Straver, M. E. et al. The 70-gene signature as a response predictor for neoadjuvant chemotherapy in breast cancer. Breast Cancer Res. Treat. 119, 551-558

Breast Cancer Research and Treatment (Impact Factor: 3.94). 02/2010; 119(3):551-558. DOI: 10.1007/s10549-009-0333-1
Source: PubMed


The 70-gene signature (MammaPrint™) is a prognostic tool used to guide adjuvant treatment decisions. The aim of this study
was to assess its value to predict chemosensitivity in the neoadjuvant setting. We obtained the 70-gene profile of stage II–III
patients prior to neoadjuvant chemotherapy and classified the prognosis-signatures. Pathological complete remission (pCR)
was used to measure chemosensitivity. Among 167 patients, 144 (86%) were having a poor and 23 (14%) a good prognosis-signature.
None of the good prognosis-signature patients achieved a pCR (0/23), whereas 29/144 patients (20%) in the poor prognosis-signature
group did (P=0.015). All triple-negative tumors (n=38) had a poor prognosis-signature. Within the non triple-negative subgroup, the response of the primary tumor remained
associated with the classification of the prognosis-signature (P=0.023). A pCR is unlikely to be achieved in tumors that have a good prognosis-signature. Tumors with a poor prognosis-signature
are more sensitive to chemotherapy.


Available from: Harm van Tinteren
The 70-gene signature as a response predictor for neoadjuvant
chemotherapy in breast cancer
Marieke E. Straver Æ Annuska M. Glas Æ Juliane Hannemann Æ Jelle Wesseling Æ
Marc J. van de Vijver Æ Emiel J. Th. Rutgers Æ Marie-Jeanne T. F. D. Vrancken Peeters Æ
Harm van Tinteren Æ Laura J. van‘t Veer Æ Sjoerd Rodenhuis
Received: 27 January 2009 / Accepted: 29 January 2009 / Published online: 13 February 2009
Ó Springer Science+Business Media, LLC. 2009
Abstract The 70-gene signature (MammaPrint
prognostic tool used to guide adjuvant treatment decisions.
The aim of this study was to assess its value to predict
chemosensitivity in the neoadjuvant setting. We obtained
the 70-gene profile of stage II–III patients prior to neoad-
juvant chemotherapy and classified the prognosis-
signatures. Pathological complete remission (pCR) was
used to measure chemosensitivity. Among 167 patients,
144 (86%) were having a poor and 23 (14%) a good
prognosis-signature. None of the good prognosis-signature
patients achieved a pCR (0/23), whereas 29/144 patients
(20%) in the poor prognosis-signature group did (P =
0.015). All triple-negative tumors (n = 38) had a poor
prognosis-signature. Within the non triple-negative sub-
group, the response of the primary tumor remained
associated with the classification of the prognosis-signature
(P = 0.023). A pCR is unlikely to be achieved in tumors
that have a good prognosis-signature. Tumors with a poor
prognosis-signature are more sensitive to chemotherapy.
Keywords Breast cancer Neoadjuvant
Chemosensitivity Predictive Gene expression signature
The mortality of breast cancer is decreasing in the devel-
oped part of the world. This is largely a result of effective
adjuvant systemic therapy [1]. An important problem of
adjuvant therapy is overtreatment, which consists of the
administration of adjuvant therapy in patients for whom
surgical resection of the tumor alone would be curative.
These patients will experience toxicity without benefiting
from the treatment. Currently, the selection of patients for
adjuvant treatment is based on tumor and patients charac-
teristics such as endocrine responsiveness, tumor grade,
lymph node status and age. One strategy to reduce over-
treatment is the use of prognostic biomarkers. Systematic
analysis of gene-expression patterns using microarray
technology has led to the discovery of prognostic gene-
expression signatures, one of which is the 70-gene prog-
nostic-signature (MammaPrint
The prognostic value of 70-gene profile has been vali-
dated in a range of series of patients [36]. These studies
M. E. Straver E. J. Th. Rutgers
M.-J. T. F. D. Vrancken Peeters
Department of Surgical Oncology, The Netherlands Cancer
Institute Antoni van Leeuwenhoek Hospital, Plesmanlaan 121,
1066 CX Amsterdam,
The Netherlands
M. E. Straver
A. M. Glas L. J. van‘t Veer
Agendia BV, Sience Park 406, 1098 XH Amsterdam,
The Netherlands
J. Hannemann J. Wesseling M. J. van de Vijver
L. J. van‘t Veer
Department of Pathology, The Netherlands Cancer
Institute Antoni van Leeuwenhoek Hospital, Plesmanlaan 121,
1066 CX Amsterdam, The Netherlands
H. van Tinteren
Department of Biometrics, The Netherlands Cancer
Institute Antoni van Leeuwenhoek Hospital, Plesmanlaan 121,
1066 CX Amsterdam, The Netherlands
S. Rodenhuis (&)
Department of Medical Oncology, The Netherlands Cancer
Institute Antoni van Leeuwenhoek Hospital, Plesmanlaan 121,
1066 CX Amsterdam, The Netherlands
Breast Cancer Res Treat (2010) 119:551–558
DOI 10.1007/s10549-009-0333-1
Page 1
confirmed that the 70-gene signature accurately discrimi-
nates between patients at high risk of distant metastasis and
death and patients with a favorable prognosis. Furthermore,
it was shown that the 70-gene signature adds independent
prognostic information to that provided by commonly used
clinicopathological factors.
