www.thelancet.com/oncology Vol 13 April 2012 375
Neoadjuvant bevacizumab, trastuzumab, and chemotherapy
for primary infl ammatory HER2-positive breast cancer
(BEVERLY-2): an open-label, single-arm phase 2 study
Jean-Yves Pierga, Thierry Petit, Thierry Delozier, Jean-Marc Ferrero, Mario Campone, Joseph Gligorov, Florence Lerebours, Henri Roché,
Thomas Bachelot, Emmanuelle Charafe-Jauff ret, Maria Pavlyuk, Sandrine Kraemer, François-Clément Bidard, Patrice Viens
Background Bevacizumab and trastuzumab are effi cacious for treatment of advanced or HER2-positive metastatic
breast cancer; however, few data exist for this regimen in infl ammatory breast cancer. In our phase 2 trial, we aimed
to assess effi cacy and safety of neoadjuvant bevacizumab combined with trastuzumab and chemotherapy in patients
with primary HER2-positive infl ammatory breast cancer.
Methods In our phase 2, multicentre, open-label, single-arm, non-comparative trial, we enrolled women (aged
≥18 years) with histologically confi rmed HER2-positive non-metastatic infl ammatory breast cancer at private or public
oncology centres in France. Before surgery, patients were treated with fl uorouracil, epirubicin, cyclophosphamide,
and bevacizumab (cycles 1–4) and docetaxel, bevacizumab, and trastuzumab (cycles 5–8) in 3-week cycles. After
surgery, patients received adjuvant radiotherapy, trastuzumab, and bevacizumab. For the primary endpoint, we
assessed the proportion of patients who achieved a pathological complete response (defi ned by central review of
surgical specimens according to Sataloff classifi cation, counting missing data as failure) and adverse events in all
enrolled patients. This study is registered with ClinicalTrials.gov, number NCT00717405.
Findings Between Oct 23, 2008, and Oct 28, 2009, we enrolled 52 patients at 21 centres. 42 (81%) of 52 patients received
all eight cycles of neoadjuvant therapy and 49 (94%) underwent surgery. After neoadjuvant therapy, 33 of 52 patients had
a pathological complete response according to central review (63·5%, 95% CI 49·4–77·5). The most common adverse
events were asthenia and nausea (both occurred in 36 [69%] of 52 patients). 25 (48%) patients had grade 3–4 neutropenia,
which was the most common grade 3–4 adverse event. Only one grade 3 or worse adverse event regarded as related to
bevacizumab was reported (hypertension, one patient). Four patients (8%) had cardiac failure.
Interpretation Neoadjuvant treatment with bevacizumab, trastuzumab, and chemotherapy was effi cacious and well
tolerated in patients with previously untreated primary infl ammatory breast cancer. Further confi rmation of use of
bevacizumab in infl ammatory breast cancer is needed.
Funding Roche (France).
Infl ammatory breast cancer is a very aggressive form of
locally advanced breast cancer, and accounts for up to
5% of all cases.1 Such cancer is characterised by the
clinical appearance of infl ammation with oedema
(termed peau d’orange) and redness of the skin.2 Typically,
infl ammatory breast cancer is diagnosed in young
women, who are more likely to have metastasis than are
those with other types of breast cancer.3 Multimodal
treatment regimens—including endocrine treatment,
chemotherapy, surgery, and radio therapy—and neo-
adjuvant approaches have greatly improved the prognosis
for infl ammatory breast cancer,4 resulting in a 5-year
survival of about 40%.5 However, women with this cancer
have worse prognosis than do those with non-
infl ammatory locally advanced breast cancer.1
Overexpression of HER2 occurs in 15–25% of patients
with breast cancer.6 HER2 positivity is generally associated
with aggressive disease; however, introduction of the
humanised monoclonal antibody trastuzumab, which
targets HER2, has improved outcomes in metastatic breast
cancer and early disease.7,8 Combination of trastuzumab
with various chemotherapy regimens for the neoadjuvant
treatment of locally advanced breast cancer, including
infl ammatory breast cancer, has also been assessed in
several clinical studies.9–13 A substudy of the NOAH trial9
assessed response to standard chemotherapy in 76 women
with infl ammatory breast cancer, 62 of whom had HER2-
positive disease. Trastuzumab was coadministered with
chemotherapy in 31 patients with HER2-positive disease,
54·8% of whom achieved a pathological complete response
(pCR), compared with 19·3% of those with HER2-positive
disease who received chemotherapy alone.10
Infl ammatory breast cancer is characterised patho-
logically by high vascularity and increased micro vessel
density because of high expression of angiogenic factors
(eg, VEGF).3 VEGF is a key mediator of angiogenesis and
is involved in endothelial and tumour cell growth and
motility and blood vessel permeability.14 Infl ammatory
breast cancer’s vascular nature probably makes it
Lancet Oncol 2012; 13: 375–84
February 28, 2012
See Comment page 324
Institut Curie, Paris, France
(Prof J-Y Pierga MD,
F-C Bidard MD); Université Paris
Descartes, Paris, France
(J-Y Pierga, F-C Bidard); Centre
Paul Strauss, Strasbourg,
France (Prof T Petit MD); Centre
François Baclesse, Caen, France
(T Delozier MD); Centre Antoine
Lacassagne, Nice, France
(Prof J-M Ferrero MD); Institut
de Cancérologie de
Saint Herblain, France
(M Campone MD); Hôpitaux
Universitaires Paris Est–Tenon,
Alliance pour la Recherche en
Cancerologie, Paris, France
(J Gligorov MD); Institut Curie
Hôpital René Huguenin,
(F Lerebours MD); Institut
Claudius Regaud, Toulouse,
France (Prof H Roché MD);
Centre Léon-Bérard, Lyon,
France (T Bachelot MD);
(Prof E Charafe-Jauff ret MD,
Prof P Viens MD); Roche,
(M Pavlyuk MD, S Kraemer MSc);
and Aix-Marseilles Université,
(P Viens, E Charafe-Jauff ret)
Prof Patrice Viens, Institut Paoli
Calmettes and Aix-Marseilles
Université, 232 Boulevard Sainte
Marguerite, 75005 13273
www.thelancet.com/oncology Vol 13 April 2012
especially amenable to anti-angiogenic treatment. Use of
bevacizumab, a VEGF-targeting monoclonal antibody,
resulted in substantially improved progression-free
survival and response in patients with advanced breast
cancer in four randomised phase 3 trials15–18 and showed
neoadjuvant activity in a pilot study in patients with
previously untreated locally advanced breast cancer or
infl ammatory breast cancer.19
Combination of bevacizumab and trastuzumab for
treatment of advanced or HER2-positive metastatic breast
cancer has been assessed in two trials.20,21 First, an open-
label, phase 1 dose-escalation trial20 of three doses of
bevacizumab in combination with trastuzumab noted
preliminary clinical responses in fi ve (56%) of nine
patients and two other patients developed stable disease.
