Review Article
Circulating tumour cells in clinical practice: Methods of detection
and possible characterization
Marianna Alunni-Fabbroni a,*, Maria Teresa Sandri b
a
Beckman Coulter Biomedical GmbH, Sauerbruchstrasse 50 – 81377 Munich, Germany
b
European Institute of Oncology, Unit of Laboratory Medicine, Milan, Italy
a r t i c l e i n f o
Article history:
Accepted 18 January 2010
Available online 29 January 2010
Keywords:
Circulating tumour cell
Enrichment method
Detection method
RT-PCR
Real time PCR
Immunomagnetic separation
Minimal residual disease
Cell characterization
a b s t r a c t
Circulating Tumour Cells (CTCs) can be released from the primary tumour into the bloodstream and may
colonize distant organs giving rise to metastasis. The presence of CTCs in the blood has been documented
more than a century ago, and in the meanwhile various methods have been described for their detection.
Most of them require an initial enrichment step, since CTCs are a very rare event. The different technologies
and also the differences among the screened populations make the clinical significance of CTCs
detection difficult to interprete. Here we will review the different assays up to now available for CTC
detection and analysis. Moreover, we will focus on the relevance of the clinical data, generated so far
and based on the CTCs analysis. Since the vast majority of data have been produced in breast cancer
patients, the review will focus especially on this malignancy.
Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction
The detection of Circulating Tumour Cells (CTCs) in peripheral
blood has been described more than a century ago by T.R. Ashworth,
an Austrian pathologist who first reported this type of cells
[1]. The presence of CTCs in the bloodstream fits very well with the
‘‘seed and soil” theory of metastasis formation: tumour cells enter
the blood circulation after detaching from the primary tumour and
can migrate to reach distant organs, where they can implant themselves
and give rise to metastasis. Though metastatic spread represents
the ultimate cause of death, the release of tumour cells may
happen also at early stages of the disease: about 30–40% of patients,
thought to have a localised disease, may in fact present occult
metastasis, probably derived from CTCs, which will be
responsible for the disease progression [2,3]. According to these
considerations, it seems to be likely that CTCs detection and analysis
may play an important role in the diagnosis and treatment of
cancer patients.
2. CTC analysis
Much of the research of the past decades has been focused on
the development of reliable methods for CTC enrichment and identification,
mainly trying to overcome severe technical limitations.
When present in the patient’s blood, CTCs are very rare events with
an expected concentration as low as one cell per 105
–107
with respect
to mononuclear cells. It is clear that enrichment steps are
necessary to increase the isolation success rate. Most of the enrichment
methods developed so far make use of specific markers in
order to identify the cells and to distinguish them from leukocytes.
Among others, the most common epithelial markers are cytokeratins
(CKs), cytoskeletal proteins always expressed in epithelial
cells, and EpCAM, a cell adhesion molecule present in epithelial
cells as well. CTCs identification must be specific and sensitive
enough to distinguish the malignant cells from all the other
circulating non-tumour hematopoietic cells. Immunomediated,
cytometric- and PCR-based strategies have been proposed to increase
the chances to distinguish a CTC from the cellular background,
with different performance characteristics. Nevertheless,
no well-designed comparative investigations have been performed
so far, therefore the comparison of results is quite difficult.
3. Enrichment process
Different approaches have been considered to overcome the
restrictions linked to the low CTC concentration in peripheral
blood. Here the most common methods such as filtration, density
gradient and immunomagnetic enrichment are shortly described.
Table 1 schematically summarises the different methods, listing
their advantages and disadvantages.
1046-2023/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.ymeth.2010.01.027
* Corresponding author. Fax: +49 89 579 589 3503.
E-mail address: malunni@beckman.com (M. Alunni-Fabbroni).
Methods 50 (2010) 289–297
Contents lists available at ScienceDirect
Methods
journal homepage: www.elsevier.com/locate/ymeth
3.1. Filtration
ISET (Isolation by Size of Epithelial Tumour cells) is a size baseddirect
method for enrichment of epithelial cells. ISET makes use of
8 lm pores filters which enable to separate small leukocytes from
the larger epithelial cells [4]. After the filtration step, the putative
CTCs can be fixed and stained for different markers such as CKs.
Unfortunately, large leukocytes can be trapped by the filter as well,
therefore contaminating the CTC fraction or small CTC can pass
through the pores therefore depleting the CTC population. This is
why this technique is generally considered not highly sensitive
and poorly specific.
3.2. Density gradient
An alternative to filtration can be the density gradient centrifugation
using Ficoll-Hypaque: because of their difference in density,
it is possible to separate CTCs and mononuclear cells (with a density
< 1.077 g/ml) from blood cells and granulocytes (with a density
> 1.077 g/ml). However, due to the migration of cells to the
plasma layer or to the presence of aggregates, CTCs can be easily
lost during this process. Moreover, the centrifugation step must
be performed immediately to prevent mixing of the different layers.
In addition, Ficoll-Hypaque can be toxic for the cells when they
are too long in contact with these chemicals, therefore limiting the
method. Density gradient, being independent from the presence of
specific markers, is considered a method feasible for the enrichment
of any kind of CTC, but because of the lack of markers also
a method with low specificity. In order to partially overcome some
drawbacks presented by the density gradient enrichment, Oncoquick
(Greiner Bio-One, Germany) is a separation method based
on density gradient having the advantage to prevent cross contamination
of the different layers thanks to a special membrane which
keeps them separate [5,6].
A step further is proposed by Stem Cell Technology (Vancouver,
Canada) which offers RosetteSep™, a density gradient method based
on negative selection [7,8]. When mixed with whole blood, an antibody
cocktail forms tetrameric complexes (immunorosettes) recognising
cell surface markers on unwanted human hematopoietic
cells. The immunorosettes can easily be removed by a density gradient
procedure while the CTCs can be collected from the interface between
the plasma and the buoyant density medium.
3.3. Immunomagnetic cell enrichment
The immunomagnetic cell enrichment is a magnetic bead-based
separationtechnology. CTCsare positively selectedbymeans of antibodies
coupled to magnetic beads targeted to epithelial markers
(suchasthe alreadymentionedCKsor EpCAM),or to tumour-specific
antigens, for examples to the CarcinoEmbryonic Antigene (CEA) or to
the Human Epidermal growth factor Receptor 2 (HER2). The CTCs,
once bound to the magnetic beads, are then separated from the leukocyte
background through a magnetic field. In order to get rid of
leukocytes, a valid alternative procedure can be a negative selection,
performed by using magnetic beads coated with anti-CD45 and antiCD61
antibodies. Both molecules recognise surface markers expressedonly
on leukocytesand megakaryocytes, respectively. In this
way the cellular population is depleted of unwanted cells, while the
epithelial cells, not expressing CD45 or CD61, are left in solution.
