Node Positive Breast Cancer Pts Research Articles

257O - Intense Dose-Dense Epirubicin (E), Paclitaxel (T), Cyclophosphamide (C) (Iddetc) is Highly Effective in Adjuvant Therapy of Node-Positive Breast Cancer Patients (Pts). an Analysis of the Gbg, Ago-B, and Noggo Study Groups

ABSTRACT Aim: In locally advanced breast cancer pts (≥pN2), adjuvant iddETC resulted in superior disease-free (DFS) and overall survival (OS) compared to conventionally dosed EC > Tq3w (AGO-ETC; Mobus JCO 2010). In the GAIN trial, iddETC was compared to a 4-drug dose-dense regimen with capecitabine (ddEC > TwX). Using the GAIN data, here we want to confirm the iddETC-results of the AGO-ETC trial with regard to DFS, OS, and toxicity. Methods: In the AGO-ETC trial (1998-2003) 681 of 1,284, and in GAIN (2004-2008) 1,498 of 3,023 pts were treated with iddETC. Pts were included in AGO-ETC if ≥pN2 and in GAIN if ≥pN1; median number of positive nodes was 8 and 5, respectively; all other inclusion criteria were comparable. Median follow-up was 62 and 74 months, respectively, in both trials primary end point was DFS. IddETC consists of a sequence of 3 x E 150mg/m2, 3 x T 225 mg/m2, and 3 x C 2,5 g/m2 q2w, with primary G-CSF support. After 40% accrual in GAIN, a reduced dose of C (2,0 g/m2) and antibiotic prophylaxis was implemented. Pts were randomly assigned to anemia prophylaxis with ESF in AGO-ETC, and to adjuvant ibandronate in GAIN (von Minckwitz JCO 2013). Results: In the iddETC-arm of AGO-ETC, 5y-DFS was 70% (95% CI 68% - 75%) and 5y-OS was 82% (95% CI, 79%to 85%). Comparing to conventional EC > Tq3w, HR for DFS was 0.69 (95% CI, 0.57 - 0.84). In GAIN, 5y-DFS was 80% (95% CI 78% - 82%) and 5y-OS was 85%. DdEC > TwX was not superior to iddETC. Evaluating GAIN pts with ≥pN2 (n = 923) separately, in order to match with the AGO-ETC inclusion criteria, results were comparable: 5yr-DFS was 75% (95% CI 72% - 78%) and 5y-OS was 85% (95% CI 83% - 88%). In AGO-ETC, 47 pts (7%) had at least one episode of febrile neutropenia during iddETC, no toxic death was observed. In GAIN, three pts (0,2%) died from therapy related septicaemia during iddETC. In both trials, paclitaxel induced neuropathy was in the expected range, and no grade 3 congestive heart failure occurred. Treatment induced MDS/AML was rare in both, AGO-ETC (0,6%) and GAIN (0,3%). Conclusions: We consistently show high efficacy of adjuvant iddETC for node-positive breast cancer pts (≥pN2) in two consecutive prospective trials: DFS and OS-results were superior to those with conventionally dosed therapy. In the reported studies, iddETC was feasible without excessive toxicity. We suggest to add iddETC to the group of preferred regimens in current guidelines. Disclosure: C. Thomssen: has received compensated honoraria for presentations and participating in advisory boards from Roche, Amgen, Pfizer; G. von Minckwitz: has received research funds from Roche, BMS, Amgen, Speaker and consultancy honoraria from Roche and Amgen; H. Tesch: has received compensated honoraria from Roche; N. Harbeck has received compensated honoraria from Roche; V. Muller: has received speaker honoraria from Amgen, Celgene, Glaxo Smith Kline, Pierre-Fabre, Roche, Jannsen-Cilag I. Bauerfeind: speakers honoraria from Amgen and Bristol Myers Squibb; S. Loibl: has received compensated honoraria for presentations and participating in advisory boards from Roche, Amgen V. Moebus: has receivedCompensated honoraria from Amgen, Jannsen-Cilag, Roche for presentations and participation in advisory boards. All other authors have declared no conflicts of interest.