The combination of bevacizumab and trastuzumab was
well tolerated with no reported grade 3–4 adverse events.
After this encouraging outcome, a phase-2 study21 was
done in 37 women with advanced HER2-positive breast
cancer. Patients received trastuzumab (4 mg/kg loading
dose followed by 2 mg/kg weekly) and bevacizumab at
the recommended phase 2 dose (10 mg/kg every 2 weeks).
The overall response rate was 54·1% (20 of 37 patients),
with one patient achieving a complete response and
19 achieving partial responses; 11 patients (29·7%) had a
best response of stable disease. Grade 3–4 drug-related
adverse events included dyspnoea, left ventricular dys-
function, hyper tension, and proteinuria.
We undertook this phase 2 study on the basis of
previous results with trastuzumab plus chemotherapy,
the reported activity of bevacizumab in infl ammatory
breast cancer, and the effi cacy of bevacizumab plus
trastuzumab in advanced and metastatic disease. We
aimed to assess the effi cacy and safety of preoperative
treatment with bevacizumab, trastuzumab, and chemo-
therapy in patients with primary infl ammatory HER2-
positive breast cancer. We also assessed the potential of
circulating tumour cells and circulating endothelial cells
as effi cacy biomarkers, since circulating tumour cell
positivity has been shown to be an independent
prognostic factor in patients with non-metastatic breast
cancer receiving either adjuvant22 or neoadjuvant
chemotherapy,23 and both circulating tumour and
circulating endothelial cells have been reported as effi cacy
biomarkers in patients with metastatic breast cancer
receiving antiangiogenic therapy plus chemotherapy.24,25
Study design and patients
In our phase 2, multicentre, open-label, non-
randomised, Simon (two-stage), single-arm, non-
comparative trial, we enrolled women (aged ≥18 years)
who had histologically confi rmed breast cancer and
confi rmed infl ammatory breast cancer. Infl ammatory
breast cancer was defi ned as T4d (any N), stage II
(infl ammation, erythema, and/or oedema localised to
<50% of the breast surface), or stage III (generalised
infl ammation and oedema on >50% of the breast
surface) by Institut Gustave-Roussy classifi cation26 or as
presence of tumour emboli in the lymph vessels of the
superfi cial derma on skin biopsy sampling. Eligible
patients had to have Eastern Cooperative Oncology
Group (ECOG) performance status of 0–2, life
expectancy of 3 months or more, HER2-positive
tumours (assessed locally but reviewed centrally after
the last inclusion), known hormone receptor status,
and no metastases. Hormone-receptor positivity was
defi ned as overexpression of 10% or more.
Patients with non-infl ammatory breast cancer with
tumoral skin permeation, bilateral breast cancer,
previous cancer history relapsing 5 years or more before
study entry, and in-situ contralateral breast carcinoma
were ineligible, as were those who had previously been
treated with chemotherapy, radiotherapy, or endocrine
therapy for breast cancer. Other exclusion criteria were
major surgery or substantial traumatic injury 28 days or
less before start of study treatment or anticipation of the
need for major surgery during study treatment; minor
surgery 24 h or less before fi rst bevacizumab infusion;
non-healing wound, active peptic ulcer or bone fracture;
or history of abdominal fi stula, gastrointestinal
perforation, or intra-abdominal abscess within 6 months
of enrolment; uncontrolled hypertension; or pregnancy.
Patients were not allowed to have received hormonal
therapy in the 3 weeks before enrolment in this study.
All patients provided signed written informed consent
before study-specifi c screening
separately for the translational research studies.
Approval for study design, conduct, and the informed
consent documentation was received on March 29, 2009,
from the independent ethics committee (Comité de
Protection des Personnes Sud Méditerranée I, Institut
Paoli Calmettes, Marseille, France).
HER2 status was assessed by investigators by immuno-
histochemistry (IHC), fl uorescence in-situ hybridisation
(FISH), or chromogenic in-situ hybrid isation (CISH),
and reanalysed centrally with IHC and silver in-situ
hybridisation (SISH) assays by an experienced pathologist
(EC-J) who was masked to patient’s data, including the
original IHC test results. HER2 status in IHC was
assessed with the A485 polyclonal antibody (Dako,
Glostrup, Denmark) on the BenchMark XT system
(Ventana Medical Systems, Tucson, AZ, USA). Biopsy
samples were graded according to the HercepTest (Dako)
scoring system (0, 1+, 2+, or 3+).
We did the SISH analyses with the INFORM HER2
technology (Ventana Medical Systems) and BenchMark
XT system according to standard procedures. For all
biopsy samples, HER2 signals and chromosome 17 were
counted in X60 tumour cell nuclei. We quantifi ed HER2
as the ratio of HER2 to chromosome 17 signals with the
conventional historical interpretation scale and also by the
www.thelancet.com/oncology Vol 13 April 2012 377
American Society of Clinical Oncology and College of
American Pathologists reporting scheme. HER2 amplifi -
cation was defi ned as the ratio of HER2 to chromosome 17
centromeric signals of X2·2.27 All tumours with six or
more HER2 signals per nuclei had a ratio of X2·2 and
therefore were classifi ed as amplifi ed tumours. All
tumours with fewer than six HER2 signals per nuclei had
an HER2 to chromosome 17 ratio of less than 1·8 and
were deemed to be non-amplifi ed.
Our trial had four planned stages (fi gure 1). In stage 1,
patients received neoadjuvant treatment with fl uor o-
uracil, epirubicin, cyclophosphamide, and bevacizumab
(cycles 1–4) followed by docetaxel, bevacizumab, and
trastuzumab (cycles 5–8). In stage 2, patients underwent
surgery (mastectomy and axillary node dissection) with
trastuzumab maintenance during the perioperative
period. In stage 3, patients received adjuvant treatment
with radiotherapy, trastuzumab, and bevacizumab.
Endocrine therapy was introduced at the investigator’s
discretion in patients with hormone receptor-positive
disease. In stage 4, patients were followed up for 5 years
after inclusion of the last patient. The study is in progress,
and we report data from stages 1 and 2 here.
Neoadjuvant chemotherapy was intravenous fl uor o-
uracil (500 mg/m²), epirubicin (100 mg/m²), and
cyclophosphamide (500 mg/m²) on day 1 of a 3-week
cycle for four cycles followed by intravenous docetaxel
(100 mg/m²) on day 1 for a further four cycles.
Premedication was given according to local practice.
In the neoadjuvant phase, patients received intravenous
bevacizumab (15 mg/kg) once every 3 weeks for eight
injections. Bevacizumab was stopped at least 4 weeks
before surgery and restarted during or after radiotherapy
once the wound was healed entirely. Adjuvant
bevacizumab (15 mg/kg) was given every 3 weeks for up
to 30 weeks. Trastuzumab was given at an initial intra-
venous loading dose of 8 mg/kg and then at a dose of
6 mg/kg once every 3 weeks from cycle 5 of neoadjuvant
treatment, and was continued in the perioperative period
for about 30 weeks after surgery. Mandatory radiotherapy
was started 2–4 weeks after surgery, if the wound had
healed properly, and was given for 4–6 weeks according
to site-standard practice.
