Systemic therapy of inflammatory breast cancer from high-dose chemotherapy to targeted therapies

The French experience


  • Patrice Viens MD,

    1. Department of Medical Oncology, Paoli-Calmettes Institute, Marseille Cancer Research Center, UMR891 Inserm, IFR137, Marseille, France
    2. University of the Mediterranean, Marseille, France
    Search for more papers by this author
  • Carole Tarpin MD,

    1. Department of Medical Oncology, Paoli-Calmettes Institute, Marseille Cancer Research Center, UMR891 Inserm, IFR137, Marseille, France
    Search for more papers by this author
  • Henri Roche MD, PhD,

    1. Department of Medical Oncology, Claudius Regaud Institute, Toulouse, France
    Search for more papers by this author
  • François Bertucci MD, PhD

    Corresponding author
    1. Department of Medical Oncology, Paoli-Calmettes Institute, Marseille Cancer Research Center, UMR891 Inserm, IFR137, Marseille, France
    2. University of the Mediterranean, Marseille, France
    • Département d'Oncologie Médicale, Institut Paoli-Calmettes, UMR891 Inserm, 232, Bd Sainte-Marguerite, 13273 Marseille Cedex 09, France
    Search for more papers by this author
    • Fax: (011) 33 4 91 22 36 70



Aggressiveness of inflammatory breast cancer (IBC) is related to its metastatic potential. The introduction of primary chemotherapy in the multimodality treatment has dramatically changed the prognosis. However, survival remains poor. Since 1995, innovative systemic therapies have been assessed in France in multicentric clinical trials, initially centered on high-dose chemotherapy (HDC) with hematopoietic stem cell transplantation (HSCT), and, more recently, on targeted therapies.


The authors present the rationale and first results of these French studies specifically dedicated to nonmetastastic IBC.


More than 380 patients have been included in 5 trials. The first 3 trials enrolled 329 women and concerned HDC (PEGASE 02, 05, 07). PEGASE 02 and PEGASE 05 showed a high pathological complete response rate (30%) after primary sequential HDC, and suggested that more than 4 cycles does not seem to provide any benefit. PEGASE 07 tested adjuvant maintenance chemotherapy after neoadjuvant HDC. Analysis is ongoing. The 2 other trials currently underway combine targeted therapies with conventional-dose chemotherapy in ERBB2-negative (Beverly 1 trial; bevacizumab) and ERBB2-positive (Beverly 2; bevacizumab and trastuzumab) IBC.


HDC with HSCT remains experimental with high pCR rates and which likely benefits to subgroups of patients that remain to be identified. Targeted therapies, such as anti-ERBB2 and antiangiogenic drugs, are being tested, and should improve survival as demonstrated in non-IBC. With emerging targeted drugs, there is hope that a cure becomes an achievable goal for more patients. Because of the rarity and the heterogeneity of disease, well-designed large-scale collaborative studies are mandatory. Cancer 2010;116(11 suppl):2829–36. © 2010 American Cancer Society.

Inflammatory breast cancer (IBC) is a rare (∼ 5% of cases) but aggressive form of breast cancer. Conversely to that of non-IBC, its incidence rose during the 1990s.1 Aggressiveness of disease is related to its rapid progression and its metastatic potential, highlighted by the high incidence of axillary lymph node involvement and distant metastases (35%) at time of diagnosis, and by the very low rate of long-term survival (<5%) when patients were historically treated with locoregional treatment alone. The introduction of primary combination chemotherapy in the multimodality treatment has dramatically changed the prognosis.2 However, and despite this progress, there has not been an important change in the survival of IBC patients in the last 30 years,3 with a 5-year survival remaining between 30 and 50%.4, 5

Improvement of the systemic treatment is a crucial objective. Because of the relative infrequency of the disease, its rapid progression, and its poor prognosis, a few clinical IBC-specific trials are available; IBC patients generally being enrolled in trials including locally advanced breast cancers. No phase 3 study exclusively dedicated to IBC patients has been published or performed and most of the current systemic regimens have been extrapolated from those used in non-IBC and from registries or retrospective noncontrolled studies. Until recently, anthracycline-based combination was the standard regimen of primary chemotherapy.5-8 Then, taxanes were introduced with some benefit in terms of response, progression-free survival, and overall survival, notably for patients with estrogen receptor-negative tumor,9, 10 making the taxane-anthracycline combination the current standard. But results are not sufficient, calling for the urgent optimization of neoadjuvant systemic treatments. Reported prognostic features11 remain contested. However, response to conventional chemotherapy—notably pathological response—seems to be a strong indicator of long-term survival,2, 9, 12-14 furthermore highlighting the importance of the neoadjuvant systemic treatment.