Pegfilgrastim (P) administration after 24, 72, or 96 hours (h) to allow dose-dense (DD) anthracycline- and taxane-based chemotherapy (CT) in breast cancer (BC) patients (pts): A single-center experience within the GIM2 randomized phase III study.

9632 Background: The GIM2 study (ClinicalTrials.gov number, NCT00433420) compared two different schedules of CT (DD versus standard duration) as adjuvant treatment of 2,091 node positive BC pts. DD...

Abstract S5-06: Epirubicin and cyclophosphamide (EC) followed by paclitaxel (T) versus fluorouracil, epirubicin and cyclophosphamide (FEC) followed by T, all given every 3 weeks or 2 weeks, in node-positive early breast cancer (BC) patients (pts). Final results of the gruppo Italiano mammella (GIM)-2 randomized phase III study

Abstract Background. Using a 2 × 2 factorial design, we compared the efficacy of two different schedules of chemotherapy (CT), dose-dense (DD) every 2 weeks versus standard duration every 3 weeks and the role of Fluorouracil (F) in addition to EC, in the adjuvant treatment of node positive BC pts. Patients and Methods. Node positive early BC pts younger than 70 years were eligible and randomly assigned to one of the following treatment arms: ARM A (4 courses of EC [90/600 mg/m2, on day 1, q 21] followed by 4 courses of T [175 mg/m2 on day 1, q 21]); ARM B (4 courses of FEC [600/90/600 mg/m2, on day 1, q 21] followed by 4 courses of T [175 mg/m2 on day 1, q 21]); ARM C (4 courses of EC [90/600 mg/m2, on day 1, q 14] followed by 4 courses of T [175 mg/m2 on day 1, q 14]); ARM D (4 courses of FEC [600/90/600 mg/m2, on day 1, q 14] followed by 4 courses of T [175 mg/m2 on day 1, q 14]). Pts enrolled in the DD arms (ARM C and D) received subcutaneous pegfilgrastim (6 mg) 24 hours after CT. HER2-positive pts enrolled after April 2006 received trastuzumab for 1 year at the end of CT. The primary study endpoint was disease free survival (DFS). Secondary endpoints included overall survival (OS) and toxicity. All analyses were conducted according to the intention-to-treat principle. Results. From April 2003 to July 2006, 2091 pts were enrolled (545 in ARM A, 544 in ARM B, 502 in ARM C, 500 in ARM D). The planned number of CT cycles was completed in 89% of pts in the DD group and in 88.1% of pts in the standard duration group. Pts treated with DD CT (ARM C + ARM D) experienced more frequently anemia (all grades: 67.2% vs 49.5%, p < .0001; G3-G4: 0.2% vs 1.4%, p = .004) and bone pain (53.5% vs 39.2%, p < .0001). Neutropenia occurred more frequently in standard duration arms (ARM A + ARM B): 66.7% vs 28.0%, p = < .0001. At a median follow up of 7.0 years, multivariate analysis, showed that DD treatments (ARM C + ARM D vs ARM A + ARM B) improved the primary endpoint, DFS (hazard ratio [HR] = 0.78; 95% CI 0.66-0.94; p = .007), and OS (HR = 0.68; 95% CI 0.52-0.87; p = .002). Conversely, the addition of F to EC did not improve clinical outcomes: the HR for DFS and OS (ARM A + ARM C vs ARM B + ARM D) were respectively 0.98 (95% CI 0.83-1.17; p = .85) and 0.93 (95% CI 0.73-1.19; p = .578). Subgroup analyses revealed that the benefits associated with DD CT were independent of hormone receptor status: in hormone receptor positive the HRs for DFS and OS were 0.80 (0.65-0.98) and 0.69 (0.51-0.93) respectively; in hormone receptor negative the HRs for DFS and OS were 0.67 (0.47-0.95) and 0.57 (0.36-0.92) respectively (p for interaction for DFS: 0.339; p for interaction for OS: 0.495). Pathological tumor size, nodal status (1-3 vs ≥ 4 positive nodes), histological grade and hormone receptor status were independently associate with DFS and OS. Conclusions. DD CT as adjuvant treatment in node positive BC pts, improves DFS and OS significantly. Conversely, the addition of F to EC does not improve clinical outcomes. (ClinicalTrials.gov number, NCT00433420). Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr S5-06.

