Efficacy and safety of neoadjuvant trastuzumab combined with paclitaxel and epirubicin

A retrospective review of the M. D. Anderson experience


  • Shaheenah Dawood MBBCh, MRCP,

    1. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    2. Department of Medical Oncology, Department of Health and Medical Services, Dubai Hospital, Dubai, United Arab Emirates
    Search for more papers by this author
  • Ana M. Gonzalez-Angulo MD,

    1. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Florentia Peintinger MD,

    1. Department of Pathology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Kristine Broglio MS,

    1. Department of Quantitative Sciences, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • William F. Symmans MD,

    1. Department of Pathology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Shu-Wan Kau MD,

    1. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Rabiul Islam MD,

    1. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Gabriel N. Hortobagyi MD,

    1. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Aman U. Buzdar MD

    Corresponding author
    1. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    • Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030
    Search for more papers by this author
    • Fax: (713) 383-0234



A previously published prospective randomized phase 3 trial showed that administration of 24 weeks of primary systemic chemotherapy (PST) with paclitaxel and FEC75 (fluorouracil, epirubicin, cyclophosphamide) concurrently with trastuzumab in patients with HER2-positive primary breast cancer resulted in a 60% pathologic complete response rate (PCR) with no associated severe cardiac toxicity. The purpose of this study was to review the efficacy and safety of a similar regimen outside the setting of a clinical trial.


Patients with HER2-positive breast cancer (defined as either immunohistochemical 3+ or fluorescence in situ hybridization-positive) that had received 24 weeks of neoadjuvant trastuzumab concurrently with taxane and anthracycline-based chemotherapy between 2004 and 2006 were included in the analysis. PST chemotherapy consisted of paclitaxel (80 mg/m2) weekly for 12 weeks followed by 4 cycles of FEC75 (500 mg/m2, 75 mg/m2, and 500 mg/m2, respectively).


Forty patients were identified. The median age was 48 years (range, 29–81). In all, 60% of patients had stage III disease and 4 had inflammatory breast cancer. The PCR rate was 55% (95% confidence interval [CI], 38.5%–70.7%). At a median follow-up of 19 months. 5 patients had a recurrence, of which 4 did not achieve a PCR. No severe cardiac events were observed.


Stage II and III HER2-positive breast cancer patients achieved a high rate of PCR with trastuzumab given concurrently with paclitaxel and FEC75 chemotherapy. No severe cardiac events were observed with the regimen. The data concur with the results of a previously published trial. Cancer 2007. © 2007 American Cancer Society.

The evolution of the multidisciplinary management approach combined with the administration of polychemotherapy, incorporating anthracyclines and taxanes, has resulted in significant reductions in the risk of death with subsequent improvements in 5-year survival rates for patients with early and advanced stage breast cancers.1, 2 The introduction of trastuzumab, a humanized monoclonal antibody directed against the extracellular domain of the HER2 receptor3 has resulted in further improvements in survival of patients whose tumors overexpress HER2 in both the metastatic3 and adjuvant setting.4–7 Cardiac toxicity is the main side effect of trastuzumab, with cardiac dysfunction observed in 27%3 of patients receiving trastuzumab concomitantly with an anthracycline-based regimen and in 1.7% to 4.1% of patients receiving trastuzumab for 1 year either after chemotherapy or starting with the taxane phase of a sequential chemotherapy regimen.4, 5

Administering primary systemic chemotherapy (PST) has several advantages including downstaging the primary tumor and thus allowing higher rates of breast-conserving surgeries, as well as providing an in vivo assessment of tumor chemosensitivity. In addition, PST allows for determination of the pathologic complete response (PCR), considered a surrogate marker for long-term disease-free survival.8 Keeping these factors in mind, and the finding that the combination of trastuzumab with an anthracycline has significant activity, we recently published the results of a phase 3 study9, 10 where patients with early-stage operable breast cancer received PST with 4 cycles of continuous infusion paclitaxel followed by 4 cycles of fluorouracil, epirubicin, cyclophosphamide (FEC) concomitantly with weekly trastuzumab. Epirubicin in this trial was given at a reduced dose of 75 mg/m2. With this regimen we achieved a 60% PCR rate with only 10% of patients developing grade 1 and 2 cardiac toxicity and only 1 patient developing symptomatic heart failure.