Since 1995, innovative systemic therapeutic approaches have been assessed in France in the context of multicentric trials, first centered on high-dose chemotherapy with hematopoietic stem cell transplantation, and more recently on molecularly targeted therapies. Here, we present the rationale and the results of these national studies.

High-Dose Chemotherapy With Hematopoietic Stem Cell Transplantation

High-dose chemotherapy (HDC) with hematopoietic stem cell transplantation (HSCT) was tested in the early 1990s in poor-prognosis breast cancer including IBC. The rationale included the disappointing results of conventional chemotherapy, the chemosensitivity of disease, the dose-response, and dose-intensity relationships demonstrated with alkylating agents.15, 16 In IBC, additional rationale came from the importance of initial response to primary chemotherapy, and the increase of pathological response rates by using semi-intensive chemotherapy without HSCT.8, 17

Publications about alkylating agent-based HDC with HSCT in IBC are rare. Initial data came from the International Bone Marrow Transplant Registry,18 and since 1997, from some pilot or phase 2 studies, where HDC was delivered as consolidation after conventional chemotherapy and then surgery, in most of cases.19-25 Globally, even if it was difficult to compare these studies, often multicentric and/or with small cohorts, with heterogeneous treatments and limited follow-up (median approximately 47 months), the 3-year or 4-year disease-free survival (DFS) rates (from 45 to 65%) and overall survival (OS) rates (from 52 to 89%) were similar to each other and were in the upper range of historical cohorts of IBC patients treated with conventional chemotherapy. More recently, Somlo et al reported on a large series of 120 women with IBC who were treated on sequentially developed single-cycle or tandem-cycle HDC trials after primary conventional-dose chemotherapy and surgery.26 With a median follow-up of 61 months, estimated 5-year relapse-free survival and OS were 44% and 64%, respectively. Positive prognostic markers were positive estrogen receptor status, low number of involved axillary lymph nodes, and tandem HDC. We retrospectively analyzed a series of 74 consecutive cases of nonmetastatic IBC treated in our institution with induction anthracycline-based chemotherapy.27 Several parameters including intensity of chemotherapy (20 patients with conventional doses and 54 with HDC and HSCT), pathological features and immunohistochemistry were evaluated for a potential prognostic role. With a median follow-up of 48 months after diagnosis, the 5-year projected DFS was 24% and OS 41%. In multivariate analysis, the strongest prognostic factor was the delivery of HDC. The 5-year DFS and OS of patients were, respectively 28% and 50% with HDC and 15% and 18% with conventional chemotherapy.

Many of these pilot studies had shown a potential benefit for HDC with HSCT in the treatment of IBC. But, like for non-IBC, the debate was confusing within the oncologists regarding the actual benefit. In the middle 1990s, consensus conferences concluded on the requirement of randomized studies larger enough to detect survival differences between 10 and 20%. In 1994, the French Federation of Cancer Centers (Federation Nationale des Centres de Lutte Contre le Cancer) and the French Bone Marrow Society (Société Française de Greffe de Moëlle) merged their efforts to create the PEGASE Group. The objective was to perform larger and multicentric randomized comparative studies of HDC with HSCT in poor-prognosis breast cancer. In nonmetastatic IBC, 3 trials were successively conducted between January 1995 and June 2005 and enrolled 329 patients: 2 phase 2 trials (PEGASE 02 and PEGASE 05) and a phase 3 trial (PEGASE 07).