Node-negative patients with a good prognosis-signature
who did not receive any adjuvant systemic therapy, had a
10-year overall survival of 89% (95%CI 0.81–0.94). These
results suggest that adjuvant therapy could be limited to
endocrine treatment for tumors with a good prognosis-sig-
nature. At present, the MINDACT trial is addressing this
question. In this trial, treatment selection based on the 70-
gene signature as compared to clinical risk assessment, may
show that it does not compromise the overall outcome [7].
Virtually all (92–98%) of the good prognosis-signature
tumors show a high expression of the estrogen receptor
(ER). Furthermore, tumors with a good prognosis signature
are usually those with lower proliferative rates. It is,
therefore, often assumed that good prognosis-signature
tumors may be less sensitive to chemotherapy than tumors
with a poor prognosis-signature. If true, the 70-gene sig-
nature would have predictive power in addition to its
prognostic value.
To analyze the predictive value of the 70-gene profile,
we determined the 70-gene signature in tumors of patients
treated with neoadjuvant chemotherapy at The Netherlands
Cancer Institute. The objectives of this study were: (1) to
analyze the association between the pathological complete
response (pCR) rate and the results of the prognostic sig-
nature test, and (2) to assess the predictive value in
different subgroups defined by the expression of hormone
receptors and the amplification of the human epidermal
growth factor receptor 2 (HER2) genes.
Fresh frozen tumor biopsies and clinical data were collected
from a consecutive series of 171 patients who received
neoadjuvant chemotherapy at The Netherlands Cancer
Institute between 2000 and 2008. Patients received neoad-
juvant chemotherapy in one of two clinical studies ongoing
or received treatment according to the standard arm of these
trials [8]. Patients with invasive breast cancer greater than
3 cm and/or involved lymph nodes were eligible for these
studies. The clinical studies were approved by the institu-
tional ethical committee and informed consent was obtained
from all patients. Prior to neoadjuvant treatment, 14-gauge
biopsies of the breast tumor were taken under ultrasound
guidance. These biopsies were snap-frozen in liquid
nitrogen and stored at -70°C. All patients of whom ade-
quate RNA could be extracted from the tumor samples were
included in the study, provided that they had undergone
surgery to determine the pathological remission status.
Clinicopathological data
Clinical data were collected from medical records, blinded
to the 70-gene prognosis-signature. Tumor size was assessed
by MRI (n = 155) when available or by mammogram and
ultrasound examination (n = 16). Nodal status prior to
neoadjuvant chemotherapy was determined by ultrasound
guided fine-needle aspiration or, when negative sentinel
node biopsy. In 16 patients with inconclusive cytological
assessment the pathological nodal status prior to neoadju-
vant chemotherapy remained unknown. Tumor grading was
defined according to the Elston and Ellis method [9].
Estrogen receptor (ER) status and progesterone receptor
status were determined by immunohistochemistry and
interpreted positive if more than 10% of the nuclei stained
positive. HER2 status was assessed by scoring the intensity
of membrane staining using immunohistochemistry.
Tumors with a score of 3? (strong homogeneous staining)
were considered HER2-positive. In case of 2? scores
(moderate homogeneous staining) chromogenic in situ
hybridization (CISH) was used to determine amplification
[10]. Amplification was defined as a gene copy number of
over five per cell. Tumors were classified in three subgroups
according to their receptor status using immunohistochem-
ical staining; (1) ER-positive and HER2-negative tumors,
(2) triple negative tumors (ER-negative, PR-negative and
HER2-negative) and (3) HER2-positive tumors.
The treatment regimen depended on the presence or
absence of HER2 amplification. Preoperative chemotherapy
for HER2-negative tumors employed one of the following
regimens: AC (six cycles of doxorubicin 60 mg/m
cyclophosphamide 600 mg/m
, q 3 weeks); dose dense (dd)
AC (AC q 2 weeks with filgrastim) [11]; AD (six cycles of
doxorubicin 50 mg/m
and docetaxel 75 mg/m
) or DC (six
cycles of docetaxel 75 mg/m
and capecitabine 2 9 dd
1,000 mg/m
orally during 14 days, q 3 weeks) [12, 13]. For
HER2-positive tumors, the regimens included ddAC and
PTC (paclitaxel 80 mg/m
, trastuzumab 2 mg/kg
and carboplatin AUC 2–3 mg/ml min times 6, q 8 weeks)
after 2005 [14, 15]. After one or three cycles (depending on
the specific protocol) the tumor response was evaluated by
contrast enhanced MRI [16]. Chemotherapy regimens were
changed to a presumably non-cross-resistant regimen when
response failure was apparent upon radiological evaluation.
After the last course of chemotherapy patients underwent
mastectomy or breast conserving surgery. Three patients
had progressive disease. In two patients, surgery was per-
formed prior to the completion of chemotherapy and one
552 Breast Cancer Res Treat (2010) 119:551–558
Page 2
patient with mastitis carcinomatosa was treated with radi-
ation therapy only. To prevent bias, these patients were kept
in the analysis despite the fact that they underwent early or
no surgery.