Abstract S2-1: The role of sentinel lymph node surgery in patients presenting with node positive breast cancer (T0-T4, N1-2) who receive neoadjuvant chemotherapy – results from the ACOSOG Z1071 trial

Abstract Background: The utility of sentinel lymph node (SLN) surgery after neoadjuvant chemotherapy (NAC) in patients presenting with node-positive breast cancer has not been determined. The American College of Surgeons Oncology Group (ACOSOG) Z0171 trial was designed to evaluate SLN surgery after NAC in women presenting with node positive disease. Methods: ACOSOG Z1071 enrolled women with clinical T0-4, N1-2, M0 breast cancer receiving NAC. At the time of surgery, all patients were to undergo SLN surgery followed by axillary lymph node dissection (ALND). The primary endpoint was false negative rate (FNR) in women with cN1 disease with 2 or more SLNs reviewed. Positive SLNs were defined as metastases >0.2mm on H&E. The protocol encouraged dual tracer technique. A Bayesian study design with a non-informative prior was chosen to assess whether the probability that the SLN surgery FNR is greater than 10%. Results: From July 2009 to July 2011, 756 patients were enrolled from 136 institutions. Fifteen women were ineligible and 33 withdrew. Of 708 evaluable pts (668 cN1, 40 cN2), 643 had a SLN identified and an ALND (607 cN1, with indeterminate SLN results in 2); 52 pts (48 cN1) had no SLN identified and had ALND; 11 underwent ALND only (all cN1), and 2 pts had SLN only (both cN1). In patients with SLN and ALND, the SLN identification rate was 92.5% (92.7% in cN1, 90% in cN2). SLN correctly identified nodal status in 84% of the 695 pts [258 of pathologically node negative and 327 of pathologically node positive; cN1: 83.8% (549/655), cN2: 90.0% (36/40)]. Of the 643 pts with a SLN identified there was a complete pathologic response in 40.3% (40.3 % for cN1 and 50% for cN2). Of the pts with a positive SLN, the SLN was the only site of disease in 40%. For pts with cN1 disease with 2+ SLNs identified with residual nodal disease, the SLN FNR was 12.8%. In pts with dual tracer technique the FNR was 11.1%. There were no FN results among pts with cN2 disease with 2+ SLNs reviewed. Of the 40 pts with a false negative SLN of the 528 cN1 patients with 2+ SLNs examined, the number of positive nodes at ALND was 1 (50.0%); 2 (25%); 3 (10.0%) and 4–9 (15.0%). Conclusions: NAC resulted in eradication of lymph node disease in 40% of node positive breast cancer patients. SLN surgery after NAC in node positive breast cancer pts correctly identified nodal status in 84% of all patients and was associated with a FNR of 12.8%. The FNR of SLN is higher than the prespecified study endpoint of 10%. Further analysis of factors associated with FNR such as clinical response, histological findings and axillary ultrasound findings is warranted prior to widespread use of SLN in these patients. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr S2-1.

18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) as early imaging biomarker of axillary sentinel lymph node (SLN) status in locally advanced node-positive breast cancer (NPBC) patients (pts) receiving neoadjuvant chemotherapy (NACT).

