xi
POSITRON
Preface
Breast Cancer
Norbert Avril, MD
Guest Editor
Breast cancer is one of the most devastating diseases
affecting women in Western countries. Although
progress has been made in the development of new
therapeutic agents, approximately 40,000 patients in
the United States alone will die of breast cancer
in 2005. The diagnosis of breast cancer—especially in
the early stages—is challenging, and mammography
and ultrasound play an important role in screening
asymptomatic women as well as in the initial workup
of suspicious breast lesions.
Positron emission tomography (PET) has become
widely available in the United States, and this first
issue of the PET Clinics is devoted to the current
status and potential future applications of molecular
PET imaging in breast cancer. Higher glucose con-
sumption in cancer cells compared with normal
tissue has made F-18 fluorodeoxyglucose (FDG)
the most commonly used PET radiopharmaceu-
tical. The metabolic information from FDG-PET pro-
vides sensitive and specific information about the
spread of disease but is limited in evaluating pri-
mary breast cancer. New and improved dedicated
PET breast scanners have both higher sensitivity
and spatial resolution. This necessitates a recon-
sideration of the potential clinical application of
FDG-PET in primary breast cancer.
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) xi
Norbert Avril, MD
Department of Nuclear Medicine
Barts and The London School of Medicine
Queen Mary, University of London
St. Bartholomew’s Hospital, West Smithfield (QE II)
London EC1A 7BE, UK
E-mail address:
n.e.avril@qmul.ac.uk
The development of integrated PET and CT scan-
ners (PET/CT) was a milestone in the acceptance of
metabolic PET imaging in the clinical workup of
cancer patient. The coregistered metabolic and ana-
tomic information from PET/CT is resulting in
improved sensitivity and specificity compared with
the single imaging procedures. FDG-PET and FDG-
PET/CT have become an integral part of routine
patient care in breast cancer within the past few
years. New applications such as radiation therapy
planning based on metabolic/anatomic informa-
tion are on the horizon but still need careful evalua-
tion and validation. PET tracers other than FDG
targeting tumor cell proliferation and angiogenesis
as well as radiolabeled receptor ligands are now in-
creasingly being used to noninvasively characterize
breast cancer.
There are further potential applications of FDG-
PET beyond diagnostic staging. Changes in tumor
glucose consumption occur early in the course of
chemotherapy and ultimately predict treatment
outcome. The use of FDG-PET for monitoring
breast cancer therapy holds promise to reduce the
number of ineffective therapies and unnecessary
side effects and to facilitate the effective evaluation
of new therapeutic approaches.
doi:10.1016/j.cpet.2005.10.001

F-18 Fluorodeoxyglucose-Positron
Emission Tomography Imaging
for Primary Breast Cancer and
Loco-Regional Staging
a,*
Norbert Avril, MD , Lee P. Adler,
& Detection of primary breast tumors
& Positron emission tomography imaging of
breast cancer
Positron emission tomography imaging
procedure and image analysis in the
breast
Positron emission tomography imaging of
the breast
Biological features of breast cancer
Breast cancer is the most common female ma-
lignancy in most European countries, North Amer-
ica, and Australia; it is less frequent in Asia and in
Africa. In Europe, one out of every 10 to 15 women
will develop breast cancer in her lifetime, and the
risk is even higher in the United States, where it is
one out of every eight women. Approximately 95%
of breast cancer cases occur sporadically without
any known genetic mutation, and the causal
mechanisms underlying this disease have yet to
be fully elucidated. Overall, 5-year survival rates
are approximately 75%, with ranges of 92% for
Stage I (pT1, pN0, M0) to 15% for Stage IV (M1)
disease [1].
The main prognostic factors in patients who have
breast cancer are lymph node status, tumor size,
histologic grade, and the presence or absence of
a
Department of Nuclear Medicine, Barts and the London School of Medicine, Queen Mary, University of
London, West Smithfield (QEII), London, EC1A 7BE, UK
b
Adler Institute for Advanced Imaging, The Pavilion, 261 Old York Road, Jenkintown, PA 19046, USA
* Corresponding author.
E-mail address: AvrilNE@upmc.edu (N. Avril).
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
1
POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 1–13
b
MD
& Loco-regional staging
& Positron emission mammography
& Current clinical status of F-18
fluorodeoxyglucose-positron emission
tomography in primary breast cancer
& References
distant metastases. Most patients who have locally
advanced disease have axillary lymph nodes in-
volved with their tumors, but a subset of patients
have large primary tumors without lymph node
involvement. For patients who have lymph node
metastases, more than four lymph nodes involved
predict poorer survival [2]. The College of Ameri-
can Pathologists has recently considered prognostic
and predictive factors in breast cancer and stratified
them into categories reflecting the strength of pub-
lished evidence [3]. Factors ranked in Category I
included tumor, node, metastases (TNM) staging;
histologic grade; histologic type; mitotic figure
counts; and hormone receptor status. Category II
factors included c-erbB-2 (Her2-neu), proliferation
markers, lymphatic and vascular channel invasion,
and p53. Factors in Category III included DNA
doi:10.1016/j.cpet.2005.09.008
2 Avril & Adler
ploidy analysis, microvessel density, epidermal
growth factor receptor, transforming growth factor-
alpha, bcl-2, pS2, and cathepsin D.
Invasive ductal carcinoma is the most common
histological type (70%–80%), followed by invasive
lobular carcinoma (6%–10%) and medullary car-
cinoma (~3%). The remaining tumors include a
variety of histological types. Invasive breast cancer
may be present as a single tumor, or as multifocal
if tumors are growing in the same quadrant of
the breast, and as multicentric if they are detected
in different quadrants. The disease can occur in
any part of the breast, but most frequently occurs
in the upper outer quadrant. Locally advanced
breast cancer remains a particular challenge, be-
cause the majority of patients who have this
diagnosis develop distant metastases despite appro-
priate therapy. Patients who have locally advanced
disease include primary tumors with direct exten-
sion to chest wall or skin (stage T4); advanced
nodal disease, such as fixed axillary nodes or
involvement of ipsilateral supraclavicular, infracla-
vicular, or internal mammary nodes; and inflam-
matory carcinomas.
Noninvasive breast cancer consists of two histo-
logical and clinical subtypes: ductal and lobular
in-situ carcinomas. The carcinoma cells are con-
fined within the terminal duct lobular unit and
the adjacent ducts, but have not yet invaded
through the basement membrane. Generally, lobu-
lar carcinoma in situ (LCIS) does not present as
a palpable tumor and is usually found incidentally
in breast biopsies, often multifocal and bilateral.
Ductal carcinoma in situ (DCIS) is increasingly
diagnosed because of microcalcifications seen on
mammograms, and is more likely to be confined to
one breast or even to one quadrant of the breast.
If breast cancer is suspected, a biopsy is necessary
to confirm the diagnosis. The National Compre-
hensive Cancer Network has published guidelines
for the work-up of women who have newly diag-
nosed breast cancer. The recommendations include
history and physical examination, diagnostic bilat-
eral mammogram and ultrasound and optional
breast MRI, pathology review, and determination
of estrogen receptor, progesterone receptor, nuclear
grade and HER-2/neu status. To evaluate distant
metastases, chest radiograph, ultrasound of the
abdomen, bone scintigraphy, and CT or MRI may
be indicated.
Detection of primary breast tumors
Improved methods to detect and diagnose breast
cancer early are required to achieve a significant
impact on morbidity and mortality. More than
80% of cancers are detected because of a suspicious
mass, either by self-examination or routine breast
examination. Clinical signs include a fixed, hard
mass; asymmetry of the breast contour; a pro-
trusion; a subtle dimpling of the skin (peau d’
orange); or a bloody nipple discharge. Depending
on the size of the breast and the density of breast
tissue, most tumors are not palpable smaller than
1 cm in diameter. Breast carcinomas are often pres-
ent as irregularly shaped, firm or hard, yet painless
nodules or masses. Physical examination typically
does not allow an accurate differentiation between
a malignant and a nonmalignant mass. Therefore,
imaging modalities are used to improve the diag-
nostic accuracy, and various new and innovative
technologies are being investigated for advancing
the early detection and diagnosis of breast cancer.
Screening-mammography allows the detection of
breast cancer earlier than breast self-examination,
and is generally credited with earlier diagnosis and
an overall improvement in survival for patients
who have newly diagnosed breast cancer. Mam-
mography localizes and assesses the extent of a le-
sion as well as identifying other suspicious masses.
Studies in large series have shown that mammog-
raphy, using mediolateral oblique and craniocau-
dal projections, is a useful tool to improve early
detection of breast cancer. Depending on the lesion
size and the radiographic appearance and breast
tissue density, sensitivity ranges from 54% to 58%
in women under age 40, and from 81% to 94% in
those over 50 [4,5]. Malignant and benign breast
lesions often display similar radiographic appear-
ance, resulting in a major limitation of mammog-
raphy [6]. A Breast Imaging Reporting and Data
System (BI-RADS) has recently been introduced to
characterize mammography [7]. The categories are
listed and explained in Box 1.
In some studies, approximately 6 to 8 out of
10 patients who have suspicious lesions in mam-
mography and who undergo surgery have benign
histology [6,8]. A recent multicenter analysis of
332,926 diagnostic mammography examinations
[9] found the positive predictive value of a biopsy
recommendation to be 31.5%, and that of a biopsy
Box 1: BI-RADS categories and definitions
0: More information is needed to give a final
mammogram report.
1: Mammogram is normal.
2: Mammogram shows benign finding.
3: Probably benign finding—short interval
follow-up suggested.
4: Suspicious abnormality—biopsy should
be considered.
5: Highly suggestive of malignancy—
appropriate action should be taken.

performed to be 39.5% This means that currently
almost two thirds have a negative biopsy. For
screening mammography, the positive predictive
value is even lower. About 10% of breast carci-
nomas are not identified in mammography, even
when they are palpable [10]. One reason is that
mammography is limited, because cancer can have
photon absorption similar to that of normal breast
tissue, especially in younger women who have radio-
graphically dense breasts. Despite these limitations,
mammography is viewed as the best tool currently
available for screening and early diagnosis.
Ultrasound has become an important imaging
modality in evaluating the breast [11]. Ultrasound
is often used in addition to mammography, and
provides differentiation between cystic lesions and
solid tumors. In younger patients who have dense
breasts, ultrasound can be superior in the detection
of breast cancer in comparison with mammog-
raphy. Breast cancer typically shows irregularly
shaped hypoechoic masses, posterior acoustical
shadowing, and ill-defined demarcation against
the surrounding tissue. Doppler ultrasound may
help distinguish benign from malignant breast dis-
ease; however, the diagnostic accuracy is often not
sufficient enough to accurately characterize abnor-
mal tissue, and to specifically exclude malignancy
[12]. Another common application of ultrasound is
to provide guidance for interventional procedures
[12,13]. Less common uses include assisting in
staging of breast cancer and evaluating patients
who have implants. Recently, there has been an
interest in using ultrasound to screen asymptomatic
women for breast cancer, as is done with mam-
mography. Further studies must be performed to
assess if this reduces mortality from breast cancer.
Although primarily used to image the female
breast, ultrasound also can be used to evaluate
breast-related concerns in men. Uses of contrast-
enhanced ultrasound are still experimental and
would add an invasive component to an otherwise
noninvasive study.
MRI has become a valuable tool in breast disease,
especially in cases that are difficult to diagnose.
Recent progress in both spatial and temporal reso-
lutions, the imaging sequences used, pharmacoki-
netic modeling of contrast uptake, and the use
of dedicated breast coils has contributed to the
advancement of this imaging technique. More re-
cently, phased-array breast coils, pulse sequences
engineered to saturate signal from fat-containing
tissue (FATSAT), and gadolinium-based contrast
agents have done so as well [14]. MRI has several
distinct advantages for breast imaging. These in-
clude three-dimensional visualization of breast tis-
sue, information about tissue vascularity, and chest
wall visualization. Moreover, MRI allows evalua-
FDG-PET Imaging for Breast Cancer 3
tion of dense breast parenchyma that often limits
the detection of breast cancer in mammography.
The use of paramagnetic contrast agents has
been found to be essential in characterizing breast
masses. Signal enhancement following injection of
intravenous (IV) contrast is a highly sensitive cri-
terion to detect breast cancer, and sensitivity is
more than 90% in most studies. Unfortunately,
the high sensitivity of MRI for invasive breast car-
cinomas is coupled with a correspondingly low
specificity [14,15]. To improve upon the disap-
pointingly low positive predictive value of enhanc-
ing lesions on MRI, high-speed dynamic imaging
of the breast must be performed in make use of the
differential washout rates of contrast agent gado-
linium (III)-diethyltriaminepentaacetic acid (Gd-
DTPA) in benign and malignant breast lesions.
Even with dynamic imaging, MRI has not proven
to be a robust technique for discriminating benign
from malignant tumors in the community setting.
MRI of the breast offers higher sensitivity for the
detection of multifocal or multicentric cancer,
which is important in selecting patients appropriate
for breast-conserving surgery. It is also a valuable
tool for the screening of patients who have a high
risk of breast cancer, or in whom there is axillary
disease or nipple discharge and conventional imag-
ing has not revealed the primary focus. Lesions
detected on MRI are frequently visible on mammog-
raphy or ultrasound in retrospect, if not prospec-
tively, allowing for subsequent biopsy. Techniques
are also now available to biopsy lesions only appar-
ent on MRI of the breast. MRI can differentiate scar
tissue from tumor, and is specifically useful in pa-
tients suspected for local recurrent disease. Studies
suggest that MRI can identify responders and non-
responders to neoadjuvant chemotherapy with
more accuracy [16]. It is the modality of choice
for the assessment of breast implants for rupture,
with accuracy higher than radiograph mammogra-
phy and ultrasound. The most important limita-
tions of MRI beside the low specificity are patient
compliance, scan time, and cost.
Positron emission tomography imaging of
breast cancer
Positron emission tomography (PET) is a noninva-
sive imaging technique that measures the concen-
tration of positron-emitting radiopharmaceuticals
within the body. Depending upon the radiolabeled
tracer used, PET can be used to determine various
physiological and biochemical processes in vivo.
PET is highly sensitive, with the capacity to detect
picomolar concentrations of radiotracer, and pro-
vides superior image resolution compared with con-
ventional nuclear medicine imaging with gamma
4 Avril & Adler
cameras. Currently, PET imaging can target several
biological features of cancer, including glucose
metabolism, cell proliferation, perfusion, and hyp-
oxia. Approximately 95% of clinical PET examina-
tions are performed in patients who have known or
suspected cancer, and virtually all of these are per-
formed with a single radiotracer, the radiolabeled
glucose analog, [F-18] 2-deoxy-2-fluoro-D-glucose
(FDG). Following malignant transformation, vari-
ous tumors are characterized by elevated glucose
consumption and subsequent increased uptake and
accumulation of FDG. PET imaging using FDG
provides more sensitive and more specific infor-
mation about the extent of disease than morpho-
logical/anatomical imaging alone. FDG-PET has
become a standard imaging procedure for staging
and restaging of many types of cancer [17]. In the
United States, PET scans performed on patients
who have head and neck cancer, follicular thyroid
cancer, solitary pulmonary nodules, lung cancer,
breast cancer, esophageal cancer, colorectal cancer,
lymphoma, melanoma, and cervical cancer are
generally considered reimbursable by third-party
payers. The metabolic activity of neoplastic tissue
assessed by PET offers additional information about
cancer biology, and can be used for the differentia-
tion between benign and malignant lesions, identi-
fication of early disease and staging of metastases,
assessment of therapeutic effectiveness, and to de-
termine tumor aggressiveness. The uptake mecha-
nism and biochemical pathway of the glucose
analog FDG has been extensively studied both in
vitro and in vivo. The transport of the radiotracer
through the cell membrane via glucose transport
proteins, particularly glucose transporter type 1
(GLUT-1), and subsequent intracellular phosphory-
lation by hexokinase (HK) have been identified as
key steps for subsequent tissue accumulation [18].
Because FDG-6-phosphate is not a suitable sub-
strate for glucose-6-phosphate isomerase, and the
enzyme level of glucose-6-phosphatase is generally
low in tumors, FDG-6-phosphate accumulates in
cells and is visualized by PET.
Positron emission tomography imaging
procedure and image analysis in the breast
To ensure a standardized metabolic state, including
low plasma glucose levels, oncology patients must
fast for at least 4 to 6 hours before administration
of FDG. Blood glucose level before tracer injection
should ideally not exceed 150 mg/100 ml. Intra-
venous administration of about 300 to 400 MBq
(~10 mCi) F-18 FDG is used in most studies; how-
ever, Adler and colleagues [19,20] reported on a
higher breast cancer detection rate using up to
750 MBq (~20 mCi) F-18 FDG. To avoid artificial
tracer retention in the axilla region, the tracer
should be injected into an arm vein contralateral
to the suspected tumor. Most of the studies re-
ported in literature are done in two-dimensional
mode data acquisition, and the influence of three-
dimensional mode on the results of breast imaging
still needs to be studied. Imaging in prone position
with both arms at the side and the breast hanging
free is recommended to avoid compression and
deformation of the breast. Data acquisition should
be started approximately 60 minutes after tracer
injection. Boerner and colleagues [21] showed in-
creasing target-to-background ratios over time, sug-
gesting a benefit to longer waiting periods between
tracer injection and data acquisition. Lower image
quality, due to radionuclide decay, has to be taken
into account, however. Attenuation correction is
recommended for optimal tumor localization as
well as subsequent quantification of regional tracer
uptake. The use of iterative reconstruction algo-
rithms results in better image quality; an increase
in diagnostic accuracy has not yet been reported.
Visual image interpretation should include analysis
of transaxial, coronal, and sagittal views. Breast can-
cer is typically present with focally increased FDG
uptake, whereas benign tumors are negative in PET
imaging. Proliferative mammary dysplasia may re-
sult in moderate but diffuse increased tracer uptake.
Attenuation-corrected PET images provide quan-
titative information about the tracer concentration
in tissue. Various approaches of different complex-
ity can be applied for quantitative PET analysis.
Standardized uptake values (SUV) are frequently
being calculated, providing a semiquantitative mea-
sure of FDG accumulation in tissue by normaliz-
ing the tissue radioactivity concentration measured
with PET to injected dose and patient’s body
weight. Quantitative methods may be used to com-
plement visual image analysis for differentiation
between benign and malignant breast tumors;
that is, by using an SUV-normalized scale for
image display [22]. In particular, SUV correction
for partial volume effects and normalization to
blood glucose has been shown to yield the highest
diagnostic accuracy for breast imaging. Corre-
sponding threshold values for optimal tumor char-
acterization have been published for various
quantification methods [22]. Dynamic data acqui-
sition allows calculation of the tracer influx con-
stant, although this procedure is more complex and
did not increase diagnostic accuracy.
Whole-body imaging can be improved by in-
travenous injection of furosemide (20–40 mg) to
reduce tracer retention in the urinary system and
butylscopolamine (20–40 mg) to reduce FDG up-
take in the bowel [23]. Image evaluation requires
an appreciation of the normal physiologic FDG
 FDG-PET Imaging for Breast Cancer 5

Fig. 1. Transaxial FDG-PET image (A), CT image (B), and fused PET/CT image (C), demonstrating a focal area of
intense FDG uptake in the left breast.