No bevacizumab treatment interruption or dose
changes were planned, unless the patient’s weight varied
by more than 10%, in which case the dose was
recalculated. Trastuzumab and bevacizumab were
discontinued defi nitively in patients with a symptomatic
decrease (or temporarily in patients with an asymptomatic
decrease to <50%) in left ventricular ejection fraction,
and B-type natriuretic peptide assessment was requested.
Epirubicin 100 mg/m2,
cyclophosphamide 500 mg/m2,
and fluorouracil 500 mg/m2 every
3 weeks for four cycles
Docetaxel 100 mg/m2 every
3 weeks for four cycles
Neoadjuvant trastuzumab 8 mg/kg loading dose
then 6 mg/kg every 3 weeks for 12 weeks
Adjuvant trastuzumab 6 mg/kg
every 3 weeks for 42 weeks
Endocrine therapy if
Neoadjuvant bevacizumab 15 mg/kg every 3 weeks for 24 weeks
Adjuvant bevacizumab 15 mg/kg
every 3 weeks for 30 weeks
Figure 1: Study design and treatment schedule
Median age, years
ECOG performance status
Lymph node involvement based on clinical staging
Scarff -Bloom-Richardson grade
Infl ammatory breast cancer classifi cation
PEV II IGR-PEV
PEV III IGR-PEV
T4d and IGR-PEV II
T4d and IGR-PEV III
52·2 (44·2–59·8; 24·6–68·3)
Data are median (IQR, range) or n (%). ECOG=Eastern Cooperative Oncology Group.
IGR=Institut Gustave-Roussy. PEV=poussée évolutive (acute exacerbation).
*Insuffi cient biopsy material was available to allow grading in four patients.
Table 1: Baseline characteristics of patients
www.thelancet.com/oncology Vol 13 April 2012
After asymptomatic decreases in left ventricular ejection
fraction, trastuzumab and bevacizumab were reinitiated
at the standard dose (no dose reduction). Patients were
subsequently monitored until the end of the study.
Bevacizumab was permanently discontinued if held for
two consecutive or two intermittent cycles. Trastuzumab
was discontinued if held for three intermittent or two
The primary endpoint was the proportion of patients
with a pCR assessed by central review of surgical
specimens according to
Secondary endpoints were the proportion of patients
with a pCR assessed by central review according to
Chevallier classifi cation29 and also as assessed by the
investigators according to Sataloff classifi cation. For
assessment of response, at least 20 tissue blocks were
prepared for every patient, according to a standardised
methodological approach.30 Haematoxylin and eosin-
stained slides of surgical resection specimens after
chemotherapy were reviewed centrally (EC-J) by
microscope. On the basis of patients’ pathological
response in the primary breast site and/or in the lymph
nodes, and in line with published data,31 we designated
patients as having a pCR when surgical samples showed
total or near-total treatment eff ect and absence of nodal
Sataloff classifi cation.28
involvement (ie, Sataloff classifi cation TA and NA or NB
or Chevalier classifi cation Ch1 and Ch2).
Additional secondary endpoints were disease-free
survival, recurrence-free interval, overall survival, safety
(adverse events coded according to the MedDRA
guidelines and their intensity graded by the National
Cancer Institute Common Terminology Criteria for
Adverse Events version 3.0), cardiac safety according to
New York Heart Association classifi cation (for patients
with cardiac failure), and translational research.
Translational research studied factors that were predictive
of effi cacy and included biology, pharmacokinetics,
pharma co genetics, pharmacogenomics, proteomics, and
imaging studies. Clinical tumour response was assessed
by the investigators. We defi ned clinical response as
absence of infl ammatory signs, fi xed lymph nodes, or
visceral dissemination of the disease.
For translational research studies, we obtained fi ve
supplementary blood samples (2×10 mL) from every
patient: one before the fi rst bevacizumab administration
(neoadjuvant period), one before the fi rst trastuzumab
administration (neoadjuvant period), one before surgery,
one before bevacizumab reintroduction (adjuvant period),
and one at the end of adjuvant treatment (fi nal visit). All
samples were analysed within 3 days at the Institut Curie
(Paris, France) by Claire Mathiot. Circulating tumour cell
and circulating endothelial cell counts were measured in
7·5 mL and 4 mL samples, respectively, with CellSearch
(Veridex LLC, Raritan, NJ, USA). HER2 immuno cyto-
fl uorescence staining (FITC-conjugated antibody HER81,
Veridex LLC) was assessed on detected circulating
tumour cells and categorised as class 0–3; as per Riethdorf
and colleagues,32 we only regarded classes 2 and 3 as
We calculated the sample size on the basis of a Simon
two-stage design for phase 2 study. From the proportion
of patients with a pCR reported in the PEGASE 02
study,33 we regarded a proportion of patients with a pCR
of 40% or more as proof of effi cacy of the combination
and of less than 20% as insuffi cient to continue the
assessment. Assuming a risk of α=0·05 (type I error)
and β=0·10 (type II error), we needed to include
45 patients (maximum) in two stages. To allow for non-
assessable data for up to 10% of patients for the primary
endpoint, we enrolled 52 patients. The fi rst stage
included 24 patients who were followed up until
assessment of pathological response. If six or more
pathological complete responses were noted, recruitment
was to be continued.
For the primary outcome, we show the proportion of
patients with a pCR on central review according to
Sataloff classifi cation with 95% CI (missing data for
patients with insuffi cient tissue for assessment or who
did not undergo surgery were regarded as failure). Only
patients with centrally confi rmed HER2 amplifi cation
Axillary lymph node response28
NANB NCND Not assessed
Sataloff classifi cation (central review)
TA 33 (63·5%)
Overall 34 (65·4%)
Sataloff classifi cation (investigator review)
TA 31 (59·6%)
TC 1 (1·9%)
Total 32 (61·5%)
Chevalier classifi cation (central review)
Grade 2 ··
Grade 3 ··
1 (1·9%) 8 (15·4%)2 (3·8%) 7 (13·5%)52
6 (11·5%) 8 (15·4%)2 (3·8%)4 (7·7%) 52
NA=evidence of treatment eff ect, no metastatic disease. NB=no nodal metastasis or treatment eff ect. NC=evidence of
treatment eff ect but nodal metastasis still present. ND=viable metastasis disease, no treatment eff ect. TA=total or near
total therapeutic primary site response. TB=>50% (but not total or near total) primary site response. TC=<50% primary
site response, but eff ect evident. TD=no primary site response.