PEGASE 02 trial was conducted between January 1995 and September 1996 through 17 anticancer centers in France and accrued 100 patients. The primary endpoint was the rate of pathological complete response (pCR) obtained after sequential HDC combining cyclophosphamide (C), doxorubicin (D), and 5-fluorouracil (5FU). At diagnosis, patients received 4 cycles of chemotherapy every 3 weeks associating C 6 g/m2 and D 75 mg/m2 at cycle 1, C 3 g/m2 and D 75 mg/m2 at cycle 2, C 3 g/m2, D 75 mg/m2 and 5FU 2500 mg/m2 at cycles 3 and 4. G-CSF was administered after each cycle. Hematopoietic stem cells (HSC) were harvested after cycle(s) 1 + /-2, and reinfused on Day 7 of cycles 3 and 4. Mastectomy with second axillary dissection (if not performed initially) was performed after HDC, followed by radiotherapy. Tamoxifen was then given to menopausal patients with hormone receptor-positive tumor. Ninety-five patients were eligible. They received 366 cycles of chemotherapy. Eighty-seven women completed the planned treatment. The relative dose-intensity was 0.97 for C and 0.96 for D. The most frequent nonhematological toxicity was vomiting and the most frequent hematological toxicity was febrile neutropenia. One patient died from septic shock during neutropenia after cycle 1. Eighty-six patients underwent mastectomy, which was performed in a median of 3.5 months after the first cycle of HDC. The pCR rate was 32%, encouraging when compared with the rate observed after standard chemotherapy.28 With a 3-year median follow-up, the 3-year DFS rate was 44% and OS 70%. This study confirmed the feasibility of sequential HDC with HSCT in IBC, and revealed interesting rates of pCR and 3-year survival. Two specific studies on quality of life29 and economic assessment30 further confirmed the feasibility of the approach. We are now analyzing the survival results with an extended follow-up and the prognostic value of molecular features assessed by immunohistochemistry.

The following phase 2 trial (PEGASE 05) was initiated on August 1997. Like in PEGASE 02 study, the aim was to increase the rate of pCR by intensifying the primary chemotherapy to improve survival. Patients received 7 cycles of primary chemotherapy. The PEGASE 02 scheme was modified as follows: 3 cycles of 100 mg/m2 of docetaxel with G-CSF every 2 weeks were integrated after the first 2 cycles of CD, and the second CD cycle was similar to the first 1 (C 6 g/m2 and D 75 mg/m2). The fifth and the sixth cycles of CD delivered C 3 g/m2 and D 75 mg/m2. HSCs were harvested after cycle(s) 1 + /-2, and reinfused after cycles 6 and 7. The primary endpoint was pCR, and secondary endpoints were 5-year DFS and OS. After the enrolment of 54 patients, the study was prematurely stopped on June 1999 because of toxicity, notably hematological with serious infectious complications (8 patients) and 2 deaths during febrile neutropenia. The pCR rate, assessed in 48 women, was 30%, similar to that reported in PEGASE 02 trial. Survival data are being analyzed.31

These 2 trials led to the following conclusions: first, increase of the pCR rate with intensified regimen and increase of 3-year survival; second, more than 4 cycles of intensified primary chemotherapy does not seem to provide any benefit; and third, because of the high rate of relapse, need for testing maintenance chemotherapy with drugs not used during induction phase. On the basis of these data, the PEGASE 07 trial was launched on 2000. In this phase 3 trial, patients were randomized at time of inclusion to receive after primary chemotherapy followed by surgery and radiotherapy, either no chemotherapy, or 4 cycles of docetaxel (85 mg/m2) and 5FU (3500 mg/m2 over 4 days) every 3 weeks as maintenance chemotherapy. Primary chemotherapy comprised 4 cycles of C-Epirubicin regimen (C 4 g/m2 and E 150 mg/m2 per cycle every 3 weeks) with harvest of HSC after cycle 1 and reinfusion after cycles 2, 3, and 4. The trial was closed on June 2005 after the expected accrual of 175 patients. The primary endpoint is the 3-year DFS. Secondary endpoints include pCR, OS, quality of life, and translational studies. First results will be available by the end of 2009.

Globally, the results of these studies, which enrolled 329 patients over a 10-year period, are interesting in term of pathological response and early event-free survival. Today, the approach remains limited to specialized centers and remains experimental. More mature follow-up of published and ongoing studies is required, as well as larger collaborative prospective trials. Another crucial objective is to define predictors for selecting patients the most susceptible to selectively benefit from HDC.32, 33 Nonetheless, the difficulty of these studies and the relatively high rate of relapse after conventional or high-dose chemotherapy impose testing in parallel with other innovative systemic therapeutic strategies such as targeted therapies.