Fig. 2. Moderate FDG uptake in a local recurrence in the left breast. Transaxial FDG-PET (A), CT (B), and fused
PET/CT (C).
6 Avril & Adler

Fig. 3. Frontal maximum-intensity projection PET image (A) and sagittal slices from a PET/CT scan (B) show a
mammographically occult 8-mm breast cancer detected incidentally by FDG-PET during a workup for presumed
recurrent breast cancer in a patient with rising carcinoembryonic antigen. An ultrasound guided biopsy demon-
strated an invasive ductal carcinoma, which was resected with clear margins.

uptake distribution and of variation between indi-
viduals, as well as consideration of artifacts and
benign conditions that can mimic malignancy.
Increased FDG uptake is found within the brain
cortex, the myocardium, and the urinary tract.
Low-to-moderate uptake is seen in the base of the
tongue, salivary glands, thyroid, liver, spleen, gas-
trointestinal tract, bone marrow, musculature, and
reproductive organs. Of particular importance is
the inconsistent amount of normal uptake in glan-
dular breast tissue. The most common normal
cause of misinterpretation is related to muscle ac-
tivity. Muscle tension may lead to increased FDG
uptake, and physical activity immediately before or
after tracer injection can lead to spurious muscle
activity. In some patients, supraclavicular uptake
has been shown to represent brown fat. Inflam-
matory and infectious processes also demonstrate
increased FDG uptake, as well as some benign
diseases, such as Paget’s disease, Graves’ disease,
granulomatous disorders, healing fractures, and post-
radiation changes.
Positron emission tomography imaging of
the breast
The first FDG imaging in breast cancer patients,
using a collimated gamma camera, was reported
in 1989 by Minn and Soini [24]. Shortly afterwards,
Kubota and coworkers [25] reported on PET imag-
ing with FDG in one case with local recurrence.
In a first series of 10 patients who had locally
advanced breast cancer, Wahl and colleagues [26]
successfully identified all breast carcinomas. Subse-
quent studies including a limited number of pa-
tients, predominantly having advanced stages of
disease, suggested a high accuracy of FDG-PET
for the detection of primary breast carcinomas
[Figs. 1–7] [19,24–26]. The largest patient group
reported to date includes 144 patients who had
185 histologically confirmed breast tumors [27].

Fig. 4. Transaxial FDG-PET (A), and fused PET/CT image (B) with a small area of increased FDG uptake in a axillary
lymph node metastasis.
 FDG-PET Imaging for Breast Cancer 7

Fig. 5. Coronal PET/CT images demonstrate a false-positive axillary lymph node finding in a patient 2 weeks
following a flu shot (arrows). There is also extensive, mild FDG uptake corresponding to metabolically active fat
in the supraclavicular soft tissues.

PET detected breast cancer with an overall sensitiv-
ity of 64.4% by conservative image reading (regard-
ing only definite FDG uptake as positive), and
80.3% by sensitive image reading (regarding equivo-
cal as well as definite FDG uptake as positive). When
applying sensitive image reading, however, speci-
ficity decreased from 94.3% to 75.5% [27,28].
Schirrmeister and coworkers [29] found similar
results in 117 patients, with a sensitivity of 93%
and specificity of 75%. The use of non-attenuation–
corrected PET imaging combined with sensitive im-
aging reading may have contributed to the higher
sensitivity and lower specificity in this study. A
recent study compared MR imaging of the breast
with FDG-PET [30] found a comparable diagnostic
accuracy (88% versus 84%) for both methods in
32 patients. The sensitivity of FDG-PET was 79%,
whereas MRI detected all primary breast carci-
nomas; however, the specificity of FDG-PET was
higher (94% versus 72%). Baslaim and coworkers
[31] evaluated the usefulness of FDG-PET for diag-
nosing and staging of inflammatory breast cancer.
All 7 patients studied presented with diffuse breast
enlargement, redness, and peau d’orange. PET
showed diffuse FDG uptake in the involved breast,
with intense uptake in the primary tumor as well as
increased FDG uptake in the skin.
The ability of PET to detect breast cancer greatly
depends on tumor size. Regarding small tumors,
only 30 (68.2%) out of 44 breast carcinomas at

Fig. 6. Mediolateral radiograph mammograms of both breasts showed heterogeneously dense breast parenchyma
with bilateral masses (arrows). Ultrasound-guided right breast biopsy showed invasive ductal carcinoma
(A, arrow), whereas left breast ultrasound-guided biopsy showed papilloma (B, arrow). PEM of right breast
showed dominant known invasive cancer (C, arrow) and smaller foci in upper breast, proving multicentric cancer
with positive lymph node (arrowhead), and increased focal activity on left side (D, arrow) that was confirmed as
DCIS at surgery.
8 Avril & Adler

Fig. 7. (A–D) Mammogram, FDG-PEM, whole body PET, and CT in a patient with DCIS.

stage pT1 (<2 cm) were correctly identified, com- cancer were identified by PET [27]. Nevertheless,
pared with 57 (91.9%) out of 62 at stage pT2 (>2– Schirrmeister and colleagues [29] found that PET
5 cm) [27]. Sensitivity for tumors less than 1 cm was twice as sensitive in detecting multifocal
(pT1a and b) was only 25%, compared with 84.4% lesions (sensitivity 63%, specificity 95%) than the
for tumors between 1 and 2 cm in diameter (pT1c). combination of mammography and ultrasound
Table 1 provides more detailed information. (sensitivity 32%, specificity 93%).
Invasive lobular carcinomas were more often The diagnosis of in-situ carcinomas has increased
false-negative (65.2%) than invasive ductal carci- over the past decade, mainly due to increased use
nomas (23.7%). These results are consistent with a of and technological improvements in mammogra-
previous report from Crippa and colleagues [32], phy. There is little information available about
who found higher glucose metabolism for invasive the ability of PET imaging to detect noninvasive
ductal carcinomas (median SUV of 5.6) versus breast cancer. Tse and coworkers [36] studied
invasive lobular carcinomas (median SUV of 3.8). 14 patients and found that one out of two false-
This is of particular importance in the clinical negative cases had predominantly intraductal can-
application of PET, because lobular carcinomas cer with microscopic invasive foci. In 12 patients,
are more difficult to diagnose by imaging proce- (10 DCIS and 2 LCIS) none out of six in-situ
dures such as mammography, sonography, and carcinomas smaller than 2 cm could be identified
MRI [33–35]. The identification of multifocal or [27]. For larger in-situ carcinomas, three (50%)
multicentric breast cancer plays an important role out of six displayed increased FDG uptake. Al-
in the decision of therapy, because it limits breast- though the number of patients studied was small,
conserving surgery. Only 9 (50%) out of 18 pa- these data suggest that PET imaging cannot con-
tients who had multifocal or multicentric breast tribute to an improved diagnosis of noninvasive
breast cancer.
Vranjesevic and colleagues [37] evaluated the
Table 1: Sensitivity of FDG-PET and the size of influence of the breast tissue density on FDG
breast cancer uptake of normal breast tissue, and found signifi-
TNM Size n Sensitivity cantly lower SUVs for primarily fatty breasts than
for dense breasts. Benign conditions of the breast
pTis 12 42% are more common, and are often difficult to differ-
pT1 <2.0 cm 44 68% entiate from breast cancer in conventional imaging
pT1a <0.5 cm 4 25%
modalities. In general, benign breast masses display
pT1b >0.5–1.0 cm 8 25%
pT1c >1.0–2.0 cm 32 84% low FDG uptake. Only 3 out of 53 benign breast
pT2 >2.0–5.0 cm 62 92% masses presented with focally increased tracer
>2.0–3.0 cm 33 94% uptake, including one rare case of a ductal ade-
>3.0–4.0 cm 15 87% noma, one case with dysplastic tissue, and one
>4.0–5.0 cm 14 93% fibroadenoma [27]. Fibroadenomas are common
pT3 >5.0 cm 14 100% benign tumors, and only 1 out of 9 displayed in-
creased tracer uptake. Moreover, dysplastic tissue
Data from Avril N, Rose CA, Schelling M, et al. Breast im-
aging with positron emission tomography and fluorine-
often accounts for false-positive results in MRI,
18 fluorodeoxyglucose: use and limitations. J Clin Oncol predominantly showing a diffuse pattern of little
2000;18(20):3495–502. or moderate FDG uptake.

Biological features of breast cancer
FDG uptake in breast cancer shows considerable
variation. To address that issue, the degree of tracer
accumulation was recently correlated with various
tumor characteristics [Table 2] [38]. Histologic sec-
tions of breast cancer specimens were analyzed for
histologic type, microscopic tumor growth pattern,
percentage of tumor cells, presence of inflamma-
tory cells, density of blood vessels, histopathologic
grading, tumor cell proliferation (mitotic rate and
antibody binding of MIB-1), expression of estrogen
and progesterone receptors, and expression of the
glucose transporter protein Glut-1. Invasive ductal
carcinomas displayed significantly higher FDG
uptake compared with invasive lobular carci-
nomas. The SUVs for clearly defined lesions were
higher compared with tumors with diffuse growth
patterns. Lower densities of blood vessels cor-
responded to higher FDG uptakes. In addition,
there was a positive correlation between FDG up-
take and tumor cell proliferation, but only a weak
relationship between FDG uptake and the percen-
tage of tumor cells. There was no relationship be-
tween FDG uptake and tumor size; axillary lymph
node status; percentage of necrotic, fibrotic and
cystic compounds; presence of inflammatory cells;
steroid receptor status; and expression of Glut-1.
Histologic and immunohistochemical tissue
analysis could not fully explain the variation of
FDG uptake in breast cancer. The study authors
concluded that the degree of metabolic changes in
malignant tumors is mainly caused by complex
interactions between the cellular energy metabo-
lism and the microenvironment of the tumor.
Hence FDG-PET uptake can not be used to predict
the biologic behaviors of breast cancer, such as
Table 2: FDG-uptake versus histology/
immunohistochemistry
SUV vs. p n
Histology (ductal versus lobular) 0.003 50
Tumor growth (nodular versus 0.007 49
diffuse)
Grading 0.69 50
% tumor cells 0.06 50
Inflammatory cells 0.74 50
Capillaries 0.08 50
Proliferation (MIB-1) 0.009 40
Estrogen receptor status 0.47 42
Progesterone receptor status 0.29 42
Glut-1 transporter protein 0.21 45
Data from Avril N, Menzel M, Dose J, et al. Glucose me-
tabolism of breast cancer assessed by 18F-FDG PET: his-
tologic and immunohistochemical tissue analysis. J Nucl
Med 2001;42(1):9–16.
FDG-PET Imaging for Breast Cancer 9
differentiation, histopathologic grading, cell prolif-
eration, or axillary lymph node status.
Loco-regional staging
The axillary lymph node status is still considered
the single most important prognostic indicator in
patients who have breast cancer. Clinical examina-
tion is generally unreliable for staging the axilla
[39]. Lack of conventional imaging techniques to
determine the axillary lymph node involvement
with acceptable accuracy has been the main reason
for axillary lymph node dissection; however, up to
70% of patients who have stage T1 and T2 tumors
have negative axillary lymph nodes [40]. The ex-
tent, morbidity, and cost of the staging procedure
of axillary lymph node dissection are often greater
than those of the surgical treatment of the primary
tumor. In anatomical based imaging modalities,
such as computed tomography, ultrasound, and
MRI, the size of a particular lymph node is of
crucial importance to determine the tumor involve-
ment. Generally, lymph node enlargement over
1 cm in diameter is the decisive criterion. In con-
trast, metabolic imaging with FDG-PET is suggested
to provide more specific information, based on de-
tecting increased glucose consumption of cancer
tissue. In 1991, Wahl and coworkers [26] studied
12 patients who had locally advanced breast cancer,
and found increased FDG uptake in axillary metas-
tases. In 50 patients, Adler and colleagues [20]
reported a sensitivity of 95%, a negative predictive
value of 95%, and an overall accuracy of 77% for
axillary PET imaging. Greco and coworkers [41]
studied 167 consecutive breast cancer patients,
and axillary involvement was detected in 68 of
72 patients, resulting in a sensitivity of 94.4% and
a specificity of 86.3%; overall accuracy of lymph
node staging with PET was 89.8%. There is some
controversy about the sensitivity of axillary PET
imaging, however [42]. It is a well-known phe-
nomenon that the true FDG uptake is underesti-
mated in small cancer deposits because of the
limited spatial resolution of current PET devices,
(approximately 6–8 mm). Therefore it cannot be
expected that PET will provide visualization of
micrometastases. Avril and colleagues [43] studied
51 patients and found overall sensitivity and speci-
ficity for detection of axillary lymph node metas-
tases to be 79% and 96%, respectively. In patients
who had primary breast tumors larger than 2 cm
(>stage pT1), the sensitivity increased to 94%, with
a corresponding specificity of 100%. PET could
not identify lymph node metastases in 4 out of
6 patients who had small primary breast cancer
(stage pT1), however, which led to a sensitivity of
only 33% in this group. Although the number of
10 Avril & Adler
patients studied is small, this study clearly points
out that the current achievable spatial resolution
of PET imaging limits the detection of microme-
tastases and small tumor-infiltrated lymph nodes.
This conclusion is also supported by others; for
example, by Schirrmeister and coworkers [29],
who studied 117 breast cancer patients and found
similar results (sensitivity 79%, specificity 92%). In
a prospective multicenter study representing the
largest patient cohort so far [44], 360 women
who had newly diagnosed invasive breast cancer
underwent PET imaging. Three experienced readers
blindly interpreted PET images, and the results
from 308 axillae were compared with histopathol-
ogy. If at least one probably or definitely abnormal
axillary focus was considered positive, the sensitiv-
ity for PET was 61% and the specificity was 80%.
Patients who had false-negative PET had signifi-
cantly smaller and fewer tumor-positive lymph
nodes than true-positive cases. Semiquantitative
analysis of axillary FDG uptake showed that a
nodal standardized uptake value (lean body mass)
of more than 1.8 had a positive predictive value of
90% but a sensitivity of only 32%. Finding two or
more intense foci of tracer uptake in the axilla was
highly predictive of axillary metastasis, but had a
sensitivity of only 27%.
Sentinel node biopsy has become accepted as a
reliable method of predicting the status of the axilla
in early stages of breast cancer [40]. There are few
studies available directly comparing the diagnostic
accuracy of PET with sentinel node biopsy in breast
cancer patients. In one study [45], 5 out of 15 pa-
tients had sentinel lymph node metastases, but
PET identified only 1 of these patients. The size
of missed metastases ranged from a small micro-
metastasis identified only by immunohistochemis-
try to an 11-mm, tumor-involved lymph node.
Another study [46] included 24 clinically node-
negative breast cancer patients who had primary
tumors smaller than 3 cm, and axillary staging by
PET was accurate in 15 of 24 patients (62.5%). PET
was false-negative in 8 of 10 node-positive patients,
and false-positive in 1 patient. The sensitivity and
specificity of FDG-PET were 20% and 93%, respec-
tively. The mean diameter of false-negative axillary
lymph node metastases was 7.5 mm, and ranged
from 1 to 15 mm. In 32 breast cancer patients who
had clinically negative axillary nodes, sentinel
lymph node biopsy was false-negative in 1 patient,
whereas PET missed lymph node metastases in
11 patients, resulting in a sensitivity of 20% [47].
These studies clearly indicate the limitation of
FDG-PET for axillary staging of small primary
tumors. FDG-PET is not accurate enough in clini-
cally node-negative breast cancer patients qualify-
ing for sentinel lymph node dissection. On the
other hand, among patients who have larger tu-
mors, sentinel biopsy can be avoided in those
who have positive FDG-PET, in whom complete
axillary lymph node dissection should be the pri-
mary procedure. FDG-PET cannot replace the axil-
lary dissection, not only because of the limited
sensitivity, but also because the number of involved
lymph nodes and extranodal extension cannot be
determined. In patients who have locally advanced
disease and who are undergoing primary chemo-
therapy, however, FDG-PET seems to be a reliable
method to determine the extent of disease.
Recently, the intrathoracic lymph node status has
been retrospectively analyzed comparing CT and
PET [48]. In 73 consecutive patients who had recur-
rent or metastatic disease, PET was able to correctly
identify 40% of the patients who had intrathoracic
lymph node metastases, resulting in a sensitivity of
85% and a specificity of 90%. Only 23% of the
patients had suspiciously enlarged lymph nodes in
CT, leading to a sensitivity of 54% and a specificity
of 85%. PET and CT were both positive in 22% of
the cases. Therefore overall diagnostic accuracy of
PET was higher (88%) than that of CT (73%).
Despite the limitation in detection of small tumor
deposits, FDG-PET is currently the most sensitive
imaging modality to detect lymph node metastases,
including parasternal and mediastinal nodes.
Positron emission mammography
Because of the limited resolution of conventional
PET scanners (4.8 to 7.1 mm in plane resolution),
the even more limited resolution achieved with
current clinical protocols (1 cm or greater), and
the loss of contrast due to scatter, conventional
PET scanners are unlikely to detect the small carci-
nomas that are detectable with other imaging mo-
dalities such as mammography and MRI. Dedicated
breast PET—positron emission mammography
(PEM)—was developed to improve both resolution
and contrast of breast lesions and thereby improve
small lesion detectability, while reducing the cost
of imaging [49]. By optimizing scanner geometry
for breast imaging, a dedicated breast PET scanner
can achieve much higher resolution than a whole
body scanner, while maintaining higher count sen-
sitivity and a marked reduction in scatter [50,51].
Radiographic mammography uses compression to
reduce the mean radiographic travel path, thereby
reducing scatter. In an analogous manner, incor-
porating compression into PET imaging reduces
the mean travel path of gamma emissions, with a
resulting reduction in scatter. Because scatter rep-
resents such a low proportion of incident pho-
tons, septa or collimation is unnecessary, and a
three-dimensional acquisition with full three-