Table 2: Pathological clinical response rate according to Sataloff classifi cation (assessed by central review
or investigators) and Chevalier classifi cation (assessed by central review) in 52 enrolled participants
www.thelancet.com/oncology Vol 13 April 2012 379
were included in the assessment of pathological response
rate. Missing data were regarded as failures. The patient
with a polyploidy IHC2+/SISH-negative did not reach
complete response and was also regarded as a failure. If
14 or more patients had a pCR by the end of the inclusion
period (ie, after the second phase), the strategy was
regarded as eff ective. We also did sensitivity analyses
excluding patients with missing data.
We analysed cell count data with descriptive analyses,
in terms of the variation between timepoints and
correlation with pCR. We assessed signifi cance between
values with the Wilcoxon signed-rank test for circulating
endothelial cell counts and a McNemar’s non-parametric
test for circulating tumour cell counts.
Analyses were done with SAS version 9.1. This study is
registered with ClinicalTrials.gov, number NCT00717405,
and EUDRACT (2008-000783-16).
Role of the funding source
The sponsor was involved in protocol development and
data analysis, and had a role in writing the report. The
sponsor was not involved in the interpretation of the
data; data for interpretation were provided as raw data
tables to PV, TB, TD, MC, EC-J, J-MF, JG, and J-YP. All
authors were given the opportunity to comment on the
draft report and saw and approved the fi nal version. PV
had fi nal responsibility for the decision to submit for
Between Oct 23, 2008, and Oct 28, 2009, we enrolled
52 patients with HER2-positive infl ammatory breast
cancer at 21 private or public oncology centres in France.
Table 1 summarises baseline characteristics. 11 (21%) of
52 patients had a cutaneous biopsy, four of whom had
lymphatic emboli in the superfi cial derma. Although the
eligibility criteria for this trial allowed for the inclusion of
patients solely on the basis of presence of tumour emboli
within the vasculature of the superfi cial derma, all
enrolled patients also had clinical symptoms of
infl ammatory breast cancer as agreed in the international
consensus report of the infl ammatory breast cancer
consortium.34 Investigator-assessed HER2 overexpression
was IHC3+ in 46 patients (88%), IHC3+/FISH-positive in
one patient (2%), IHC3+/CISH-positive in one patient
(2%), FISH-positive in two patients (4%), and
IHC2+/FISH-positive in two patients (4%). Central HER2
review was undertaken in 46 patients and was IHC3+ in
fi ve patients (11%), IHC3+/CISH-positive in one patient
(2%), IHC3+/SISH-positive in 34 patients (74%), IHC2+/
SISH-positive in two patients (4%), IHC3+/SISH-negative
in three patients (7%; two patients had polyploid tumours),
and IHC2+/SISH-negative in one patient (2%; patient
had a polyploid tumour). We did not centrally review
HER2 expression in six patients because of insuffi cient
material. All patients had adequate haematological, liver,
renal, coagulation, and cardiac function and a left
ventricular ejection fraction of 55% or more at inclusion.
42 (81%) of 52 patients received all eight cycles of
neoadjuvant therapy. 45 (87%) patients received eight
cycles of bevacizumab and 50 (96%) patients received at
least four preoperative cycles of trastuzumab; 49 (94%)
patients underwent mastectomy and three patients (6%)
did not undergo surgery in the study.
17 (71%) of 24 patients had a pCR in the fi rst stage of
the study, and thus recruitment into the second stage was
continued. 33 of 52 patients had a pCR (63·5%, 95% CI
49·4–77·5; table 2). In the investigator-based assessment,
Abdominal pain upper
refl ux disease
Mucosal infl ammation
7 7 (13%)
6 6 (12%)
Data are n (%) or n. *Grade missing for some events; patients could have more
than one event of diff erent grades. †Reported at any time.
Table 3: Adverse events occurring in more than 10% of patients by grade
www.thelancet.com/oncology Vol 13 April 2012
35 of 52 patients had a pCR (67·3%, 53·6–81·0). Thus,
because the proportion of patients with a pCR was higher
than the 40% cutoff used in the statistical assumption,
we regarded the combination as eff ective. The sensitivity
analysis under taken after central review also showed a
pCR in 33 of 47 patients with available data (70·2%,
56·1–84·4); the remaining two patients from the 49 who
underwent mastectomy were regarded as having
treatment failure in the analysis because of an absence of
assessable tissue. Much the same results were noted in
the hormone receptor-positive subgroup (11 of 18 patients
[61·1%, 95% CI 35·8–86·4] had a pCR).
At the end of neoadjuvant therapy, 44 (98%) of
45 assessable patients had a clinical response (ie, an
absence of infl ammatory signs, fi xed lymph nodes, or
visceral dissemination of the disease). One patient had
stable disease. Data for disease-free survival, recurrence-
free interval, and overall survival are immature at present.
During the neoadjuvant phase, all patients had at least
one adverse event; 38 patients (73%) had at least one
grade 3 or worse adverse event, of whom ten (26%) had
an adverse event deemed related to bevacizumab. Table 3
shows adverse events occurring in more than 10% of
patients. The only grade 3 or worse adverse event
deemed related to bevacizumab was hypertension (in
one patient). 25 serious adverse events were reported in
16 (31%) of 52 patients, including febrile neutropenia or
febrile bone marrow aplasia (in ten patients [19%]), anal
abscess (in two patients [4%]), and ejection fraction
decrease, pyrexia, appendicitis, septic shock, back pain,
hyperthermia, malaise, metrorrhagia, jugular vein
thrombosis, hand-foot syndrome, and vomiting (one
patient each [2%]). Four patients (8%) had cardiac
failure. In one patient, ejection fraction decreased from
70% to 45% after four cycles of fl uorouracil, epirubicin,
and cyclophosphamide; the latest documented ejection
fraction showed recovery to 60%. One patient had an
ejection fraction of 44% decreasing to 38% in conjunction
with a tachyarrhythmia (atrial fi brillation) and another
patient had an ejection fraction that decreased from
70% to 49% on study, followed by recovery. One other
patient had an ejection fraction of 50% during the study
that recovered to 65%; this patient had an associated
underlying pericardial eff usion. Of the four cardiac
failures, two were New York Heart Association grade 2
(two symptomatic cardiac failures) and two were New
York Heart Association grade 1 (two left ventricular
dysfunction without clinical symptoms). One patient
received granulocyte colony-stimulating factor after
onset of neutropenia and the patient fully recovered.
Three (6%) of 52 patients prematurely discontinued
bevacizumab because of adverse events (one each of skin
Before cycle 1Before cycle 5
Circulating tumour cells per 7·5 mL
Log circulating endothelial cells per 4 mL
Before cycle 1Before cycle 5Before adjuvant
Figure 2: Circulating tumour cell (A) and circulating endothelial cell (B) counts in individual patients during treatment
We calculated p values with McNemar’s test (A) or the Wilcoxon signed-rank test (B). *Cycle 5 versus baseline (data for 45 patients). †End of treatment versus cycle 5
(data for 39 patients). ‡Before adjuvant treatment versus end of treatment (data for 35 patients).