The clear advantage of this approach is that detection and
counting of the enriched CTCs are relatively easy, without the
requirement of any cellular lysis. However, as it will be discussed
later, false positive selection may occur due to expression of the
epithelial markers in non-epithelial cells [5]. In addition, false negative
selection may be due to the heterogeneity of tumour cells
which may variably or not at all express a given marker [9].
Based on this technology, the Magnetic Activated Cell Sorting
System (MACSÒ
, Miltenyi Biotec GmbH, Germany) [10] or the Dynal
Magnetic BeadsÒ
(Invitrogen) [11] are technologies both enabling
to catch CTCs by immunomagnetic labelling with
microbeads. As described before, the beads, coated with antibodies
specific for epithelial or tumour markers, bind to epithelial derived
Table 1
Summary of different CTC enrichment approaches.
Enrichment method Advantages Disadvantages References
Size based ISET Easy and cheap
Feasible with EpCam positive and
negative tumour cells
Low specificity
Loss of small CTC which can pass
through the pores
Enrichment of large leukocytes
[4]
Density gradient Easy and cheap
Feasible for EpCAM positive and
negative tumour cells
Feasible for negative selection
Low specificity
Cross contamination with blood
mononuclear cells possible
OncoQuick Density gradient based
Feasible for EpCAM positive and
negative tumour cells
Cross contamination reduced because
of additional barrier
Low specificity [5,6]
Rosette Sept Good clean up of unwanted hemapoietic cells Cross contamination possible [7,8]
Immunomagnetic based MACS/Dynal
Magnetic Beads/
Easy Sep
Flexible
Cell integrity preserved
False positive due to the expression
of the same antigens on non-tumour cells
False negative due to loss of antigens on CTCs
[10–12]
AdnaTest Recognition of fixed markers (EpCAM, MUC1)
Downstream analysis (RT-qPCR of MUC1,
HER2 and GA73.3–2)
Possibility to characterize for stem cell and Epithelial
Mesenchimal Transition
No flexibility
False positive due to the expression of the
same antigens on non-tumour cells
False negative due to loss of antigens on CTCs
[13]
CellSearch Semi-automated
Combination of positive (anti-EpCAM) and
negative (anti-CD45) selection
FDA approved
Only EpCAM positive CTCs detected
False positive due to the expression of the
same antigens on non-tumour cells
False negative due to loss of antigens on CTCs
[14]
CTC-Chip Good enrichment grade 98% cell viability
Further analysis possible
Only EpCAM positive CTCs detected
Clinical validation not yet available
Not yet on the market
[64]
290 M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297
circulating tumour cells. After the separation in the magnetic field,
the cells can be isolated for further analysis. In a similar way works
EasySepÒ
(Stem Cell Technologies, vancouver Canada) [12]. This is
an immunomagnetic cell selection method where cells are targeted
for positive or negative selection using monoclonal antibodies directed
against specific cell surface markers. The labelled cells are
then interacting with the magnetic nanoparticles and separated
with the use of a magnet.
The AdnaTest (AdnaGen AG, Germany) is based on the use of
specific antibodies against epithelial and tumour markers, such
as EpCAM and Mucin 1 (MUC1), associated to magnetic beads
[13]. This test allows a step further: once collected and pooled together,
the CTCs are directly lysed, the mRNA isolated, and RTqPCR
is performed for the analysis of the selected markers
MUC1, HER2 and the surface glycoprotein GA73.3.–2, in this way
genetically identifying the CTCs.
Finally Veridex (Johnson and Johnson, Raritan, NJ) offers the CellSearch
System™, a semi-automated analyzer enriching the CTCs
with ferrofluid nanoparticles coated with anti-EpCAM antibodies.
The enriched population is then stained with specific markers able
to discriminate between epithelial cells and contaminating leukocytes.
The enriched and correctly labelled cells are then counted as
circulating tumour cells using the CellSpotter analyzer (Veridex), a
four-colour semi-automated fluorescent microscope [14].
One general limitation to keep in mind when considering the
enrichment of CTCs, is that it is still not available a ‘‘universal
marker” which could be used independently from the type of tumour
involved. Tumour markers such as HER2, mammoglobin,
CEA, can directly distinguish tumour cells from non-malignant epithelial
cells or leukocytes but their use is limited by the general
heterogeneity tumour cells are known to show. This limitation applies
not only to different genetic characteristics among different
tumour tissues, but also to different genetic characteristics within
the same tumour tissue, making the identification and enrichment
of CTCs based on these markers potentially challenging.
The alternative would be to identify CTCs on the base of epithelial
markers like EpCAM and CKs. Nevertheless there are some limitations
also in this case. CKs can be found sometime expressed
even in leukocytes when these are in an activated state [15]. Moreover,
the EpCAM marker can be down-regulated in malignant epithelial
cells when these go through the so called Epithelial to
Mesenchimal Transition (EMT), typical of the secondary metastatic
phase. Cells undergoing this phase lose epithelial characteristics
like the expression of specific markers [16–18].
In conclusion, tumour and epithelial markers, widely used in
different enrichment methods, are not necessary able to detect
all the different types of CTCs.
4. Identification process
Once collected, CTCs can be further characterized, in order to
establish their origin and their genetic profile. Most of the available
protocols, are focused either on the nucleic acid content or on the
protein level. A summary of the different methods is presented in
Table 2; advantages and disadvantages of the different methods
are listed as well.
4.1. PCR-based analysis
This is certainly an extremely powerful technique for CTC genetic
screening and in several studies PCR has been shown to be
more sensitive than immunocytochemistry [19,20]. Specificity is
achieved through the design of oligonucleotide primers targeted
to the gene(s) of interest. Reverse transcription PCR (RT-PCR) is
Table 2
Summary of different CTC identification approaches.
Identification process Advantages Disadvantages References
PCR- based analysis RT-PCR High sensitivity RNA degradation
False positive results due to unspecific
amplification, contaminations, pseudogenes
No distinction between viable and
non-viable cells
False negative results due to low
expression level
[19–27]
RT-qPCR High sensitivity
Quantitative
No visualisation of CTCs
No further analysis possible
[28–32]
Cytometric analysis FAST Scan analysis of large
volume of sample
Cell loss minimised
Quick analysis
(up to 300,000 cells/s)
Lack of validation studies in
clinical settings
[33,34]
LSC Fast
High specificity
Technically challenging
Low sensitivity
[35,36]
Flow cytometry High specificity
Multiple parameters
Low sensitivity [37,38]
CellSearch Semi-automated
High sensitivity
CTC quantification
Reproducible
Recognition of fixed marker
(EpCAM, CKs, CD45)
FDA approved
Only EpCAM+
/CK+
/CD45À
CTCs detected
Subjective images interpretation
No further analysis possible
[14,42–45]
CTC-chip 98% Cell viability
High detection rate
Further analysis possible
Only EpCam positive CTCs detected
Not commercially available
Lack of validation studies in clinical settings
[46]
EPISPOT Analysis only on viable cells
High sensitivity
CTC isolation not possible, therefore
further analysis not possible
Need of active protein secretion
Technically challenging
[47,48]
FISH Genetic analysis Further analysis not possible [4,51,52]
M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297 291
generally the method of choice to amplify target mRNAs in CTCs
[21–24]. Many ‘‘single-institution” reports on the prognostic impact
of data obtained with the PCR-based approach are available,
but multicenter validation studies have usually not yet been performed.