1129 Background: Use of FDG-PET in the early evaluation of NACT for NPBC is currently under investigation. The aim of this study is to assess FDG-PET in NPBC pts receiving NACT in order to identify a subset of pts that can be spared axillary lymph node dissection (ALND) in case of response. Methods: All pts (period 2009-2011) had [cT2 (≥3cm)-T4, pN1-3], and no prior BC treatment. All pts received 8 cycles (cy) of NACT +/- trastuzumab. FDG-PET was performed at baseline and after 2 cy. After NACT, mastectomy with SLN and ALND was peformed. As primary endpoint, the change in maximum standardized uptake value (SUVmax) in the primary tumor (PT) and ALNs was compared with pathological response (Jonckheere-Terpstra test). Logistic regression was used to determine the predictive value of a 50% reduction in SUVmax on pathological response and SLN status. Results: Forty-one pts were evaluable for the primary endpoint and 31 pts had successful SN procedures. The median age was 49.8 years (range 27-75). Overall, 9.8% (4/41; 95% CI 2.7 – 23.1) of pts had progressive disease (PD), 7.3% (3/41 ; 95% CI 1.5 – 19.9 ) stable disease (SD), 53.7% (22/41; 95% CI 37.4 – 69.3) partial response (PR), and 21.9% (9/41; 95% CI 10.6 – 37.6) a complete pathological response (pCR). A linear trend existed between pCR and higher decreases in SUVmax of the PT site (p=0.008), but not with lymph node SUVmax (p=0.294). The odds of achieving a pCR, increased significantly when SUVmax of the PT decreased with >50% (OR 10.7; 95% CI 1.2 - 98.0; p=0.03), but not lymph node SUVmax (OR 4.75; 95% CI 0.82 - 27.5; p=0.08). Achieving a true negative SLN status was more likely with >50% reductions in PT SUVmax (OR 41.2; 95% CI 4.0 - 421.9; p= 0.002). A reduction of >50% in nodal SUVmax was not predictive of negative SLN status (OR 3.33; 95% CI 0.66 - 16.8; p=0.1). Conclusions: Early changes in PT metabolism imaged with FDG-PET after only 2 cy of NACT is a promising biomarker in the management of the axilla in NPBC. Molecular imaging of PT biology and to a lesser extent of axillary lymph nodes using FDG-PET warrants further study.

Determinants of outcome in elderly patients with positive sentinel lymph nodes

614 Background: Older women are less likely to receive standard of care treatment for breast cancer. This undertreatment may be linked to the perception that elderly patients (pts) may not tolerate or derive benefit from treatment that is often viewed as “too toxic.” Methods: From July 1997 to July 2003, 1,470 pts with invasive cancer with positive sentinel lymph nodes (PSLN) by intraoperative frozen section or final pathologic exam by hematoxylin-eosin and/or immunohistochemistry (IHC). We examined sociodemographic, pathologic, and therapeutic variables that affected the outcome of elderly pts ≥70 years old. A comorbidity score was assigned using Adult Comorbidity Evaluation-27 system. Chi-square, Fisher's exact Wilcoxon tests, and multivariate logistic regression analysis were used. Results: The median age was 53 years (range 21–89) and median tumor size 1.6 cm (range 0.1–11). 1,265 (86.1%) pts were <70 years old and 205 (13.9%) pts were ≥70 years old. 183 (12.5%) pts had IHC only PSLN, 1,021 (69.5%) had 1–3 PSLN, and 266 (18%) ≥4 PSLN. Breast conservation therapy (BCT) was performed in 59.7% of pts. Elderly pts were more likely to have moderate to severe comorbidities (46% vs. 11%, p < 0.0001) and BCT (67% vs. 59%, p = 0.026), compared to pts <70 years old. Elderly pts were less likely to undergo completion ALND (73% vs. 89%, p < 0.0001), adjuvant chemotherapy (43% vs. 90%, p < 0.0001) and radiotherapy following BCT (82% vs. 90%) compared to pts <70 years old. The 5-year disease-free survival (DFS) was not significantly different for elderly pts compared to non-elderly pts (87.7% vs. 91.9%, p = 0.21), on the other hand the 5-year overall survival was significantly worse for elderly pts (80.4% vs. 93.2%, p < 0.0001), a difference that was mainly due to a significantly higher 5-year cumulative incidence of death due to other causes (13.2% vs. 1.9%, p < 0.0001). On multivariate analysis, ER or PR positive status, IHC only PSLN and T1 tumors were the only factors independently associated with improved odds of 5-year DFS. Conclusions: Tumor rather than patient factors were the primary determinants of breast cancer outcomes in our cohort of node positive breast cancer pts. Elderly breast cancer pts with an estimated life expectancy beyond 5 years should receive the same standard of care therapy as their younger counterparts. No significant financial relationships to disclose.

