paclitaxel had an improvement of the re- The ability of primary systemic therapy in increasing the sponse with CMF. Three women (1%) had progressive disease feasibility rate of breast-saving procedures was not counter- while on therapy. balanced by an increased risk of in-breast recurrence. In fact, Pathologic findings after primary systemic therapy. After 31 months after randomization, the total risk of recurrence primary doxorubicin/paclitaxel ! CMF, six patients refused was even lower, albeit nonstatistically significant because surgery (two after achieving a clinically complete response), of the low number of events, after PTS and conservative sur- two were not operated because of progressive disease; in two gery [total risk: 1.4%, 95% confidence interval (95% CI), cases, no reasons for missing surgery were provided. Pathologic 0-3.4; tumor size V 4.0 cm, 0.7%; 95% CI, 0-2.2; tumor findings are available for overall 438 patients (98%). Complete size > 4.0 cm, 2.4%; 95% CI, 0-7.0] compared with the risk absence of cancer cells was reported in breast specimens of observed after conservative surgery followed by adjuvant 75 patients (17%). In 27 additional cases (6%) residual in situ systemic therapy (total risk, 3.6%; 95% CI, 0.8-6.3; tumor size carcinoma only was described, so that 23% of patients had V 4.0 cm, 2.8%; 95% CI, 0.8-3.6; tumor size > 4.0 cm, 3.1%; no residual invasive cancer in the breast. Considering also the 95% CI, 0.3-5.9). pathologic findings in the axillary lymph nodes, overall 89 of The administration of primary chemotherapy also contrib- 438 patients (20%) had no evidence of invasive cancer after uted to a significant downstaging of pathologic nodal status: primary doxorubicin/paclitaxel ! CMF. 263 (60%) patients were pathologically node negative com- Univariate and multivariate analysis of clinical and pathologic pared with 39% in the primary surgery group (P < 0.001). Of response. The likelihood of achieving a response to primary note, fewer positive nodes were found in patients who received doxorubicin/paclitaxel ! CMF was assessed as a function of primary chemotherapy, as shown in Fig. 2B. main pretreatment variables (age, clinical tumor size, clinical