Nevertheless, a variety of biomarkers, both for epithelial
cells and for breast cancer cells have been used in many different
studies (a comprehensive overview of markers and detection rate
has been previously reviewed in [25–27]). In order to increase
the sensitivity and the specificity of the assay, multiplex RT-PCR
approaches have been often established, giving therefore the
chance to screen at the same time for more than one single marker.
There are some technical limitations though which may challenge
the performance of the PCR-based assays: (1) The need to perform
cellular lysis, which prevents from cell counting; (2) The possibility
of false positive results due to illegitimate gene transcription in
non-tumour cells; (3) The amplification of cell free nucleic acids
potentially present in blood; (4) The possibility of false positive results
derived from the use of unspecific markers; (5) The presence
in the blood of non-malignant epithelial cells, released for example
after an invasive procedure; (6) The possibility of false negative results
due to low PCR sensitivity connected to limited expression of
the tumour marker or to the presence of PCR inhibitors.
Some of the drawbacks can be overcome by combining more
than one marker together and by choosing only the cells which
are positive (or negative) for all of them in combination. Concerning
the possibility to get false positive results because of illegitimate
gene transcription in non-tumour cells, a big improvement
in the CTC analysis has been achieved with the introduction of
the quantitative real time PCR (qPCR) [28–31]. The quantification
of markers present simultaneously in tumour and non-tumour
cells allows the discrimination between authentic CTCs and contaminant
cells. Moreover false positive results can be skipped with
an appropriate primer design to avoid the amplification of processed
pseudogenes, present in the genome [32].
4.2. Cytometric methods
Cytometric methods isolate and count cancer cells using monoclonal
antibodies directed against different antigens, the most
common being once more CKs and EpCAM. As lysis is not needed,
the cell integrity is always preserved, therefore CTCs can be further
scanned and characterized. Also in this case, the major limitation
connected to the methodology is linked to the relative low specificity.
For example, the most widely used anti-CK antibodies show to
bind not specifically also to hematologic cells. This limit can be
partly overcome combining the existing antibodies to an antiCD45
antibody, in this way making possible the leukocytes detection.
Concerning EpCAM, as already discussed, not all CTCs express
this marker, since it can be down-regulated in malignant cells during
the EMT. Not even its association with tumour markers can
help to bypass completely this limitation because of their expression
variability in tumour cells, with the consequent high risk to
generate false negative results.
Nevertheless, once the cells have been immunofluorescently labelled,
there are different methods to reliably scan them.
FAST (Fiber-optic Array Scanning Technology) [33,34] is a scanning
technology characterized by a large field of view, therefore
allowing the analysis of large volumes of sample without any purification
step and minimising the risk of cell loss. Moreover, it is a
very fast scanning technology with up to 300.000 cells scanned
per second. At the moment though, validation studies in clinical
settings are still missing.
The Laser Scanning Cytometer (LSC, Compucyte Corporation,
Cambridge MA) makes possible to automatically scan and relocate
epithelial positive cells immunolabelled for multiple markers such
as EpCAM combined with the lymphocyte marker CD45 [35,36].
Also flow cytometry can be a valid method to identify CTCs in a
high specific way since multiple parameters can be considered
simultaneously, such as size, viability, DNA content, expression of
different markers [37,38]. Nevertheless both flow cytometry and
LSC, despite their high specificity, show low sensitivity therefore
high amount of samples need to be processed every time in order
to detect few CTCs.
Finally, additional scanning systems are available on the market
such as ACIS (Automated Cellular Imaging System, DAKO, Denmark)
and ARIOL (Applied Imaging Corp. San Jose, CA). Both allow
an automated and fast cell analysis, based on the morphological
evaluation of putative CTCs [39–41].
In the last years many efforts have been devoted to the development
of automated techniques offering at the same time the enrichment,
the staining and the scanning of the samples. The CellSearch
System™, previously described, enriches the CTCs with ferrofluid
nanoparticles coated with anti-EpCAM antibodies. The enriched Ep-
CAm+
cell population is then stained with phycoerythryn conjugated
antibodies directed against CK8, 18 and 19, with allophycocyanin
conjugated antibodies specificfor leukocytes (anti-CD45 antibodies)
andwiththe nucleardye40
,6-diamino-2-phenylindole (DAPI)forthe
nucleic acids staining. The resulting CK+
/DAPI+/
CD45À
cells are
counted as CTCs using the CellSpotter analyzer (Veridex), a four-colour
semi-automated fluorescent microscope [14]. After the scanning,
the images taken by the microscope are displayed in a photo
gallery which is then further analysed by an algorithm to finally
qualify the cells as CTCs (Fig. 1). To date this is the only FDA approved
method to monitor patients with advanced breast [42,43], prostate
[44] and colon [45] cancer. The permeabilization of the cells, necessary
for CKs staining, is preventing any further use of the CTCs. When
assays like colony formation are desired, Veridex offers the Cell Profile™
kit which collects viable EpCAM positive cells before the permeabilization
step.
Alternative to the CellSearch System™, the new technology
named ‘‘CTC-chip” consists of an array of 78.000 microposts coated
with anti-EpCAM antibodies. Whole blood is pumped through the
chip and EpCAM positive cells are captured and detected by cameras
recognising their morphology, their viability and the expression
of tumour markers [46].The limitation of this system is that
the CTC-chip is suitable only for EpCAM positive cells. Moreover,
validation studies in clinical trials are also still missing.
Finally, another antibody-based approach is EPISPOT (Epithelial
Immunospot), an immunological assay based on the ELISPOT (Enzyme-Linked
Immunosorbent assay) technology [47,48]. With EPISPOT
it is possible to identify and count only viable and not
apoptotic cells. The assay is in fact based on the identification of
cells able to secrete proteins like MUC1 and CK19 in short term culture.
The information is important since only viable cells would be
in principle the one generating secondary tumours. Like with the
CTC-chip though, clinical trials are still missing.
5. Genetic characterization
Once the CTCs have been enriched and isolated, the next desirable
step is certainly their genetic characterization. Genetic analysis
might help to understand, for example, if the isolated CTCs
show still malignant characteristics or how genetically similar to
the primary tumour the CTCs may be. This information would certainly
allow to better monitoring the development of the disease
giving the opportunity to analyse relevant gene segments, which
might have an influence on the tumour development and possibly
on the following therapeutic approaches.