RNA extraction and gene expression analysis
RNA isolation and amplification were performed as pre-
viously described [17]. One 5-lm tissue section of the
biopsy was hematoxylin and eosin stained to monitor the
tumor cell percentage of the tissue. Only specimens with at
least 50% tumor cells were further analyzed. To assess the
mRNA expression level of the 70 genes, RNA was
hybridized to a custom-designed array (MammaPrint
FDA 510(K) cleared), blinded to clinical data, at Agendia’s
ISO17025-certfified and CLIA accredited laboratories.
Tumors were classified as good prognosis-signature (low
risk) or poor prognosis signature (high risk) as described
previously [18].
Assessment of tumor response
Pathological complete response (pCR) was used as the
outcome measure. It was defined as the absence of invasive
carcinoma in both the breast and axilla at microscopic
examination of the resection specimen, regardless of the
presence of carcinoma in situ [19]. Furthermore, the
response of the primary tumor in the breast was assessed
separately. The response of the primary tumor was defined
as a pCR when no residual tumor cells were seen at
microscopy or as a ‘near pCR’ (npCR) when small num-
bers of scattered tumor cells or tumor cells in an area of
less than 2 mm in diameter were present.
Statistical analysis
Analyses were performed using SPSS version 15.1 (SPSS
Inc, Chicago, IL). The differences in patients and tumor
characteristics between the 70-gene poor and good prog-
nosis signature were tested using Fisher’s Exact test and
students t test. We used the Fisher’s Exact test to assess the
association between the response of the tumor and the
outcome of the 70-gene profile. Disease-free survival
curves were calculated using the Kaplan–Meier method
and compared using the log-rank test. P-values reported are
two sided.
The 70-gene profile was analyzed in 171 patients who
received neoadjuvant chemotherapy. Three of these 171
patients were excluded due to clinical reasons. Two
patients did not undergo surgery of the primary tumor and
in one patient the treatment with chemotherapy was dis-
continued early due to major toxicity. In one patient the
RNA quality was insufficient to perform gene profiling.
Thus, the predictive value of the 70-gene prognostic sig-
nature could be evaluated in 167 patients. (Fig. 1).
Patient and tumor characteristics are presented in
Table 1. Among the 167 patients, 23 (14%) had a good
prognosis-signature, whereas 144 (86%) patients had a
poor prognosis-signature. Tumors with a poor prognosis-
signature were of higher grade, and were more often
classified as triple-negative or HER2 positive tumors.
Consequently, more patients in the poor prognosis-signa-
ture group were treated with a trastuzumab based regimen.
The overall pCR rate (absence of invasive tumor in both
breast and axilla) was 17% (29/167). None of the patients
with a good prognosis-signature achieved a pCR (0/23).
The pCR rate in the poor prognosis-signature group was
Fig. 1 The patients selection and the classification of the 70-gene
signature. The pathological response in breast and axilla is determined
for both risk groups. A poor prognosis profile was significantly
associated with the pathological complete remission (pCR) of both
the breast and axilla (P = 0.015) and with response (near pCR or
pCR) of the primary tumor separately (P = 0.008)
Breast Cancer Res Treat (2010) 119:551–558 553
Page 3
20% (29/144; P = 0.015; Fig. 1). Furthermore, we asses-
sed chemosensitivity by separately analyzing the
pathological response of the primary tumor in the breast
(pCR and near-pCR). In the good prognosis-signature
group, 2 of the 23 patients achieved a near-pCR of the
primary tumor (9%). In the poor prognosis-signature group
the pathological response was 37% (53/144; P = 0.008).
Figure 2 shows the relation between the classification of
the 70-gene profile as a continuous variable and pCR.
Patients with a low MammaPrint Index have a higher
probability to achieve a pCR.
The pathological response (breast and axilla) was also
analyzed in subgroups that were characterized according to
receptor status of the tumor (Table 2). In ER-positive
(HER2-), triple-negative and HER2-positive tumors the
pCR rates were different; 3% (3/88), 34% (13/38) and 32%
Table 1 The patient and
tumor characteristics and the
association with the 70-gene
The nodal status was
determined prior to
chemotherapy (CT) with
ultrasound guided fine-needle
aspiration (FNA) and sentinel
node biopsy (SNB). pNX: no
pathological nodal status
ER, estrogen receptor; PR,
progesterone receptor; HER2,
human epidermal growth factor
receptor; NS, not significant
(P [ 0.05)
cyclophosphamide doxorubicine
(AC). Antracycline-taxane:
AC? capecitabine docetaxel
(CD) or cyclophosphamide/
docetaxel (AD). PTC:
carboplatin. BCT: breast
conserving surgery
Due to progression mastitis
carcinomatosa only radiation
n = 167
No. (%)
Good prognosis
n = 23
No. (%)
Poor prognosis
n = 144
No. (%)
P value
Mean age 46 46 46 NS
Range 23–68 31–58 23–68
Menopausal status
Pre-menopausal 119 18 (78) 101 (71) NS
Post-menopausal 39 5 (22) 34 (24)
Unknown 9 0 (0) 9 (5)
T stage (prior to CT)
T1 9 2(9) 7 (5) NS
T2 87 13(57) 74(51)
T3 62 7(30) 55 (38)
T4 9 1(4) 8(6)
pN stage (prior to CT)
pN0 (SNB-) 30 5 (22) 25 (17) NS
pN1 (SNB?/FNA?) 110 15 (65) 95(66)
pN3 (sub/supraclavicular) 11 0 (0) 11 (8)
pNX (cN0) 16 3 (13) 13 (19)
Ductal 131 14 (61) 117 (81) 0.008
Lobular 20 8 (35) 12 (8)
Others 16 1 (4) 15 (11)
I 7 3 (13) 6 (3) 0.011
II 88 16 (70) 72(50)
III 66 3 (13) 63 (44)
missing 6 1(4) 5 (3)
Subtype (based on receptor status)
ER? (ER? , HER2-) 88 21 (91) 67 (47) \0.001
TN (ER-, PR-, HER2-) 38 0 (0) 38 (26)
HER2? 41 2 (9) 39(27)
Primary systemic therapy
Antracycline-like 85 10(43) 75 (52) 0.036
Antracycline-taxane 31 9 (39) 22 (15)
Capectabine/docetaxel 14 2(9) 12 (8)
PTC 37 2(9) 35 (25)
Surgery breast
No surgery
1 0 1 (1) NS
Mastectomy 81 10 (44) 71 (49)
BCS 85 13 (56) 72 (50)
554 Breast Cancer Res Treat (2010) 119:551–558
Page 4
(13/41), respectively, (P \ 0.001). Among the ER-positive
(HER2-) patients, 21/88 (24%) were classified as having a
good prognosis-signature and 67/88 patients (76%) were
classified as having a poor prognosis-signature (Fig. 3).