Clin Cancer Res 2005;11(24) December 15, 2005 8718 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Sequential Doxorubicin/Paclitaxel and CMF nodal status, ER and progesterone receptor status, and tumor grade). Table 4 shows the univariate analysis for clinical and Ta b l e 3 . Clinical response of primary breast cancer pathologic complete response, respectively. ER and progesterone after both ATand CMF receptor status and tumor grade were all significantly associated with the likelihood of achieving clinical and pathologic Response Response after CMF response. The logistic regression analysis revealed that ER status after AT CCR PR MR NR PD was the only significant variable independently capable of CCR 100 100 0 influencing treatment outcome, with an odds ratio for negative PR 120 57 62 1 versus positive ER of 2.1 (95% CI, 1.36-3.23; P = 0.0008) for MR 84 18 33 32 1 clinically complete response, and of 5.77 (95% CI, 3.49-9.52; NR 68 9 12 22 25 0 P < 0.0001) for the achievement of pathologic response. PD 1 1 Total 184 (49%) 107 (29%) 54 (14%) 25 (7%) 3 (1%) Discussion Abbreviations: AT, doxorubicin and paclitaxel; CCR, clinical complete The present first report on the ECTO trial shows that response; PR, partial response; MR, minor response; NR, no response; PD, inclusion of paclitaxel in the doxorubicin/paclitaxel ! CMF progressive disease. is feasible, well tolerated, and endowed of marked antitumor activity. The tolerability of doxorubicin ! CMF was of the same type and incidence already reported in the single-institution of peripheral neuropathy induced by paclitaxel (7). Peripheral experience (2), and the toxicity of doxorubicin/paclitaxel ! neuropathy was reversible in all patients and lasted in median CMF was similar, except for the significantly higher incidence 3 months. Severe neutropenia and febrile neutropenia were infrequent with and without paclitaxel. The ECTO study called for administration of full doses of all drugs based on counts on the day of treatment. This approach allowed for delivery of full doses to the vast majority of patients without compromis- ing on planned dose intensity and requiring administration of granulocyte colony-stimulating factor to a very small proportion of women. The approach was different from that in other investigations of doxorubicin/paclitaxel, in which dose reduc- tions were based on nadir counts and led to a marked decrease of dose intensity (22). The good hematologic tolerability of doxorubicin/paclitaxel ! CMF is in contrast with that reported in other trials of sequential regimens inclusive of docetaxel, in which high rates of severe neutropenia and febrile neutropenia were reported in spite of prophylactic granulocyte colony- stimulating factor administration to all patients, and toxic deaths did occur (23, 24). In the ECTO trial, the most careful monitoring was applied to the cardiac function because of the higher-than-expected cardiac toxicity reported in early studies of doxorubicin/ paclitaxel (7, 8). LVEF was similarly affected in patients receiving single-agent doxorubicin in arm A for a total dose of 300 mg/m2 and in those treated with doxorubicin/ paclitaxel in arms B and C for a total dose of 240 mg/m2 of doxorubicin. The decrease was almost always non symptomatic and was reversible in the vast majority of patients after delivery of chemotherapy. Symptomatic cardiac events (grade 3) were rare, had similar incidence with both regimens (0.5% in arm A versus 0.3% in arms B and C), were res- ponsive to cardiac medications, and were all observed during or soon after chemotherapy. At median 31 months after randomization, the incidence of grade 3 events is lower than the one reported with other anthracycline-containing regimens for early breast cancer (5, 25, 26). Although longer follow-up Fig. 2. Effects of primary systemic chemotherapy on locoregional disease. A, rate only will rule out that the regimen does not entail increased of breast conserving surgery and initial tumor size. Black columns are for patients cardiac risk at later times, the early findings reported here are randomized to surgery as first treatment modality; light gray columns are for patients randomized to primary systemic therapy before surgery. Difference between reassuring. groups is statistically significant (P < 0.001, m2 test). B, rate of patients with indicated The proportion of women achieving eradication of invasive axillary nodes involved by tumor at histology. Black columns are for patients carcinoma in the breast after four cycles of chemotherapy has randomized to surgery as first treatment modality; light gray columns are for patients randomized to primary systemic therapy before surgery. Difference between been in the low range of 3% to 14% in spite of the much higher groups is statistically significant (P < 0.001, m2 test). proportion of them having major clinical response (15–18).

www.aacrjournals.org 8719 Clin Cancer Res 2005;11(24) December 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Cancer Therapy: Clinical