dimensional reconstruction is performed. An added
benefit of compression is that images can be pro-
duced with a correspondence to radiograph mam-
mograms, so PEM can be used to interrogate the
functional status of abnormalities detected by mam-
mography. Final output resolution of commercially
available equipment is approximately 1.5 mm in
plane. Resolution orthogonal to the imaging planes
is dependant on the distance between the flat plate
detectors (ie, the compressed thickness of the breast)
and is generally not as good as in-plane resolution
[52]. The dramatic improvement in resolution
achieved results in the ability to detect smaller
objects than would be detectable using a whole-
body scanner (assuming identical object-to-back-
ground activity ratios) or to detect small objects
with lower object-to-background activity ratios
than would be detectable with whole body PET [19].
A recent multicenter study [52,53] demonstrated
that high-resolution FDG-PEM detects in-situ com-
ponents of cancers better than any other modality.
This fact has been documented in retrospective
surgical studies [52] as well as prospective imaging
studies [53]. In the last-cited study, 93 consecutive
subjects who had biopsy-proven breast cancer or
suspicious breast lesions were recruited from four
sites [53]. Only evaluable cases (eg, with proof of
pathology) were shown to a blinded panel of imag-
ing specialists for review. Patients who had Type I
or poorly controlled Type II diabetes were excluded
from participating in the study. Other diabetic
patients were included, with the prospective deci-
sion that diagnostic performance of the PEM scans
would be subsequently analyzed to determine
whether diagnostic accuracy would differ in certain
subsets of patients (eg, those who had diabetes or
other medical conditions). Among index cancers in
the 77 evaluable subjects reviewed, 39 (93%) out
of 42 were PEM-positive. Including incidental
lesions, 43 (90%) out of 48 malignancies were
PEM-positive, including 10 (91%) out of 11 lesions
of DCIS and 33 (89%) out of 37 invasive carcino-
mas. Three index malignancies (a 3-mm Grade II/
III infiltrating and intraductal carcinoma, a 6-mm
Grade I/III tubular carcinoma, and a 10-mm
Grade I/III invasive ductal carcinoma with a positive
axillary lymph node) were occult on FDG-PEM, but
were visible mammographically. One contralateral
25-mm invasive lobular carcinoma was visible mam-
mographically, but was only recognized by one of
three PEM readers. Three DCIS foci (Grade II/III)
were visible on FDG-PEM, but were mammographi-
cally occult. The combination of mammography,
ultrasound, and FDG-PEM depicted 47 (98%) out
of 48 cancers, with one case of contralateral Paget’s
disease caused by Grade II DCIS missed on all
imaging modalities. FDG-PEM improved sensitivity
FDG-PET Imaging for Breast Cancer 11
when added to mammography, ultrasound, and
clinical examination, without reducing accuracy.
One of three lesions with atypical ductal hyper-
plasia showed intense FDG uptake. Five (12%) of
41 other proven benign lesions showed intense
FDG uptake, including two fibroadenomas, two
fibrocystic changes, and one fat necrosis in a parti-
cipant who had transverse rectus abdominis muscle
(TRAM) flap reconstruction. The first two readers
agreed on overall PEM assessment (ie, recommen-
dation for biopsy or not) for 83% of the index
lesions and 80% of the incidental lesions. Examin-
ing individual reader performance, at least one
reader missed a malignancy in 8 index lesions
and 3 incidental lesions. Examining all readings
of index lesions, the average area under the receiver
operating characteristic (ROC) curve was 0.91.
Qualitatively high and asymmetric FDG uptake,
as compared with radiograph mammograms, was
the finding most predictive of malignancy. Al-
though promising, additional prospective studies
are needed to determine the clinical role of FDG-
PEM imaging in the future.
Current clinical status of F-18
fluorodeoxyglucose-positron emission
tomography in primary breast cancer
Breast cancer often displays only moderately in-
creased FDG uptake, and considering the limited
spatial resolution of FDG-PET, metabolic imaging
results in a low sensitivity to detect small breast
carcinomas, micrometastases, and small-tumor–
infiltrated lymph nodes. The restricted sensitivity of
FDG-PET does not allow the screening of asymptom-
atic women for breast cancer. Moreover, negative PET
results in patients who have suspicious breast masses
or abnormal mammography do not exclude breast
cancer. Therefore, PET imaging may not be used as a
routine application for evaluation of primary breast
tumors, and currently cannot significantly reduce
unnecessary invasive procedures in patients sus-
pected of having breast cancer. Advances in technol-
ogy such as the development of dedicated breast
imaging devices (eg, PEM) may improve the detec-
tion of primary tumors with PET in the future. In
patients who have locally advanced breast cancer,
PET accurately determines the extent of disease, par-
ticularly the loco-regional lymph node status. In
smaller tumors, the sentinel node biopsy has become
the standard of care in many institutions.
References
[1] Sant M, Allemani C, Berrino F, et al. Breast car-
cinoma survival in Europe and the United States.
Cancer 2004;100(4):715–22.
12 Avril & Adler
[2] Banerjee M, George J, Song EY, et al. Tree-based
model for breast cancer prognostication. J Clin
Oncol 2004;22(13):2567–75.
[3] Fitzgibbons PL, Page DL, Weaver D, et al. Prog-
nostic factors in breast cancer. College of Ameri-
can Pathologists Consensus Statement 1999.
Arch Pathol Lab Med 2000;124(7):966–78.
[4] Sickles EA. Breast masses: mammographic evalua-
tion. Radiology 1989;173:297–303.
[5] Kopans DB, Feig SA. False positive rate of screen-
ing mammography. N Engl J Med 1998;339(8):
562–4.
[6] Kopans DB. The positive predictive value of mam-
mography. AJR Am J Roentgenol 1992;158:521–6.
[7] Liberman L, Menell JH. Breast imaging reporting
and data system (BI-RADS). Radiol Clin North
Am 2002;40(3):409–30, v.
[8] Meyer JE, Eberlein TJ, Stomper PC, et al. Biopsy
of occult breast lesions. Analysis of 1261 abnor-
malities. JAMA 1990;263:2341–3.
[9] Sickles EA, Miglioretti DL, Ballard-Barbash R,
et al. Performance benchmarks for diagnostic
mammography. Radiology 2005;235(3):775–90.
[10] Bird RE, Wallace TW, Yankaskas BC. Analysis
of cancer missed at screening mammography.
Radiology 1992;184:613–7.
[11] Fine RE, Staren ED. Updates in breast ultra-
sound. Surg Clin North Am 2004;84(4):1001–34,
v–vi.
[12] Mendelson EB. Problem-solving ultrasound.
Radiol Clin North Am 2004;42(5):909–18, vii.
[13] Simmons R. Ultrasound in the changing ap-
proaches to breast cancer diagnosis and treat-
ment. Breast J 2004;10(Suppl 1):S13–4.
[14] Lee CH. Problem solving MR imaging of
the breast. Radiol Clin North Am 2004;42(5):
919–34, vii.
[15] Hylton N. Magnetic resonance imaging of the
breast: opportunities to improve breast cancer
management. J Clin Oncol 2005;23(8):1678–84.
[16] Chen X, Moore MO, Lehman CD, et al. Com-
bined use of MRI and PET to monitor response
and assess residual disease for locally advanced
breast cancer treated with neoadjuvant chemo-
therapy. Acad Radiol 2004;11(10):1115–24.
[17] Rohren EM, Turkington TG, Coleman RE.
Clinical applications of PET in oncology. Radi-
ology 2004;231(2):305–32.
[18] Brown RS, Wahl RL. Overexpression of Glut-1
glucose transporter in human breast cancer.
An immunohistochemical study. Cancer 1993;
72(10):2979–85.
[19] Adler LP, Crowe JP, al-Kaisi NK, et al. Evaluation
of breast masses and axillary lymph nodes with
[F-18] 2- deoxy-2-fluoro-D-glucose PET. Radiol-
ogy 1993;187(3):743–50.
[20] Adler LP, Faulhaber PF, Schnur KC, et al. Axillary
lymph node metastases: screening with [F-18]
2-deoxy-2-fluoro-D-glucose (FDG) PET. Radiol-
ogy 1997;203(2):323–7.
[21] Boerner AR, Weckesser M, Herzog H, et al. Opti-
mal scan time for fluorine-18 fluorodeoxy-
glucose positron emission tomography in breast
cancer. Eur J Nucl Med 1999;26(3):226–30.
[22] Avril N, Bense S, Ziegler SI, et al. Breast imaging
with fluorine-18-FDG PET: quantitative image
analysis. J Nucl Med 1997;38(8):1186–91.
[23] Stahl A, Weber WA, Avril N, et al. Effect of
N-butylscopolamine on intestinal uptake of
fluorine-18-fluorodeoxyglucose in PET imaging
of the abdomen. Nuklearmedizin 2000;39(8):
241–5.
[24] Minn H, Soini I. F-18 fluorodeoxyglucose scintig-
raphy in diagnosis and follow up of treatment in
advanced breast cancer. Am J Clin Pathol 1989;
91(5):535–41.
[25] Kubota K, Matsuzawa T, Amemiya A, et al. Imag-
ing of breast cancer with F-18 fluorodeoxy-
glucose and positron emission tomography. J
Comput Assist Tomogr 1989;13(6):1097–8.
[26] Wahl RL, Cody RL, Hutchins GD, et al. Primary
and metastatic breast carcinoma: initial clinical
evaluation with PET with the radiolabeled
glucose analogue 2-[F-18]-fluoro-2-deoxy- D-
glucose. Radiology 1991;179(3):765–70.
[27] Avril N, Rose CA, Schelling M, et al. Breast
imaging with positron emission tomography and
fluorine-18 fluorodeoxyglucose: use and limita-
tions. J Clin Oncol 2000;18(20):3495–502.
[28] Avril N, Dose J, Jänicke F, et al. Metabolic
characterization of breast tumors with positron
emission tomography using F-18 fluorodeoxy-
glucose. J Clin Oncol 1996;14(6):1848–57.
[29] Schirrmeister H, Kuhn T, Guhlmann A, et al.
Fluorine-18 2-deoxy-2-fluoro-D-glucose PET in
the preoperative staging of breast cancer: com-
parison with the standard staging procedures.
Eur J Nucl Med 2001;28(3):351–8.
[30] Goerres GW, Michel SC, Fehr MK, et al. Follow-
up of women with breast cancer: comparison
between MRI and FDG PET. Eur Radiol 2003;
13(7):1635–44.
[31] Baslaim MM, Bakheet SM, Bakheet R, et al. 18-
Fluorodeoxyglucose-positron emission tomogra-
phy in inflammatory breast cancer. World J Surg
2003;27(10):1099–104.
[32] Crippa F, Seregni E, Agresti R, et al. Association
between F-18 fluorodeoxyglucose uptake and
postoperative histopathology, hormone receptor
status, thymidine labelling index and p53 in
primary breast cancer: a preliminary observation.
Eur J Nucl Med 1998;25(10):1429–34.
[33] Krecke KN, Gisvold JJ. Invasive lobular car-
cinoma of the breast: mammographic findings
and extent of disease at diagnosis in 184 patients.
AJR Am J Roentgenol 1993;161(5):957–60.
[34] Gilles R, Guinebretière J-M, Lucidarme O, et al.
Nonpalpable breast tumors: diagnosis with con-
trast-enhanced subtraction dynamic MR imaging.
Radiology 1994;191:625–31.
[35] Paramagul CP, Helvie MA, Adler DD. Invasive
lobular carcinoma: sonographic appearance and
role of sonography in improving diagnostic
sensitivity. Radiology 1995;195(1):231–4.

[36] Tse NY, Hoh CK, Hawkins RA, et al. The ap-
plication of positron emission tomographic
imaging with fluorodeoxyglucose to the evalua-
tion of breast disease. Ann Surg 1992;216(1):
27–34.
[37] Vranjesevic D, Schiepers C, Silverman DH, et al.
Relationship between 18F-FDG uptake and
breast density in women with normal breast
tissue. J Nucl Med 2003;44(8):1238–42.
[38] Avril N, Menzel M, Dose J, et al. Glucose me-
tabolism of breast cancer assessed by 18F-FDG
PET: histologic and immunohistochemical tissue
analysis. J Nucl Med 2001;42(1):9–16.
[39] Sacre RA. Clinical evaluation of axillar lymph
nodes compared to surgical and pathological
findings. Eur J Surg Oncol 1986;12(2):169–73.
[40] Veronesi U, Paganelli G, Galimberti V, et al.
Sentinel-node biopsy to avoid axillary dissection
in breast cancer with clinically negative lymph-
nodes. Lancet 1997;349(9069):1864–7.
[41] Greco M, Crippa F, Agresti R, et al. Axillary lymph
node staging in breast cancer by 2-Fluoro-2-
deoxy-D-glucose-positron emission tomography:
Clinical evaluation and alternative management.
J Natl Cancer Inst 2001;93(8):630–5.
[42] Torrenga H, Licht J, van der Hoeven JJ, et al. Re:
axillary lymph node staging in breast cancer by
2-fluoro-2-deoxy-D-glucose-positron emission
tomography: clinical evaluation and alternative
management. J Natl Cancer Inst 2001;93(21):
1659–61.
[43] Avril N, Dose J, Jänicke F, et al. Assessment of
axillary lymph node involvement in breast
cancer patients with positron emission tomogra-
phy using radiolabeled 2-(fluorine-18)-fluoro-
2-deoxy-D-glucose. J Natl Cancer Inst 1996;
88(17):1204–9.
[44] Wahl RL, Siegel BA, Coleman RE, et al. Prospec-
tive multicenter study of axillary nodal staging by
positron emission tomography in breast cancer:
a report of the staging breast cancer with PET
Study Group. J Clin Oncol 2004;22(2):277–85.
FDG-PET Imaging for Breast Cancer 13
[45] Kelemen PR, Lowe V, Phillips N. Positron emis-
sion tomography and sentinel lymph node dis-
section in breast cancer. Clin Breast Cancer 2002;
3(1):73–7.
[46] Fehr MK, Hornung R, Varga Z, et al. Axillary
staging using positron emission tomography in
breast cancer patients qualifying for sentinel
lymph node biopsy. Breast J 2004;10(2):89–93.
[47] Barranger E, Grahek D, Antoine M, et al. Evalua-
tion of fluorodeoxyglucose positron emission
tomography in the detection of axillary lymph
node metastases in patients with early-stage
breast cancer. Ann Surg Oncol 2003;10(6):
622–7.
[48] Eubank WB, Mankoff DA, Takasugi J, et al.
18fluorodeoxyglucose positron emission tomog-
raphy to detect mediastinal or internal mammary
metastases in breast cancer. J Clin Oncol 2001;
19(15):3516–23.
[49] Murthy K, Aznar M, Thompson CJ, et al. Results
of preliminary clinical trials of the positron
emission mammography system PEM-I: a dedi-
cated breast imaging system producing glucose
metabolic images using FDG. J Nucl Med 2000;
41(11):1851–8.
[50] Bergman AM, Thompson CJ, Murthy K, et al.
Technique to obtain positron emission mam-
mography images in registration with x-ray
mammograms. Med Phys 1998;25(11):2119–29.
[51] Weinberg I, Majewski S, Weisenberger A, et al.
Preliminary results for positron emission mam-
mography: real-time functional breast imaging in
a conventional mammography gantry. Eur J Nucl
Med 1996;23(7):804–6.
[52] Weinberg IN, Beylin D, Zavarzin V, et al. Posi-
tron emission mammography: high-resolution
biochemical breast imaging. Technol Cancer Res
Treat 2005;4(1):55–60.
[53] Tafra L, Cheng Z, Uddo O, et al. Pilot clinical
trial of FDG positron emission mammography
(PEM) in the surgical management of breast
cancer. Am J Surg 2005;190:628–32.

Diagnosis of Recurrent and
Metastatic Disease Using F-18
Fluorodeoxyglucose-Positron
Emission Tomography
William B. Eubank, MD
& Loco-regional recurrence
& Intrathoracic lymphatic recurrences
& Distant metastases
& Impact of F-18 fluorodeoxyglucose-
positron emission tomography on
patient management
Breast cancer is the most common non-skin can-
cer, and the second leading cause of cancer death in
women [1]. There are 40,000 women per year
dying of breast cancer in the United States, and
most breast cancer victims die of progressive meta-
static disease [1]. Because optimal treatment of pa-
tients who have recurrent breast cancer depends on
knowing the true extent of disease, accurate staging
of these patients is an important public health
problem. This is a useful application of FDG-PET
when it is performed to complement conventional
imaging (CI) such as CT, MRI, and bone scintigra-
phy. The additional metabolic information pro-
vided by FDG-PET increases the accuracy of detecting
recurrent or metastatic lesions [2–12]. This is par-
ticularly true in the evaluation of anatomic regions
that have been previously treated by surgery or
radiation [13], in which the discrimination be-
tween post-treatment scar and recurrent tumor
can be problematic. FDG-PET can also significantly
impact the choice of treatment, especially in pa-
This work was supported in part by National Institutes of Health (NIH) grants RO1CA42045, RO1CA72064,
RO1CA90771, and S10RR177229.
Department of Radiology, Puget Sound Veterans Administration Health Care System, 1660 South Columbian
Way, Seattle, WA 98108-1597, USA
E-mail address: weubank@u.washington.edu
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
15
POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 15–24
& Summary
& Acknowledgments
& References
tients who have more advanced or recurrent disease
[14,15].
The recognition that breast cancer is a systemic
disease, even in its early stages, led to the cur-
rent approach to treatment, which combines local
measures such as surgery and radiotherapy with
systemic treatment [16]. For most clinical trial stud-
ies, local failure is defined as any recurrence of
tumor in the ipsilateral chest wall or mastectomy
scar; regional failure is defined as any recurrence of
tumor in the ipsilateral supraclavicular, infracla-
vicular, axillary, or internal mammary nodes; and
recurrence of tumor in any other site is considered
as distant failure [17]. In general, systemic therapy
is used at almost all disease stages; however, iso-
lated loco-regional disease or single sites of meta-
static recurrence are also treated with surgery and
radiation therapy [18,19]. It is hoped that the
potential of FDG-PET to provide more accurate
and earlier detection of breast cancer recurrences
will translate into more effective treatment strate-
doi:10.1016/j.cpet.2005.09.002
16 Eubank

gies and better health outcomes for these patients
in the future.
Loco-regional recurrence
Recurrence in the breast, skin of the breast, axillary
nodes, chest wall, and supraclavicular nodes are the
most common sites of first loco-regional recurrence
after primary surgical resection [20,21]. The shift
toward breast-conserving surgery and local radia-
tion therapy for early breast cancer in recent years
has heightened concern over loco-regional recur-
rence [22]. The incidence of loco-regional recur-
rence after breast conservation treatment ranges
from 5% to 22% [23]. Independent risk factors asso-
ciated with loco-regional recurrence in this group of
patients include positive margins at surgical resec-
tion, tumors with extensive intraductal component,
high grade ductal carcinoma in situ (DCIS), patient
age under 40 years, and absence of radiation after
breast conservation therapy [23,24].
Among patients treated with mastectomy, axillary
node dissection, and adjuvant chemotherapy, the
most common sites of loco-regional recurrence are
the chest wall (68% of loco-regional recurrences)
and supraclavicular nodes (41% of loco-regional
recurrences) [Fig. 1] [25]. Recurrent disease at
both of these sites is associated with poor progno-
sis in terms of survival after recurrence [26–28].
Factors that predict an increased risk of chest wall
or supraclavicular node recurrence include four
or more positive axillary nodes, tumor size 4 cm
or larger, and extranodal extension of 2 mm or
more [25]. Supraclavicular node recurrence is tech-
nically considered Stage IV disease, and is generally
considered a harbinger to more widely dissemi-
nated disease; however, patients who have supra-
clavicular node involvement as the sole site of
disseminated disease may benefit from aggressive
local radiotherapy.
One clinical situation where FDG-PET has been
shown to be helpful is in the evaluation of pre-
viously treated patients who have symptoms of
brachial plexopathy. This debilitating condition
can be secondary to tumor recurrence in the axilla
or chest wall, or caused by scarring of tissue neigh-
boring the brachial plexus from previous surgery
or radiation. Because the signs and symptoms of
loco-regional recurrence often overlap with the side
effects of treatment [29,30] and patients who have
tumor recurrence may benefit from surgical resec-
tion [31,32], it is important to distinguish one from
the other. Hathaway and colleagues [33] showed
the value of combining the functional information
of FDG-PET and the anatomic information from
dedicated MRI to decide whether patients would
benefit from further surgery. Other studies [34]
have confirmed these early findings.
Intrathoracic lymphatic recurrences
Lymphatic drainage to the internal mammary
nodal chain is an important pathway of spread of
disease, both at the time of initial diagnosis and

Fig. 1. A 49-year-old woman who underwent bilateral mastectomies and reconstruction with implants 7 years
before recurrence of disease. She originally had extensive lobular carcinoma in situ with two small (2–3 mm) foci
of invasive lobular carcinoma in the right breast. Sentinel node and limited low axillary lymph node dissection
revealed no pathologically involved nodes, and surgical margins were negative. She received no adjuvant
treatment, but opted for prophylactic left mastectomy. At time of recurrence, she presented with palpable left
chest wall mass (biopsy-proven invasive ductal carcinoma) and axillary nodes. Coronal FDG-PET image (A) shows
recurrent chest wall mass (arrowhead; standardized uptake value [SUV] = 8.2) lateral to breast implant, uptake in
several axillary nodes (small arrows; maximum SUV = 4.5), and left supraclavicular node (long arrow; SUV = 2.1). At
more anterior level, coronal FDG-PET image (B) shows uptake in a left internal mammary node (long arrow;
SUV = 2.0) in addition to uptake in left axillary nodes (short arrows) and the left breast (arrowhead).
 FDG-PET Diagnosis 17

after primary treatment of breast cancer. Data from
sentinel node lymphoscintigraphy series in patients
who had early breast cancer at the author’s institu-
tion reveal that overall prevalence of drainage to
the internal mammary nodes is 17% [35]. This is a
similar prevalence to that shown in early extended
radical mastectomy series [36–38], in which metas-
tasis to internal mammary nodes occurred in close
to one in five women who had operable (Stage II–-
III) breast cancer. Metastasis to internal mammary
nodes can occur from tumor located anywhere
in the breast; however, in the series of the author
and colleagues, internal mammary drainage was
significantly less frequent in tumors located in the
upper outer quadrant (10%) compared with the
other three quadrants and subareolar portion of
the breast (17%–29%) [35]. Metastasis to the in-
ternal mammary and axillary nodes usually occurs
synchronously but infrequently (4%–6% incidence)
may be isolated to the internal mammary chain
[36,39]. The prognosis of patients who have in-
ternal mammary and axillary nodal metastasis is
significantly worse compared with patients who
have only axillary node disease [40,41], suggesting
that internal mammary nodal chain is a conduit for
more widespread dissemination of disease.
The importance of internal mammary nodal de-
tection and treatment remains controversial [42].
Unlike axillary nodes, internal mammary nodes are
not routinely biopsied as part of an individual
patient’s staging work-up, and their status is gener-
ally unknown. There has been reluctance to biopsy
internal mammary nodes because: (1) early radio-
therapy trials (before the era of routine adjuvant
chemotherapy) failed to demonstrate a clear bene-
fit in survival with internal mammary chain radia-
tion, and (2) there is a relatively high complication
risk (pneumothorax and bleeding) associated with
internal mammary nodal sampling. A recent large,
prospective, randomized, radiotherapy trial [43]
has shown a benefit in systemic relapse-free and
overall survival from aggressive regional nodal irra-