For a PCR-based genetic analysis of single CTCs, AmpliGrid
(Beckman Coulter Genomics, Munich, Germany) can be an ideal
platform [49,50]. This is a PCR-based chip for direct analysis of sin-
292 M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297
gle CTCs which allows to genetically determinate possible differences
among the isolated CTCs. After deposition on the reaction
site, single CTCs can be analysed by multiplex RT-PCR or RT-qPCR,
unravelling for example cellular heterogeneity, epigenetic features,
genetic fingerprint or microRNA levels (Fig. 2). Preliminary data
obtained in breast cancer patients showed that it is possible to
genetically characterize isolated CTCs by RT-qPCR run on AmpliGrid,
detecting specific tumour markers such as Keratin 5 and 18
(Görner and Bonizzi, data not shown). Nevertheless while the
AmpliGrid platform is already commercially available, its use for
CTCs genetic analysis is under evaluation and large-scale clinical
trials are still missing.
FISH (Fluorescent In Situ Hybridization) has been proposed as a
valid method for CTC genotyping [4,51,52]. FISH is a fluorescent
technology thought to detect and localise the presence or absence
of specific DNA sequences on chromosomes or their alteration due
to gene translocation or amplification.
An alternative method frequently used in genetic characterization
and therefore potentially useful for CTC analysis is CGH (Comparative
Genomic Hybridization) [53]. This is a cytogenetic method
for the analysis of changes in copy number in a given DNA.
Finally, immunostaining can be a valid approach for detection of
tumour markers. A cellular protein can be detected by the corresponding
specific antibody making possible to unravel the presence
or absence of the protein, its sub-cellular localisation,
changes in its expression or finally its degradation.
The different discussed methods for enrichment and analysis of
CTCs are summarised in Fig. 3.
6. Clinical data
CTCs have been evaluated in different malignancies, however
the larger number of data are available from breast cancer. A summary
of the clinical study results is presented in Table 3.
6.1. Prognosis
Despite all the technical limitations, numerous clinical studies
have been published in a relatively short period of time. However,
it is quite difficult to draw firm conclusions: because of different
assays, alternative definition criteria for CTCs, different clinical
populations, limited number of patients, different stages of the disFig.
1. CTCs identification using the CellSearch system. After the scan performed by the automated fluorescence microscope, the software present on a computer screens a
gallery of images to be interpreted: The cytockeratin-PE positive, DAPI positive and CD45 negative images are counted as CTCs.
M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297 293
Fig. 2. Deposition of a single CTC on the AmpliGrid chip. (A) AmpliGrid chip with 48 reaction sites for single cell deposition and PCR analysis. (B) A single CTC is deposited
with a capillary on the reaction site of the Ampligrid chip. CTCs are normally resuspended in physiological buffer and the volume of deposition is approximately 20 nl. (C) A
single CTC can be easily visualised on the reaction site when the DNA is stained with Hoechst dye.
Fig. 3. Overview of the different methods for enrichment and analysis of CTCs. In order to run any kind of analysis, CTCs must be as a first step enriched from the whole blood.
Different protocols based on filtration, density gradient or immunomagnetic beads (IMB) can be followed in order to isolate these rare cells. The CTCs can be processed then
by RT-PCR or real time PCR for messenger quantification, by FISH and by CHG for analysing chromosome sequence and number aberrations, by immunostaining for protein
detection or by EPISPOT for detection of viable cells. CTCs can also be analysed single wise by FISH or by RT-PCR and real time PCR for example when placed on the AmpliGrid
slide.
294 M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297
ease, the comparison among different studies and publication results
quite challenging. However, keeping these limitations in
mind, it’s worth to note that the vast majority of studies find an
association between the presence of CTCs and the clinical outcome
parameters such as the Disease Free Survival (DFS, i.e.the time
span between the end of a therapy and the relapse), the Progression
Free Survival (PFS, i.e.in the metastatic the time span between
the end of a therapy and the progression) or the Overall Survival
(OS, i.e. the time span between the end of the therapy and the final
patient’s death).
6.1.1. Metastatic breast cancer
The study of Cristofanilli et al. [42] showed that the presence of
CTCs, detected using the Veridex CellSearch System™
, in patients
with metastatic breast cancer has an adverse prognostic impact.
The authors analysed the blood samples of 177 patients beginning
a new therapy. Based upon a training set, the threshold for elevated
CTCs was defined as P5 CTCs per 7.5 ml of whole blood. Elevated
CTCs at baseline (i.e. before starting any kind of therapy) predicted
an extremely short median PFS and OS (2.7 and 10 months, respectively),
in comparison to low/negative CTCs corresponding to PFS
and OS of 7 and >18 months, respectively (P < 0.001). Even more
interesting, CTC values, obtained after one cycle of therapy, predicted
in which patients the therapy was likely to fail. Patients
with elevated CTCs after one cycle of therapy had median PFS
and OS of approximately 2.1 months and 8.2 months, respectively,
when measured from baseline. In contrast, median PFS and OS of
7.0 months and 22 months, respectively, were observed in a patient
group with <5 CTC at the first follow-up visit. These differences
were statistically significant (P = 0.001) for patients
receiving chemotherapy. The data were further confirmed by an
independent study [54] in a group of 80 metastatic breast cancer
patients. In this study the prevalence of metastatic breast cancer
patients with CTCs P5 per 7.5 ml was 61%; baseline CTCs value
were predictive of response to therapy, and changes of CTCs level
during treatment were found to be associated with response to
therapy. Moreover, baseline value of CTCs and CTCs variations
(both at 1 month and during follow-up) were the most significant
predictors of PFS.
6.1.2. Early breast cancer
In patients with early breast cancer undergoing adjuvant chemotherapy,
CK19 mRNA-positive cells detected by means of RTPCR
were shown to be an independent prognostic factor for both
PFS and OS, both in node-negative and node-positive patients.
Stathopoulou et al. [55] analysed the blood of 148 stage I and stage
II patients before the administration of adjuvant chemotherapy:
CK19 mRNA-positive cells were detected in 29.7% of the patients,
and their presence had a prognostic implication, because they were
significantly associated with reduced PFS (P = 0.0007) and OS
(P = 0.01). The same findings were later confirmed by an independent
study run in a group of 167 node-negative breast cancer patients
undergoing chemotherapy [23]: the detection of CK19
mRNA-positive cells was shown to be an independent predictive
and prognostic factor for poorer clinical outcome. Furthermore
the same group confirmed these results in an extended cohort of
women (444 patients) presenting node-negative and node-positive
breast cancer [56]: all patients received chemotherapy +/À hormone
therapy. The percentage of patients with detectable CTCs
was 40.8% (181/444). The adverse prognostic impact of CTCs detection
was demonstrated in the triple-negative and HER2-positive
but not in Oestrogen Receptor positive/HER2 negative tumours.