and obtained almost doubling of the rate of clinical complete Ta b l e 4 . Likelihood of clinical and pathologic response (27% after doxorubicin/paclitaxel versus 49% after complete response of breast tumor CMF), including patients who only had minor or no response to doxorubicin/paclitaxel. Another preoperative trial has Category % CCR P %PCR P addressed the question of duration or sequencing with an ER status elegant design whereby, after receiving four cycles of cyclo- Positive 42 0.005 12 <0.001 phosphamide, doxorubicin, vincristine, and prednisolone, Negative 60 42 responding patients were randomly assigned to continuing PR status with additional four cycles of the same cyclophosphamide, Positive 43 0.02 13 <0.001 doxorubicin, vincristine, and prednisolone regimen or four Negative 55 37 cycles of docetaxel (27). Switching to docetaxel led to signif- Tu m o r g r a d e icant increase of clinical and pathologic complete response rate Low + intermediate 43 0.006 19 0.006 (27) and to significantly improved disease-free and overall High 58 30 survival at three and at 5 years of follow-up (33, 34), docu- menting that sequential administration of non–cross-resistant regimens is prevailing over duration of a nontaxane regimen in NOTE: Univariate analysis. Abbreviations: CCR, clinical complete response; PCR, pathologic complete driving to improved response and efficacy (33, 34). response; PR, progesterone receptor. The ECTO study also provides a relevant qualification of the tumor response after doxorubicin/paclitaxel ! CMF. Of note, 42% of patients with ER negative tumors had pathologic complete response versus 12% in the ER-positive group and in Many trials of primary chemotherapy in the 1990s attempted multivariate analysis ER status emerged as the only indepen- at improving the rate of pathologic complete response by dent variable significantly associated with likelihood of designing new regimens that included taxanes and/or new achieving a clinically complete response (odds ratio, 2.1), and modalities of drug administration (23, 24, 27), while also most importantly a pathologic complete response (odds ratio, prolonging the total duration of treatment. This report shows 5.77; P < 0.0001). The association between ER status and that doxorubicin/paclitaxel ! CMF leads to major clinical likelihood of pathologic complete response is in agreement response in 78% (complete in 49%), an in-breast pathologic with other reports, indicating that ER-negative tumors are more complete response rate of 23% and an in-breast plus axilla sensitive to chemotherapy than ER-positive ones (23, 24). In pathologic complete response rate of 20%. These results are of the ECTO trial, this different sensitivity according to ER status the same order of magnitude of those reported by other studies was remarkably high, but it should be pointed out that that included preoperative taxanes (23, 24, 28) and raise the pathologic complete response rate in ER-positive tumors question whether the improved rate of pathologic complete should not be discounted as negligible. On the contrary, the response should be attributed to the introduction of the different pathologic complete response rate in the two subsets taxanes, the duration of preoperative treatment, or the of patients justifies the design of different drug approaches sequencing of non–cross-resistant regimens. Two adjuvant according to hormonal receptor status. studies showed that adding four cycles of thrice-weekly The present report points out to an immediate advantage of paclitaxel after standard doxorubicin and cyclophosphamide the preoperative regimen when comparing surgical findings led to improved efficacy (29, 30). Because of the study design, after doxorubicin/paclitaxel ! CMF to those in patients neither trial allows for discriminating whether the advantage undergoing surgery first (arms A and B). Tumor size was depended on the merits of the taxane or on the more prolonged decreased enough to allow for breast sparing procedures in duration of the chemotherapy in the experimental arm. Such a 65% of patients versus 34% in the adjuvant arms. Importantly, question is addressed in the ECTO study that compares two at a median follow-up in excess of 24 months from surgery, regimens of identical duration and similar feasibility, one with such an improvement in breast conservation was not at the and one without paclitaxel, but efficacy outcomes of the ECTO cost of an increased risk of local relapses (35). trial will be analyzed after follow-up longer than now available. In a French study (31), doxorubicin and paclitaxel were given Appendix 1: List of Active ECTO Participating preoperatively for either four or six cycles. The longer Centers administration led to an improved rate of clinical and histologic response (31). The finding would be in keeping with the Coordinating Center, Istituto Nazionale Tumori, Milan, Italy interpretation that treatment duration is factoring into the (G. Bonadonna, L. Gianni, M. Zambetti, M. Greco, S. Pilotti, L. antitumor activity of the doxorubicin/paclitaxel regimen and is Mariani, C. Dipace, and P. Valagussa); Hospital Vall D’Hebron, in line with the results of the first report of the activity of Barcelona, Spain (J. Baselga, B. Queralt, and D. Sabadell); doxorubicin/paclitaxel in women with metastatic breast cancer Frauenklinik vom Roten Kreuz, Munich, Germany (W. Eiermann (7). In that trial, patients received up to eight cycles of and G. Raab); Istituto Valenciano de Oncologia, Valencia, Spain doxorubicin/paclitaxel, and continuous delivery of single-agent (V. Guillem Porta, A. Llombart Cussac, and C. Vazquez); N.N. paclitaxel contributed to increasing complete responses from Petrov Research Institute of Oncology, St. Petersburg, Russia 26% to the remarkable rate of 42% (7, 32). However, the ECTO (V. Semiglazov and A. Bozhok); Hospital Clinico Universitario trial also suggests a role of the sequential application of non– de Valencia, Valencia, Spain (J. Garcia-Conde, A. Lluch, and A. cross-resistant regimens. Indeed, clinical response to CMF Martinez-Agullo´); Central Clinical Hospital named after N.A. contributed to improving the overall response in 151 women Semashko, Moscow, Russia (M. Byakhov); Azienda Ospedaliera