Toxicity and health-related quality of life (HRQoL) in node-negative breast cancer (BC) patients (pts) receiving adjuvant treatment with TAC (docetaxel, doxorubicin, cyclophosphamide) or FAC (5-fluorouracil, doxorubicin, cyclophosphamide): Impact of adding prophylactic growth factors (GF) to TAC. GEICAM Study 9805

604 Background: Node-positive BC pts treated with TAC have a longer disease-free and overall survival but a higher rate of febrile neutropenia (FN) compared with pts treated with FAC. We evaluated the toxicity and HRQoL of both regimens and assessed the impact of adding prophylactic granulocyte GF to TAC.Methods: 1059 pts with operable, high-risk (St Gallen, 1998), node-negative BC, 18–70 yrs old, were randomized to FAC (F 500 mg/m2, A 50 mg/m2, C 500 mg/m2) or TAC (T 75 mg/m2, A 50 mg/m2, C 500 mg/m2) day 1 every 3wks for 6 cycles. After enrolling 224 pts, the study was amended to require prophylactic GF for pts subsequently treated with TAC, but not FAC. The EORTC QLQ-C30 was administered at baseline, after each chemotherapy cycle and 6, 12 and 24 months after completion of chemotherapy. Rates of FN (fever ≥gr 2 with gr 4 neutropenia), severe asthenia, gr 3–4 diarrhea, and gr 2–4 mucositis were analyzed. Results: 530 pts received TAC, 114 without GF [TAC] and 416 with GF [TAC+G]; 520 received FAC. During chemotherapy, HRQoL was worse with TAC than with FAC (p=0.008), with a maximum difference of 8.0 points during cycle 4. However, no statistically significant differences in HRQoL were observed between TAC+G and FAC, with 6.1 points as the highest difference during cycle 3. During the follow-up period, no statistically significant differences in HRQoL were observed among TAC alone, TAC+G and FAC. The incidence[95%CI] of: FN was 2.3%[1.0–3.6%] for FAC, 24.6%[16.7–32.5%] for TAC, and 5.8%[3.53–8.01%] for TAC+G; gr 3–4 diarrhea 0.8% FAC, 7.0% TAC, and 2.6% TAC+G (p=0.0415); severe asthenia 1.8% FAC, 20.2% TAC, and 5.5% TAC+G (p<0.0001); gr 2–4 mucositis 24.4% FAC, 35.1% TAC, and 23.3% TAC+G (p=0.011). Conclusions: TAC transiently worsens HRQoL compared with FAC during chemotherapy. However, BC pts treated with TAC plus GF support experience the same HRQoL as pts treated with FAC. The addition of GF to TAC significantly reduces the incidence of FN, gr 3–4 diarrhea, severe asthenia, and gr 2–4 mucositis compared with TAC alone. No significant financial relationships to disclose.

Spanish Breast Cancer Research Group (GEICAM) population-based study on breast cancer outcomes (El Álamo project): Analysis of operable, node positive breast cancer patients