None of the ER-positive patients with a good prognosis-
signature achieved a pCR and all three patients who did, had
a tumor with a poor prognosis-signature. Of the HER2-
positive patients, 2/41 (5%) were classified as good prog-
nosis-signature and 39/42 (95%) as having a poor prognosis-
signature. Both HER2-positive patients with a good
prognosis-signature were also ER-positive and both did not
achieve a pCR. Among the HER2-positive patients with a
poor prognosis-signature the pCR rate was 33% (13/39).
None of the triple negative tumors in this study (n = 38) had
a good prognosis-signature. We, therefore separately ana-
lyzed the predictive power of the 70 gene signature in the
subgroup comprising all ER-positive and/or HER2-positive
tumors. Thus, we excluded the patients with triple negative
tumors which all had a poor prognosis-signature. In this
non-triple negative subgroup the pCR rate in good and
poor prognosis-signature tumors was 0% (0/23) and 15%
(16/106), respectively, (P = 0.07). When the response of the
primary tumor was assessed separately, a significant asso-
ciation between the response and the result of the 70-gene
profile could be shown. Pathological response, pCR or near-
pCR of the primary tumor, was achieved in 9% (2/23) of the
patients with a good prognosis-signature tumor and in 32%
(34/106) of the patients having a poor prognosis-signature
tumor (P = 0.023).
After a median follow up of 25 months (range: 5–91),
17 patients had a relapse. These included local recurrences
in 2 patients and distant metastases in 15 patients. None of
the patients with a good prognosis-signature had a relapse.
The disease-free survivals are shown in Fig. 4 (P = 0.066).
To prospectively assess the predictive value of a prognostic
marker, large and logistically challenging clinical trials
are required, that—in case of node-negative breast
Fig. 2 The association between
the classification of the 70-gene
prognostic signature and the
pathological response.
Classification of the 70 gene
prognosis signature of each
sample is plotted. Tumors are
ordered by their correlation to
the average profile of the good
prognosis group. Patients with a
pCR are indicated by black dots
and patients without a pCR are
indicated by white dots
Table 2 The response in three subtypes characterized by receptor
(n = 88)
No. (%)
(n = 38)
No. (%)
(n = 41)
No. (%)
(n = 167)
No. (%)
Response breast ? axilla
pCR 3 (3) 13 (34) 13 (32) 29 (17)
Response breast
pCR 6 (7) 17 (45) 17 (42) 40 (24)
npCR 7 (8) 2 (5) 6 (15) 15 (9)
The response rate differed significantly between the subtypes.
(P \ 0.001)
ER, estrogen receptor; HER2, human epidermal growth factor
receptor 2; TN, triple negative (estrogen, progesterone and HER2
negative); pCR, pathological complete remission; npCR, near pCR
Four patients did not receive trastuzumab
Breast Cancer Res Treat (2010) 119:551–558 555
Page 5
cancer—may take decades to accumulate sufficient events
for a useful analysis. An alternative and more rapid
approach is to evaluate the predictive value of a prognostic
marker for chemosensitivity in the neoadjuvant setting.
Here, the pathological response can be used as endpoint,
since the achievement of pCR has gained wide acceptance
as a predictor of a good long-term prognosis [20, 21].
In this report a series of 167 patients with stage II and III
primary invasive breast cancer who received neoadjuvant
chemotherapy is described. Although several types of
chemotherapy were used, most patients were treated with
an anthracycline-based regimen. The proportion of tumors
with a good 70-gene prognosis-signature was 14%, which
is less than the percentage (38 and 41%) reported in earlier
series of node-positive patients. This is likely the result of
the inclusion criteria which intentionally selected clinically
higher-risk patients, with larger tumors and/or axillary
lymph node involvement, for neoadjuvant chemotherapy.