Clin Cancer Res 2005;11(24) December 15, 2005 8720 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Sequential Doxorubicin/Paclitaxel and CMF

S. Maria della Misericordia, Udine, Italy (M. Mansutti and G. Centrum Onkologii, Bydgoszcz, Poland (J. Tujakowski); Faculty Gentile); Hospital de San Pau, Barcelona, Spain (J. Lopez-Lopez Hospital, Ostrava-Poruba, Czech Republic (P. Vodvarka); and C. Alonso); Policlinico Monteluce, Perugia, Italy (M.A. Krakowski Szpital Specjalistyczny im. Ludwiga Rydgiera, Colozza); Oncology Clinical Dispenser, Moscow, Russia Krakow, Poland (P. Koralewski); Katedra Onkologii-Szpital (V. Borisov); Hospital Universitario 12 de Octubre, Madrid, Kliniczny nr 1, Poznan, Poland (K. Roznowski); St. Elisabeth Spain (H. Corte´s-Funes); Klinika Nowotworow Sutka, Warsaw, Oncology Institute, Bratislava, Slovakia (S. Spanik); Cancer Poland (T. Pienkowski); Ospedale S. Bortolo, Vicenza, Italy Research RAMS-Dept. of Clinical Pharmacology, Moscow, (V. Fosser); Latvian Cancer Center, Riga, Latvia (J. Berzins); Russia (A. Garin); and Cancer Research RAMS/Department of Klinika Chemioterapii/Instytut Onkologii, Krakow, Poland Cancer Chemotherapy, Moscow, Russia (V. Gorbounova). (M. Pawlicki); Istituto San Raffaele, Milano, Italy (E. Villa); Regional Oncology Clinic, Murmansk, Russia (V. Zinykevich); Hospital of Oncology-Clinicum of the University of Tartu, Acknowledgments Tartu, Estonia (P. Padrik); Ospedale S. Chiara, Trento, Italy (E. Galligioni); Semmelweis Medical School, Budapest, Hungary We thank all the patients who have participated in our clinical trial; the many asso- ciates, in particular medical oncologists, surgeons, radiation therapists, pathologists, (M. Dank); LBI for Applied Cancer Research-Kaiser Franz Josef research nurses and data managers for their cooperation during the study; and the Spital, Wien, Austria (C. Dittrich); Uzsoki Hospital, Budapest, members of the International Advisory Board: Drs. John Bryant, GabrielHortobagyi, Hungary (T. Nagyka´lnai); Oddzial Chemioterapii-Regionalne Larry Norton, Abraham Recht, and William Wood.