Patients who receive PST but do not attain a PCR have various prognoses depending on the extent of the response to chemotherapy. We recently developed a method of dividing such patients into separate prognostic groups based on the Residual Cancer Burden (RCB).11 This index, using pathologic measurements of residual primary tumor and nodal metastases, essentially divides patients into 4 prognostic groups: RCB 0 where a PCR is attained, RCB I with minimal residual disease and considered low risk for recurrence, RCB II with intermediate residual disease and considered intermediate risk for recurrence, and RCB III considered to be chemoresistant and at high risk for recurrence. The purpose of this study was to assess the efficacy, in terms of the PCR rate and RCB category, and safety of a preoperative trastuzumab/anthracycline chemotherapy regimen given outside the controlled environment of a clinical trial.


Patient Population

Our breast cancer database was searched to identify patients who had histologically confirmed, HER2-positive (defined as either immunohistochemical 3+ or fluorescence in situ hybridization [FISH]-positive), nonmetastatic invasive tumor who had received PST, with the regimen described below, concomitantly with trastuzumab. Patient clinical information, tumor characteristics, response rates, and cardiac toxicity information were recorded. The Institutional Review Board approved this retrospective review.


Paclitaxel was administered weekly at a dose of 80 mg/m2/week for a total of 12 weeks. This was followed by 4 cycles FEC75 (fluorouracil 500 mg/m2 intravenously, epirubicin 75 mg/m2, cyclophosphamide 500 mg/m2, given on the first day of each cycle) administered every 3 weeks. Trastuzumab was administered as a loading dose of 4 mg/kg intravenously over 90 minutes on the first day and then subsequently given weekly at a dose of 2 mg/kg over 30 minutes for a total of 24 weeks, concomitantly with the above chemotherapy. After completion of PST, patients underwent adequate surgery (lumpectomy or mastectomy) followed by radiation therapy and appropriate endocrine therapy if required. In cases where trastuzumab was continued after surgery it was administered at a dose of 4 mg/kg intravenously every 3 weeks for an additional 6 months.

Pathologic Analysis

PCR was defined as no evidence of invasive residual disease in the breast and axilla. A dedicated breast cancer pathologist evaluated all available specimens to record the variables necessary to calculate RCB. Variables included: 1) primary tumor bed area; 2) overall cancer cellularity as a percentage of the primary tumor bed area; 3) percentage of in situ disease; 4) number of lymph nodes involved with disease; and 5) diameter of largest metastases within the involved lymph node. These variables were then entered into the RCB calculator, available on the M. D. Anderson website (http://www.mdanderson.org/breastcancer_RCB), which then classified patients into 4 categories: RCB class 0, RCB class I, RCB class II, or RCB class III.

Cardiac Toxicity

When available, information on ejection fraction pre- and postchemotherapy was recorded. Severe cardiac events were defined as a symptomatic drop in ejection fraction of any percentage or congestive heart failure.

Statistical Analysis

Recurrence was defined as any recurrence of disease in either locoregional or distant sites. Recurrence-free survival (RFS) was defined as the time from diagnosis to the time of first recurrence or last follow-up. Percentages of PCR and RCB were reported with their exact binomial 95% confidence intervals (CIs).


Between January 2004 and June 2006, 40 patients were treated with the regimen described above. This number excludes patients treated on the clinical trial.9, 10 Patient pretreatment characteristics are listed in Table 1. The median age was 48 years (range, 29–81). All but 2 patients had FISH testing to confirm HER2 positivity. Fifty-two percent of patients had estrogen receptor-positive disease. Thirty-eight patients had invasive ductal carcinoma and 2 patients had invasive lobular disease. The mean breast tumor size was 4 cm. More than half (60%) had stage III disease, with 4 patients presenting with inflammatory breast cancer (IBC). Postchemotherapy 31 (77.5%) patients underwent mastectomies, whereas 9 (22.5%) patients had segmental mastectomies. Sentinel lymph node biopsy was performed in 8 (20%) patients and axillary lymph node dissection in 32 (80%) patients. An average of 15 lymph nodes was dissected in patients who underwent an axillary lymph node dissection.