Moreover, Apostolaki et al. [57] demonstrated that HER2 mRNA-positive
CTCs were detected in 21% (45/214) of Stage I–II breast cancer
patients after the completion of adjuvant treatment, and this fact
was associated with reduced disease DFS in node-negative, but
not in node-positive patients. Furthermore, no association was
found between presence of HER2 mRNA-positive CTC and OS. Similar
results were found by Xenidis et al. [58], who demonstrated
that, after the completion of adjuvant chemotherapy, CK19 positive
mRNA CTCs were found in 32.7% (143/437) of the patients, and their
detection was significantly associated with increased risk of disease
recurrence and death owing to disease progression.
As part of the SUCCESS trial (a large translational research project
run in Germany) the amount of CTCs, isolated with the CellSearch
system was evaluated in patients before, during and at
the end of adjuvant chemotherapy. Using a cut-off of >1 CTC/
23 ml of blood, CTCs were detected in 9.5% and 8.7% of 1500
node-positive and high-risk node-negative women, respectively.
After a 12-month median follow-up, the detection of >1 CTC/
22.5 ml after the end of chemotherapy was associated with a shorter
DFS and OS [59].
6.2. Follow-up and therapy monitoring
Different studies showed that presence or absence of CTCs
determined during the follow-up of patients receiving chemotherapy
for advanced breast cancer seems to be more informative than
conventional imaging methods (such as radiologic assessment,
including CT or PET) for response evaluation [42,43]. An analysis
Table 3
Clinical relevance of CTC detection in breast cancer patients.
Study Tumour stage Method Number of
patients
CTC detection
rate (%)
Clinical results
Cristofanilli et al. [42] Metastatic breast cancer CellSearch 177 49 CTCs P 5/7.5 ml associated with reduced PFS
(P < 0.001) and OS (P < 0.001)
Nolè et al. [54] Metastatic breast cancer CellSearch 80 61 CTCs P 5/7.5 ml associated with reduced PFS (P = 0.002)
Budd et al. [60] Metastatic breast cancer CellSearch 138 25.4 CTC evaluation after 4 weeks from the start of
chemotherapy are better predictor of OS compared to
radiologic response evaluation performed at 10 weeks
De Giorgi et al. [61] Metastatic breast cancer CellSearch 115 21 Univariate analysis: CTCs levels at 9–12 weeks (P < 0.0001)
and FDG-PET/CT (P 6 0.001) associated with OS
Multivariate analysis: only CTCs levels at 9–12 weeks
(P < 0.001) associated with OS
Pierga et al. [63] Locally advanced breast cancer CellSearch 118 27 CTCs P 1/22.5 ml associated with reduced DFS (0.017)
Stathopoulou et al. [55] Early breast cancer (stage I–II) RT-PCR (CK19) 148 29.7 Reduced DFS (P = 0.0007) and OS (P = 0.01)
Xenidis et al. [23] Early breast cancer (stage I–II) RT-PCR (CK19) 167 21.6 Reduced DFS (P < 0.00001) and OS (P = 0.008)
Ignatiadis et al. [56] Early breast cancer (stage I–III) RT-PCR (CK19) 444 40.8 Reduced DFS (P < 0.001) and OS (P = 0.001)
Apostolaki et al. [57] Early breast cancer (stage I–II) RT-PCR (HER2) 214 21 Reduced DFI (P = 0.006), no association with OS (P = 0.2)
Xenidis et al. [58] Early breast cancer (stage I-III) RT-PCR (CK19) 437 32.7 Reduced DFS (P < 0.001) and OS (P = 0.001)
Rack et al. [59] Early breast cancer (stage I–III CellSearch 1500 9 CTCs > 1/22.5 ml associated with reduced DFS (P = 0.04)
and OS (P = 0.03)
M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297 295
performed in a group of 138 out of the 177 patients included in the
study published by Cristofanilli et al. [42], demonstrated that CTCs
count at 4 weeks after the start of chemotherapy was a more robust
and earlier predictor of survival, compared to imaging studies
done before and after a median of 10 weeks from the initiation of
the chemotherapy treatment [60]. Recently De Giorgi et al. [61]
compared the prognostic value of CTC, determined at 9–12 weeks
after the start of treatment, and of FDG-PET/CT monitoring during
systemic therapy in 102 patients with metastatic breast cancer
(MBC). The authors found that in these patients CTCs could accurately
predict prognosis, beyond functional response assessed by
FDG-PET/CT: the latter deserving a predictive role in patients with
less that 5 CTC during follow-up. A very recent study [62] confirmed
that CTC counting in patients with metastatic breast cancer
was strongly correlated with radiological signs of disease progression,
and this holds true also for the results of CTC determined 7–
9 weeks before imaging. Taking all these data together, it could be
suggested that CTC evaluation during therapy for metastatic breast
cancer may emerge as a clinically useful tool in conjunction with
standard imaging methods to better assess treatment benefit.
To provide definite evidence on the importance of CTCs determination
and in the management of patients with advanced breast
cancer, the American South West Oncology Group has recently
started the SWOG 0500 trial. In this study, patients undergoing
first-line chemotherapy for a newly diagnosed metastatic disease
are tested for CTCs: women showing elevated CTCs (P5/7.5 ml of
whole blood) before starting the chemotherapy, are re-tested after
3 weeks. If CTCs are still elevated, the patients are randomized to
continue the same therapy or to switch to a different chemotherapy
regimen. The clinical endpoints will be the comparison between
the DFS and OS of the two groups of patients, those
continuing on the same therapy and those switching to another
regimen, to see if the early change in therapy based on CTC count
will improve patient’s outcome.
Another possible application based on CTCs is to monitor minimal
residual disease in patients with locally advanced breast cancer
treated with neoadjuvant chemotherapy. A recent paper [63]
examined 118 patients undergoing primary chemotherapy. The
authors showed that at least 1 CTC per 22.5 ml of whole blood
was present in 23% of pre-chemotherapy blood samples and in
17% of post-chemotherapy blood samples. Persistence of CTCs at
the end of neoadjuvant chemotherapy was not correlated with
treatment response; however their pre- or post-chemotherapy
detection was an independent prognostic factor for shorter DFS.
These results are in line with what reported from the first analysis
of the SUCCESS trial, in which the presence of >1 CTC at the end of
the adjuvant treatment was associated with earlier relapse. However,
we still do not know if other therapeutic options offered to
women with persistence of CTC at the end of therapy would have
any impact on their outcome, as no study has yet addressed this
issue.
6.3. CTC as a ‘‘real-time” biopsy
Apart from detection or counting of CTCs, one of the most interesting
and possibly challenging application will be the phenotypical
and molecular characterization of CTC, which will allow to
obtain more detailed information on CTC biology. In this light, CTCs
may represent a ‘‘real time biopsy”: especially in the metastatic
setting, CTCs may better represent tumour genetics than primary
tumours. It has been shown that expression of clinically and therapeutically
relevant markers such as ER, PG and HER2 can differ
between a primary tumour and its relative metastasis [52,64–
66]. For example, Meng et al. [52] studied CTCs HER2 status in
24 patients with HER2 negative primary tumours who developed
recurrent disease. Nine patients showed HER2 gene amplification
on CTCs, and 4 of them were treated with trastuzumab-containing
therapy: one had complete response and 2 had partial response.