We observed that tumors with a poor prognostic 70-gene
signature are more likely to achieve a complete response
than those with a good prognostic signature. Even if the
triple-negative tumor subgroup was excluded, which is
usually associated with a relatively high pCR rate and in
this current study with an invariably poor prognosis 70-
gene profile, the response to chemotherapy remained sig-
nificantly higher in the poor prognosis-signature tumors.
This strongly suggests that the 70-gene profile has pre-
dictive power with respect to chemosensitivity.
The absence of ER expression and poor differentiation,
tumor characteristics more often seen in the poor signature
group, are generally believed to be associated with a higher
likelihood of response to chemotherapy [1]. Molecular
subtypes such as luminal, basal, ERBB2 and normal-like
subtypes, differ markedly with respect to prognosis, with a
basal-like subtype having a worse prognosis than a luminal
subtype [22]. To a degree, these molecular subtypes can be
also be distinguished using immunohistochemistry [23].
Carey et al. [24] classified tumors of a series of patients
undergoing neoadjuvant therapy in basal-like, luminal and
ERBB2-like molecular subtypes using immunohistochem-
ical staining of the hormone and HER2 receptors. A
significantly different response to neoadjuvant chemother-
apy was observed in these subtypes. Basal-like tumors (i.e.
triple negative) had a higher pCR rate compared to luminal
subtypes which express the estrogen receptor. Our study
confirms these findings, with a higher pCR rate in triple
negative tumors as compared to ER-positive tumors. Since
Fig. 3 The distribution of the
three different subtypes
characterized by receptor status
within good and poor prognosis
patients. All patients with triple
negative tumors had a tumor
with a poor prognosis-signature.
We, therefore, separately
analyzed the predictive power
of the 70 gene signature in the
subgroup including ER-positive
and HER2-positive tumors.
Response of the breast is
defined pathological complete
remission (pCR) or near pCR
Fig. 4 The disease-free survival of the good and poor prognosis
patients separately. The median follow up was 25 months and 17
relapses occurred in the poor prognosis group
556 Breast Cancer Res Treat (2010) 119:551–558
Page 6
all triple negative tumors had a poor prognosis-signature, it
was expected that the pCR rate would be higher in the poor
signature group. Nevertheless, the predictive value
remained after exclusion of the triple negative tumors,
suggesting additional predictive value for the 70-gene
Another prognostic gene expression test, the 21-gene
recurrence score (Oncotype DX
assay) has been corre-
lated with pCR in the neoadjuvant setting [25].
Subsequently, Paik et al. [26] retrospectively assessed its
predictive value for ER-positive patients in the adjuvant
setting. They showed that the neoadjuvant result was
confirmed in the adjuvant setting, as a high recurrence
score was associated with relatively greater benefit from
adjuvant cyclophosphamide, methotrexate, 5-flourouracil
chemotherapy in patients with ER-positive, lymph node-
negative disease.
Despite the proven predictive value of pCR in neoad-
juvant trials, there is no consensus on the measurement of
this important endpoint. Absence of tumor cells in both the
axilla and breast is the most stringent definition and was
therefore primarily used in this study. On the other hand,
we believe that a reduction of tumor load of the primary
tumor to single scattered tumor cells may also imply
chemosensitivity. To gain more insight in chemosensitivity
we separately assessed the responses of the primary tumors
according to this frequently used definition. However,
longer follow-up will be required to confirm that the
eradication of a large number of tumor cells is associated
with long-term survival.
In conclusion, this study shows that tumors with a poor
70-gene signature are more likely to achieve a pCR,
whereas tumors with a good prognosis-signature are not.
This finding has several important clinical implications.
Currently, the 70 gene profile is used to select patients for
adjuvant chemotherapy whereby chemotherapy is fre-
quently withheld for tumors with a good prognosis-
signature [27]. Our results add justification to this policy;
not only is the absolute risk of relapse lower in these
patients, but their tumors are also less sensitive to che-
motherapy. For the good prognosis signature group, this
may lead to a significantly lower proportional risk reduc-
tion than suggested in the Oxford overview or in the
frequently used Adjuvant! Online program [28]. We also
believe that the stratification of subjects according to
the 70-gene profile could be helpful in controlled trials
investigating the effectiveness of new drugs or new com-
binations of drugs.
Conflicts of interest This study was financially supported by
Agendia, the commercial company that markets the 70-gene signature
as Mammaprint. L. van ‘t Veer is a shareholder in and (part time)
employed by Agendia. A.M. Glas is employed by Agendia.
1. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG)
(2005) Effects of chemotherapy and hormonal therapy for early
breast cancer on recurrence and 15-year survival: an overview of
the randomised trials. Lancet 365:1687–1717
2. van ‘t Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao
M, Peterse HL, van der Kooy K, Marton MJ, Witteveen AT,
Schreiber GJ, Kerkhoven RM, Roberts C, Linsley PS, Bernards
R, Friend SH (2002) Gene expression profiling predicts clinical
outcome of breast cancer. Nature 415:530–536
3. Bueno-de-Mesquita JM, Linn SC, Keijzer R, Wesseling J, Nuyten
DS, van Krimpen C, Meijers C, de Graaf PW, Bos MM, Hart AA,
Rutgers EJ, Peterse JL, Halfwerk H, de Groot R, Pronk A, Floore
AN, Glas AM, van’t Veer LJ, van de Vijver MJ (2008) Validation
of 70-gene prognosis signature in node-negative breast cancer.