Baseline Before cycle 5 4 weeks before
4 weeks after
Circulating tumour cells
Patients with available data
One or more cells per 7·5 mL
Five or more cells per 7·5 mL
Circulating endothelial cells
Patients with available data
Median cells per 4 mL
8 (4–13, 1–727)
22 (55%) 20 or more cells per 4 mL 8 (16%)
Data are n, n (%), or median (IQR, range).
Table 4: Circulating tumour cell and circulating endothelial cell counts
www.thelancet.com/oncology Vol 13 April 2012 381
toxicity, bone marrow failure, and colitis). Two patients
(4%) discontinued all treatment prematurely because of
toxicity (one because of persistent thrombocytopenia and
one because of grade 4 febrile neutropenia). No deaths
related to treatment occurred. Reported postoperative
event rates were much the same as those reported with
neoadjuvant chemotherapy without bevacizumab. The
most common postoperative complications were seroma
(in 14 [29%] of 49 patients) and wound-healing
complications (in fi ve [10%] of 49 patients). No bleeds
occurred that required surgery.
In the translational research study, we assessed
potential relations between circulating tumour cell
counts, circulating endothelial cell counts, and pCR
according to investigator review of pCR. At baseline,
18 (35%) of 52 patients had one or more circulating
tumour cell per 7·5 mL (range 1–92 cells). Circulating
tumour cells positivity and circulating endothelial cell
counts did not correlate with clinical–pathological charac-
teristics at baseline. The circulating tumour cell detection
rate decreased signifi cantly during the fi rst cycles of
chemotherapy plus bevacizumab (p=0·03), and remained
low and stable throughout the following cycles (fi gure 2,
table 4). We noted a progressive and signifi cant increase
in the circulating endothelial cell count until presurgery
bevacizumab discontinuation (fi gure 2).
There was no correlation between circulating tumour
cell detection and pCR at any timepoint. A circulating
endothelial cell count of fewer than 20 cells per 4 mL (as
in the criteria suggested by ATHENA24) after the fi rst
four cycles of therapy was positively associated with pCR
(23 [48%] of 48 patients assessed; p=0·01). The clinical
response rate was 98%, meaning that correlation testing
could not be done for this endpoint. In terms of
lymphatic dissemination of cancer cells, we noted no
signifi cant correlation between circulating tumour cell
positivity and axillary lymph node status either at
baseline (clinical nodal status, p=0·33) or after treatment
(pathological nodal status, p=0·18). Of 48 patients with
circulating endothelial cell values at baseline and cycle 5,
eight presented with hypertension. A non-parametric
Wilcoxon test showed no signifi cant correlation between
the variation in circulating endothelial cell counts
(cycle 5–baseline) and hypertension (p=0·91).
In 18 patients with circulating tumour cells, HER2
immuno cytofl uorescence status was regarded as strong
(either class 2 or 3 staining) in 15 patients (and negative
in two and non-assessable in one) and was globally
homogeneous on visual examination. During treatment,
HER2 immuno fl uorescence was stable over time; we
noted a change in only two patients (one became HER2
negative and one became HER2 positive).
In the BEVERLY-2 study, we showed that bevacizumab
plus fl uorouracil, epirubicin, and cyclophosphamide,
followed by docetaxel and trastuzumab, is an active
treatment regimen in HER2-positive infl ammatory
breast cancer (panel). Conventionally, patients with
infl ammatory breast cancer are less likely to respond to
treatment and have a poorer prognosis than are those
with other breast cancers. However, advances have been
made in the treatment of patients with infl ammatory
cancer in the past 15 years. In the phase 2 PEGASE 02
study,33 patients received neoadjuvant chemotherapy
with cyclo phosphamide, doxorubicin, and fl uorouracil.
Although the clinical response rate was high (90%),
pCR was achieved in only 32% of patients, which
nevertheless compared favourably with traditional
chemo therapy regimens (pCR 0–25%).33 This outcome
showed the chemosensitivity of infl ammatory breast
cancer. However, much the same proportion of patients
achieved a pCR in PEGASE 07 (33 [38%] of 87 patients)
in patients receiving primary high-dose chemo therapy
(epirubicin and cyclophosphamide with autologous
peripheral blood stem-cell transplantation) followed by
surgery, radiotherapy, and observation.35 In a randomised
study13 of patients with infl ammatory breast cancer
assigned to receive chemotherapy alone or chemotherapy
plus trastuzumab, fi ve of 19 (26%) patients who received
chemotherapy alone achieved a pCR compared with
15 of 23 (65%) who received chemotherapy and
trastuzumab (p=0·02). After treatment of a second
cohort with chemotherapy and trastuzumab, 27 of
45 (60%) patients had achieved a pCR.13 Much the same
result was reported in a cohort of patients with
infl ammatory breast cancer who were treated with
chemotherapy plus trastuzumab (10 [63%] of 16 patients
achieved a pCR).36 The proportion of patients with a pCR
that we noted in our study—63·5%—was high and was
much the same as previous studies of trastuzumab
combination therapy in locally advanced breast cancer
(including infl ammatory breast cancer).13,36
Analysis of these data suggests that bevacizumab and
trastuzumab plus chemotherapy is a positive step in terms
of achievement of pCR. Although survival data are not
available for our study so far, previous studies have linked
pCR with signifi cant improvements in long-term out-
comes in the setting of HER2-positive primary breast
cancer, including disease-free survival and overall
survival.37 In the TECHNO trial,37 patients achieving a pCR
had signifi cantly improved 3-year disease-free survival
and overall survival compared with those who did not
achieve a pCR. Addition of trastuzumab to neoadjuvant
chemotherapy in NOAH9 further improved 3-year event-
free survival in patients with HER2 positive infl ammatory
breast cancer (hazard ratio 0·27, 95% CI 0·11–0·65).
Although toxicity was generally consistent with reports
from previous studies, cardiac toxicity in our study was
higher than has been reported in phase 3 trials of
bevacizumab plus chemotherapy in patients with
metastatic breast cancer.38,39 However, toxicity in our
study was lower than that reported in the TORI-B03 study40
of fi rst-line bevacizumab plus trastuzumab in
www.thelancet.com/oncology Vol 13 April 2012
HER2-expressing unresectable locally advanced breast
cancer or metastatic breast cancer. As in TORI-B03,40 most
cases were mild-to-moderate and were resolved. Cardiac
toxicity was not reported in the GeparQuinto41 study and
no factors have yet been identifi ed that are predictive for
cardiac toxicity with this regimen. Further study is needed
to assess the combination’s safety. The ongoing
phase 3 studies BETH (NCT00625898) and AVEREL
(NCT00391092) are presently assessing bevacizumab and
trastuzumab in HER2-positive breast cancer. The phase 3
NeoALLTO study42 excluded patients with infl ammatory
breast cancer and although about 40% of patients in
NeoSphere43 had locally advanced breast cancer (32%) or
infl ammatory breast cancer (6%), no response analysis has
been undertaken in this subpopulation.