Other similar reports strongly suggest that in a small group of patients
HER2 status may change at disease progression [67,68]: the
major caveat in analysing these studies relies on the fact that low
number of patients are included. Only large clinical trials will be
able to verify if acquisition of HER2-positive CTC in patients with
HER2 negative primary tumours can be considered predictive of response
to trastuzumab-based therapy.
Gaining or losing some markers, potential targets for specific
therapy, may lead to a change of the following therapy. On this
base, Smirnov et al. [69], looking at gene expression profiling by
means of the microarray technology, studied CTCs from patients
with breast, colon and prostate metastatic cancer: from the generated
profiles, they selected genes expressed in the CTCs of all the
three cancers, and proposed a list of genes CTC-associated.
Recently the concept of cancer stem cells responsible for tumour
origin, maintenance and resistance to treatment has gained
prominence in the field of breast cancer research [70]. It is supposed
that these cells represent a minor subset of cells in the tumour,
and have different characteristics compared to the more
differentiated tumour cells. In particular while cancer therapies
are now effective in debulking the tumour mass, they may be
non-effective in producing long-term remission, maybe because
they do not affect the cancer stem cell population. Recently, it
has been reported that a subpopulation of CTCs from patients with
metastatic breast cancer express putative stem cell progenitor phenotype
[71]. The authors found that most patients in an advanced
disease phase show a CK+
/CD44+
/CD24À/low
phenotype, while in a
separate group of patients, were found CTCs distinguished by an
ALDH1high
/CD24À/low
phenotype. These finding may represent a
first clinical step toward a better characterization of CTCs, but further
molecular studies are needed to define the possible prognostic
impact, and more importantly, to define targets which may render
these cells sensitive to specific therapy and therefore susceptible of
definitive elimination.
Acknowledgments
We thank the members of our laboratories for the research
work, in particular Karin Görner (Beckman Coulter Biomedical
GmbH) and Giuseppina Bonizzi (IEO-IFOM). We thank also Claudia
Holzhauer (Beckman Coulter Biomedical GmbH) for critically
reviewing the paper.
References
[1] T.R. Ashworth, Aust. Med. J. 14 (1869) 146–149.
[2] S.E. Clare, S.F. Sener, W. Wilkens, R. Goldschmidt, D. Merkel, D.J. Winchester,
Ann. Surg. Oncol. 4 (1997) 447–451.
[3] S. Braun, K. Pantel, P. Muller, W. Janni, F. Hepp, C.R. Kentenich, S. Gastroph, A.
Wischnik, T. Dimpfl, G. Kindermann, G. Riethmuller, G. Schlimok, N. Engl. J.
Med. 342 (2000) 525–533.
[4] G. Vona, A. Sabile, M. Louha, V. Sitruk, S. Romana, K. Schütze, F. Capron, D.
Franco, M. Pazzagli, M. Vekemans, B. Lacour, C. Bréchot, P. Paterlini-Bréchot,
Am. J. Pathol. 156 (2000) 57–63.
[5] R. Gertler, R. Rosenberg, K. Fuehrer, M. Dahm, H. Nekarda, J.R. Siewert, Recent
Results Cancer Res. 162 (2003) 149–155.
[6] V. Muller, N. Stahmann, S. Riethdorf, T. Rau, T. Zabel, A. Goetz, F. Janicke, K.
Pantel, Clin. Cancer Res. 11 (2005) 3678–3685.
[7] B. Naume, E. Borgen, S. Tøssvik, N. Pavlak, D. Oates, J.M. Nesland, Cytotherapy 6
(2004) 244–252.
[8] G.M. Hayes, R. Busch, J. Voogt, I.M. Siah, T.A. Gee, M.K. Hellerstein, N. Chiorazzi,
K.R. Rai, E.J. Murphy, Leuk. Res. (2009).
[9] J.C. Goeminne, T. Guillaume, M. Symann, Ann. Oncol. 11 (2000) 785–792.
[10] I. McNiece, R. Briddell, G. Stoney, B. Kern, K. Zilm, D. Recktenwald, S. Miltenyi, J.
Hematother. 6 (1997) 5–11.
[11] A.A. Neurauter, M. Bonyhadi, E. Lien, L. Nøkleby, E. Ruud, S. Camacho, T.
Aarvak, Adv. Biochem. Eng. Biotechnol. 106 (2007) 41–73.
[12] C.E. Peters, S.M. Woodside, A.C. Eaves, Methods Mol. Biol. 302 (2005) 95–116.
[13] S. Lankiewicz, B.G. Rivero, O. Böcher, Mol. Biotechnol. 34 (2006) 15–27.
296 M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297
[14] W.J. Allard, J. Matera, M.C. Miller, M. Repollet, M.C. Connelly, C. Rao, A.G. Tibbe,
J.W. Uhr, L.W. Terstappen, Clin. Cancer Res. 10 (2004) 6897–6904.
[15] R. Jung, K. Petersen, W. Krüger, M. Wolf, C. Wagener, A. Zander, M. Neumaier,
Br. J. Cancer 81 (1999) 870–873.
[16] K. Polyak, R.A. Weinberg, Nat. Rev. Cancer 9 (2009) 265–273.
[17] J.P. Thiery, J.P. Sleeman, Nat. Rev. Mol. Cell Biol. 7 (2006) 131–142.
[18] J. Yang, R.A. Weinberg, Dev. Cell 14 (2008) 818–829.
[19] B.M. Smith, M.J. Slade, J. English, H. Graham, M. Lüchtenborg, H.D. Sinnett, N.C.
Cross, R.C. Coombes, J. Clin. Oncol. 18 (2000) 1432–1439.
[20] A.C. Lambrechts, A.J. Bosma, S.G. Klaver, B. Top, L. Perebolte, L.J. van’t Veer, S.
Rodenhuis, Breast Cancer Res. Treat. 6 (1999) 219–231.
[21] P. Bossolasco, C. Ricci, G. Farina, D. Soligo, D. Pedretti, A. Scanni, G.L. Deliliers,
Cancer Detect. Prev. 26 (2002) 60–63.
[22] V.V. Iakovlev, R.S. Goswami, J. Vecchiarelli, N.C. Arneson, S.J. Done, Breast
Cancer Res. Treat. 107 (2008) 145–154.
[23] N. Xenidis, M. Perraki, M. Kafousi, S. Apostolaki, I. Bolonaki, A. Stathopoulou, K.
Kalbakis, N. Androulakis, C. Kouroussis, T. Pallis, C. Christophylakis, K.
Argyraki, E.S. Lianidou, S. Stathopoulos, V. Georgoulias, D. Mavroudis, J. Clin.