Breast Cancer Res Treat. doi:10.1007/s10549-008-0191-2
4. Buyse M, Loi S, Van’t Veer L, Viale G, Delorenzi M, Glas AM,
d’Assignies MS, Bergh J, Lidereau R, Ellis P, Harris A, Bogaerts
J, Therasse P, Floore A, Amakrane M, Piette F, Rutgers E,
Sotiriou C, Cardoso F, Piccart MJ (2006) Validation and clinical
utility of a 70-gene prognostic signature for women with node-
negative breast cancer. J Natl Cancer Inst 98:1183–1192
5. Mook S, Schmidt MK, Viale G, Pruneri G, Eekhout I, Floore A,
Glas AM, Bogaerts J, Cardoso F, Piccart-Gebhart MJ, Rutgers
ET, van’t Veer LJ (2008) The 70-gene prognosis-signature pre-
dicts disease outcome in breast cancer patients with 1–3 positive
lymph nodes in an independent validation study. Breast Cancer
Res Treat. doi:10.1007/s10549-008-0130-2
6. van de Vijver MJ, He YD, van’t Veer LJ, Dai H, Hart AA,
Voskuil DW, Schreiber GJ, Peterse JL, Roberts C, Marton MJ,
Parrish M, Atsma D, Witteveen A, Glas A, Delahaye L, van der
Velde T, Bartelink H, Rodenhuis S, Rutgers ET, Friend SH,
Bernards R (2002) A gene-expression signature as a predictor of
survival in breast cancer. N Engl J Med 347:1999–2009
7. Cardoso F, Van’t Veer L, Rutgers E, Loi S, Mook S, Piccart-
Gebhart MJ (2008) Clinical application of the 70-gene profile: the
MINDACT trial. J Clin Oncol 26:729–735
8. Hannemann J, Oosterkamp HM, Bosch CA, Velds A, Wessels
LF, Loo C, Rutgers EJ, Rodenhuis S, van de Vijver MJ (2005)
Changes in gene expression associated with response to neoad-
juvant chemotherapy in breast cancer. J Clin Oncol 23:3331–
9. Elston CW, Ellis IO (1991) Pathological prognostic factors in
breast cancer. I. The value of histological grade in breast cancer:
experience from a large study with long-term follow-up. Histo-
pathology 19:403–410
10. Hauser-Kronberger C, Dandachi N (2004) Comparison of chro-
mogenic in situ hybridization with other methodologies for HER2
status assessment in breast cancer. J Mol Histol 35:647–653
11. Citron ML, Berry DA, Cirrincione C, Hudis C, Winer EP, Gra-
dishar WJ, Davidson NE, Martino S, Livingston R, Ingle JN,
Perez EA, Carpenter J, Hurd D, Holland JF, Smith BL, Sartor CI,
Leung EH, Abrams J, Schilsky RL, Muss HB, Norton L (2003)
Randomized trial of dose-dense versus conventionally scheduled
and sequential versus concurrent combination chemotherapy as
postoperative adjuvant treatment of node-positive primary breast
cancer: first report of Intergroup Trial C9741/Cancer and Leu-
kemia Group B Trial 9741. J Clin Oncol 21:1431–1439
12. Lebowitz PF, Eng-Wong J, Swain SM, Berman A, Merino MJ,
Chow CK, Venzon D, Zia F, Danforth D, Liu E, Zujewski J
(2004) A phase II trial of neoadjuvant docetaxel and capecitabine
for locally advanced breast cancer. Clin Cancer Res 10:6764–
Breast Cancer Res Treat (2010) 119:551–558 557
Page 7
13. O’Shaughnessy JA, Blum JL (2006) Capecitabine/taxane com-
bination therapy: evolving clinical utility in breast cancer. Clin
Breast Cancer 7:42–50
14. Burris HIII, Yardley D, Jones S, Houston G, Broome C,
Thompson D, Greco FA, White M, Hainsworth J (2004) Phase II
trial of trastuzumab followed by weekly paclitaxel/carboplatin as
first-line treatment for patients with metastatic breast cancer. J
Clin Oncol 22:1621–1629
15. Perez EA, Suman VJ, Rowland KM, Ingle JN, Salim M, Loprinzi
CL, Flynn PJ, Mailliard JA, Kardinal CG, Krook JE, Thrower
AR, Visscher DW, Jenkins RB (2005) Two concurrent phase II
trials of paclitaxel/carboplatin/trastuzumab (weekly or every-3-
week schedule) as first-line therapy in women with HER2-over-
expressing metastatic breast cancer: NCCTG study 983252. Clin
Breast Cancer 6:425–432
16. Loo CE, Teertstra HJ, Rodenhuis S, van de Vijver MJ, Hanne-
mann J, Muller SH, Peeters MJ, Gilhuijs KG (2008) Dynamic
contrast-enhanced MRI for prediction of breast cancer response
to neoadjuvant chemotherapy: initial results. AJR Am J Roent-
genol 191:1331–1338
17. Weigelt B, Glas AM, Wessels LF, Witteveen AT, Peterse JL,
van’t Veer LJ (2003) Gene expression profiles of primary breast
tumors maintained in distant metastases. Proc Natl Acad Sci U S
A 100:15901–15905
18. Glas AM, Floore A, Delahaye LJ, Witteveen AT, Pover RC,
Bakx N, Lahti-Domenici JS, Bruinsma TJ, Warmoes MO, Ber-
nards R, Wessels LF, van’t Veer LJ (2006) Converting a breast
cancer microarray signature into a high-throughput diagnostic
test. BMC Genomics 7:278
19. Mazouni C, Peintinger F, Wan-Kau S, Andre F, Gonzalez-Ang-
ulo AM, Symmans WF, Meric-Bernstam F, Valero V, Hortobagyi
GN, Pusztai L (2007) Residual ductal carcinoma in situ in
patients with complete eradication of invasive breast cancer after
neoadjuvant chemotherapy does not adversely affect patient
outcome. J Clin Oncol 25:2650–2655
20. Rastogi P, Anderson SJ, Bear HD, Geyer CE, Kahlenberg MS,
Robidoux A, Margolese RG, Hoehn JL, Vogel VG, Dakhil SR,