Table 1. Characteristics of Patients Receiving Neoadjuvant Trastuzumab (Herceptin) (n = 40)
CharacteristicNo. of patients
Median age, y (range)48 (29–81)
Menopause status
Estrogen receptor status
Breast tumor size, cm
 Median (range)3.3 (0–7)
Clinical stage
Tumor stage at presentation
 T4d (inflammatory)4
Node stage at presentation


Table 2 summarizes response rates. All patients had an objective response to therapy. Twenty-two patients (55%, 95% CI, 38.5%–70.7%) achieved a PCR (Fig. 1). A partial response was observed in 18 patients (45%, 95% CI, 29.3%–61.5%). Of the 18 patients who had a partial response, 2 patients had no residual disease in the breast and 5 had isolated tumor cells in either the breast or the lymph nodes. Specimens to calculate RCB were available for 38 patients (95%) (Fig. 2). Twenty-two patients (57.9%, 95% CI, 40.8%–73.7%) were categorized as RCB class 0, 4 patients (10.5%, 95% CI, 2.9%–24.8%) as RCB class I, 11 patients (28.9%, 95% CI, 15.4%–45.9%) as RCB class II, and 1 patient (2.6%, 95% CI, 0.7%–13.8%) as RCB class III.

Figure 1.

Pathologic complete response (PCR) by stage (n = 22).

Figure 2.

Residual cancer burden (n = 38).

Table 2. Response Rate According to Estrogen Status
 PCR (%)Partial response (%)
  1. PCR indicates pathologic complete response.

Total22 (55)18 (45)
Estrogen receptor-positive10 (48)11 (52)
Estrogen receptor-negative12 (63)7 (37)


At a median follow-up 19 months (range, 7–31 months) 5 (12%) patients have recurred. The median RFS has not yet been attained. Four of the patients who recurred did not achieve a PCR, 1 had primary inflammatory breast cancer, and 1 has died. Brain was the first site of metastases in 4 patients, whereas liver/bone was the first site in 1 patient. At 2 years the RFS was 74.5% (95% CI, 55.2%–100%).

Safety and Tolerance

Thirty-nine (97.5%) patients completed 24 weeks of chemotherapy with concomitant trastuzumab, with 1 patient abandoning treatment because of poor tolerance to FEC. One patient developed febrile neutropenia during FEC administration requiring prophylactic G-CSF with no dose modification necessary.

During this review it was noted that several patients had cardiac evaluation performed outside the institution with ejection fraction values pre- and postchemotherapy with trastuzumab available in only 21 patients. Five patients had an asymptomatic drop of ejection fraction of greater than 10%, which on follow-up reverted to pretreatment values. No patient experienced symptomatic congestive heart failure. At the time of this analysis 21 patients continued to receive trastuzumab postsurgery, with 50% completing 1 year of therapy. No adverse symptomatic cardiac events have been observed thus far.


The objective of this retrospective analysis was to determine the efficacy and safety of adding trastuzumab to sequential PST composed of paclitaxel and FEC75 to patients with nonmetastatic breast cancer outside the setting of a clinical trial. In this study we attained a PCR rate of 55%, similar to the PCR rate (60%) achieved in our earlier phase 3 randomized study.9

Previous studies incorporating trastuzumab into primary systemic chemotherapeutic regimens have achieved PCR rates ranging from 12% to 23%.12–14 The higher rate observed in our study may be explained by the longer (24 weeks) duration of preoperative trastuzumab administered, sequential use of PST, and the combination of trastuzumab with the anthracycline-based regimen FEC75. When the results of our earlier phase 39, 10 study were pooled with this retrospective study, a total of 85 patients who received neoadjuvant trastuzumab concomitant with paclitaxel and FEC75 were obtained, of which more than half (52 patients) achieved PCR (61.1%, 95% CI, 40.0%–71.6%). In addition, high PCR rates were observed regardless of stage of disease.

In our earlier study9 paclitaxel was administered every 3 weeks as a continuous infusion. Greene et al.15 compared weekly paclitaxel to a 3-weekly regimen for a total of 12 weeks followed by 4 cycles of FAC (fluorouracil, doxorubicin, cyclophosphamide). Higher PCR rates were observed with the weekly paclitaxel schedule (28% vs 16%; P = .02). Despite the finding that all the patients in this study received weekly paclitaxel, PCR rates did not increase further than that observed in our earlier study,9, 10 which may be explained by the finding that a larger proportion of patients in this study had more advanced disease than in our earlier study.9, 10 This suggests that the duration of therapy, the use of an anthracycline, method of paclitaxel administration, and, most important, the use of trastuzumab are the determining factors for PCR.