Oncol. 24 (2006) 3756–3762.
[24] J.Y. Wang, C.H. Wu, C.Y. Lu, J.S. Hsieh, D.C. Wu, S.Y. Huang, S.R. Lin, World J.
Surg. 30 (2006) 1007–1013.
[25] K. Pantel, R.H. Brakenhoff, B. Brandt, Nat. Rev. Cancer. 8 (2008) 329–340.
[26] K. Pantel, C. Alix-Panabières, S. Riethdorf, Nat. Rev. Clin. Oncol. 6 (2009) 339–
351.
[27] B. Mostert, S. Sleijfer, J.A. Foekens, J.W. Gratama, Cancer Treat. Rev. 35 (2009)
463–474.
[28] T. Nolan, R.E. Hands, S.A. Bustin, Nat. Protoc. 1 (2006) 1559–1582.
[29] A. Stathopoulou, A. Gizi, M. Perraki, S. Apostolaki, N. Malamos, D. Mavroudis, V.
Georgoulias, E.S. Lianidou, Clin. Cancer Res. 9 (2003) 5145–5151.
[30] A. Stathopoulou, M. Ntoulia, M. Perraki, S. Apostolaki, D. Mavroudis, N.
Malamos, V. Georgoulias, E.S. Lianidou, Int. J. Cancer 119 (2006)
1654–1659.
[31] I.H. Benoy, H. Elst, P. Van Dam, S. Scharpé, E. Van Marck, P.B. Vermeulen, L.Y.
Dirix, Clin. Chem. Lab. Med. 44 (2006) 1082–1087.
[32] L. Xi, D.G. Nicastri, T. El-Hefnawy, S.J. Hughes, J.D. Luketich, T.E. Godfrey, Clin.
Chem. 53 (2007) 1206–1215.
[33] H.B. Hsieh, D. Marrinucci, K. Bethel, D.N. Curry, M. Humphrey, R.T. Krivacic, J.
Kroener, L. Kroener, A. Ladanyi, N. Lazarus, P. Kuhn, R.H. Bruce, J. Nieva,
Biosens. Bioelectron. 21 (2006) 1893–1899.
[34] R.T. Krivacic, A. Ladanyi, D.N. Curry, H.B. Hsieh, P. Kuhn, D.E. Bergsrud, J.F.
Kepros, T. Barbera, M.Y. Ho, L.B. Chen, R.A. Lerner, R.H. Bruce, Proc. Natl. Acad.
Sci. USA 101 (2004) 10501–10504.
[35] K. Pachmann, J.H. Clement, C.P. Schneider, B. Willen, O. Camara, U. Pachmann,
K. Höffken, Clin. Chem. Lab. Med. 43 (2005) 617–627.
[36] K. Pachmann, R. Dengler, K. Lobodasch, F. Fröhlich, T. Kroll, M. Rengsberger, R.
Schubert, U. Pachmann, J. Cancer, Res. Clin. Oncol. 134 (2008) 59–65.
[37] I. Cruz, J. Ciudad, J.J. Cruz, M. Ramos, A. Gómez-Alonso, J.C. Adansa, C.
Rodríguez, A. Orfao, Am. J. Clin. Pathol. 123 (2005) 66–74.
[38] S.J. Simpson, M. Vachula, M.J. Kennedy, H. Kaizer, J.S. Coon, R. Ghalie, S.
Williams, D. Van Epps, Exp. Hematol. 23 (1995) 1062–1068.
[39] S.K. Kraeft, R. Sutherland, L. Gravelin, G.H. Hu, L.H. Ferland, P. Richardson, A.
Elias, L.B. Chen, Clin. Cancer Res. 6 (2000) 434–442.
[40] J.L. Mansi, W.E. Mesker, T. McDonnell, A.M. Van Driel-Kulker, J.S. Ploem, R.C.
Coombes, J. Immunol. Meth. 112 (1988) 105–111.
[41] W.E. Mesker, J.M. vd Burg, P.S. Oud, C.F. Knepflé, M.C. Ouwerkerk-Velzen, N.W.
Schipper, H.J. Tanke, Cytometry 17 (1994) 209–215.
[42] M. Cristofanilli, G.T. Budd, M.J. Ellis, A. Stopeck, J. Matera, M.C. Miller, J.M.
Reuben, G.V. Doyle, W.J. Allard, L.W. Terstappen, D.F. Hayes, N. Engl. J. Med.
351 (2004) 781–791.
[43] M. Cristofanilli, D.F. Hayes, G.T. Budd, M.J. Ellis, A. Stopeck, J.M. Reuben, G.V.
Doyle, J. Matera, W.J. Allard, M.C. Miller, H.A. Fritsche, G.N. Hortobagyi, L.W.
Terstappen, J. Clin. Oncol. 23 (2005) 1420–1430.
[44] D.C. Danila, G. Heller, G.A. Gignac, R. Gonzalez-Espinoza, A. Anand, E. Tanaka,
H. Lilja, L. Schwartz, S. Larson, M. Fleisher, H.I. Scher, Clin. Cancer Res. 13
(2007) 7053–7058.
[45] S.J. Cohen, C.J.A. Punt, N. Iannotti, B.H. Saidman, K.D. Sabbath, N.Y. Gabrail, J.
Picus, M. Morse, E. Mitchell, M.C. Miller, G.V. Doyle, H. Tissing, L.W.
Terstappen, N.J. Meropol, J. Clin. Oncol. 19 (2008) 3213–3221.
[46] S. Nagrath, L.V. Sequist, S. Maheswaran, D.W. Bell, D Irimia, L. Ulkus, Smith,
M.R. Kwak, S. Digumarthy, A. Muzikansky, P. Ryan, U.J. Balis, R.G. Tompkins,
DA. Haber, M. Toner, Nature 450 (2007) 1235–1239.
[47] J.-P.V. Alix-Panabières, J.P. Vendrell, O. Pellé, X. Rebillard, S. Riethdorf, V.
Müller, M. Fabbro, K. Pantel, Clin. Chem. 53 (2007) 537–539.
[48] C. Alix-Panabières, J.P. Brouillet, M. Fabbro, H. Yssel, T. Rousset, T. Maudelonde,
G. Choquet-Kastylevsky, J.P. Vendrell, J. Immunol. Meth. 299 (2005) 177–188.
[49] N.E. Holodick, K. Repetny, X. Zhong, T.L. Rothstein, Eur. J. Immunol. 39 (2009)
2383–2394.
[50] J. Waak, S.S. Weber, A. Waldenmaier, K. Görner, M. Alunni-Fabbroni, H. Schell,
D. Vogt-Weisenhorn, T.T. Pham, V. Reumers, V. Baekelandt, W. Wurst, P.J.
Kahle, FASEB J. 23 (2009) 2478–2489.