IBC is highly angiogenic and overexpresses the ERBB2 and EGFR tyrosine kinase receptors more frequently than non-IBC.34 These observations inevitably led to test in IBC patients the biological therapies targeting the corresponding pathways and recently approved in combination with chemotherapy for the treatment of non-IBC.

Trastuzumab is a recombinant humanized monoclonal antibody directed against ERBB2. Its introduction into the treatment of ERBB2-positive metastatic35, 36 and early stage breast cancer37-40 improves survival of patients. The high rate of pCR (60%) reported by Buzdar et al in early stage breast cancer after delivery of trastuzumab associated with an anthracycline/taxane-based regimen of primary chemotherapy41 prompted several studies at assessing its association with neoadjuvant conventional-dose chemotherapy in ERBB2-positive locally advanced breast cancers (LABC) including IBC. High rates of pCR were observed, ranging from 17 to 55%.42-50 In a substudy50 of the NOAH (Neoadjuvant Herceptin) study,49 76 women with IBC, including 31 ERBB2-positive patients were assessed: the pCR rate was 55% for patients who received trastuzumab and standard chemotherapy compared with 19% for patients who received chemotherapy only. Although these series are limited in size, and given the survival advantage observed in early stage breast cancer, these results strongly argue for use of trastuzumab in treatment of patients with ERBB2-positive IBC. The interest for targeting ERBB2 in IBC was further confirmed by the activity of lapatinib—a dual inhibitor of both ERBB2 and EGFR approved in the treatment of ERBB2-positive metastatic breast cancer after failure of trastuzumab51—in 2 phase 2 trials,52, 53 with 95% of clinical response after a combination of lapatinib and paclitaxel in a series of 21 chemotherapy-naive ERBB2-positive IBC patients.53

The other targeted therapies recently assessed in IBC are antiangiogenic drugs, notably bevacizumab. Bevacizumab, a recombinant humanized monoclonal antibody to VEGF, improves the response rate when delivered in combination with chemotherapy in previously heavily pretreated metastatic breast cancer patients,54 and it improves the response and progression-free survival rates when associated with first-line chemotherapy in metastatic patients.55 Bevacizumab was then tested in combination with chemotherapy in the neoadjuvant setting with promising results in terms of feasibility and efficacy.56-58 The feasibility of combining bevacizumab and anthracycline-based chemotherapy was recently reported in the adjuvant59 and neoadjuvant setting60, 61 of breast cancer patients. To date, studies with bevacizumab in IBC patients are scarcer than those with trastuzumab.60-62 Wedam et al treated 21 patients with LABC (1 patient) and IBC (20 patients) with neoadjuvant bevacizumab followed by bevacizumab plus docetaxel and doxorubicin chemotherapy.60 The clinical partial response rate was 67%. The pCR rate was not available. Antiangiogenic and antitumor effects were documented on tumor biopsies63 and dynamic contrast-enhanced magnetic resonance imaging64 by serial assessments performed prior, during, and after the neoadjuvant treatment.