PCR is considered an important endpoint in trials assessing efficacy of primary systemic regimens, as it has been shown in several studies to correlate with improved long-term outcomes.8, 16 Kuerer et al.8 showed that in patients with locally advanced breast cancer PCR predicted for improved 5-year overall and disease-free survival rates (89% and 87%, respectively) compared with patients who did not achieve a PCR (64% and 58%, respectively). In the study by Hurley et al.,12 48 patients with HER2-positive locally advanced breast cancer or IBC received PST with 12 weeks of docetaxel, cisplatin, and trastuzumab followed by surgery and adjuvant chemotherapy with doxorubicin and cyclophosphamide. They attained a PCR rate of 17%, not very different from an earlier study that used the same regimen without trastuzumab (20%).17 However, they observed that the addition of trastuzumab substantially improved 4-year progression-free (81% vs 65%) and overall survival (86% vs 73%), which agrees with the impressive survival benefit seen with the addition of trastuzumab to adjuvant chemotherapy.4–7 The high PCR (55%) rate seen in our study occurred despite the finding that more than half (57%) the patients had stage III disease, and 4 had IBC, which may, on further follow-up, translate to even better survival rates than those observed so far in this group of patients. In our previously mentioned clinical trial, only 26% had stage III breast cancer; because the PCR rate has been reported to be inversely related to initial tumor size, the high PCR rate in this more recent series is somewhat unexpected, and might be associated with the use of weekly paclitaxel in this series.

Several studies8, 18 have also shown an inverse relation between long-term outcomes of patients who receive PST that do not achieve a PCR and the amount of residual disease left behind, essentially dividing them into different prognostic groups. To better define these groups, Symmans et al.11 reviewed postoperative pathology specimens of 432 patients that had received PST chemotherapy with either FAC alone or paclitaxel and FAC. Variables including size and cellularity of breast residual tumor and number and size of residual nodal metastases were used to calculate RCB that classified patients into 4 classes (0-III) ranging from no evidence of disease to presence of bulky disease. RCB was found to be a better predictor of identifying high-risk patients than standard AJCC (6th ed.) clinical staging with class III representing high-risk chemoresistant patients. In our study the majority of patients who did not achieve a PCR were classified as RCB class II and I with only 1 patient classified as RCB class III, reflecting the efficacious nature of the regimen used.

Through the use of gene expression profiling with Affymetrix U133A gene chips, performed on 21 samples obtained in our earlier phase 3 randomized study, we showed that clinical variables that have traditionally been used to predict PCR such as estrogen receptor status, grade, tumor size, and nodal status lost their predictive value when trastuzumab was added to primary systemic chemotherapy.19 This concurs with our results, as we achieved similar PCR rates in both estrogen receptor-positive and -negative patients.

Although serial cardiac evaluation information was not available for all patients in this retrospective study, the low incidence of symptomatic cardiac events is similar to that observed in our previous randomized trial.8 Encouraged by the safety observed in our earlier study,8 21 patients went on to receive adjuvant trastuzumab with no symptomatic cardiac events reported so far. The recent adjuvant trials have raised the question about the optimal duration of trastuzumab, with 3 trials4–6 using 1 year of trastuzumab and 1 trial7 administering only 9 weeks of trastuzumab. A fourth pilot study from the Eastern Cooperative Oncology Group (ECOG)20 that randomized patients to either a short course (10 weeks) of trastuzumab versus a long course (52 weeks) of trastuzumab did not show any difference in disease-free or overall survival at a median follow-up of 64 months. Randomized trials with sufficient power are needed to evaluate whether a shorter, less cardiotoxic course of trastuzumab will be as efficacious.

In conclusion, the administration of PST with paclitaxel and FEC chemotherapy concurrently with trastuzumab in our cohort produced high PCR rates with no observed symptomatic cardiac events. The majority of patients who did not achieve a PCR had minimal to moderate amounts of residual disease. Although this regimen requires evaluation in a large prospective randomized trial to further define cardiac safety, it may be considered in individual patients with locally advanced or IBC to improve the chances of achieving a PCR and long-term outcomes.


Supported in part by the Susan G. Komen Foundation and the Nellie B. Connally Fund for Breast Cancer Research.