Fig. 2. A 46-year-old woman who presented with sternal pain 5 years after undergoing left mastectomy for
invasive ductal carcinoma (T3, N0) followed by chemotherapy and radiation. Anterior coronal FDG-PET image (A)
shows intense uptake in the left parasternal region (long arrow) and sternum (short arrows; maximum SUV = 9.4)
consistent with disease in the internal mammary node chain with local spread to the sternum. Physiologic uptake
in the anterior heart (arrowhead) is also present. A posterior coronal FDG-PET image of the chest (B) shows linear
uptake along the supradiaphragmatic region (arrow; SUV = 4.1) consistent with pleural metastasis and axial
FDG-PET (C ) and CT (D) of the upper chest show a small lung metastasis (arrows; SUV = 2.0) in addition to sternal
uptake (arrowheads).
18 Eubank

diation (including internal mammary field) fol-
lowing lumpectomy or mastectomy, even in pa-
tients who had fairly limited spread to the axilla.
These data suggest that eradication of residual loco-
regional metastasis (including internal mammary
nodal disease) has a strong systemic effect.
FDG uptake in the internal mammary nodal chain
has been anecdotally reported in some of the stud-
ies that have focused on detection of primary
tumor or axillary staging [44,45]. In one study of
85 patients who underwent FDG-PET before axil-
lary node dissection [44], 12 (14%) had uptake
in the internal mammary region, but there was
no histological confirmation of these nodes. The
author and colleagues’ experience with imaging
patients who have locally advanced breast cancer
shows that the prevalence of internal mammary
FDG uptake can be as high as 25%, and that the
presence of internal mammary FDG uptake pre-
dicts treatment failure patterns of disease consis-
tent with internal mammary nodal involvement
and progression [Fig. 2] [46]. A preliminary study
by Bernstein and coworkers [47] showed the fea-
sibility of detecting internal mammary nodal me-
tastases in early-stage patients using FDG-PET.
FDG-PET may prove to be an ideal method of non-
invasively staging this important nodal region, and
may aid in the selection of patients who would
potentially benefit most from directed internal mam-
mary nodal radiotherapy; however, further work
needs to be done to confirm FDG-PET findings
with histopathology.
Neoplastic spread to mediastinal nodes is also
common in patients who have advanced disease,
and the mediastinum is a common site of recurrence
in patients who have undergone axillary node dis-
section and radiation [Fig. 3]. As with internal
mammary nodes, mediastinal nodes are rarely
sampled in breast cancer patients. CT, the conven-
tional method of staging these nodes, relies on
size criteria to determine the presence or absence
of disease. This method has been proven signifi-
cantly less accurate than FDG-PET in patients who
have non-small–cell lung cancer in which histologic
analysis is used as the gold standard [48,49]. In the
author and colleagues’ retrospective series of 73 pa-
tients who had recurrent or metastatic breast cancer
and who underwent both FDG-PET and chest CT
[50], FDG uptake in mediastinal or internal mam-
mary nodes was two times more prevalent than
suspiciously enlarged nodes by CT, suggesting that
PET is a much more sensitive technique at detecting
nodal disease. In the subset of patients who had
confirmation, the sensitivity of FDG-PET was sig-
nificantly higher (85%) than CT (50%), with nearly
the same level of specificity (90% for PET and
83% for CT). Ten of 33 (30%) patients suspected
of having only loco-regional recurrence by CI and
clinical examination had mediastinal or internal
mammary FDG uptake. Risk factors associated
with mediastinal or internal mammary FDG uptake
in these patients were recurrent chest wall invasion
and three or more positive axillary nodes.
Distant metastases
The skeleton is the most common site of distant
metastasis in breast cancer; nearly 70% of patients
who have advanced disease have bone metastasis
[51]. Bone scintigraphy is considered the most sen-
sitive method of detecting and determining the
extent of skeletal metastases; however, purely osteo-

Fig. 3. A 65-year-old man with history of multiple chest wall recurrences of estrogen receptor-positive invasive
ductal carcinoma of the right breast. He had undergone multiple chest wall resections, radiation, chemotherapy,
and hormonal therapy for these episodes of recurrent disease. Anterior coronal image from FDG-PET (A)
performed 11 years after original diagnosis of breast cancer shows a focus of uptake in the right anterior chest
(arrow; SUV = 6.0) consistent with pathologic involvement of an internal mammary node. A more posterior
coronal image (B) shows two foci of uptake (arrows) in the right hilum (SUV = 6.4) and the pretracheal region
of the mediastinum (SUV = 5.0).
 FDG-PET Diagnosis 19

Fig. 4. A 43-year-old woman with bone-dominant metastatic breast cancer. The posterior and anterior projections
of bone scan (A) show widespread foci of uptake in the axial skeleton and ribs consistent with metastatic disease.
Anterior (B) and posterior (C ) coronal images from FDG-PET performed 1 week after the bone scan show much
more extensive involvement of the axial skeleton, with a predominant intramedullary uptake pattern in the spine.
The pattern of uptake in the ribs is also different compared with the bone scan. The bone scan shows discrete foci
in multiple ribs bilaterally that are not as apparent on FDG-PET; these sites represent areas of active cortical bone
remodeling from either sclerotic metastases or pathologic fractures. The rib activity in the FDG-PET scan is more
diffuse and less discrete, consistent with intramedullary metastases.

lytic lesions or metastases confined to the marrow
cavity may be difficult to detect on bone scintigra-
phy, because of a lack of sufficient osteoblastic re-
sponse [52]. Retrospective studies comparing the
sensitivity of bone scintigraphy to FDG-PET in the
detection of skeletal metastases in patients who
have advanced disease have shown conflicting
results [3,4,9,53–56]. Some studies have shown
FDG-PET to be equal or superior to planar bone
scintigraphy in the detection of skeletal metastases
20 Eubank

[53,54], whereas others have shown FDG-PET to be
less sensitive [3,4,9,55] on a lesion-based analysis.
The results from a study by Cook and colleagues
[56], in which skeletal metastases from breast can-
cer patients were categorized into osteoblastic,
osteolytic, or mixed disease subsets based on plain
film or CT findings, provides an explanation for
these conflicting reports. In the study of 23 breast
cancer patients who had known skeletal metasta-
ses and who underwent both bone scintigraphy
and FDG-PET, FDG-PET detected more lesions
than bone scintigraphy, except in a subgroup of
patients who had osteoblastic metastases. More-
over, the level of FDG uptake in osteolytic lesions
was significantly greater compared with osteoblas-
tic lesions, and the prognosis of patients who had
osteolytic-predominant disease was significantly
worse. A more recent prospective study compar-
ing the sensitivity of FDG-PET with bone single
photon emission computed tomography (SPECT)
in 15 patients [57] showed a clear superiority of
bone SPECT over FDG-PET in sensitivity in detect-
ing skeletal metastases (85% for SPECT versus
17% for FDG), and corroborated the findings of
Cook and coworkers regarding the improvement in
detection of the subset of osteolytic lesions by
FDG-PET. These data clearly show a complemen-
tary nature of bone scintigraphy and FDG-PET in
the evaluation of skeletal metastases in breast can-
cer patients [Fig. 4]. These results also suggest that
FDG-PET and bone scan should not be considered
substitutes for each other for bone metastasis stag-
ing in breast cancer. In the author’s center, bone
scintigraphy remains one of the routine studies in
breast cancer metastatic staging, with FDG-PET to
help clarify staging in the case of difficult or equivo-
cal conventional staging. Evolving data suggest that
[F-18]-fluoride PET may provide similar and likely
improved bone metastasis detection in breast can-
cer and other tumors compared with bone scin-
tigraphy [58,59], and may play a role in breast
cancer bone metastasis staging in the future.
Several studies have shown that whole-body
FDG-PET is a highly accurate method compared
with CI for restaging patients who have previously
undergone primary treatment for breast cancer. Re-
sults of these studies are summarized in Table 1.
One advantage of FDG-PET is that large areas can
be surveyed; it is therefore able to evaluate sites of
recurrent disease that can be extensive and sepa-
rated by large anatomic distances. Several investiga-
tions have shown the added benefit of FDG-PET to
CI in asymptomatic patients who had elevated
tumor marker serum levels and negative or equivo-
cal CI [4,7,8]. Although the ability of FDG-PET to
successfully detect skeletal metastases compared
with bone scintigraphy has been mixed in these
series, as previously discussed, FDG-PET has been
proven significantly more accurate in the evalua-
tion of nodal disease, and equal or superior to CI
for visceral metastases [3,9,50]. In a retrospective
study of 61 patients, Vranjesevic and colleagues
[60] showed that FDG-PET is significantly more
accurate than a combination of CI studies in pre-
dicting the outcome (disease-free survival) of pa-
tients being re-evaluated after primary treatment of
breast cancer. These data support the practice at the
author’s institution of using FDG-PET as a comple-
ment to CI for the evaluation of patients who have
suspected recurrent disease.

Table 1: Studies comparing the ability of whole-body FDG-PET with conventional imaging to detect
loco-regional and distant recurrences in patients who have previously undergone primary treatment
for breast cancer
Number of Confirmed positive/ FDG-PET Sensitivity FDG-PET Specificity
Study patients negative cases (TP/TP+FN)a (TN/TN+FP)a
Bender et al [2] 75 60/15 95% (41/43) 96% (213/221)
Moon et al [3] 57 29/28 93% (27/29) 79% (22/28)
Lonneux et al [4] 39b 33/6 94% (31/33) 50% (3/6)
Kim et al [5] 27 17/10 94% (16/17) 80% (8/10)
Lin et al [6] 36 11/25 85% (23/27) 96% (85/89)
Liu et al [7] 30b 28/2 89% (25/28) 50% (1/2)
Suarez et al [8] 38b 26/12 92% (24/26) 75% (9/12)
Gallowitsch et al [9] 62 34/28 97% (33/34) 82% (23/28)
Siggelkow et al [10] 57 31/26 81% (25/31) 96% (25/26)
Kamel et al [11] 60 43/17 89% (24/27) LRR 84% (16/19) LRR
100% (26/26) DM 97% (30/31) DM

Abbreviations: DM, distant metastases; FN, false negative; FP, false positive; LRR, loco-regional recurrence; TN, true
negative; TP, true positive.
a
Values calculated on patient analysis except in [2] and [6], in which values are calculated on lesion analysis.
b
Patients were mostly or all asymptomatic with elevated tumor markers.

Impact of F-18 fluorodeoxyglucose-positron
emission tomography on patient
management
A distinguishing feature of the management of
breast cancer is the need for a multidisciplinary
approach for optimal local and regional control
of the disease. Many strategies and agents offer
patients who have advanced breast cancer a thera-
peutic benefit, including surgery, radiation, che-
motherapy, and hormonal therapy. Choosing the
most appropriate therapy depends primarily on ac-
curately defining the extent of disease. As of October,
2002, PET has been approved for payment by the
Center for Medicare and Medicaid Services in the
United States for staging or restaging of patients
who have recurrent or metastatic disease, especially
when conventional staging studies are equivocal.
Despite this approval, there is relatively little in-
sight into which patients who have recurrent or
metastatic breast cancer benefit the most from
FDG-PET. A previous investigation [14] showed
that FDG-PET had a major impact on the manage-
ment of patients who had breast cancer and who
underwent restaging for local or distant recurrences.
FDG-PET changed the clinical stage in 36% of pa-
tients and induced changes in therapy in 58% of
the patients. In the author and colleagues’ retro-
spective study of 125 patients who had advanced
breast cancer and who were undergoing CI and
FDG-PET for staging [15], the extent of disease
was changed in 67% of patients (increased in
43% and decreased in 24%), and the therapeutic
plan was altered in 32% of patients based on FDG-
PET findings. FDG-PET altered therapy most fre-
quently in two subgroups of patients: (1) patients
who had suspected or proven loco-regional re-
currence, under consideration for aggressive local
therapy (FDG-PET altered treatment in 44%), and
(2) patients who had known metastases being
evaluated for response to therapy (FDG-PET altered
treatment in 33%). These results indicate that FDG-
PET should not be used as the sole restaging tool
in patients who have recurrent or metastatic dis-
ease, but should be used to answer specific ques-
tions that will likely impact their management.
Future prospective trials using oncologist-directed
questionnaires will help to further define the role
and provide data for the cost-benefit analysis of
FDG-PET in staging patients who have advanced
breast cancer.
There have been a few preliminary studies show-
ing promise for FDG-PET/CT in the evaluation of
patients who have recurrent or metastatic breast
cancer [61–63]. Because of the accurate registration
of metabolic and morphologic images from these
integrated systems, the accuracy in localizing FDG
FDG-PET Diagnosis 21
uptake is improved; this can potentially lead to an
improvement in specificity (reduction in false posi-
tive results) and staging of disease compared with
separate evaluation of CT and PET data. PET/CT
also shows promise in radiation treatment plan-
ning by modifying target tissue volumes selected
for irradiation, based on the combination of mor-
phologic and metabolic data [64]. The impact of
PET/CT on patient outcomes needs to be addressed
in future studies.
Summary
FDG-PET is a useful tool in the evaluation of pa-
tients who have suspected or proven breast cancer
recurrence. In the clinical setting, it should be used
as a complement to conventional staging studies
and not as a replacement. FDG-PET is generally
more sensitive in the detection of loco-regional
and distant metastases than conventional imaging,
and is therefore a more accurate method of deter-
mining the true extent of disease. One exception
is the detection of sclerotic bone metastases; these
lesions are often not metabolically active enough
for FDG-PET detection, but are readily detected
by bone scintigraphy. To have the greatest impact
on patient management, FDG-PET should be used
to answer specific clinical questions. In the clinical
setting of an asymptomatic patient who has ris-
ing tumor markers and negative or equivocal find-
ings on clinical examination or CI, FDG-PET has
the ability to detect otherwise occult disease. Pre-
liminary investigations show that FDG-PET has
the greatest impact on the choice of treatment in pa-
tients who have suspected or proven loco-regional
recurrence and who are being considered for ag-
gressive curative treatment, and in the evaluation
of treatment response in patients who have meta-
static disease.
Acknowledgments
The authors wish to acknowledge Erin Schubert for
help with figure reproduction.
References
[1] Greenlee RT, Murray T, Bolden S, et al. Cancer Sta-
tistics, 2000. CA Cancer J Clin 2000;50(1):7–33.
[2] Bender H, Kirst J, Palmedo H, et al. Value of
18
fluoro-deoxyglucose positron emission tomog-
raphy in the staging of recurrent breast carci-
noma. Anticancer Res 1997;17(3B):1687–92.
[3] Moon DH, Maddahi J, Silverman DHS, et al.
Accuracy of whole-body [fluorine-18]-FDG-PET
for the detection of recurrent or metastatic breast
carcinoma. J Nucl Med 1998;39(3):431–5.
22 Eubank
[4] Lonneux M, Borbath I, Berliere M, et al. The place
of whole-body FDG-PET for the diagnosis of
distant recurrence of breast cancer. Clin Positron
Imaging 2000;3(2):45–9.
[5] Kim TS, Moon WK, Lee DS, et al. Fluorodeoxy-
glucose positron emission tomography for de-
tection of recurrent or metastatic breast cancer.
World J Surg 2001;25(7):829–34.
[6] Lin WY, Tsai SC, Cheng KY, et al. Fluorine-18
FDG-PET in detecting local recurrence and dis-
tant metastases in breast cancer–Taiwanese ex-
periences. Cancer Invest 2002;20(5–6):725–9.
[7] Liu CS, Shen YY, Lin CC, et al. Clinical impact
of [18F]FDG-PET in patients with suspected
recurrent breast cancer based on asymptomati-
cally elevated tumor marker serum levels: a pre-
liminary report. Jpn J Clin Oncol 2002;32(7):
244–7.
[8] Suarez M, Perez-Castejon MJ, Jimenez A, et al.
Early diagnosis of recurrent breast cancer with
FDG-PET in patients with progressive elevation
of serum tumor markers. Q J Nucl Med 2002;
46(2):113–21.
[9] Gallowitsch HJ, Kresnik E, Gasser J, et al. [18F]-
fluorodeoxyglucose positron-emission tomogra-
phy in the diagnosis of tumor recurrence and
metastases in the follow-up of patients with breast
carcinoma: a comparison to conventional imag-
ing. Invest Radiol 2003;38(5):250–6.
[10] Siggelkow W, Zimny M, Faridi A, et al. The value
of positron emission tomography in the fol-
low-up for breast cancer. Anticancer Res 2003;
23(2C):1859–68.
[11] Kamel EM, Wyss MT, Fehr MK, et al. [18F]-
fluorodeoxyglucose positron emission tomogra-
phy in patients with suspected recurrence of breast
cancer. J Cancer Res Clin Oncol 2003;129(3):
147–53.
[12] Eubank WB, Mankoff DA, Vesselle HJ, et al.
Detection of locoregional and distant recurrences
in breast cancer patients by using FDG-PET.
Radiographics 2002;22(1):5–17.
[13] Eubank WB, Mankoff DA, Schmiedl UP, et al.
Imaging of oncologic patients: benefit of com-
bined CT and [F-18]-fluorodeoxyglucose posi-
tron emission tomogoraphy scan interpretation
in the diagnosis of malignancy. Am J Roentgenol
1998;171(4):1103–10.
[14] Yap CS, Seltzer MA, Schiepers C, et al. Impact of
whole-body 18FDG-PET on staging and manag-
ing patients with breast cancer: the referring
physician’s perspective. J Nucl Med 2001;42(9):
1334–7.
[15] Eubank WB, Mankoff DA, Bhattacharya M, et al.
Impact of [F-18]-Fluorodeoxyglucose PET on
defining the extent of disease and management
of patients with recurrent or metastatic breast
cancer. Am J Roentgenol 2004;83(2):479–86.
[16] Hortobagyi GN. Developments in chemotherapy
of breast cancer. Cancer 2000;88(Suppl 12):3073–9.
[17] Taghian A, Jeong J, Mamounas E, et al. Patterns
of locoregional failure in patients with operable
breast cancer treated by mastectomy and adju-
vant chemotherapy with or without tamoxifen
and without radiotherapy: results from five
National Surgical Adjuvant Breast and Bowel
Project randomized clinical trials. J Clin Oncol
2004;22(21):4247–54.
[18] Schwaibold F, Fowble BL, Solin LJ, et al. The
results of radiation therapy for isolated local
regional recurrence after mastectomy. Int J Radiat
Oncol Biol Phys 1991;21(2):299–310.
[19] Probstfeld MR, O’Connell TX. Treatment of
locally recurrent breast carcinoma. Arch Surg
1989;124(10):1127–9 [discussion: 1130].
[20] Valagussa P, Bonadonna G, Veronesi U. Patterns
of relapse and survival following radical mas-
tectomy. Analysis of 716 consecutive patients.
Cancer 1978;41(3):1170–8.
[21] Kamby C, Rose C, Ejlertsen B, et al. Stage and
pattern of metastases in patients with breast
cancer. Eur J Cancer Clin Oncol 1987;23(12):
1925–34.
[22] Schmolling J, Maus B, Rezek D, et al. Breast
preservation versus mastectomy—recurrence and
survival rates of primary breast cancer patients
treated at the UFK Bonn. Eur J Gynaecol Oncol
1997;18(1):29–33.
[23] Huston TL, Simmons RM. Locally recurrent
breast cancer after conservation therapy. Am J
Surg 2005;189(2):229–35.
[24] Leborgne F, Leborgne JH, Ortega B, et al. Breast
conservation treatment of early stage breast can-
cer: patterns of failure. Int J Radiat Oncol Biol
Phys 1995;31(4):765–75.
[25] Katz A, Strom EA, Buchholz TA, et al. Locore-
gional recurrence patterns after mastectomy and
doxorubicin-based chemotherapy: implications
for postoperative irradiation. J Clin Oncol 2000;
18(15):2817–27.
[26] Aristei C, Marsella AR, Chionne F, et al. Regional
node failure in patients with four or more
positive lymph nodes submitted to conservative
surgery followed by radiotherapy to the breast.
Am J Clin Oncol 2000;23(3):217–21.
[27] Nikkanen TA. Recurrence of breast cancer. A re-
trospective study of 569 cases in clinical Stages I–
III. Acta Chir Scand 1981;147(4):239–45.
[28] Kiricuta IC, Willner J, Kolbl O, et al. The prog-
nostic significance of the supraclavicular lymph
node metastases in breast cancer patients. Int J
Radiat Oncol Biol Phys 1994;28(2):387–93.
[29] Iyer RB, Fenstermacher MJ, Libshitz HI. MR
imaging of the treated brachial plexus. Am J
Roentgenol 1996;167(1):225–9.
[30] Bagley FH, Walsh JW, Cady B, et al. Carcinoma-
tous versus radiation-induced brachial plexus
neuropathy in breast cancer. Cancer 1978;41(6):
2154–7.
[31] Toi M, Tanaka S, Bando M, et al. Outcome of
surgical resection for chest wall recurrence in
breast cancer patients. J Surg Oncol 1997;37(1):
853–63.
[32] Hathaway CL, Rand RP, Moe R, et al. Salvage