We are unable to compare studies in infl ammatory
breast cancer with those in non-infl ammatory breast
cancer. In non-infl ammatory breast cancer studies, few
patients undergo complete mastectomy and identifi cation
of complete responses is often more straightforward in
patients with a partial mastectomy. Treatment of HER2-
positive breast cancer is evolving, with randomised
studies suggesting a trend towards increased rates of
pCR with trastuzumab in combination with lapatinib
and other tyrosine-kinase inhibitors.42,44–47 However,
because these studies were done in metastatic breast
cancer and advanced breast cancer, whether this
combination is a valid treatment strategy for infl ammatory
breast cancer remains to be seen.
Our study is the fi rst to report circulating tumour cell
detection specifi cally in patients with non-metastatic
infl ammatory breast cancer. Circulating tumour cell
count, via CellSearch, is approved by the US Food and
Drug Administration for management of metastatic
breast cancer, but is investigational in non-metastatic
disease. Circulating tumour cell detection rates (one or
more cells per 7·5 mL) in the neoadjuvant setting were
23% in REMAGUS0223 and 22% in GeparQuattro,32 with
4% crossing the threshold (≥5 cells per 7·5 mL) for
positivity in REMAGUS02 and 3% in GeparQuattro. In
our study, the detection rate at baseline of 35% was
higher than either of these non-infl ammatory breast
cancer studies; seven (13%) of 52 patients in BEVERLY-2
had a baseline circulating tumour cell count of fi ve or
more cells per 7·5 mL, although they had no distant
metastases. This diff erence may be a result of the high
vascularisation and metastatic potential of infl ammatory
breast cancer. Concentrations of circulating tumour cells
dropped between baseline and cycle 5 (p=0·03). This
large and early decrease diff ers from reports in two
previous neoadjuvant studies without bevacizumab, in
which positivity for circulating tumour cells was stable or
slightly decreased during treatment.23,32 This decrease
was consistent with reports from ATHENA24 and
IC 2006-0425 in metastatic breast cancer, suggesting it is
related to bevacizumab use. In agreement with previous
reports from the REMAGUS02,23 and GeparQuattro,32
detection of circulating tumour cells was not correlated
with pCR at any timepoint.
Baseline circulating endothelial cell counts and their
increase before the introduction of trastuzumab at cycle 5
were much the same as those reported in patients with
metastatic breast cancer who were treated with an
equivalent regimen (chemotherapy and bevacizumab) in
ATHENA.24 This increase and the reported kinetics of
circulating endothelial cells overall might be attributable
to the eff ect of bevacizumab, the chemotherapy, or their
combination. In our study, a circulating endothelial cell
count of fewer than 20 cells per 4 mL before cycle 5 was
signifi cantly associated with attainment of pCR. This
correlation suggests circulating endothelial cells are a
promising indicator of effi cacy of the chemotherapy,
bevacizumab, and trastuzumab neoadjuvant regimen in
patients with infl ammatory breast cancer; this fi nding
should be confi rmed in independent trials. Follow-up is
required to allow investigation of the correlation between
circulating tumour cell counts, circulating endothelial
cell counts, and survival endpoints in this study.
Costa and colleagues48 suggested that neoadjuvant
chemotherapy achieves equivalent responses in patients
with infl ammatory breast cancer or locally advanced
breast cancer when compared with operable breast
cancer. However, no recent randomised phase 3 trials
devoted to infl ammatory breast cancer have been
published to support this suggestion. Several diff erent
options could be proposed to confi rm our results
including a randomised phase 2 trial, a comparison with
a theoretical pCR rate expected in our study with validated
nomograms, or a proof-of-concept study in which
bevacizumab is introduced if initial response is absent
by clinicoradiological, circulating tumour-circulating
endothelial cell counts, and/or FDG-PET assessment.
Panel: Research in context
We systematically reviewed PubMed, Web of Knowledge, and ClinicalTrials.gov with the
keywords “infl ammatory breast cancer” between Jan 1, 2002, and Jan 1, 2007, for articles
published in English. Our search identifi ed no publications or planned/ongoing clinical trials
of trastuzumab and bevacizumab in non-metastatic HER2-positive infl ammatory breast
cancer. For non-metastatic infl ammatory breast cancer, the present standard of care is a
neoadjuvant combination of anthracyclines and a taxane, with additional trastuzumab in
HER2-positive patients. These fi ndings, together with the potentially important role of
neoangiogenesis in infl ammatory breast cancer, led us to initiate the BEVERLY-2 study.
In our phase 2 study, we showed a high rate of pathological complete response and provide
evidence for the effi cacy of the combination of a taxane-anthracycline chemotherapy
backbone with trastuzumab and bevacizumab in this rare disease. When combined with the
results of the infl ammatory breast cancer subgroup analysis within the NOAH study,10 our
data suggest that the combination of a taxane-anthracycline chemotherapy backbone with
trastuzumab should be a standard treatment for non-metastatic infl ammatory breast cancer
for patients with HER2-positive disease. The importance of bevacizumab within this
combination needs further confi rmation.
www.thelancet.com/oncology Vol 13 April 2012 383
Our study has limitations. Because this trial was non-
randomised and non-comparative, assessment of the
contribution of bevacizumab in this setting was diffi cult.
In the absence of a standardised pCR defi nition, the
standards systems used in France (Sataloff and Chevalier
classifi cations) were used, although they are not
commonly used in some other countries. Survival data
are not available and longer follow-up is needed to fully
assess the benefi t of this treatment regimen in the long
term. Nevertheless, our study showed that the combination
of bevacizumab, trastuzumab, and chemo therapy was
effi cacious and well tolerated in patients with previously
untreated primary infl ammatory breast cancer; follow-up
is needed to assess long-term toxicity and eff ects on
survival. Phase 3 data are also needed to confi rm the long-
term clinical effi cacy of this treatment strategy.
J-YP, TD, MC, JG, HR, and PV recruited and managed patients in the
study and J-YP, MC, JG, FL, HR, and TB were involved in data collection.
J-YP, TD, MP, SK, and PV participated in protocol development. J-YP,
TD, J-MF, HR, EC-J, MP, SK, F-CB, and PV analysed and interpreted the
data. J-YP, MP, SK, and PV prepared the fi rst draft of the manuscript. All
authors were given the opportunity to comment on the draft manuscript
and saw and approved the fi nal version of the report.
Confl icts of interest
J-YP and JG have had consultant or advisory roles for and have received
honoraria and research funding from Roche. TP has had consultant or
advisory roles for Roche and Novartis and has received honoraria from
Roche. TB and PV have had consultant or advisory roles for and have
received research funding from Roche. MP and SK are employees of
Roche. All other authors declared that they have no confl icts of interest.