In this context, it was important to set up phase 2 studies in IBC, excluding LABC, to test the impact of combining anthracyclines and taxane-based chemotherapy with active targeted therapies in the neoadjuvant treatment. In France, 2 phase 2 national multicentric trials, Beverly 1 and Beverly 2, conducted by P. Viens have been recently launched to assess the introduction of bevacizumab in the primary treatment of ERBB2-negative (Beverly 1) and ERBB2-positive (Beverly 2 in combination with trastuzumab) nonmetastatic IBC patients. The inclusion and exclusion criteria are the same (except the ERBB2 status) in the 2 studies exclusively dedicated to IBC patients. The endpoints are also similar. The primary endpoint is the rate of pCR on surgical samples after the neoadjuvant treatment. The secondary endpoints include the rate of DFS and OS at 3 and 5 years, the toxicity notably cardiac, and many translational research studies aimed at identifying markers predictive of efficacy and/or toxicity (circulating tumor and endothelial cells, pharmacokinetics, pharmacogenetics, pharmacogenomics, proteomics, and imaging). Beverly 2 is promoted by Roche SAS. It will assess the combination of bevacizumab with standard trastuzumab and primary chemotherapy in ERBB2-positive IBC. The rationale for combining the 2 antibodies comes from preclinical and clinical data. In a human breast cancer xenograft model, the antitumor activity was enhanced when trastuzumab was combined with bevacizumab.65 The combination of the 2 monoclonal antibodies, bevacizumab and trastuzumab was administered for the first time to human subjects in a phase 1 study including metastatic breast cancer patients.66 The activity of the association was then confirmed in a phase 2 trial,67 with objective responses in 20 of 37 (54%) assessable patients with an advanced ERBB2-positive breast cancer, as well as an acceptable toxicity profile, notably cardiac. The Beverly 2 design is as follows (Fig. 1). During the first phase (neoadjuvant treatment: 24 weeks), the patients receive 4 cycles of FEC100 (5FU 500 mg/m2, epirubicin 100 mg/m2, cyclophosphamide 500 mg/m2) plus bevacizumab (15 mg/kg) every 3 weeks, followed by 4 cycles of docetaxel (100 mg/m2) plus bevacizumab (15 mg/kg) and trastuzumab (6 mg/kg) every 3 weeks. The second phase is the surgery with mastectomy and axillary lymph node dissection, performed at least 4 weeks after the last infusion of bevacizumab. Trastuzumab is maintained during surgery every 3 weeks. The third phase is the adjuvant treatment, which includes radiation therapy (during approximately 6 weeks) plus bevacizumab (every 3 weeks restarted during the radiotherapy period and then continued for 30 additional weeks) and trastuzumab (every 3 weeks continued during the radiotherapy period and then for 30 additional weeks). If the tumor is hormone receptor-positive, hormone therapy is introduced after the end of radiotherapy (scheme left at the investigator's discretion). Each patient will receive 8 cycles of chemotherapy (neoadjuvant), 18 cycles of trastuzumab (neoadjuvant/adjuvant), and 18 cycles of bevacizumab (neoadjuvant/adjuvant). The hypothesis is that this antiangiogenic targeted therapy will improve the clinical outcome, by increasing the pCR rate from 20% to 40%. A total of 50 patients will be included over a 24-month recruitment period with a Simon 2-step design. The first patient was included on September 2008 and the study is ongoing with a recruitment above that expected. Beverly 1 the same trial, but dedicated to ERBB2 negative IBC. It is promoted by the French Federation of Cancer Centers (French Federation Nationale des Centres de Lutte Contre le Cancer). Designed to assess the combination of bevacizumab with standard primary chemotherapy, its scheme is the same than that of Beverly 2, but without trastuzumab. The hypothesis is that bevacizumab will increase the pCR rate from 20% to 30%. A total of 100 patients will be included over an 18-month recruitment period with a Simon 2-step design. The first patient was included on January 2009 and the study is ongoing with the expected recruitment rate.

Figure 1.

The design of the Beverly 2 trial is illustrated.


Multimodal chemotherapy has improved prognosis of IBC. Current standard regimens include anthracycline and taxane, combined with trastuzumab for ERBB2-positive tumors. However, there is still significant room for improvement in survival of patients. In France, since 1995, more than 380 patients with nonmetastatic IBC have been included in national multicentric clinical trials testing first the value of HDC with HSCT, and, more recently, the value of targeted therapies in the neoadjuvant and adjuvant treatment. Enrolment is still ongoing for the later. HDC with HSCT remains an experimental approach with high pCR rates and which likely benefits subgroups of patients, who remain to be identified. Another major research direction is based on a better understanding of the molecular features of IBC, a crucial objective. It has already allowed for the development of therapies targeting ERBB2-dependent pathways and angiogenesis with promising results. With the large number of new potential therapeutic targets that will be identified in the near future and the large number of emerging novel drugs, there is hope that cure becomes an achievable goal. But we must also keep in mind the great heterogeneity of IBC at the biological level with the existence of molecular subtypes similar to those described in non-IBC.68, 69 This heterogeneity will have to be taken into account in the design of future clinical and translational studies. In this context, it is the responsibility of the scientific community (medical, translational, and fundamental) to design large-scale collaborative studies centered on this rare, heterogeneous, but aggressive disease.


Our work is supported by Institut Paoli-Calmettes, Inserm, Institut National du Cancer (Inca 2007 Translational Research), Association pour le Recherche contre le Cancer (ARC 2002). P. Viens is member of advisory committee for GSK France, Roche France, Sanofi Aventis France.