6045 Background: El Álamo project is a retrospective analysis of 15381 patients (pts) diagnosed of breast cancer (BC) across 43 of the main Spanish hospitals. Methods: Two cohorts were analysed: Álamo I (AI) consisted of 4532 pts diagnosed with BC between 1990 and 1993; Álamo II (AII) consisted of 10849 pts diagnosed between 1994 and 1999. AI and AII data were collected by end of year (yr) 2000 and 2003, respectively. 158 items per pt were recorded, including demographic, therapeutic and outcome data. Pts with operable, node-positive (NP) disease were selected. Results: In AI project, 51% of all pts presented with NP disease, vs. 45% in AII. Median age at diagnosis was 56 yr (AI) vs. 57 yr (AII); 63.1% vs. 63.5% of pts were postmenopausal. Percentage of pts with 1–3 NP increased from 52.2% (AI) to 54.5% (AII); whereas it decreased for pts with 4–9 NP (30% in AI vs. 26.7% in AII); proportion of pts with 3 10 NP was similar in both groups (15.5% in AI vs. 15.7% in AII). Regarding treatment, 99% of pts in both AI and AII received adjuvant therapy (AT). AT in AI group were: chemotherapy (CT) (70.4%), hormone therapy (HT) (60.6%) and radiotherapy (RT) (44.7%), usually in combination. AT in AII group were: CT (82.2%), HT (75.5%) and RT (59.2%), usually in combination. 30 pts (2.2%) were treated with high dose CT in AI vs. 296 pts (9.1%) in AII. 51.3% of all pts in AI received an anthracycline containing regimen as CT vs. 61.6% in AII. BC recurrence rate has diminished from 46.4% (AI) to 29.6% (AII). The respective nine-year survival rates for AI and AII were 58.9% (CI95%; 56.3%–61.4%) and 68.1% (CI95%; 65.5%–70.7%).Conclusions: The relative population of operable, NP BC pts has decreased by 6% in Spain between 1990 and 1997. This seems to be the main cause of outcome improvement in this group of pts, although a contribution of therapy cannot be ruled out. No significant financial relationships to disclose.

Dose-dense sequential chemotherapy with epirubicin(E), paclitaxel (T) and cyclophosphamide (C) (ETC) is superior to conventional dosed chemotherapy in high-risk breast cancer patients (≥ 4 +LN). First results of an AGO-trial

513 Background: Dose-dense chemotherapy will become a new treatment option in breast cancer patients (pts) (INT C9741). Here we report the initial results of a second prospective randomized trial that compares dose-dense vs. conventionally scheduled sequential chemotherapy in high-risk node positive pts. Methods: A multi-center phase-III trial of the German “Arbeitsgemeinschaft fuer Gynaekologische Onkologie (AGO)” recruited 1284 pts from 12/98 until 4/03. Pts below 65 years of age were eligible if they had at least 4 involved axillary lymph nodes. In the experimental arm, pts were assigned to receive three courses each of epirubicin(150 mg/m2), paclitaxel (225 mg/m2) and cyclophosphamide (2500 mg/m2) at 2 weeks interval (q2wks) (ETC) with G-CSF support (5μg/kg/SC day 3–10). A second randomization ± Epoetin-alpha was performed (150 IU/kg/SC three times weekly). In the standard arm 4 courses of conventional dosed epirubicin/cyclophosphamide (90/600 mg/m2) followed by 4 courses of paclitaxel (175 mg/m2) (q3wks) were given. Results: Median patient age was 50 years. Risk factors were well balanced in both arms. 59% and 41% of the pts had 4–9 and ≥10 infiltrated nodes. The median number of positive nodes was 8. In this second interim analysis 1169 pts were fully evaluable. No unusual toxicities were seen, especially no severe cardiotoxicity. Hematological toxicity (anemia, neutropenia and thrombopenia) was more frequent in the ETC arm (p<0,0001) and varied distinctly between drugs. The incidence was highest during treatment with cyclophosphamide and lowest during treatment with paclitaxel. 7% of pts in the ETC arm vs. 2% in the standard arm were hospitalized for febrile neutropenia (p<0,0001). There were no treatment-related deaths during therapy. 82% of pts in the ETC arm vs. 90% in the standard arm received the planned number of cycles. The number of dose reductions was relatively small, ranging from 6,5% on the ETC arm to 2% on the standard arm. At a median follow-up of 28 months 94 pts in the ETC arm vs. 127 pts in the standard arm had relapsed. This difference was highly significant in the two-sided log-rank test (p=0,0009) with a HR of 0,64 (95% CI: 0,49–0,83). Estimated 3-year RFS is 80% in the ETC arm vs. 70% in the standard arm. 103 pts have died (43 events vs. 60 events) favoring again the ETC arm (two-sided log-rank test: p=0,030; HR: 0,65 (95% CI: 0,44–0,96). Estimated 3-year OAS is 90% in the ETC arm vs. 87% in the standard arm. There was a highly significant difference in anemia grade 1–4 as well as in the number of RBC's transfusions between ETC ± Epoetin alpha (p<0,0001). No difference in DFS (p=0,84) or OAS (p=0,98) was found between ETC ± Epoetin alpha. Conclusions: Evidence to date indicates that the dose-dense ETC-regimen as adjuvant therapy in high-risk breast cancer pts is well tolerated and significantly improves DFS and OAS. Confirming the results of INT C9741, dose-dense sequential regimens may become a future standard option in the adjuvant treatment of node positive breast cancer pts. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Bristol-Myers Squibb; Amgen; Pharmacia; Ortho-Biotech