References 1. Norton L, Simon R. The Norton-Simon hypothesis Presence of a growth-stimulating factor in serum fol- docetaxel for node positive breast cancer. N Engl J revisited. CancerTreat Rep1986;70:163 ^ 9. lowing primary tumor removal in mice. Cancer Res Med 2005;352:2302^13. 2. Bonadonna G, Zambetti M,Valagussa P. Sequential 19 8 9;4 9 :19 9 6 ^ 20 01. 26. Zambetti M, Moliterni A, Materazzo C, et al. Long-term or alternating doxorubicin and CMF regimens in breast 14. Fisher B, Saffer E, Rudock C, Coyle J, Gunduz N. cardiac sequelaeinoperable breast cancer patients given cancer with more than three positive nodes. Ten-year Effect of local or systemic treatment prior to primary adjuvant chemotherapy with or without doxorubicin results. JAMA1995;273:542 ^ 7. tumor removal on the production and response to a andbreastirradiation. JClin Oncol 2001;19:37^43. 3. Bonadonna G, Zambetti M, Moliterni A, Gianni L, serum growth-stimulating factor in mice. Cancer Res 27. Smith IC, Heys SD, Hutcheon AW, et al. Neoadju- Valagussa P. Clinical relevance of different sequencing 1989;49:2002^ 4. vant chemotherapy in breast cancer: significantly of doxorubicin and cyclophosphamide, methotrexate, 15. Fisher B, Brown A, Mamounas E, et al. Effect of pre- enhanced response with docetaxel. J Clin Oncol and fluorouracil in operable breast cancer. JClin Oncol operative chemotherapy on local-regional disease in 2002;20:1456 ^ 66. 2004;22:1614^ 20. women with operable breast cancer: findings from 28. Green MC, Buzdar AU, SmithT, et al.Weekly (wkly) 4. Poole CJ, Earl HM, Dunn JA, et al. NEAT (National National Surgical Adjuvant Breast and Bowel Project paclitaxel (P) followed by FAC as primary systemic Epirubicin AdjuvantTrial) and SCTBG BR9601 (Scot- B-18. J Clin Oncol 1997;15:2483 ^ 93. chemotherapy (PSC) of operable breast cancer tish Cancer Trials Breast Group) phase III adjuvant 16. Fisher B, BryantJ,Wolmark N, et al. Effect ofpreoper- improves pathologic complete remission (pCR) rates breast trials show a significant relapse-free and overall ative chemotherapy on the outcome of women with when compared to every 3-week (Q 3 wk) P therapy survival advantage for sequential ECMF. Proc Am Soc operablebreastcancer.JClin Oncol1998;16:2672^85. (tx) followed by FAC- final results of a prospective Clin Oncol 2003;22:4. (abstract 13). 17. Wolmark N,Wang J, Mamounas E, Bryant J, Fisher B. phase III randomized trial. Proc Am Soc Clin Oncol 5. Holmes FA,Walters RS,Theriault RL, et al. Phase IItrial Preoperative chemotherapy in patients with operable 2002;21:35a (abstr 135). of taxol, an active drug in the treatment of metastatic breast cancer: nine-year results from National Surgical 29. Henderson IC, Berry DA, Demetri GD, et al. breast cancer. JNatl Cancer Inst1991;83:1797 ^ 805. Adjuvant Breast and Bowel Project B-18. J Natl Can- Improved outcomes from adding sequential Paclitaxel 6. Gianni L, Munzone E, Capri G, et al. Paclitaxel in cer Inst Monogr 2001:96 ^ 102. but not from escalating Doxorubicin dose in an adju- metastatic breast cancer: a trial of two doses by a 3- 18. Bonadonna G,Valagussa P,Brambilla C, et al. Prima- vant chemotherapy regimen for patients with node- hour infusion in patients with disease recurrence after ry chemotherapy in operable breast cancer: eight-year positive primary breast cancer. J Clin Oncol 2003;21: prior therapy with anthracyclines. J Natl Cancer Inst experience at the Milan Cancer Institute. J Clin Oncol 976 ^ 83. 1998;16:93^100. 1995;87:1169^75. 30. Mamounas EP, Bryant J, Lembersky BC, et al. Pacli- 7. Gianni L, Munzone E, Capri G, et al: Paclitaxel by 19. Taves DR. Minimization: a new method of assigning taxel after doxorubicin plus cyclophosphamide as patients to treatment and control groups. Clin Pharma- 3-hour infusion in combination with bolus doxorubicin adjuvant chemotherapy for node-positive breast col Ther 1974;15:443 ^ 53. in women with untreated metastatic breast cancer: cancer: results from NSABP B-28. J Clin Oncol 2005; high antitumor efficacy and cardiac effects in a dose- 20. CommonToxicity Criteria. http://ctep.cancer.gov/ 23:3686 ^ 96. reporting/ctc.html. finding and sequence-finding study. J Clin Oncol 31. Romieu G, Tubiana-Hulin M, Fumoleau P, et al. A 21. Hosmer DW, Jr., Lemeshow S. Applied logistic 19 9 5;13 : 26 8 8 ^ 9 9. multicenter randomized phase II study of 4 or 6 cycles regression. NewYork: JohnWiley & Sons, Inc.; 1989. 