surgery for locally advanced and locally recurrent
breast cancer. Arch Surg 1994;129(6):582–7.
[33] Hathaway PB, Mankoff DA, Maravilla KR, et al.
The value of combined FDG-PET and magnetic
resonance imaging in the evaluation of suspected
recurrent local-regional breast cancer: prelimi-
nary experience. Radiology 1998;210(3):807–14.
[34] Ahmad A, Barrington S, Maisey M, et al. Use of
positron emission tomography in evaluation of
brachial plexopathy in breast cancer patients. Br J
Cancer 1999;79(3–4):478–82.
[35] Byrd DR, Dunnwald LK, Mankoff DA, et al.
Internal mammary lymph node drainage pat-
terns with breast cancer documented by breast
lymphoscintigraphy. J Surg Oncol 2001;8(3):
234–40.
[36] Donegan WL. The influence of untreated internal
mammary metastases upon the course of mam-
mary cancer. Cancer 1977;39(2):533–8.
[37] Veronesi U, Cascinelli N, Greco M, et al. Prog-
nosis of breast cancer patients after mastectomy
and dissection of internal mammary nodes. Ann
Surg 1985;202(6):702–7.
[38] Lacour J, Le M, Caceres E, et al. Radical mas-
tectomy versus radical mastectomy plus internal
mammary dissection. Ten year results of an in-
ternational cooperative trial in breast cancer.
Cancer 1983;51(10):1941–3.
[39] Noguchi M, Ohta N, Thomas M, et al. Risk of
internal mammary lymph node metastases and
its prognostic value in breast cancer patients.
J Surg Oncol 1993;52(1):26–30.
[40] Veronesi U, Marubini E, Del Vecchio M, et al.
Local recurrences and distant metastases after
conservative breast cancer treatments: partly
independent events. J Natl Cancer Inst 1995;
87(1):19–27.
[41] Cody 3rd HS, Urban JA. Internal mammary node
status: a major prognosticator in axillary node-
negative breast cancer. Ann Surg Oncol 1995;
2(1):32–7.
[42] Freedman GM, Fowble BL, Nicolaou N, et al.
Should internal mammary lymph nodes in
breast cancer be a target for the radiation on-
cologist? Int J Radiat Oncol Biol Phys 2000;46(4):
805–14.
[43] Ragaz J, Olivotto IA, Spinelli JJ, et al. Locoregional
radiation therapy in patients with high-risk
breast cancer receiving adjuvant chemother-
apy: 20-year results of the British Columbia ran-
domized trial. J Natl Cancer Inst 2005;97(2):
116–25.
[44] Schirrmeister H, Kuhn T, Guhlmann A, et al. [F-
18]-2-deoxy-2-fluoro-D-glucose PET in the pre-
operative staging of breast cancer: comparison
with the standard staging procedures. Eur J Nucl
Med 2001;28(3):351–8.
[45] Greco M, Crippa F, Agresti R, et al. Axillary lymph
node staging in breast cancer by 2-fluoro-2-deoxy-
D-glucose-positron emission tomography: clinical
evaluation and alternative management. J Natl
Cancer Inst 2001;93(8):630–5.
FDG-PET Diagnosis 23
[46] Bellon JR, Livingston RB, Eubank WB, et al.
Evaluation of the internal mammary (IM) lymph
nodes by FDG-PET in locally advanced breast
cancer (LABC). Am J Clin Oncol 2004;27(4):
407–10.
[47] Bernstein V, Jones A, Mankoff DA, et al. Assess-
ment of internal mammary lymph nodes by
fluorodeoxyglucose positron emission (FDG-
PET) in medial hemisphere breast cancer [ab-
stract]. J Nucl Med 2000;41:289P.
[48] Vansteenkiste JF, Stroobants SG, De Leyn PR,
et al. Lymph node staging in non-small–cell lung
cancer with FDG-PET scan: a prospective study
on 690 lymph node stations from 68 patients.
J Clin Oncol 1998;16(6):2142–9.
[49] Scott WJ, Gobar LS, Terry JD, et al. Mediastinal
lymph node staging of non-small–cell lung
cancer: a prospective comparison of computed
tomography and positron emission tomography.
J Thorac Cardiovasc Surg 1996;111(3):642–8.
[50] Eubank WB, Mankoff DA, Takasugi J, et al.
18
Fluorodeoxyglucose positron emission tomog-
raphy to detect mediastinal or internal mammary
metastases in breast cancer. J Clin Oncol 2001;
19(15):3516–23.
[51] Coleman RE, Rubens RD. The clinical course of
bone metastases from breast cancer. Br J Cancer
1987;55(1):61–6.
[52] Nielsen OS, Munro AJ, Tannock IF. Bone me-
tastases: pathophysiology and management pol-
icy. J Clin Oncol 1991;9(3):509–24.
[53] Ohta M, Tokuda Y, Suzuki Y, et al. Whole body
PET for the evaluation of bony metastases in
patients with breast cancer: comparison with
99Tcm-MDP bone scintigraphy. Nucl Med Com-
mun 2001;22(8):875–9.
[54] Yang SN, Liang JA, Lin FJ, et al. Comparing
whole body [18F]-2-deoxyglucose positron emis-
sion tomography and technetium-99m methy-
lene diphosphonate bone scan to detect bone
metastases in patients with breast cancer. J Can-
cer Res Clin Oncol 2002;128(6):325–8.
[55] Kao CH, Hsieh JF, Tsai SC, et al. Comparison and
discrepancy of [18F]-2-deoxyglucose positron
emission tomography and Tc-99m MDP bone
scan to detect bone metastases. Anticancer Res
2000;20(3B):2189–92.
[56] Cook GJ, Houston S, Rubens R, et al. Detection
of bone metastases in breast cancer by 18FDG-
PET: differing metabolic activity in osteoblastic
and osteolytic lesions. J Clin Oncol 1998;16(10):
3375–9.
[57] Uematsu T, Yuen S, Yukisawa S, et al. Compari-
son of FDG-PET and SPECT for detection of bone
metastases in breast cancer. Am J Roentgenol
2005;84(4):1266–73.
[58] Schirrmeister H, Guhlmann A, Kotzerke J, et al.
Early detection and accurate description of extent
of metastatic bone disease in breast cancer with
fluoride ion and positron emission tomography.
J Clin Oncol 1999;17(8):2381–9.
[59] Schirrmeister H, Glatting G, Hetzel J, et al.
24 Eubank
Prospective evaluation of the clinical value of
planar bone scans, SPECT, and [18]F-labeled
NaF PET in newly diagnosed lung cancer. J Nucl
Med 2001;42(12):1800–4.
[60] Vranjesevic D, Filmont JE, Meta J, et al. Whole-
body 18F-FDG-PET and conventional imaging for
predicting outcome in previously treated breast
cancer patients. J Nucl Med 2002;43(3):325–9.
[61] Pelosi E, Messa C, Sironi S, et al. Value of
integrated PET/CT for lesion localisation in can-
cer patients: a comparative study. Eur J Nucl Med
Mol Imaging 2004;31(7):932–9.
[62] Buck AK, Wahl A, Eicher U, et al. Combined
morphological and functional imaging with
FDG-PET/CT for restaging breast cancer—impact
on patient management [abstract]. J Nucl Med
2003;44:78P.
[63] Tatsumi M, Cohade C, Mourtzikos KA, et al.
Initial experience with FDG-PET-CT in the
evaluation of breast cancer [abstract]. J Nucl
Med 2003;44:394P.
[64] Ciernik F, Dizendorf E, Baumert BG, et al.
Radiation treatment planning with an integrated
positron emission and computer tomography
(PET/CT): A feasibility study. Int J Radiat Oncol
Biol Phys 2003;57(3):853–63.

Detection of Bone Metastases in
Breast Cancer by Positron
Emission Tomography
Holger Schirrmeister, MD
& F-18 fluoride positron emission
tomography
& F-18 fluorodeoxyglucose
& F-18 fluoride positron emission tomography
versus F-18 fluorodeoxyglucose-positron
emission tomography: which radiotracer
should be used in breast cancer?
Carcinoma of the breast is the most prevalent
cancer in women in the United States and Western
Europe, and is commonly associated with meta-
static bone disease. The incidence of breast cancer
has been increasing in Western Europe and the
United States, but the number of deaths resulting
from breast cancer has remained relatively stable
[1]. This is probably related to more effective treat-
ment strategies and to the detection of disease at
early stages.
Although the incidence of bone metastases is as
low as 1% to 2% at the time of primary diagnosis
[2], bone metastases are found in one third of all
patients who have recurrent disease [3]. In a study
including 587 patients who died from breast can-
cer [4], 69% had radiological evidence of skeletal
metastases before death. There are two different
types of bony metastases—osteolytic and osteoblas-
tic disease—depending on their radiographic ap-
pearance. Patients who have breast cancer can
have either osteolytic or osteoblastic metastases, or
can present with mixed lesions containing both
features [5]. Interestingly, patients who have bone
metastases only survive 2 to 3 years longer than
patients who have visceral involvement. The distri-
Clinic of Nuclear Medicine, University of Kiel, Arnold-Heller Str. 9, Kiel 24105, Germany
E-mail address: hschirrmeister@nuc-med.uni-kiel.de
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
25
POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 25–32
& Summary
& References
bution of bone metastases is a relevant prognostic
factor. Yamashita and colleagues [6,7] reported a
significantly longer survival in patients who had
bone metastases exclusively to the upper regions
of the skeleton than in patients who had metastases
to the pelvis and the lower extremities. Further-
more, presence of osteoblastic features in bony
metastases was also reported to be associated with
prolonged survival [6,7].
It has been estimated that bone metastases be-
come visible on plain radiographs after 30% to
50% of bone mineral loss [8]. Secondary formation
of bone also occurs as response to osteolytic de-
struction; hence, even osteolytic bone metastases
can be detected with bone scintigraphy several
months before they are apparent on plain radio-
graphs [9,10]. Bone scintigraphy is not perfect in
detecting bone metastases, however; compara-
tive studies with MRI revealed significantly more
vertebral metastases not detected on conventional
planar bone scintigraphy [11,12]. Nevertheless,
conventional bone scintigraphy remains the most
suitable technique for whole-body screening, be-
cause whole-body MRI is currently not practical
for routine whole-body surveys, and its sensitivity
doi:10.1016/j.cpet.2005.09.005
26 Schirrmeister

in detecting extra-vertebral metastases is not yet bone scintigraphy ranged between 80% and 90% in
sufficiently assessed. the skull, thorax, and the extremities. The most
Positron emission tomography (PET) is a rela- probable explanation for these findings is that the
tively new imaging technique for noninvasive as- detection of vertebral bone metastases was limited
sessment of several metabolic pathways. During the by superimposed tracer retention in soft tissue and
last decade, this method evolved from a research other parts of the skeleton. This drawback of con-
tool to a standard imaging modality for routine ventional gamma camera imaging is generally not
staging of several types of malignant tumors. PET present with tomographic imaging. Hence, in con-
using the glucose analog 2-(fluoro-18)-2-deoxy-D- trast to conventional bone scintigraphy, the sensi-
glucose (FDG) enables detection of primary breast tivity of F-18 fluoride PET was independent from
cancer and its metastases on the basis of increased the anatomical localization of skeletal lesions [16].
glucose metabolism in neoplastic tissues. Char- In another prospective study [17], F-18 fluoride
acterization of bone metastases is also possible PET was performed in 34 patients who had breast
with F-18 sodium fluoride, which shows ostoblastic cancer and high risk of metastatic bone disease.
bone reaction. The potential of PET with FDG F-18 fluoride PET revealed more metastases than
and F-18 fluoride in detecting bone metastases conventional bone scintigraphy in 11 of 17 patients
originating from breast cancer is discussed in who had bony metastases. Three patients were true-
this article. positive on F-18 fluoride PET, but had no signs of
metastatic bone disease on normal bone scintigra-
phy [Figs. 1, 2]. In another patient who had known
F-18 fluoride positron emission tomography metastases, an osteolytic lesion was depicted with
Skeletal uptake of the positron emitter F-18 fluo- F-18 fluoride PET that had been missed by bone
ride is based on the exchange of hydroxyl ions in
the hydroxyapatite crystal. Hence, the uptake of
F-18 fluoride is an indicator of bone mineraliza-
tion [13]. F-18 fluoride was introduced into clini-
cal practice by Blau in 1962 [14]. Subsequently,
F-18 fluoride was approved by the Food and
Drug Administration (FDA) as a radiotracer for
bone scintigraphy. One decade later, this radio-
pharmaceutical was replaced by Tc-99m-labeled
polyphosphonates, because conventional gamma
cameras performed better in terms of sensitivity
and spatial resolution of the 140 keV photons
from Tc-99m compared with the 511 keV photons
from F-18 fluoride.
In comparing the pharmacokinetics of Tc-99m
labeled polyphosphonates and F-18 fluoride, it is
important to note that the uptake of F-18 fluoride
into bone is approximately twofold higher and the
blood clearance is significantly faster compared
with technetium labeled polyphosphonates. This
leads to an increased bone-to-background ratio
for F-18 fluoride. The improved sensitivity and bet-
ter spatial resolution of modern PET scanners allow
performing highly sensitive and specific whole-
body screening for bone metastases [15–19]. In a
pilot study [16], F-18 fluoride PET revealed two
times more bone metastases than bone scintigra-
phy in patients who had osteolytic disease from
thyroid cancer or osteoblastic metastases from pros-
tate cancer. An important finding in this study
was that the sensitivity of planar bone scintigra-
phy was highly variable, depending on the skele- Fig. 1. Normal bone scintigraphy in a patient with
tal region. The sensitivity of bone scintigraphy was breast cancer (pT2,N0). Bone scintigraphy was per-
as low as 20% in the thoracic spine and 40% formed because of increasing serum alkaline phospha-
in the lumbar spine; in contrast, the sensitivity of tase and calcium levels.
 PEM Detection of Breast Cancer Bone Metastases 27

Fig. 2. F-18 fluoride PET of the same patient as presented in Fig. 1. PET imaging was obtained 2 hours after
injection of 370 MBq F-18 fluoride. Two osteolytic metastases are present in the thoracic and lumbar spine.
Maximum intensity projection images (left). Sagittal sections of the thoracic and lumbar spine (middle). Corre-
sponding sagittal CT images (right).

scintigraphy. Because of the risk for pathologic
bone fracture, this metastasis was surgically stabi-
lized. In this highly selected series, clinical manage-
ment was influenced in 4 out of 34 patients by the
results from F-18 fluoride PET.
Along with the twofold higher sensitivity in de-
tecting bone metastases, there is also a twofold
higher rate of benign lesions visible on F-18 fluo-
ride PET [16]. In a recent review article [20], it was
assumed that this would lead to an increased rate of
false-positive results. This assumption, however, is
not supported by the current literature, in which
the specificity of F-18 fluoride PET was 90% and
higher in several studies [16–19].
It is important to note that uptake of F-18 fluo-
ride is not tumor-specific, and that numerous
benign lesions are detected by F-18 fluoride PET.
Differentiation between benign and malignant
bony lesions is not possible by means of quantita-
tive analysis of bone metabolism [21]. Neverthe-
less, the superior spatial resolution of modern PET
scanners allows for exact anatomical localization of
lesions and a better differentiation between benign
[22] and malignant lesions [23] compared with
conventional gamma cameras.
In breast cancer, both osteolytic and osteoblastic
bone metastases have been reported to demon-
strate focally increased F-18 fluoride uptake [15].
In the experience of the author’s group, osteolytic
metastases were often characterized as photopenic
lesions surrounded by a rim of increased activity
[Fig. 3]. This pattern was observed exclusively
in osteolytic metastases and not in benign le-
sions [23].
Both osteolytic metastases and benign lesions
appear as regions with focally increased tracer
uptake. Therefore, knowledge of the typical appear-
ance of benign lesions becomes crucial for differ-
entiation between benign lesions and metastases.
Arthritis of the articular facets and endplate frac-
tures account for approximately 80% of all benign
abnormalities detected with F-18 fluoride PET [22].
These abnormalities have a very characteristic F-18
fluoride uptake pattern [Fig. 4]. Osteophytes are

Fig. 3. F-18 fluoride PET. Typical pattern of osteolytic (left) and osteoblastic metastases (right).
28 Schirrmeister

Fig. 4. F-18 fluoride PET. Typical pattern of benign lesions.

characterized as lesions with variable tracer uptake was approximately sevenfold higher in osteolytic
located at the superior and inferior edges of adja- bone metastases compared with osteoblastic me-
cent endplates, whereas increased F-18 fluoride tastases [Figs. 5, 6]. In the entire patient group,
uptake at adjacent surfaces of intervertebral joints FDG-PET detected more bone metastases than
is typical for intervertebral arthritis. On F-18 fluo- bone scintigraphy. Bone scintigraphy, however,
ride PET, lesions neither located at joint surfaces was more sensitive in a subgroup of patients who
nor showing the typical pattern of endplate frac- had predominantly osteoblastic disease.
tures, osteophytes, or serial rib fractures are suspi- Ohta and colleagues [25] compared the accuracy
cious for metastatic disease [23]. of conventional bone scintigraphy with FDG-PET
in 51 breast cancer patients. In this series, the sen-
sitivity and specificity of FDG-PET were 77.7% and
97.6%, respectively, compared with 77.7% and
F-18 fluorodeoxyglucose
80.3% for bone scintigraphy. A similar sensitivity
The glucose analog 2-(fluoro-18)-2-deoxy-D-glucose and an even higher specificity were found for FDG-
(FDG) is currently the most commonly used radio- PET compared with bone scintigraphy in a series of
tracer in oncologic PET imaging. This surro- 48 patients who had biopsy-proven breast cancer
gate marker for glucose metabolism enters the cell and suspected bone metastases [26]. Comparable
via the glucose transporter membrane proteins results for FDG-PET and bone scintigraphy were
Glut-1 and Glut-5. FDG is then posphorylated by also reported in a series of 24 breast cancer patients
the enzyme hexokinase. In contrast to glucose,
FDG-6-phospate is not a substrate for the glucose-
6-phophate isomerase, and is therefore trapped
within tumor cells.
Although the exact uptake mechanism of FDG in
bone metastases is not yet known, it is assumed
that FDG is taken up directly into the tumor tissue
of skeletal metastases and not into the surrounding
remodeled bone. There are only a few reports
addressing the accuracy of FDG-PET for detecting
skeletal metastases of breast cancer. This is proba-
bly related to the low incidence of bone metastases
at initial diagnosis, and the absence of a reliable
gold standard.
In a series of 23 breast cancer patients who had
metastatic bone disease, Cook and coworkers [24]
evaluated FDG-PET in osteoblastic and osteolytic Fig. 5. FDG-PET in a patient with recurrent breast
bone metastases by semiquantitative standardized cancer presenting with increased FDG uptake in a
uptake values (SUV). In this series, the FDG uptake small osteolytic metastasis (L-4).