Dose-dense sequential chemotherapy with epirubicin(E), paclitaxel (T) and cyclophosphamide (C) (ETC) is superior to conventional dosed chemotherapy in high-risk breast cancer patients (≥ 4 +LN). First results of an AGO-trial

513 Background: Dose-dense chemotherapy will become a new treatment option in breast cancer patients (pts) (INT C9741). Here we report the initial results of a second prospective randomized trial that compares dose-dense vs. conventionally scheduled sequential chemotherapy in high-risk node positive pts. Methods: A multi-center phase-III trial of the German “Arbeitsgemeinschaft fuer Gynaekologische Onkologie (AGO)” recruited 1284 pts from 12/98 until 4/03. Pts below 65 years of age were eligible if they had at least 4 involved axillary lymph nodes. In the experimental arm, pts were assigned to receive three courses each of epirubicin(150 mg/m2), paclitaxel (225 mg/m2) and cyclophosphamide (2500 mg/m2) at 2 weeks interval (q2wks) (ETC) with G-CSF support (5μg/kg/SC day 3–10). A second randomization ± Epoetin-alpha was performed (150 IU/kg/SC three times weekly). In the standard arm 4 courses of conventional dosed epirubicin/cyclophosphamide (90/600 mg/m2) followed by 4 courses of paclitaxel (175 mg/m2) (q3wks) were given. Results: Median patient age was 50 years. Risk factors were well balanced in both arms. 59% and 41% of the pts had 4–9 and ≥10 infiltrated nodes. The median number of positive nodes was 8. In this second interim analysis 1169 pts were fully evaluable. No unusual toxicities were seen, especially no severe cardiotoxicity. Hematological toxicity (anemia, neutropenia and thrombopenia) was more frequent in the ETC arm (p<0,0001) and varied distinctly between drugs. The incidence was highest during treatment with cyclophosphamide and lowest during treatment with paclitaxel. 7% of pts in the ETC arm vs. 2% in the standard arm were hospitalized for febrile neutropenia (p<0,0001). There were no treatment-related deaths during therapy. 82% of pts in the ETC arm vs. 90% in the standard arm received the planned number of cycles. The number of dose reductions was relatively small, ranging from 6,5% on the ETC arm to 2% on the standard arm. At a median follow-up of 28 months 94 pts in the ETC arm vs. 127 pts in the standard arm had relapsed. This difference was highly significant in the two-sided log-rank test (p=0,0009) with a HR of 0,64 (95% CI: 0,49–0,83). Estimated 3-year RFS is 80% in the ETC arm vs. 70% in the standard arm. 103 pts have died (43 events vs. 60 events) favoring again the ETC arm (two-sided log-rank test: p=0,030; HR: 0,65 (95% CI: 0,44–0,96). Estimated 3-year OAS is 90% in the ETC arm vs. 87% in the standard arm. There was a highly significant difference in anemia grade 1–4 as well as in the number of RBC's transfusions between ETC ± Epoetin alpha (p<0,0001). No difference in DFS (p=0,84) or OAS (p=0,98) was found between ETC ± Epoetin alpha. Conclusions: Evidence to date indicates that the dose-dense ETC-regimen as adjuvant therapy in high-risk breast cancer pts is well tolerated and significantly improves DFS and OAS. Confirming the results of INT C9741, dose-dense sequential regimens may become a future standard option in the adjuvant treatment of node positive breast cancer pts. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Bristol-Myers Squibb; Amgen; Pharmacia; Ortho-Biotech