The trial was sponsored by Roche (France). Editorial assistance (funded
by Roche [France]) was provided by Rachel Edwards (Prism Ideas,
1 Dawood S, Ueno NT, Valero V, et al. Diff erences in survival among
women with stage III infl ammatory and noninfl ammatory locally
advanced breast cancer appear early: a large population-based study.
Cancer 2011; 117: 1819–26.
2 Anderson WF, Schairer C, Chen BE, Hance KW, Levine PH.
Epidemiology of infl ammatory breast cancer (IBC). Breast Dis
2005–2006; 22: 9–23.
3 Lerebours F, Bieche I, Lidereau R. Update on infl ammatory breast
cancer. Breast Cancer Res 2005; 7: 52–58.
4 Dushkin H, Cristofanilli M. Infl ammatory breast cancer.
J Natl Compr Canc Netw 2011; 9: 233–40.
5 Cristofanilli M, Valero V, Buzdar AU, et al. Infl ammatory breast
cancer (IBC) and patterns of recurrence: understanding the biology
of a unique disease. Cancer 2007; 110: 1436–44.
6 Rayson D, Richel D, Chia S, Jackisch C, van der Vegt S, Suter T.
Anthracycline-trastuzumab regimens for HER2/neu-overexpressing
breast cancer: current experience and future strategies. Ann Oncol
2008; 19: 1530–39.
7 Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al, and the
Herceptin Adjuvant (HERA) Trial Study Team. Trastuzumab after
adjuvant chemotherapy in HER2-positive breast cancer.
N Engl J Med 2005; 353: 1659–72.
8 Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy
plus a monoclonal antibody against HER2 for metastatic breast
cancer that overexpresses HER2. N Engl J Med 2001; 344: 783–92.
9 Gianni L, Eiermann W, Semiglazov V, et al. Neoadjuvant
chemotherapy with trastuzumab followed by adjuvant trastuzumab
versus neoadjuvant chemotherapy alone, in patients with
HER2-positive locally advanced breast cancer (the NOAH trial):
a randomised controlled superiority trial with a parallel
HER2-negative cohort. Lancet 2010; 375: 377–84.
10 Baselga J, Semiglazov V, Manikhas GM, et al. Effi cacy of
neoadjuvant trastuzumab in patients with infl ammatory breast
cancer: data from the NOAH (NeOAdjuvant Herceptin) phase III
trial. Eur J Cancer Suppl 2007; 5: 193 (abstr 2030).
11 Petrelli F, Borgonovo K, Cabiddu M, Ghilardi M, Barni S.
Neoadjuvant chemotherapy and concomitant trastuzumab in breast
cancer: a pooled analysis of two randomized trials. Anticancer Drugs
2011; 22: 128–35.
12 Wildiers H, Neven P, Christiaens MR, et al. Neoadjuvant
capecitabine and docetaxel (plus trastuzumab): an eff ective
non-anthracycline-based chemotherapy regimen for patients with
locally advanced breast cancer. Ann Oncol 2011; 22: 588–94.
13 Buzdar AU, Valero V, Ibrahim NK, et al. Neoadjuvant therapy with
paclitaxel followed by 5-fl uorouracil, epirubicin, and
cyclophosphamide chemotherapy and concurrent trastuzumab in
human epidermal growth factor receptor 2-positive operable breast
cancer: an update of the initial randomized study population and
data of additional patients treated with the same regimen.
Clin Cancer Res 2007; 13: 228–33.
14 Ferrara N. Vascular endothelial growth factor: basic science and
clinical progress. Endocr Rev 2004; 25: 581–611.
15 Gray R, Bhattacharya S, Bowden C, Miller K, Comis RL.
Independent review of E2100: a phase III trial of bevacizumab plus
paclitaxel versus paclitaxel in women with metastatic breast cancer.
J Clin Oncol 2009; 27: 4966–72.
16 Miles DW, Chan A, Dirix LY, et al. Phase III study of bevacizumab
plus docetaxel compared with placebo plus docetaxel for the
fi rst-line treatment of human epidermal growth factor receptor
2-negative metastatic breast cancer. J Clin Oncol 2010; 28: 3239–47.
17 Robert NJ, Diéras V, Glaspy J, et al. RIBBON-1: randomized,
double-blind, placebo-controlled, phase III trial of chemotherapy
with or without bevacizumab for fi rst-line treatment of human
epidermal growth factor receptor 2-negative, locally recurrent or
metastatic breast cancer. J Clin Oncol 2011; 29: 1252–60.
18 Brufsky A, Bondarenko I, Smirnov V, et al. RIBBON-2:
a randomized, double-blind, placebo-controlled, phase III trial
evaluating the effi cacy and safety of bevacizumab in combination
with chemotherapy for second-line treatment of HER2-negative
metastatic breast cancer. Cancer Res 2009; 69 (suppl 3): abstr 42.
19 Wedam SB, Low JA, Yang SX, et al. Antiangiogenic and antitumor
eff ects of bevacizumab in patients with infl ammatory and locally
advanced breast cancer. J Clin Oncol 2006; 24: 769–77.
20 Pegram M, Yeon C, Durna LN, et al. Phase I combined biological
therapy of breast cancer using two humanized monoclonal
antibodies directed against HER2 proto-oncogene and vascular
endothelial growth factor (VEGF). Breast Cancer Res Treat 2004;
88 (suppl 1): abstr 3039.
21 Pegram M, Chan D, Dichmann R, et al. Phase II combined
biological therapy targeting the HER2 proto-oncogene and the
vascular endothelial growth factor using trastuzumab (T) and
bevacizumab (B) as fi rst line treatment of HER2-amplifi ed breast
cancer. Breast Cancer Res Treat 2006; 100 (suppl 1): abstr 301.
22 Rack BK, Schindlbeck C, Andergassen A, et al. Use of circulating
tumor cells (CTC) in peripheral blood of breast cancer patients
before and after adjuvant chemotherapy to predict risk for relapse:
the SUCCESS trial. Proc Am Soc Clin Oncol 2010;
28 (suppl): abstr 1003.
23 Pierga JY, Bidard FC, Mathiot C, et al. Circulating tumor cell
detection predicts early metastatic relapse after neoadjuvant
chemotherapy in large operable and locally advanced breast
cancer in a phase II randomized trial. Clin Cancer Res 2008;
24 Bidard FC, Mathiot C, Degeorges A, et al. Clinical value of
circulating endothelial cells and circulating tumor cells in
metastatic breast cancer patients treated fi rst line with bevacizumab
and chemotherapy. Ann Oncol 2010; 21: 1765–71.