Fig. 6. FDG-PET in a patient with recurrent breast cancer presenting with multiple osteoblastic bone metastases.
Intense increased FDG uptake is present in a liver metastasis, whereas the bone metastases show only little
FDG uptake.
by Kao and coworkers [27]. In their study, FDG-
PET was less sensitive but significantly more speci-
fic than bone scintigraphy.
One limitation of the studies described above is
that planer bone scintigraphy was compared with
FDG-PET tomographic images. In general, tomo-
graphic imaging methods are more sensitive than
planar imaging, because lesions are detectable with-
out superposition by other anatomical structures.
For example, small lung metastases are seen on CT
but frequently not on plain radiographs, and ver-
tebral bone metastases are seen on single photon
emission computed tomography (SPECT), CT, or
MRI, despite negative planar bone scintigraphy or
plain radiographs. Accordingly, Kosuda and col-
leagues [28] demonstrated a significantly improved
sensitivity for detection of bone metastases by
SPECT imaging compared with planar bone scintig-
raphy. The improved sensitivity of bone scintigra-
phy by using SPECT imaging was also found in a
mixed series of 174 patients by Han and coworkers
[29], and in another prospective series of 103
patients who had lung cancer [19].
The first study comparing bone SPECT and FDG-
PET was recently published by Uematsu and co-
workers [30]. In this study, 15 patients who had
known metastatic bone disease were included. CT,
MRI, and the subsequent clinical follow-up were
used as references. In a lesion-by-lesion analysis,
the sensitivity of SPECT was 95%, and only 17% for
FDG-PET. The specificity was not statistically sig-
nificantly different between both imaging methods
(SPECT 99%, PET 100%).
The results reported by Cook and colleagues [24]
and the studies comparing bone scintigraphy with
FDG-PET [25–27,30] indicate that FDG-PET might
not be more sensitive than bone scintigraphy but
is probably more specific. Because FDG-PET and
bone scintigraphy have a different sensitivity for
detecting oteoblastic and osteolytic bone metas-
tases, the imaging modalities should be regarded
as complementary, but cannot replace each other.
PEM Detection of Breast Cancer Bone Metastases 29
F-18 fluoride positron emission tomography
versus F-18 fluorodeoxyglucose-positron
emission tomography: which radiotracer
should be used in breast cancer?
A prospective comparison of FDG-PET and F-18
fluoride PET in detecting bone metastases from
breast cancer has not yet been published. A re-
cent study [19] evaluated the accuracy and cost-
effectiveness of F-18 fluoride PET in 103 patients
who had initial diagnosis of lung cancer. In this
study, FDG-PET was performed for whole-body
staging in 41 patients, 10 of whom had bone me-
tastases. One of these patients had disseminated
bone metastases, but was negative on FDG-PET.
Furthermore, the number of bone metastases was
underestimated in an additional 4 patients. The
results of this study are in line with the results of
the study reported by Uesuda and coworkers [30],
who compared bone SPECT and FDG-PET. Both
studies indicated a significantly better sensitivity by
using bone-specific radiotracers in combination
with tomographic imaging (SPECT or PET). An in-
teresting finding of that study was that 2 patients
who had highly suspicious vertebral metastases on
MRI were negative on F-18 fluoride PET. Skeletal
metastases were ruled out by clinical follow-up in
1 patient and by biopsy in the other patient. F-18
fluoride PET might therefore become the first imag-
ing method that is sensitive and specific enough for
confirmation of suspicious lesions that are seen
on MRI, but are negative on currently available
conventional imaging methods.
One important advantage of FDG over F-18 fluo-
ride is that the FDG uptake is not limited to bone
metabolism. Several studies have shown that FDG-
PET reliably detects primary tumors [31–36], and is
currently the most sensitive imaging technique for
noninvasive assessment of axillary lymph nodes
and visceral metastases. Furthermore, Eubank and
colleagues [37] have shown that FDG-PET is accu-
rate in evaluating the internal lymph nodes. FDG
30 Schirrmeister

Fig. 7. FDG-PET (left) and bone scintigraphy (right) was performed for restaging in a patient with recurrent
breast cancer. FDG-PET shows multiple lymph node metastases but fewer bone metastases compared with
bone scintigraphy.

PET was superior to conventional imaging and as
sensitive as MRI in the differentiation between uni-
focal and multifocal disease [38]. In a series of 117
patients who had suspected breast cancer [36],
whole-body FDG-PET was as accurate as the cur-
rently employed imaging methods in identification
of the primary tumor, and significantly more accu-
rate in the detection of lymph node and distant
metastases. In that series, only 2 patients had bone
metastases. Both were true-positive on both bone
scintigraphy and FDG-PET. Because of the low inci-
dence of bone metastases, a potentially lower sen-
sitivity in detecting osteoblastic bone metastases
with FDG-PET seems less relevant at initial staging.
Regarding the specificity in evaluation of the
skeleton, there are some rare benign bone tumors
and inflammatory lesions that demonstrate in-
creased FDG uptake. In contrast, the uptake of
F-18 fluoride is not limited to malignant bone
lesions. In turn, F-18 PET fluoride requires specific
experience with this method. This limitation might
be overcome by using coregistered PET and CT im-
ages (PET/CT). Recently Even-Sapir and coworkers
[39] found a significantly higher accuracy of F-18
fluoride PET/CT compared with F-18 fluoride PET
in a mixed series of 44 cancer patients.
Summary
FDG-PET might be less sensitive than F-18 fluoride
PET in detecting bone metastases from breast can-
cer, but FDG-PET allows accurate assessment of the
primary tumor, lymph nodes, and visceral metas-
tases in one single imaging study. The uptake of
FDG is more specific for malignant lesions than
bone metabolic radiotracers. Hence, FDG-PET in

Fig. 8. F-18 fluoride PET of the same patient as presented in Fig. 7. F-18 fluoride PET was performed because
the patient complained of neck stiffness and a suspicious lesion was seen on MRI in the second cervical vertebral
body. The F-18 fluoride PET scan confirms the bone metastasis by intense increased bone metabolism and reveals
an additional, previously unsuspected, bone metastasis in first cervical vertebral body (arrows).

combination with bone scintigraphy might become
the standard of care for staging and restaging of
the skeleton in breast cancer. F-18 fluoride PET is
currently the most sensitive whole-body imaging
modality for bone metastases, but is associated
with a significant learning curve. In breast cancer,
the the use of F-18 fluoride PET should be based
on individual cases [Figs. 7, 8] depending on the
expected change of therapy management.
References
[1] Newcomb PA, Lantz PM. Recent trends in breast
cancer incidence, mortality, and mammography.
Breast Cancer Res Treat 1993;28(2):97–106.
[2] Perez DJ, Powles TJ, Milan J, et al. Detection of
breast carcinoma metastases in bone: relative
merits of x-rays and skeletal scintigraphy in breast
cancer. Lancet 1983;10;2(8350):613–6.
[3] Kunkler IH, Merrick MV. The value of non-
staging skeletal scintigraphy. Clin Radiol 1986;
37(6):561–2.
[4] Coleman RE, Rubens RD. the clinical course of
bone metastases from breast cancer. Br J Cancer
1987;55:61–6.
[5] Hortobagyi GN. Bone metastases in breast cancer
patients. Semin Oncol 1991;18(Suppl 5):11–5.
[6] Yamashita K, Koyama H, Inaji H. Prognostic sig-
nificance of bone metastasis from breast cancer.
Clin Orthop 1995;312:89–94.
[7] Yamashita K, Ueda T, Komatsubara Y, et al.
Breast cancer with bone-only metastases. Visceral
metastases-free rate in relation to anatomic dis-
tribution of bone metastases. Cancer 1991;68(3):
634–7.
[8] Bachman AS, Sproul EE. Correlation of radio-
graphic and autopsy findings in suspected me-
tastases in the spine. Bull N Y Acad Med 1940;
44:169–75.
[9] Silberstein ER, Saenger EL, Tofe AJ, et al. Imaging
of bone metastases with 99mTc-Sn-HEDP (diphos-
phonate), 18F and skeletal radiography. Radi-
ology 1973;107:551–5.
[10] Green D, Jeremy R, Towson J, et al. The role of
fluorine 18 scanning in the detection of skeletal
metastases in early breast cancer. Aust N Z J Surg
1973;43:251–4.
[11] Frank JA, Ling A, Patronas NJ, et al. Detection of
malignant bone tumors: MR imaging vs scin-
tigraphy. AJR Am J Roentgenol 1990;155:1043–8.
[12] Gosfield E, Alavi A, Kneeland B. Comparison of
radionuclide bone scans and magnetic resonance
imaging in detecting spinal metastases. J Nucl
Med 1993;34:2191–8.
[13] Hawkins RA, Choi Y, Huang SC, et al. Evaluation
of skeletal kinetics of fluorine 18-fluoride ion
with PET. J Nucl Med 1992;33:633–64.
[14] Blau M, Nagler W, Bender MA. Fluorine-18 a new
isotope for bone scanning. J Nucl Med 1962;3:
332–4.
[15] Petren-Mallmin M, Andreasson I, Ljunggren O,
PEM Detection of Breast Cancer Bone Metastases 31
et al. Skeletal metastases from breast cancer:
uptake of F-18 fluoride measured with positron
emission tomography in correlation with CT.
Skeletal Radiol 1998;27(2):72–6.
[16] Schirrmeister H, Guhlmann CA, Elsner K, et al.
Planar bone imaging vs. 18F-PET in patients with
cancer of the prostate, thyroid and lung. J Nucl
Med 1999;40(10):1623–9.
[17] Schirrmeister H, Guhlmann CA, Kotzerke J, et al.
Early detection and accurate description of ex-
tent of metastatic bone disease in breast cancer
with 18F-fluoride ion and positron emission to-
mograpy. J Clin Onol 1999;17(8):2381–9.
[18] Schirrmeister H, Glatting G, Hetzel J, et al. Pro-
spective evaluation of the clinical value of pla-
nar bone scans, SPECT and F-18 labelled NaF
PET in newly diagnosed lung cancer. J Nucl Med
2001;42(12):1800–4.
[19] Hetzel M, Arslandemir C, König HH, et al. F-18
NaF PET for detection of bone metastases in lung
cancer. Accuracy, cost-effectiveness, and impact
on patient management. J Bone Miner Res 2003;
18(12):2206–14.
[20] Buscombe JR, Holloway B, Roche N, et al.
Position of nuclear medicine modalities in the
diagnostic work-up of breast cancer. Q J Nucl
Med Mol Imaging 2004;48(2):109–18.
[21] Hoh CK, Hawkins RA, Dahlbom M, et al. Whole
body skeletal imaging with 18F fluoride ion and
PET. J Comput Assist Tomogr 1993;17(1):34–41.
[22] Schirrmeister H, Diederichs CG, Rentschler M,
et al. Die Positronenemissionstomographie des
Skelettsystems mit 18FNa: Häufigkeit, Befund-
muster und Verteilung benigner Veränderungen
[Positron-emission tomography of the skeletal sys-
tem using 18FNa: the incidence, pattern of the find-
ings and distribution of benign changes]. Fortschr
Röntgenstr 1998;169(3):310–4 [in German].
[23] Schirrmeister H, Guhlmann CA, Elsner K, et al.
Die Positronenemissionstomographie des Ske-
lettsystems mit 18FNa: Häufigkeit, Befundmuster
und Verteilung von Skelettmetastasen [Positron
emission tomography of the skeletal system
using 18FNa: frequency, distribution and appear-
ance of skeletal metastases]. Rontgenpraxis 1999;
52(1):19–25 [in German].
[24] Cook GJ, Houston S, Rubens R, et al. Detection
of bone metastases in breast cancer by 18FDG
PET: differing metabolic activity in osteoblastic
and osteolytic lesions. J Clin Oncol 1998;16(10):
3375–9.
[25] Ohta M, Tokuda Y, Suzuki Y, et al. Whole body
PET for the evaluation of bony metastases in
patients with breast cancer: comparison with
99mTc-MDP bone scintigraphy. Nucl Med Com-
mun 2001;22(8):875–9.
[26] Yang SN, Liang JA, Lin FJ, et al. Comparing
whole body 18F–2-deoxyglucose positron emis-
sion tomography and technetium-99m methy-
lene diphophonate bone scan to detect bone
metastases in patients with breast cancer.
J Cancer Res Clin Oncol 2002;128(6):325–8.
32 Schirrmeister
[27] Kao CH, Hsieh JF, Tsai SC, et al. Comparison and
discrepancy of 18F-deoxyglucose-positron emis-
sion tomography and Tc-99m MDP bone scan to
detect bone metastases. Anticancer Res 2000;
20(3B):2189–92.
[28] Kosuda S, Kaji T, Yokoyama H, et al. Does bone
SPECT actually have lower sensitivity for detect-
ing vertebral metastasis than MRI? J Nucl Med
1996;37(6):975–8.
[29] Han LJ, Au-Yong TK, Tong WC, et al. Compari-
son of bone single-photon emission tomogra-
phy and planar imaging in the detection of
vertebral metastases in patients with back pain.
Eur J Nucl Med 1998;25(6):635–8.
[30] Uematsu T, Yuen S, Yukisawa S, et al. Compari-
son of FDG PET and SPECT for detection of
bone metastases in breast cancer. AJR Am J
Roentgenol 2005;184(4):1266–73.
[31] Adler LP, Crowe JP, Al-Kaisi NK, et al. Evaluation
of breast masses and axillary lymph nodes with
F-18 2-deoxy-3-fluoro-D-glucose. Radiology
1993;187(3):743–50.
[32] Avril N, Dose J, Jänicke F, et al. Assessment
of axillary lymph node involvment in breast
cancer patients with positron emission tomogra-
phy using radiolabeled 2-(fluorine-18)-deoxy-
2-fluoro-D-glucose. J Natl Cancer Inst 1996;
88(17):1204–9.
[33] Crippa F, Agresti R, Seregni E, et al. Prospective
evaluation of fluorine-18-FDG PET in presurgical
staging of the axilla in breast cancer. J Nucl Med
1998;39(1):4–8.
[34] Moon DH, Maddahi J, Silverman DHS, et al.
Accuracy of whole-body fluorine-18 FDG PET for
the detection of recurrent or metastatic breast
carcinoma. J Nucl Med 1998;39(3):431–5.
[35] Eubank WB, Mankoff DA, Vesselle HJ, et al. De-
tection of locoregional and distant recurrences
in breast cancer patients by using FDG PET.
Radiographics 2002;22(1):5–17.
[36] Schirrmeister H, Kuhn T, Guhlmann A, et al.
Fluorine-18 2-deoxy-2-fluoro-D-glucose PET in
the preoperative staging of breast cancer: com-
parison with the standard staging procedures.
Eur J Nucl Med 2001;28(3):351–8.
[37] Eubank WB, Mankoff DA, Takasugi J, et al.
18fluorodeoxyglucose positron emission tomog-
raphy to detect mediastinal or internal mam-
mary metastases in breast cancer. J Clin Oncol
2001;19(15):3516–23.
[38] Rieber A, Schirrmeister H, Gabelmann A, et al.
Pre-operative staging of invasive breast cancer
with MR mammography and/or PET: boon or
bunk? Br J Radiol 2002;75(898):789–98.
[39] Even-Sapir E, Metser U, Flusser G, et al. Assess-
ment of malignant skeletal disease: initial experi-
ence with 18F-fluoride PET/CT and comparison
between 18F-fluoride PET and 18F-fluoride PET/
CT. J Nucl Med 2004;45(2):272–8.
 33

POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 33–38

Instrumentation for Positron

Emission Mammography
Christopher J. Thompson, DSc FCCPM

& Feasibility and promise of early & Recent clinical findings

instruments & Future prospects for positron emission
& Evolution of positron emission mammography
mammography instruments & Acknowledgments
& Early clinical findings & References

The use of positron emission tomography (PET)
to provide images of the glucose metabolism of
malignant disease is becoming more and more
widespread, and whole body PET (WB-PET) images
are now often used to choose appropriate therapy
for cancer patients. The first use of F-18 fluorodeoxy-
glucose-positron emission tomography (FDG-PET)
for the study of breast cancer was reported by Wahl
and colleagues [1], and since then many other re-
ports have been made [2,3], to cite the first few only.
When combined with radiograph CT, PET images
provide the equivalent of a ‘‘metabolic contrast
agent,’’ which serves to highlight the abnormal glu-
cose metabolism in tumors. Combined PET/CT
scanners now outsell conventional PET scanners,
and they have become the standard of care in
many centers.
PET scanners have a limited spatial resolution
compared with structural imaging modalities such
as CT and MRI. Because the identification of small
tumors leads to earlier diagnosis and treatment,
much effort has gone into trying to improve the
spatial resolution of PET. There are both instru-
mental and fundamental factors that degrade spa-
tial resolution in PET. The fundamental limit is
caused by the distance positrons move away from
the parent nucleus before they lose energy and
annihilate with an electron in tissue. Another limi-
tation is the noncollinearity of the pair of gamma
rays that do not travel away from the point of
annihilation at exactly 180°, because of the energy
of the electron at the time of annihilation. Al-
though this appears as a fundamental limit in con-
ventional PET, it is much less of a problem in an
instrument that has its detectors much closer to
each other than the separation needed in WB-
PET. The major instrumental limitation is the size
of the detectors. Making the detectors smaller
improves the resolution at the expense of increased
complexity and resulting cost; however, if detectors
are placed closer, fewer detectors are required, so
instruments made especially for breast imaging can
be made to provide higher spatial resolution than
those from WB-PET. The factors that limit spatial
resolution (SR) in PET can be combined as the sum
of independent variables in the formula:
ðSRÞ2 ¼ R*½PR 2 þ ð0:005*DSÞ2 þ ðCW=2Þ2 þ BE2

In this equation, R is a factor that relates to the
reconstruction method and filter, PR is the effective

The development and evaluation of the PEM-1 scanner was funded by grants from the National Cancer
Institute of Canada’s Canadian Breast Cancer Research Initiative #6139 and #9232 in 1997 and 1999.
Departments of Neurology and Neurosurgery, Medical Physics, and Biomedical Engineering, Suite #798,
Montréal Neurological Institute, McGill University, 3801 University Street, Montreal, Québec, Canada H3A 2B4
E-mail address: Christopher.Thompson@McGill.Ca