8. Gehl J, Boesgaard M, Paaske T, Vittrup Jensen B, of Adriamycin/Taxol (paclitaxel) (AT) as neoadjuvant 22. Biganzoli L, CuferT, Bruning P, et al. Doxorubicin Dombernowsky P. Combined doxorubicin and pacli- treatment of breast cancer. Ann Oncol 2002;13 Suppl and paclitaxel versus doxorubicin and cyclophospha- taxel in advanced breast cancer: effective and cardio- 5:33 ^ 4 (abstr 118). toxic. Ann Oncol 1996;7:687 ^ 93. mide as first-line chemotherapy in metastatic breast cancer: the European Organization for Research and 32. Gianni L, Capri G. Experience at the Istituto Nazio- 9. Gianni L,Vigano L, Locatelli A, et al. Human pharma- Treatment of Cancer 10961Multicenter Phase III Trial. naleTumori with paclitaxel in combination with doxo- cokinetic characterization and in vitro study of the JClinOncol2002;20:3114^21. rubicin in women with untreated breast cancer. Semin interaction between doxorubicin and paclitaxel in Oncol 1997;24 Suppl 3:S1^3. 23. Bear HD, Anderson S, Brown A, et al. National Sur- patients with breast cancer. J Clin Oncol 1997;15: 33. HutcheonAW,HeysSD,SarkarTK.Aberdeen 906^15. gical Adjuvant Breast and Bowel Project Protocol B- 27.The effect on tumor response of adding sequential Breast Group. Neoadjuvant docetaxel in locally 10. Minotti G, SaponieroA, Licata S, et al; Paclitaxel and preoperative docetaxel to preoperative doxorubicin advanced breast cancer. Breast Cancer Res Treat docetaxel enhance the metabolism of doxorubicin to and cyclophosphamide: preliminary results from 2003;79 Suppl 1:S19^ 24. toxic species in human myocardium. Clin Cancer Res National Surgical Adjuvant Breast and Bowel Project 34. Hutcheon AW, Heys SD, SarkarTK, et al. Docetaxel 2001;76:1511^ 5. Protocol B-27. J Clin Oncol 2003;21:4165 ^ 74. primary chemotherapy in breast cancer: a five year 11. Gianni L, Dombernowsky P, Sledge G, et al. Cardiac 24. von Minckwitz G, Raab G, CaputoA, et al. Doxorubi- update of the Aberdeen trial. Breast Cancer ResTreat function following combination therapy with paclitaxel cin with cyclophosphamide followed by docetaxel 2003;82 Suppl.1:S9 (abstr 11). and doxorubicin: an analysis of 657 women with every 21days compared with doxorubicin and doce- 35. Eiermann W, Sabadell D, Baselga J, et al. European advancedbreastcancer.AnnOncol2001;13:1067 ^ 73. taxel every14days as preoperative treatment in opera- cooperative trial in operable breast cancer (ECTO): no 12. Skipper HE. Kinetics of mammary tumor cell growth ble breastcancer: the GEPARDUOstudyof theGerman increased risk of local breast tumor recurrence (LBR) andimplications for therapy.Cancer1971;28:1479 ^99. Breast Group. JClin Oncol 2005;23:2676 ^ 85. as first and only event after primary systemic therapy 13. Fisher B, Gunduz N, Coyle J, Rudock C, Saffer E. 25. Martin M, PienkowskyT, Mackey J, et al. Adjuvant (PST).Proc AmSoc Clin Oncol 2003;22:10(abstr 37).

www.aacrjournals.org 8721 Clin Cancer Res 2005;11(24) December 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Feasibility and Tolerability of Sequential Doxorubicin/Paclitaxel Followed by Cyclophosphamide, Methotrexate, and Fluorouracil and Its Effects on Tumor Response as Preoperative Therapy

Luca Gianni, José Baselga, Wolfgang Eiermann, et al.

Clin Cancer Res 2005;11:8715-8721.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/11/24/8715

Cited articles This article cites 31 articles, 14 of which you can access for free at: http://clincancerres.aacrjournals.org/content/11/24/8715.full#ref-list-1

Citing articles This article has been cited by 13 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/11/24/8715.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/11/24/8715. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.