Tamkus D, King KM, Pajon ER, Wright MJ, Robert J, Paik S,
Mamounas EP, Wolmark N (2008) Preoperative chemotherapy:
updates of National Surgical Adjuvant Breast and Bowel Project
Protocols B-18 and B-27. J Clin Oncol 26:778–785
21. van der Hage JA, van de Velde CJ, Julien JP, Tubiana-Hulin M,
Vandervelden C, Duchateau L (2001) Preoperative chemotherapy
in primary operable breast cancer: results from the European
Organization for Research and Treatment of Cancer trial 10902. J
Clin Oncol 19:4224–4237
22. Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees
CA, Pollack JR, Ross DT, Johnsen H, Akslen LA, Fluge O,
Pergamenschikov A, Williams C, Zhu SX, Lonning PE, Borre-
sen-Dale AL, Brown PO, Botstein D (2000) Molecular portraits
of human breast tumours. Nature 406:747–752
23. Nielsen TO, Hsu FD, Jensen K, Cheang M, Karaca G, Hu Z,
Hernandez-Boussard T, Livasy C, Cowan D, Dressler L, Akslen
LA, Ragaz J, Gown AM, Gilks CB, van de Rijn M, Perou CM
(2004) Immunohistochemical and clinical characterization of the
basal-like subtype of invasive breast carcinoma. Clin Cancer Res
24. Carey LA, Dees EC, Sawyer L, Gatti L, Moore DT, Collichio F,
Ollila DW, Sartor CI, Graham ML, Perou CM (2007) The triple
negative paradox: primary tumor chemosensitivity of breast
cancer subtypes. Clin Cancer Res 13:2329–2334
25. Gianni L, Zambetti M, Clark K, Baker J, Cronin M, Wu J,
Mariani G, Rodriguez J, Carcangiu M, Watson D, Valagussa P,
Rouzier R, Symmans WF, Ross JS, Hortobagyi GN, Pusztai L,
Shak S (2005) Gene expression profiles in paraffin-embedded
core biopsy tissue predict response to chemotherapy in women
with locally advanced breast cancer. J Clin Oncol 23:7265–
26. Paik S, Tang G, Shak S, Kim C, Baker J, Kim W, Cronin M,
Baehner FL, Watson D, Bryant J, Costantino JP, Geyer CE Jr,
Wickerham DL, Wolmark N (2006) Gene expression and benefit
of chemotherapy in women with node-negative, estrogen recep-
tor-positive breast cancer. J Clin Oncol 24:3726–3734
27. Bueno-de-Mesquita JM, van Harten WH, Retel VP, van’t Veer
LJ, van Dam FS, Karsenberg K, Douma KF, van Tinteren H,
Peterse JL, Wesseling J, Wu TS, Atsma D, Rutgers EJ, Brink G,
Floore AN, Glas AM, Roumen RM, Bellot FE, van Krimpen C,
Rodenhuis S, van de Vijver MJ, Linn SC (2007) Use of 70-gene
signature to predict prognosis of patients with node-negative
breast cancer: a prospective community-based feasibility study
(RASTER). Lancet Oncol 8:1079–1087
28. Ravdin PM, Siminoff LA, Davis GJ, Mercer MB, Hewlett J,
Gerson N, Parker HL (2001) Computer program to assist in
making decisions about adjuvant therapy for women with early
breast cancer. J Clin Oncol 19:980–991
558 Breast Cancer Res Treat (2010) 119:551–558
Page 8
    • "In such applications, most of these features are irrelevant with a relatively small sample size (tens or at most hundreds). These situations arise particularly in the bioinformatics field involving the analysis of gene expression and proteomic profiles for different purposes such as disease diagnosis, prognosis and treatment response prediction [72] [78]. The second challenge concerns the problem of processing simultaneously mixed-type and heterogeneous data (qualitative, quantitative, interval, etc.) which are present almost in all daily produced datasets (for instance most of the UCI repository datasets are of mixed type). "
    [Show abstract] [Hide abstract] ABSTRACT: The present paper describes a new feature weighting method based on a membership margin. Distinctive properties of the proposed method include its capability to process problems characterized by mixed-type data (quantitative, qualitative and interval) as well as a huge number of features. The key idea is to map simultaneously all the features of different types into a common space; the membership space. Once all features are represented in a homogeneous space, a feature weighting task can be performed in unified way. This weighting approach is integrated here within a fuzzy classifier through a fuzzy rule weighted concept in order to improve its performance. Each antecedent fuzzy set in the fuzzy if–then rule is weighted to characterize the importance of each proposition and therefore its corresponding feature. Weight estimation process is based on membership margin maximization to estimate a fuzzy weight of each feature in the membership space. Experiments on low and high dimensional real-world datasets demonstrate that the proposed approach can improve significantly the performance of the fuzzy rule-based as well as other state of the art classifiers and can even outperform classical feature weighting approaches. In particular, we show that this approach can yield meaningful results on two real-world applications for cancer prognosis and industrial process diagnosis.