25 Pierga J-Y, Hajage D, Bachelot T, et al. High independent prognostic
and predictive value of circulating tumor cells compared with
serum tumor markers in a large prospective trial in fi rst-line
chemotherapy for metastatic breast cancer patients. Ann Oncol 2011;
published online June 3. DOI:10.1093/annonc/mdr263.
26 Valero V, Buzdar AU, Hortobagyi GN. Infl ammatory breast cancer:
clinical features and the role of multimodality therapy. Breast J 1996;
Articles Download full-text
www.thelancet.com/oncology Vol 13 April 2012
27 Wolff AC, Hammond ME, Schwartz JN, et al, and the American
Society of Clinical Oncology, and the College of American
Pathologists. American Society of Clinical Oncology/College of
American Pathologists guideline recommendations for human
epidermal growth factor receptor 2 testing in breast cancer.
J Clin Oncol 2007; 25: 118–45.
28 Sataloff DM, Mason BA, Prestipino AJ, Seinige UL, Lieber CP,
Baloch Z. Pathologic response to induction chemotherapy in locally
advanced carcinoma of the breast: a determinant of outcome.
J Am Coll Surg 1995; 180: 297–306.
29 Chevallier B, Bastit P, Graic Y, et al. The Centre H. Becquerel
studies in infl ammatory non metastatic breast cancer. Combined
modality approach in 178 patients. Br J Cancer 1993; 67: 594–601.
30 Sahoo S, Lester SC. Pathology of breast carcinomas after
neoadjuvant chemotherapy: an overview with recommendations
on specimen processing and reporting. Arch Pathol Lab Med 2009;
31 Penault-Llorca F, Abrial C, Raoelfi ls I, et al. Comparison of the
prognostic signifi cance of Chevallier and Sataloff ’s pathologic
classifi cations after neoadjuvant chemotherapy of operable breast
cancer. Hum Pathol 2008; 39: abstr 1221.
32 Riethdorf S, Müller V, Zhang L, et al. Detection and HER2
expression of circulating tumor cells: prospective monitoring in
breast cancer patients treated in the neoadjuvant GeparQuattro
trial. Clin Cancer Res 2010; 16: 2634–45.
33 Viens P, Palangié T, Janvier M, et al. First-line high-dose sequential
chemotherapy with rG-CSF and repeated blood stem cell
transplantation in untreated infl ammatory breast cancer: toxicity
and response (PEGASE 02 trial). Br J Cancer 1999; 81: 449–56.
34 Dawood S, Merajver SD, Viens P, et al. International expert panel
on infl ammatory breast cancer: consensus statement for
standardized diagnosis and treatment. Ann Oncol 2011; 22: 515–23.
35 Viens P, Palangie T, Ferrero J-M, et al. 5FU-docetaxel maintenance
regimen after high-dose of pre-operative epirubicin-cyclophosphamide
for treatment of infl ammatory breast cancer: results of the
FNCLCC-Pegase 07 trial. 32nd Annual San Antonio Breast Cancer
Symposium; San Antonio, TX, USA; Dec 9–13, 2009: abstr 2081.
36 Dawood S, Gong Y, Broglio K, et al. Trastuzumab in primary
infl ammatory breast cancer (IBC): high pathological response rates
and improved outcome. Breast J 2010; 16: 529–32.
37 Untch M, Fasching PA, Konecny GE, et al. Abstract P1-11-03:
pathological complete response after neoadjuvant chemotherapy
+ trastuzumab treatment predicts survival and detects a patient
subgroup at high need for improvement of anti-HER2 therapy.
Three year median follow-up data of the TECHNO trial.
Cancer Res 2011; 70 (suppl 2): abstr S3–1.
38 Miller K, Wang M, Gralow J, et al. Paclitaxel plus bevacizumab
versus paclitaxel alone for metastatic breast cancer. N Engl J Med
2007; 357: 2666–76.
39 Miller KD, Chap LI, Holmes FA, et al. Randomized phase III trial of
capecitabine compared with bevacizumab plus capecitabine in
patients with previously treated metastatic breast cancer.
J Clin Oncol 2005; 23: 792–99.
40 Hurvitz S, Pegram M, Lin L, et al. Final results of a phase II trial
evaluating trastuzumab and bevacizumab as fi rst line treatment of
HER2-amplifi ed advanced breast cancer. Cancer Res 2010;
69: abstr 6094.
41 von Minckwitz G, Eidtmann H, Loibl S, et al. Integrating
bevacizumab, everolimus, and lapatinib into current neoadjuvant
chemotherapy regimen for primary breast cancer. Safety results of
the GeparQuinto trial. Ann Oncol 2011; 22: 301–06.
42 Baselga J, Bradbury I, Eidtmann H, et al. Lapatinib with
trastuzumab for HER2-positive early breast cancer (NeoALTTO):
a randomised, open-label, multicentre, phase 3 trial. Lancet 2012;
43 Gianni L, Pienkowski T, Im YH, et al. Effi cacy and safety of
neoadjuvant pertuzumab and trastuzumab in women with locally
advanced, infl ammatory, or early HER2-positive breast cancer
(NeoSphere): a randomised multicentre, open-label, phase 2 trial.
Lancet Oncol 2012; 13: 25–32.
44 Blackwell KL, Burstein HJ, Storniolo AM, et al. Randomized study
of lapatinib alone or in combination with trastuzumab in women
with ErbB2-positive, trastuzumab-refractory metastatic breast
cancer. J Clin Oncol 2010; 28: 1124–30.
45 Guarneri V, Frassoldati A, Bottini A, et al. Final results of a phase II
randomized trial of neoadjuvant anthracycline-taxane chemotherapy
plus lapatinib, trastuzumab, or both in HER2-positive breast cancer
(CHER-LOB trial). Proc Am Soc Clin Oncol 2011;
29 (suppl): abstr 507.
46 Blackwell K, Burstein H, Sledge G, et al. Updated survival analysis
of a randomized study of lapatinib alone or in combination with
trastuzumab in women with HER2-positive metastatic breast cancer
progressing on trastuzumab therapy. Cancer Res 2010; 69: abstr 61.
47 Callahan RD, Patel R, Chan D, et al. Phase II trial of presurgical
treatment with trastuzumab (H) or lapatinib (Ty) or the
combination of trastuzumab and lapatinib (H+Ty), followed by six
cycles of docetaxel (T) and carboplatin (C) with trastuzumab (TCH)
or lapatinib (TCTy) or the combination of trastuzumab and
lapatinib (TCHTy) in patients with HER2+ breast cancer.
34th San Antonio Breast Cancer Symposium; San Antonio, TX,
USA; Dec 6–10, 2011. Abstract P1-11-12.
48 Costa SD, Loibl S, Kaufmann M, et al. Neoadjuvant chemotherapy
shows similar response in patients with infl ammatory or locally
advanced breast cancer when compared with operable breast
cancer: a secondary analysis of the GeparTrio trial data. J Clin Oncol
2010; 28: 83–91.