1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cpet.2005.09.004
pet.theclinics.com
34 Thompson
positron range in tissue 0.005 = 1/2*tan(θ/2), θ is
the mean angle of noncollinearity, DS is the detec-
tor separation, CW is the crystal width, and BE
combines the effects of light sharing and under-
sampling of the image.
One of the design factors in all PET scanners is
a trade-off between spatial resolution, sensitivity,
and cost. In the equation above, the factor CW
is the one that balances the cost versus the spa-
tial resolution, and the one over which scanner
manufacturers have the most control. In WB-PET,
moving the detectors closer together is not possi-
ble, because the patient would no longer fit in
the scanner; however, there is an option for
doing this in instruments designed exclusively for
breast imaging.
The idea of a dedicated PET scanner for breast
imaging was first proposed by Weinberg in 1993, in
a US patent application [4], and a successful pro-
posal for a small business incentive for research
(SBIR) grant from the US National Institutes of
Health (NIH).
Feasibility and promise of early instruments
Through a collaboration with Weinberg, the first
experiments to examine the concept were published
in 1993 [5], and the name ‘‘positron emission mam-
mography’’ (PEM) was coined to represent this
technique. The concept was to place two planar
detectors capable of detecting the 511 keV annihi-
lation photons in a conventional mammography
unit. Placing the breast on a ‘‘magnification table’’
sometimes used in these instruments provides
the possibility of having one detector between the
radiograph tube and the compression plate, and
another between the lower aspect of the breast
and the Radiograph sensor. The two detectors
move out of the radiograph field for conventional
mammography, and move back over and under
the breast for the PEM acquisition [6]. This concept
predates PET/CT by several years [7], but the goal
was very much the same as that of PET/CT as it
has evolved today: to provide a coregistered anato-
mical and functional image in the same procedure
with minimal movement of the patient. The de-
sign of the instrument is illustrated in Fig 1.
An important finding of this first PEM paper [5]
was that a small hyperactive region was just as
visible in a superposition of a few near vertical pro-
jections as it was in fully reconstructed tomo-
graphic images. The experiments were performed
in a 15-slice brain scanner on a box phantom con-
taining four tubes of various sizes with either no
activity or 9.3 times the background. The images
were made over different times, so that each con-
secutive image contains half the counts of the pre-
x-ray
source
co-registration
tool
upper
detector
compression
compression breast
paddle
45 cm
lower
magnification detector
table
This side faces
chest wall rails
x-ray film cassette
holder
45 cm
Fig. 1. Layout of PEM-1 breast imaging system. The
two detectors, mounted on rails, are moved out of the
radiographic field for conventional mammography,
and back over and under the breast during the acqui-
sition of a metabolic image.
vious one. This paper provided the basic estimate
of both the signal-to-noise ratio and count-rate that
could be expected from a clinical PEM instrument.
Encouraged by these results, the author’s group
applied for funding from the Canadian Breast
Cancer Research Initiative. We then built and per-
formed a preliminary clinical trail of an instrument
known as ‘‘PEM-1.’’ Because we had concluded that
back-projection images were sufficient to identify
regions of higher-than-surrounding uptake, we did
not perform any sophisticated reconstruction, pre-
ferring to opt for an almost real-time image display
of the PEM image. The goal we set was to perform
the clinical trial using only 75 MBq (2 mCi) of
FDG, and an imaging time of 2 minutes per breast,
because this is about the time it takes to develop a
radiograph film in an automatic film processor.
Evolution of positron emission
mammography instruments
When performing WB-PET scans, it is a simple
matter to overscan the regions most likely to harbor
metastases, and to overlap the bed positions to
compensate for the reduced sensitivity toward the
axial ends of each set of slices caused by the fall-off
in three dimensional (3D) sensitivity in the scan-
ner. When imaging the breast with PET detectors in
a mammographic configuration, this is not possi-
ble, as illustrated in Fig. 2. Even when the detectors
are placed very close to the chest wall, some shield-
ing is required. This is a serious problem when
imaging small breasts, and for investigations close

to the chest wall. PET imaging in the 3D mode is
always less sensitive toward the axial ends of the
field of view, because fewer detector pairs can be in
coincidence in these regions.
A variety of different PEM geometries have been
proposed, and several of these are currently in use
[Fig. 3]. Almost all have a larger field of view than
the first prototype. Complete coverage of the breast
while the breast remains in place has become
the norm. Fig. 3 shows the detector arrangement
in each device. Part A is the classic two-parallel
crystal arrays coupled to position-sensitive photo-
multipliers (PS-PMTs). One detector is moveable to
allow positioning the breast and for variable com-
pression to suit the patient’s anatomy. This geome-
try was first used in the PEM-1 scanner [8–12], and
later by Smith and coworkers of the Thomas Jeffer-
son Laboratory [13]. Part B shows a pair of linear
arrays of detectors that scan across the breast dur-
ing the examination [14]. This technique is used
in the only commercially available PEM scanner
at present, the PEM Flex from Naviscan PET Sys-
tems in Rockville, Maryland. Part C is a boxlike
detector array that surrounds the breast, and should
Fig. 2. Geometry for conventional radiographic mam-
mography and PEM using two planar detectors. The
radiograph beam is centered over the edge of the
image sensor near the chest wall, and collimated to
expose only the compressed breast. The response of
the sensor to the radiograph beam (plotted below the
geometry) is essentially constant over the entire sur-
face. In contrast, the response of the PEM detectors is
triangular, with its peak in the geometric center of the
detector pair.
Instrumentation for PEM 35
Fig. 3. Four geometries used or proposed for PEM.
(A) Two small planar detectors with variable separa-
tion. (B) Two multi-element detectors that move inside
an enclosure during imaging. (C ) Rectangular box
structure with moveable detectors. (D) Cylindrical crys-
tal array surrounding pendulous breast.
allow a more complete reconstruction of the activ-
ity within the breast. This is the system proposed by
Huber and colleagues [15] of the Berkeley National
Laboratory in Berkeley, California. A similar con-
cept encloses the breast in a small circular array
of detectors with the breast pendant through the
hole [16].
All of these instruments use very finely pixilated
detectors and a compact geometry designed to
reduce the blurring associated with the noncollin-
earity of the annihilation photons. Of special in-
terest is the proposal by Huber and coworkers [15],
which also encodes the depth at which each
gamma ray is detected in the crystal. This allows
for a very compact geometry while avoiding the
blurring associated with very oblique gamma ray
penetration of the detector.
Early clinical findings
In 1999, the author’s group published the results of
the first clinical trial of any PEM instrument [12].
During the clinical trial of the PEM-1 scanner, we
studied 14 patients, 10 of whom had various breast
cancers confirmed by pathological investigation of
the surgically excised specimens. Only 5 of these
had a clearly focal uptake (with a mean contrast of
5.8:1 with respect to the surrounding breast tissue).
Three other patients were considered PEM-positive
on the basis of a significant count-rate asymmetry,
after accounting for factors like isotope decay and
volume of breast tissue in the field of view, and de-
tector separation.
As an example of a typical patient from that
study, consider the case of a 75-year-old lady
who presented with a suspicious mammogram of
her right breast, and was a candidate for lumpec-
tomy. She had a cranio-caudal (CC) PEM scan of
her right breast starting 60 minutes after the admin-
36 Thompson

Fig. 4. PEM-1 images showing focal uptake in a 1.5 cm inter-ductal carcinoma (upper row) compared with normal
uptake in the contra-lateral breast (lower row). These images were scaled to the equivalent counting time and
isotope decay, showing an asymmetry in the overall uptake.

istration of 75 MBq of 18F-FDG. A CC scan of the
left breast was then acquired starting 70 minutes
postinjection. She underwent a right-breast seg-
mental mastectomy 4 days after the PEM study.
The pathologist’s report described a 1.5 × 1.5 ×
1.3 cm intraductal carcinoma and infiltrating duc-
tal carcinoma, with a histological grade of 3 on the
Bloom and Richardson scale. The appearance of the
PEM images is shown in Fig. 4. Seven images are
presented for each breast, with the right breast
displayed on the top row. The images are displayed
as if the breast were pendant, so that the chest wall
would be above the images presented. The left-
most image corresponds with the upper aspect
of the breast on the CC mammogram. The focal
activity visible near the top (chest wall) in the
upper row is most intense in image 4, but extends
further toward the upper breast surface. The lower
row of images from the normal breast is more
uniform, and was read as normal.
The correlation between the PEM and mammo-
gram is illustrated in Fig. 5. The PEM image is first
‘‘windowed’’ such that only the suspicious region is
visible. The mammogram, following digitization
with a video frame digitizer, is then displayed.
The images are rescaled and fused by aligning the
top-right and bottom-left corners of a digital repre-
sentation of a wire grid with its image, which is
visible on the mammogram. A fused image is made
by displaying alternate pixels from the mammo-
gram and scaled PEM image in differing color
scales. This technique is now used to display PET/
CT images, although the image fusion software of
the author’s group predates PET/CT.
Recent clinical findings
The first report of clinical results from PEM Flex was
published by Weinberg and coworkers earlier this
year [14]. They reported on 94 cases performed at

Fig. 5. Correlation of the intensity-thresholded PEM-1 image and a hyperdense region of the mammogram using
the coregistration technique developed for this instrument. The field of view of the mammogram is bigger than
that of the PEM-1 detectors, and in the active PEM region, alternate pixels are assigned to PEM and radiograph
images in a manner identical to that commonly used for PET/CT image display.

Fig. 6. PEM image from the Naviscan imager. In addi-
tion to one large and four smaller interductal lesions
in the outer aspect of the breast, there is another
hypermetabolic region at the posterior wall (arrow),
suggestive of axillary node involvement. (From Wein-
berg IN, Beylin D, Zavarzin V, et al. Positron emission
mammography: high-resolution biochemical breast
imaging. Technol Cancer Res Treat 2005;4:55–60;
with permission.)
four different sites during the first year of use of the
instrument. Analysis of these cases showed a sensi-
tivity of 93% and a specificity of 83%.
Fig. 6 is an example of a PEM image from this
instrument. Unlike the PEM-1 scanner, the field of
view is much larger, 24 × 18 cm rather than 5.5 ×
6.0 cm. The device also employs a sophisticated
limited-angle reconstruction algorithm, which pro-
duces much clearer 3D images. The five smaller
lesions would not have been visible on a conven-
tional PET scan, or on the PEM-1 instrument. The
PEM Flex also has the ability of scanning closer to
the chest wall.
Future prospects for positron emission
mammography
In the United States, Medicare approval for use
of PET imaging of breast cancer was given in late
2003 [17]. Unlike the approvals granted for PET in
other forms of malignant disease, the approval docu-
ment refers to three specific areas: initial staging
of axillary lymph nodes, detection of loco-regional
recurrence or distant metastasis/recurrence, and
evaluating response to treatment. Although this is
intended for the use of conventional WB-PET scan-
ners, there is an opportunity for the assessment of
dedicated PEM imaging instruments within the
guidelines of these new regulations. The ’’response
to treatments’’ aspect of the approval document is
Instrumentation for PEM 37
perhaps the most relevant to PEM. The technology
assessment report from the Blue Cross and Blue
Shield Association [17], which forms the basis of
the review of the utility of FDG-PET in breast can-
cer, contains over 70 references on the use of
PET and single photon imaging agents such as
99m
Tc sestamibi in the detection and staging of
breast cancer.
The major limitations of early PEM instruments
were their small field of view and reduced sensi-
tivity near the chest wall. The field of view was
limited by the PS-PMTS available in the early 1990s.
Recent demands for these devices and their deploy-
ment in small-animal PET scanners has led to much
improvement, both in uniformity and in ability
to image much closer to the edge without serious
distortion. Now instruments such as the PEM Flex
are able to capitalize on the advances in photon
sensors. The problem of sensitivity loss near the
edges of the field of view is intrinsic to 3D PET
acquisitions. During early PEM development, this
was not considered seriously enough. Now that it
has been widely recognized, the situation has been
improved through the use of higher density, thin-
ner shielding, better PMTs, and improved geome-
try. This results in the effective field of view being at
least 3 cm closer to the chest in the latest instru-
ments than in the prototype developed by the
author’s group.
Acknowledgments
The author gratefully acknowledges support from
the Natural Science and Engineering Council of
Canada over the last 15 years. PEM research at
the McGill University teaching hospitals has been
a team effort. The work of Dr. R Lisbona, Chairman
of the Radiology Department, and Dr. Antoine
Loutfi of the Surgical Oncology unit of the McGill
University Health Centre: Royal Victoria Hospital
played vital roles in the success of this project.
Students Kavita Murthy, Yani Picard, James Robar,
Alanah Bergman, Raymond Clancy, Marie-Anne
Aznar, Martin Hinse, and Nan Zhang all contrib-
uted to the development of PEM under the au-
thor’s supervision.
References
[1] Wahl RL, Cody RL, Hutchins GD, et al. Primary
and metastatic breast carcinoma: initial clinical
evaluation with PET with the radiolabeled glu-
cose analogue 2-[F-18]-fluoro-2-deoxy-D-glucose.
Radiol 1991;179:765–70.
[2] Adler LP, Crowe JP, Al-Kaisi NK, et al. Evaluation
of breast masses and axillary lymph nodes with
[F-18] 2-deoxy-2- fluoro-D-glucose PET. Radiol
1993;187:743–50.
38 Thompson
[3] Avril N, Schelling M, Dose J, et al. Utility of PET
in breast cancer. Clin Positron Imaging 1999;2:
261–71.
[4] Weinberg IN. US patent # 5,252,830. Dedicated
apparatus and method for emission mammogra-
phy. October 12, 1993.
[5] Thompson CJ, Murthy K, Weinberg IN, et al.
Feasibility study for positron emission mammog-
raphy. Med Phys 1994;21:529–37.
[6] Thompson CJ, Murthy K, Picard Y, et al. Positron
emission mammography (PEM): a promising tech-
nique to detect breast cancer. IEEE Trans Nucl Sci
1995;NS42:1012–7.
[7] Bergman A, Thompson CJ, Murthy K, et al.
Technique to obtain positron emission mammo-
graphy images in registration with X-ray mam-
mograms. Med Phys 1998;25:2119–29.
[8] Townsend DW, Beyer T, Kinahan PE, et al.
The SMART scanner: a combined PET/CT tomo-
graph for clinical oncology. RSNA, Chicago,
December 1998.
[9] Townsend DW, Nutt R. US patent #6,490,476.
Combined PET and X-Ray CT tomograph and
method for using same. December 3, 2002.
[10] Robar JL, Thompson CJ, Murthy K, et al. Con-
struction and calibration of detectors for high
resolution metabolic breast cancer imaging. Nucl
Instrum Methods Phys Res A 1997;392:402–6.
[11] Murthy K, Bergman AM, Thompson CJ, et al.
Positron emission mammographic instrument:
initial results. Radiology 2000;215:280–5.
[12] Murthy K, Aznar M, Thompson CJ, et al.
Preliminary clinical evaluation of an instrument
for positron emission mammography (PEM-I).
J Nucl Med 2000;41:1851–8.
[13] Smith MF, Majewski S, Weisenberger AG, et al.
Analysis of factors affecting positron emission
mammography (PEM) image formation. IEEE
Trans Nucl Sci 2003;50:53–9.
[14] Weinberg IN, Beylin D, Zavarzin V, et al. Posi-
tron emission mammography: high-resolution
biochemical breast imaging. Technol Cancer Res
Treat 2005;4:55–60.
[15] Huber JS, Choong WS, Wang J, et al. Develop-
ment of the LBNL positron emission mammog-
raphy camera. IEEE Trans Nucl Sci 2003;50:
1650–3.
[16] Karimian A, Thompson CJ, Sarkar S, et al. CYBPET:
a cylindrical PET system for breast imaging.
Nucl Instrum Methods Phys Res A 2005;545(2):
427–35.
[17] FDG positron emission tomography for evaluat-
ing breast cancer. Blue Cross and Blue Shield
Technology Evaluation Center Report: 18:14.
Available at: http://www.bcbs.com/tec/Vol18/
18_14.pdf. Accessed August 11, 2005.

FDG-PET and PET/CT in Radiation
Therapy Simulation and
Management of Patients Who Have
Primary and Recurrent Breast Cancer
a,*
Dwight E. Heron, MD , Sushil Beriwal,
& Conventional radiation therapy planning
& Image guidance for radiation therapy
planning
& Radiation therapy planning: use of F-18
fluorodeoxyglucose-positron emission
tomography and positron emission
tomography/computed tomography
& F-18 fluorodeoxyglucose-positron emission
tomography for radiation therapy of the
breast and loco-regional lymph nodes
& F-18 fluorodeoxyglucose-positron emission
tomography/computed tomography for
radiation treatment planning
Precise anatomical information regarding the
location and extent of tumor tissue is essential for
radiation treatment planning. In the past, anatomi-
cal imaging has greatly improved the accuracy of
delineating target structures and is currently the
basis of modern three dimensional (3D)-based ra-
diation treatment [1]. Plain radiography, CT, and
MRI provide structural or morphological informa-
tion based on tissue density, size, vascularity, and
fat or water content, and are routinely being used
for radiation treatment planning. The excellent spa-
tial resolution of anatomical imaging techniques
allows the detection of subcentimeter lesions
a
Department of Radiation Oncology, University of Pittsburgh School of Medicine, University of Pittsburgh
Cancer Institute, UPMC Shadyside Hospital, 5150 Centre Avenue, #545, Pittsburgh, PA 15232, USA
b
Department of Nuclear Medicine, Barts and the London School of Medicine, Queen Mary, University of
London, West Smithfield (QEII), London, EC1A 7BE, UK
* Corresponding author.
E-mail address: herond2@upmc.edu (D.E. Heron).
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
39
POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 39–49
a b
MD , Norbert Avril, MD
& Current limitations of F-18
fluorodeoxyglucose-positron emission
tomography and positron emission
tomography/computed tomography for
radiation therapy planning
& Radiosurgery treatment planning and
positron emission tomography/computed
tomography
& Future perspectives
& Summary
& References
(eg, in the lungs); however, definition of tumor
involvement, in lymph nodes for example, based on
the increased size is more difficult. Some enlarged
lymph nodes may be reactive, whereas smaller nodes
may harbor metastatic foci. MRI is better at outlining
soft tissues, particularly in the brain, and has been
valuable in complementing CT-based radiation treat-
ment planning [2].
Over recent years, functional imaging with posi-
tron emission tomography (PET) has gained in-
creased importance in determining biological or
molecular abnormalities in specific tumors. PET
using [F-18] fluorodeoxyglucose (FDG) allows the
doi:10.1016/j.cpet.2005.09.003
40 Heron et al
Fig. 1. 2D tangential beam showing the isodose lines
in the central axis.
characterization of suspicious masses as well as de-
termination of the spread of disease to loco-regional
lymph nodes and distant sites [3]. The metabolic
information derived from FDG-PET can influence
the radiation treatment planning in several ways.
First, FDG-PET may reveal a tumor target that was
previously not detected by anatomical imaging.
Second, FDG-PET may detect additional tumor re-
gions outside the tumor volume defined by CT or
MRI. Third, FDG-PET may show subregions or foci
with increased or altered biological activity within
the gross tumor volume that could be preferentially
targeted and treated with escalated radiation doses.
Therefore, the more precise 3D delineation of tumor
volumes could translate into improved loco-regional
control with radiation therapy.
Breast cancer is the most common malignancy
among women in the United States, with approxi-
mately 211,240 cases expected in 2005 [4]. Breast
irradiation was first used to minimize the risk of
loco-regional recurrence in patients who had lo-
cally advanced cancers [5]. Beginning in the late
1970s, breast irradiation has evolved into the main-
stay of therapy in patients treated with breast
conservation surgery. In recent years, breast conser-
vation therapy consisting of a segmental mastec-
tomy and definitive breast irradiation has assumed
an increasing role in the management of patients
who have both invasive and noninvasive breast
cancer. No fewer than six prospective, randomized
controlled trials have shown equivalent outcomes
of breast conservation therapy compared with mas-
tectomy [6–11]. Distant metastases of breast cancer
are frequently found in lymph nodes, lungs, liver,
and bones.
In the metastatic setting, radiation therapy is cur-
rently used for treatment of soft tissue and bone
metastasis as well as for intracranial disease. Pallia-
tive radiation has been shown to be effective in im-
proving the quality of life and providing relief of
symptoms, especially the pain of bone metastasis.
Conventional radiation therapy planning
For the last half a century, breast irradiation has
been performed through parallel opposing tangen-
tial beams with wedges, with matching anterior
supraclavicular and posterior axillary fields when-
ever appropriate. Radiation simulation, the process
of selecting the area to be treated and beam se-
lection, is routinely performed through fluoro-
scopic placement of the tangential beams, based
upon clinically determined breast tissue borders.
Treatment planning has been single-slice, two-
dimensional (2D) dosimetry, applying wedge beam
data onto manually obtained breast central axis con-
tours [Fig. 1].
Although this unsophisticated technique initially
produced a breakthrough in breast conservation
therapy, a number significant limitations of this
technique are now apparent, including excess car-
diac and lung sequelae of breast irradiation [12]. In
addition, patients desire better cosmetic results and
shorter treatment times. Increased scatter radiation
to contralateral breast from wedges may increase
the incidence of second malignant neoplasm [13].
Furthermore, some have suggested that poor dose
distribution may contribute to late in-breast tumor
failures. Data from Bhatnagar and colleagues [14]
suggest significant reduction of scatter radiation to
the contralateral breast with the use of intensity-
modulated radiation therapy (IMRT). As a result,
newer techniques using CT-based treatment plan-
ning have been used to create ‘‘beam’s eye’’ views to
avoid these underlying critical structures [Fig. 2].
,
Fig. 2. Beam s eye view of 3D partial wide tangent
covering chest wall and internal mammary nodes
(blue) and sparing heart (green).