[51] S. Meng, D. Tripathy, E.P. Frenkel, S. Shete, E.Z. Naftalis, J.F. Huth, P.D. Beitsch,
M. Leitch, D. Hoover Euhus, B. Haley, L. Morrison, T.P. Fleming, D. Herlyn, L.W.
Terstappen, T. Fehm, T.F. Tucker, N. Lane, J. Wang, J.W. Uhr, Clin. Cancer Res. 10
(2004) 8152–8162.
[52] S. Meng, D. Tripathy, S. Shete, R. Ashfaq, B. Haley, S. Perkins, P. Beitsch, A. Khan,
D. Euhus, C. Osborne, E. Frenkel, S. Hoover, M. Leitch, E. Clifford, E. Vitetta, L.
Morrison, D. Herlyn, L.W. Terstappen, T. Fleming, T. Fehm, T. Tucker, N. Lane, J.
Wang, J. Uhr, Proc. Natl. Acad. Sci. USA 101 (2004) 9393–9398.
[53] O. Schmidt-Kittler, T. Ragg, A. Daskalakis, M. Granzow, A. Ahr, T.J. Blankenstein,
M. Kaufmann, J. Diebold, H. Arnhold, P. Muller, J. Bischoff, D. Harich, G.
Schlimok, G. Riethmuller, R. Eils, C.A. Klein, Proc. Natl. Acad. Sci. USA 100
(2003) 7737–7742.
[54] F. Nolé, E. Munzone, L. Zorzino, I. Minchella, M. Salvatici, E. Botteri, M. Medici,
E. Verri, L. Adamoli, N. Rotmensz, A. Goldhirsch, M.T. Sandri, Ann. Oncol. 19
(2008) 891–897.
[55] A. Stathopoulou, I. Vachonikolis, D. Mavroudis, M. Perraki, Ch. Kouroussis, S.
Apostolaki, N. Malamos, S. Kakolyris, A. Kotsakis, N. Xenidis, D. Reppa, V.
Georgoulias, J. Clin. Oncol. 20 (2002) 3404–3412.
[56] M. Ignatiadis, N. Xenidis, M. Perraki, S. Apostolaki, E. Politaki, M. Kafousi, E.N.
Stathopoulos, A. Stathopoulou, E. Lianidou, G. Chlouverakis, C. Sotiriou, V.
Georgoulias, D. Mavroudis, J. Clin. Oncol. 25 (2007) 5194–5202.
[57] S. Apostolaki, M. Perraki, A. Pallis, V. Bozionelou, S. Agelaki, P. Kanellou, A.
Kotsakis, E. Politaki, K. Kalbakis, A. Kalykaki, L. Vamvakas, V. Georgoulias, D.
Mavroudis, Ann. Oncol. 18 (2007) 851–858.
[58] N. Xenidis, M. Ignatiadis, S. Apostolaki, M. Perraki, K. KalbakisK, S. AgelakiS,
E.N. Stathopoulos, G. Chlouverakis, E. Lianidou, S. Kakolyris, V. Georgoulias, D.
Mavroudis, J. Clin. Oncol. 27 (2009) 2177–2184.
[59] B. Rack, C. Schindlbeck, A. Schneeweiss, J. Hilfrich, R. Lorenz, M.W. Beckmann,
K. Pantel, W. Lichtenegger, H.L. Sommer, W.J. Janni, Proc. Am. Soc. Clin. Oncol.
26 (2008) (abstract 503).
[60] G.T. Budd, M. Cristofanilli, M.J. Ellis, A. Stopeck, E. Borden, M.C. Miller, J.
Matera, M. Repollet, G.V. Doyle, L.W. Terstappen, D.F. Hayes, Clin. Cancer Res.
12 (2006) 6403–6409.
[61] U. De Giorgi, V. Valero, E. Rohren, M. Mego, G.V. Doyle, M.C. Miller, N.T. Ueno,
B.C. Handy, J.M. Reuben, H.A. Macapinlac, G.N. Hortobagyi, M. Cristofanilli,
Ann. Oncol. (2009) (Epub ahead of print).
[62] M.C. Liu, P.G. Shields, R.D. Warren, P. Cohen, M. Wilkinson, Y.L. Ottaviano, S.B.
Rao, J. Eng-Wong, F. Seillier-Moiseiwitsch, A.M. Noone, C. Isaacs, J. Clin. Oncol.
(2009) (Epub ahead of print).
[63] J.Y. Pierga, F.C. Bidard, C. Mathiot, E. Brain, S. Delaloge, S. Giachetti, P. de
Cremoux, R. Salmon, A. Vincent-Salomon, M. Marty, Clin. Cancer Res. 14 (2008)
7004–7010.
[64] M. Tewes, B. Aktas, A. Welt, S. Mueller, S. Hauch, R. Kimmig, S. Kasimir-Bauer,
Breast Cancer Res. Treat. 115 (2009) 581–590.
[65] T. Fehm, S. Becker, S. Duerr-Stoerzer, K. Sotlar, V. Mueller, D. Wallwiener, N.
Lane, E. Solomayer, J. Uhr, Breast Cancer Res. 9 (2007) R74.
[66] E. Munzone, F. Nolè, L. Zorzino, M. Medici, I. Minchella, M.C. Cassatella, C.
Casadio, M. Picozzi, L. Adamoli, A. Goldhirsch, M. Sandri, J. Clin. Oncol. 26
(Suppl.) (2008) (abstract 11017).
[67] D.F. Hayes, T.M. Walker, B. Singh, E.S. Vitetta, J.W. Uhr, S. Gross, C. Rao, G.V.
Doyle, L.W. Terstappen, Int. J. Oncol. 21 (2002) 1111–1117.
[68] S. Meng, D. Tripathy, S. Shete, R. Ashfaq, H. Saboorian, B. Haley, E. Frenkel, D.
Euhus, M. Leitch, C. Osborne, E. Clifford, S. Perkins, P. Beitsch, A. Khan, L.
Morrison, D. Herlyn, L.W. Terstappen, N. Lane, J. Wang, J. Uhr, Proc. Natl. Acad.
Sci. USA 103 (2006) 17361–17365.
[69] D.A. Smirnov, D.R. Zweitzig, B.W. Foulk, M.C. Miller, G.V. Doyle, K.J. Pienta, N.J.
Meropol, L.M. Weiner, S.J. Cohen, J.G. Moreno, M.C. Connelly, L.W. Terstappen,
S.M. O’Hara, Cancer Res. 65 (2005) 4993–4997.
[70] B.J. Morrison, C.W. Schmidt, S.R. Lakhani, B.A. Reynolds, J.A. Lopez, Breast
Cancer Res. 10 (2008) 210.
[71] P.A. Theodoropoulos, H. Polioudaki, S. Agelaki, G. Kallergi, Z. Saridaki, D.
Mavroudis, V. Georgoulias, Cancer Lett. (2009) (Epub ahead of print).
M. Alunni-Fabbroni, M.T. Sandri / Methods 50 (2010) 289–297 297