    No preview · Article · Nov 2015 · Information Sciences
  • Source
    • "Rather, the results, so far, are more consistent with the underlying mechanisms of drug resistance being unequally distributed between the subclasses . Similarly, other gene expression signatures have revealed correlations to outcome but to be modest with respect to predicting drug resistance (Albain et al., 2010; Paik et al., 2006; Straver et al., 2010). To explore potential mechanisms of anthracycline resistance in breast cancer, we have undertaken a different approach, searching for defects in key functional pathways regulating vital cellular functions like growth arrest, DNA repair and apoptosis (Lonning et al., 2013). "
    [Show abstract] [Hide abstract] ABSTRACT: Chemoresistance is the main obstacle to cancer cure. Contrasting studies focusing on single gene mutations, we hypothesize chemoresistance to be due to inactivation of key pathways affecting cellular mechanisms such as apoptosis, senescence, or DNA repair. In support of this hypothesis, we have previously shown inactivation of either TP53 or its key activators CHK2 and ATM to predict resistance to DNA damaging drugs in breast cancer better than TP53 mutations alone. Further, we hypothesized that redundant pathway(s) may compensate for loss of p53-pathway signaling and that these are inactivated as well in resistant tumour cells. Here, we assessed genetic alterations of the retinoblastoma gene (RB1) and its key regulators: Cyclin D and E as well as their inhibitors p16 and p27. In an exploratory cohort of 69 patients selected from two prospective studies treated with either doxorubicin monotherapy or 5-FU and mitomycin for locally advanced breast cancers, we found defects in the pRB-pathway to be associated with therapy resistance (p-values ranging from 0.001 to 0.094, depending on the cut-off value applied to p27 expression levels). Although statistically weaker, we observed confirmatory associations in a validation cohort from another prospective study (n = 107 patients treated with neoadjuvant epirubicin monotherapy; p-values ranging from 7.0 × 10(-4) to 0.001 in the combined data sets). Importantly, inactivation of the p53-and the pRB-pathways in concert predicted resistance to therapy more strongly than each of the two pathways assessed individually (exploratory cohort: p-values ranging from 3.9 × 10(-6) to 7.5 × 10(-3) depending on cut-off values applied to ATM and p27 mRNA expression levels). Again, similar findings were confirmed in the validation cohort, with p-values ranging from 6.0 × 10(-7) to 6.5 × 10(-5) in the combined data sets. Our findings strongly indicate that concomitant inactivation of the p53- and pRB- pathways predict resistance towards anthracyclines and mitomycin in breast cancer in vivo. Copyright © 2015 The Authors. Published by Elsevier B.V. All rights reserved.
    Full-text · Article · May 2015 · Molecular oncology
  • Source
    • "Similar results have also been reported on Mammaprint Ò which suggest that adjuvant chemotherapy improves the prognosis of the patients at high risk, but not those at low risk by Mammaprint Ò [7]. Although we have yet to prove the usefulness of Curebest TM 95GC in this regard, we have preliminary results that Curebest TM 95GC, like Oncotype DX Ò [33] and Mammaprint Ò [34], can predict response to neoadjuvant chemotherapy with a statistical significance [28], and that a difference in DRFS seen in patients treated with adjuvant hormonal therapy alone between high risk and low risk disappears when they are treated with neoadjuvant chemotherapy and adjuvant hormonal therapy [28] . These results seem to suggest a possibility that Curebest TM 95GC is also useful in the prediction of efficacy of adjuvant chemotherapy for ER-positive tumors. "
    [Show abstract] [Hide abstract] ABSTRACT: Accurate prediction of recurrence risk is of vital importance for tailoring adjuvant chemotherapy for individual breast cancer patients. Although recurrence risk has been assessed by means of examination of histological data and biomarkers (ER, PR, HER2, Ki67), such conventional examinations are not accurate enough to select subsets of patients who are at sufficiently low risk of recurrence to be spared adjuvant chemotherapy without comprising the prognosis. In the past two decades or so, comprehensive gene expression analysis technology has rapidly developed and made it possible to construct recurrence prediction models for breast cancer based on multi-gene expression in tumor tissues. These models include MammaPrint, Oncotype DX, PAM50 ROR, GGI, EndoPredict, BCI, and Curebest 95GC. In clinical practice, these multi-gene classifiers are mostly used for ER-positive and node-negative breast cancer patients for whom deciding the indication of adjuvant chemotherapy based on conventional histological examination findings alone is often difficult. This article briefly reviews these multi-gene expression-based classifiers with special emphasis on Curebest™ 95GC, which was developed by us for ER-positive and node-negative breast cancer patients.
    Full-text · Article · Feb 2015 · Breast Cancer
Show more