These beams eye views allow the radiation oncolo-
gist to examine the target and nearby critical struc-
tures from the perspective of the beam’s path.
Image guidance for radiation therapy
planning
CT-based, 3D treatment planning represents an
evolution from 2D planning, and provides volu-
metric information that permits a greater assess-
ment of the beam’s path and subsequent dose to
target tissues and normal structures [Fig. 3]. CT-
based approaches also permit virtual simulation, a
process in which the CT data set is acquired rapidly,
thus significantly reducing the time the patient is
spending on the simulation table. Treatment of
chest wall and regional nodes poses significant chal-
lenges to radiation oncologists, particularly in the
postmastectomy setting. The range of body habi-
tus and close proximity of the internal mammary
nodes to the heart often necessitates individualized
treatment planning, with complex field arrange-
ments. 3D treatment planning with volumetric
information helps in clinical decision-making
Fig. 3. 3D radiation treatment plan showing isodose
distribution covering the breast in axial and sagit-
tal planes.
FDG-PET and PET/CT Radiation Therapy Planning 41
when selecting field arrangements for treatment of
these patients, and has become standard of care.
Radiation therapy planning: use of F-18
fluorodeoxyglucose-positron emission
tomography and positron emission
tomography/computed tomography
Currently, CT-assisted volume definition remains
the gold standard for external beam radiation ther-
apy with curative intent [1]. Accurate localization
of the gross tumor volume (GTV) is critical to
optimizing the therapeutic ratio, by sparing normal
tissues and maximizing coverage of tumor volumes
in conformal radiation techniques. Using current
available anatomical imaging for treatment plan-
ning has several limitations. These include the
visualization and precise delineation of tumors in
certain areas of the body, such as the head and
neck, but also in postsurgical distorted anatomy,
and in residual scar tissue after chemotherapy,
specifically in previously irradiated areas. The in-
corporation of metabolic FDG-PET imaging in
radiation treatment planning has raised hopes for
further improvement, which would lead to so-
called ‘‘multidimensional conformal radiotherapy.’’
FDG-PET is complementary to conventional meth-
ods of staging in that it has been shown to have
both higher sensitivity and specificity in detecting
nodal metastases and distance in a variety of ma-
lignancies, including breast cancer, when compared
with CT alone [15]. The sensitivity of FDG-PET in
the detection of axillary lymph node metastasis
from breast cancer varies from 20% to 100%
when compared with sentinel lymph node (SLN)
biopsy or axillary node dissection [16–22]. This
variation in the sensitivity of FDG-PET to detect
tumor-involved lymph nodes depends primarily
on the tumor stage, and subsequently on the extent
of disease. Step sectioning and immunohistochemi-
cal staining of sentinel lymph nodes detected
micrometastases in up to 45% of cases [21]. FDG-
PET can miss micrometastases in lymph nodes
when the number of tumor cells is too small to
provide a signal above the background activity.
Avril and coworkers [17] studied 51 patients and
found an overall sensitivity and specificity for de-
tection of axillary lymph node metastases of 79%
and 96%, respectively. In patients who had primary
breast tumors larger than 2 cm (>stage pT1), the
sensitivity increased to 94%, with a corresponding
specificity of 100%; however, in patients who had
small primary breast cancer (stage pT1), the sensi-
tivity of FDG-PET was only 33%. In a prospective
multicenter study including 360 women [23] who
had newly diagnosed invasive breast cancer, the
largest patient group so far, the sensitivity for
42 Heron et al
FDG-PET was 61%, with a corresponding specificity
of 80%. Despite the limitations of FDG-PET dis-
cussed above, specifically in detecting micrometas-
tases, its accuracy for detecting tumor-involved
axillary lymph nodes is superior to those of CT or
MRI alone.
Studies have consistently demonstrated that
FDG-PET is better than CT in the detection of
internal mammary as well as mediastinal lymph
nodal metastases [Fig. 4]. In 73 consecutive pa-
tients who had recurrent or metastatic disease,
PET was able to correctly identify 40% of the pa-
tients who had intrathoracic lymph node metas-
tases, resulting in a sensitivity of 85% and a
specificity of 90% [24]. Only 23% of the patients
had suspiciously enlarged lymph nodes in CT, lead-
ing to a sensitivity of 54% and a specificity of 85%.
Therefore, the overall diagnostic accuracy of PET
(88%) was higher compared with that of CT (73%)
[24]. In locally advanced breast cancer patients, the
prevalence of internal mammary FDG uptake can
be as high as 25% [25]. The potential importance
of internal mammary nodal drainage has also been
brought to the forefront with the widespread use
of axillary sentinel node mapping, which fre-
quently shows internal mammary lymph drainage
on lymphoscintigraphy. There is some speculation
that a subset of patients who have ‘‘sternal metas-
tases’’ may in fact be internal mammary node fail-
ures with erosion of the sternum. If detected at
Fig. 4. (A–C) PET, CT, and PET/CT demonstrating FDG-avid left chest wall metastasis.
initial diagnosis, this area could be effectively tar-
geted with conformal radiation techniques.
Several reports have suggested that FDG-PET is
superior to conventional imaging for detection of
distant metastatic disease. The use of FDG-PET
in patients who have locally advanced breast can-
cer frequently results in the detection of unexpected
distant metastases, and has been shown to substan-
tially change the clinical management [25]. Recently,
Dose and colleagues [26] compared FDG-PET
with chest radiograph, bone scintigraphy, and ultra-
sound of the abdomen for detection of metastatic
disease in 50 breast cancer patients. FDG-PET iden-
tified metastatic disease with a sensitivity and a speci-
ficity of 86% and 90%, respectively, as compared
with 35% and 95%, respectively, for conventional
imaging procedures. In another retrospective analy-
sis of 62 patients [27], sensitivity and specificity for
detecting local recurrences or distant metastases
were 97% and 82%, respectively, for FDG-PET, ver-
sus 84% and 60% for conventional imaging.
The skeleton is a common site for distant metas-
tases in breast cancer. Bone scintigraphy is an estab-
lished screening method for bony metastases, and
allows determining the extent of disease; however,
tracer uptake in bone scintigraphy reflects osteo-
blastic activity, and is therefore limited in the detec-
tion of osteolytic lesions. Cook and coworkers [28]
studied 23 breast cancer patients who had known
skeletal metastases, and FDG-PET was able to de-

tect more lesions than bone scintigraphy overall.
Higher FDG uptake was observed for osteolytic
than for osteoblastic bone metastases, and FDG-
PET also detected more osteolytic lesions. On the
other hand, PET detected fewer bone metastases
than bone scintigraphy in patients who had osteo-
blastic metastases. In a larger series of 48 breast
cancer patients [29], a total of 127 bone lesions,
including 105 metastatic and 22 benign lesions,
were found by either FDG-PET or bone scintigra-
phy. The sensitivity and diagnostic accuracy of
FDG-PET were 95.2% and 94.5%, respectively, ver-
sus 93.3% and 78.7% for bone scintigraphy. Al-
though the overall sensitivity for bone scintigraphy
and PET are reported as comparable, bone scin-
tigraphy seems to be superior in the detection of
osteoblastic disease, whereas FDG-PET is superior
in osteolytic metastases, suggesting a complemen-
tary role for the imaging procedures.
F-18 fluorodeoxyglucose-positron emission
tomography for radiation therapy of the
breast and loco-regional lymph nodes
Breast cancer may be unique in that there are
numerous loco-regional treatments using various
combinations of surgical techniques and radiother-
apy regimens. Surgery may vary from excisional
biopsy with microscopically involved margins, to
lumpectomy, which usually indicates a more gen-
erous resection (again with or without positive
margins), to quadrantectomy, simple mastectomy,
modified radical mastectomy, and even to radical
mastectomy. There are several accepted radiation
therapy techniques, ranging from treatment of the
entire breast with or without a boost to the lum-
pectomy cavity following breast conservation sur-
gery, to more comprehensive techniques including
the regional nodes [30]. These nodal areas include
the axilla, supraclavicular region, and the upper
internal mammary region. Similarly, following the
mastectomy, radiotherapy may be administered to
just the chest wall, or to the chest wall with or
without one or more of the regional nodal areas.
Currently, FDG-PET cannot provide additional
information in the radiation treatment planning
of primary breast cancer following breast conser-
ving surgery, because there is no macroscopic (re-
sidual) disease. In the setting of local recurrence,
FDG-PET may aid in determining more precisely
the extent of tumor tissue, and may identify addi-
tional lymph node involvement in the axilla or
internal mammary chain, as well as assess meta-
static tumor to the chest wall.
Treatment-related toxicities in breast cancer pa-
tients treated with radiation therapy have improved
over the last 2 decades. Nevertheless, meta-analysis
FDG-PET and PET/CT Radiation Therapy Planning 43
has suggested more late cardiac death in patients
treated with radiation therapy compared with those
treated surgery alone [31]. This effect appeared to
be more pronounced in younger women who
had left-sided breast cancer and who were treated
with anthracycline-based chemotherapy. The excess
cardiac mortality is thought to be due to a com-
bined toxicity of chemotherapy and radiation of
the left ventricle and major coronary vessels. The
introduction of intensity-modulated radiation ther-
apy (IMRT) technology in breast treatment has the
potential to provide uniform dose coverage to the
left breast while limiting dose to the underlying
heart and lungs [Fig. 5]; however, this issue is
complicated by the respiratory motion of the
chest wall and physiologic cardiac motion. These
motions may cause geometric deviations of the
heart from the intended location, causing unde-
sired higher doses to the heart. The use of four-
dimensional (4D) CT and beam gating systems in
radiation therapy has proven effective in limiting
the motion impacts to radiation treatments in lung
cancer [32]. Similar techniques may be used to
manage chest wall and cardiac motion by the selec-
tion of phases that minimize the overlap of the
treatment portal with the heart and lungs.
The accurate evaluation and diagnosis of internal
mammary metastases is of great importance in
radiation treatment planning. The diagnosis of
internal mammary metastases can facilitate directed
radiation or surgical intervention, thus allowing
for treatment planning tailored more closely to
each individual patient’s disease and risk. Similarly,
chemotherapy choices may also be affected, par-
ticularly in the patient who appears to have node-
negative disease by conventional staging of the
axilla, but is found to have internal mammary
involvement. The routine use of radiation to treat
the internal mammary lymph nodes in women
who have locally advanced breast cancer remains
controversial; however, internal mammary disease
has been associated with higher rates of distant me-
tastases and lower overall survival [33]. Although
older extended radical mastectomy series did not
Fig. 5. (A, B) IMRT plan showing homogenous dose
to the breast in axial and sagittal plane with sparing
of heart.
44 Heron et al
demonstrate a survival advantage to internal mam-
mary dissection, they did demonstrate that for
medially located tumors the incidence of involve-
ment was up to 30% to 40%, especially when the
axilla was also positive for metastases [34,35]. In
subset analyses, there was a suggestion of a poten-
tial disease-free advantage in patients who had
medially located primary tumors and axillary me-
tastases, those most likely to have internal mam-
mary involvement when these areas were subjected
to treatment. Moreover, a recent large retrospective
review from the British Columbia Cancer Agency
[36] found a poorer overall survival and distant
metastases-free survival in high-risk women who
had medial tumors, again indicating the presence
of untreated internal mammary disease. Despite an
overall trend to less extensive surgery, interest in the
detection and treatment of internal mammary nodes
has resurfaced in recent years. Two large random-
ized trials published in the New England Journal of
Medicine [37,38] demonstrated a survival benefit to
postmastectomy chest wall and regional nodal
radiation that included the internal mammary
nodes. The relative contribution of internal mam-
mary treatment, however, remains unclear.
FDG-PET can contribute in significant ways to the
clinical management and radiation planning of
patients who have suspected loco-regional recur-
rences. The most common sites of loco-regional
recurrence among patients treated with mastec-
tomy, axillary node dissection, or radiation therapy
are the chest wall and supraclavicular nodes. Be-
cause it provides biological and functional in-
formation, FDG-PET often is complementary to
conventional staging methods such as physical ex-
amination or cross-sectional imaging (CT or MR),
which rely more on changes in morphology to
detect disease recurrence. This is particularly true
in the evaluation of anatomic regions that have
been previously treated by surgery or radiation,
where the discrimination between post-treatment
scar and recurrent tumor is often problematic. Bra-
chial plexopathy, for example, is difficult to assess
on conventional anatomical imaging. FDG-PET
and MRI were compared in 10 patients who had
clinical findings suggestive of breast cancer me-
tastases [39]. Out of 9 patients who had local-
regional breast cancer metastases, MRI was positive
in 5 patients and indeterminate in 4 patients,
whereas FDG-PET was positive in all patients. Simi-
lar results were found in 19 breast cancer patients
who had symptoms referable to the brachial plexus,
and it was concluded that FDG-PET may particu-
larly be useful in distinguishing between radiation-
induced and metastatic plexopathy [40]. Van Oost
and colleagues [25] recently corroborated the need
for a more sensitive staging tool in patients who
have first-episode loco-regional recurrence. In their
study of 175 patients, they found that 16% had
distant metastases at the time of loco-regional recur-
rence, and 24% developed distant metastases within
18 months of confirmation of recurrent disease.
They estimated that FDG-PET would upstage and
likely change the therapeutic plan in up to 29%
of patients who have negative conventional stag-
ing studies.
F-18 fluorodeoxyglucose-positron emission
tomography/computed tomography for
radiation treatment planning
Combining two imaging devices, PET and CT, into
one integrated PET/CT scanner has substantially
broadened the clinical use of metabolic PET imag-
ing, specifically in oncology, and particularly for
radiation treatment planning [41–46]. PET/CT is
unique because it allows the acquisition of spatially
and temporally registered PET and CT data in one
imaging procedure, providing combined anatomi-
cal and functional imaging information that allows
accurate tissue characterization and determination
of the exact localization and extent of tumor tissue.
The diagnostic accuracy of PET/CT is generally
higher compared with either imaging procedure
performed separately, because of the exact ana-
tomical correlation of abnormal FDG uptake and
the functional correlation of suspected morphologi-
cal abnormalities [47]. The accuracy of FDG-PET/CT
is specifically higher in the neck and abdomen,
because it distinguishes physiological from patho-
logical uptake by confirmation of metabolic abnor-
mality with anatomic correlation. Several reports
indicate a distinct advantage of PET/CT in breast
cancer as well [47].
Fused FDG-PET and CT images provide radiation
oncologists with two pieces of critical information
with a single study: the extent of viable tumor, and
its exact location. Initial studies in patients who
had various tumor types have confirmed that
using FDG-PET/CT both in pretreatment planning
and in follow-up evaluations has a significant
impact on radiotherapy management in up to
56% of patients [46]. In 24 patients who were
planned with 3D conformal radiation therapy for
lung cancer, FDG-PET clearly altered the radiation
therapy volume in 14 (58%) patients. FDG-PET
also helped to distinguish tumor from atelectasis
in 3 patients. Unsuspected nodal disease was de-
tected in 10 patients, and 1 patient had a separate
tumor focus detected within the same lobe of the
lung [45]. Similarly, Heron and colleagues [42]
showed that in head and neck cancer, FDG-PET/CT
simulation provided valuable information that re-
sulted in greater delineation of normal tissues from

tumor-bearing areas at high risk for recurrence. In
this study of 21 consecutive patients, the FDG-PET
portion of PET/CT demonstrated the primary tu-
mor in all cases, whereas the coregistered CT did
not delineate the primary in three cases. In 8 pa-
tients, additional areas of disease were identified by
FDG-PET. Because of the often poor anatomical
delineation of head and neck tumors, the volumes
for the primaries were significantly larger on CT
treatment planning than on FDG-PET, which pro-
vided more precise metabolic definition of the
tumor extent. In contrast, the volumes for nodal
regions were smaller on CT because FDG-PET de-
monstrated tumor outside the CT-defined target
volumes. The additional metabolic information
from FDG-PET resulted in a more accurate charac-
terization of the extent of disease for both primary
tumors and nodal regions, with the potential clini-
cal implication of improved radiation treatment.
Ciernik and coworkers [46] investigated the useful-
ness of FDG-PET/CT imaging for target volume
definition in 39 patients who had solid tumors.
The gross tumor volume increased by 25% or more
because of FDG-PET in 17% of cases who had head-
and-neck and lung cancer, and in 33% of cases who
had cancer of the pelvis. The gross tumor volume
was reduced 25% or more in 33% of patients who
had head-and-neck cancer, in 67% who had lung
cancer, and in 19% who had cancer of the pelvis.
Overall, in 56% of cases, gross tumor volume delin-
eation was changed significantly if information from
metabolic PET imaging was used in the planning
process. The modification of the gross tumor volume
translated into altered planning target volume (PTV)
changes exceeding 20% in 46% of cases.
An important implication from improved target
definition is avoiding geographic misses and unnec-
essary exposure of nontarget tissue to radiation. In
addition, improved target definition may also
allow dose escalation in 3D and IMRT. The authors
found a close correlation between the metabolic
activity on pretreatment FDG-PET and the sub-
sequent risk of treatment failure (unpublished
data, 2005). If this observation is confirmed in
larger trials, the intensity of FDG uptake in tumor
tissue could guide and tailor the radiation dose
delivered by IMRT, and therefore improve radiation
treatment outcomes.
Current limitations of F-18
fluorodeoxyglucose-positron emission
tomography and positron emission
tomography/computed tomography for
radiation therapy planning
There are important differences between the cur-
rently used anatomical-imaging–based radiation
FDG-PET and PET/CT Radiation Therapy Planning 45
treatment planning and potential FDG-PET treat-
ment planning. Although FDG-PET offers unique
metabolic information about the tumor activity,
the limited spatial resolution of PET compared
with CT or MRI may not characterize small lesions
with sufficient accuracy. Current PET technologies
do not allow identifying micrometastases, which
limits its use for narrowing the treatment volume
in lymph nodes areas at risk. The specificity of
FDG-PET is in the range of 80% to 95%, and gen-
erally higher than that of CT or MRI.; however,
the uptake of FDG is not specific for tumor tissue,
and granulomatous or acute inflammatory pro-
cesses can result in false-positive PET findings. An
important limitation of FDG-PET in radiation treat-
ment planning is the precise delineation of tumor
tissue based on the PET images. Unlike anatomical
imaging, the size of metabolic abnormalities var-
ies depending on the scaling of the PET display.
Whereas tumors often have well-defined ana-
tomic borders on CT images, the edges of tumors
on FDG-PET imaging appear indistinct to the con-
touring physician. Some have arbitrarily defined
the FDG-avid volume as the region encompassed
by the 40% to 50% [45] intensity level relative
to the tumor maximum, whereas others have nor-
malized to the FDG uptake in the liver without
background subtraction [42]; hence the FDG-avid
volume may have significant interobserver variabil-
ity. The use of standardized uptake value (SUV)
normalized parametric images may help to over-
come this problem. Major limitations of FDG-PET
for radiation treatment planning are the limited
anatomical landmarks, which are necessary for ex-
act tumor target definition on the treatment plan-
ning system.
As discussed above, the use of coregistered
FDG-PET and CT images obtained from combined
PET/CT scanners may greatly improve the use of
PET for radiation treatment planning, by provid-
ing the metabolic tumor activity and the precise
anatomical information. The sequential nature of
the PET/CT acquisition does not completely pre-
vent misregistration between the PET and the
CT portion of the examination. In most PET/CT
scanners, a CT scan is acquired first from the base
of the skull though the pelvis, followed within
approximately 20 to 60 seconds by PET data acqui-
sition. Because a PET emission scan takes between
3 and 5 minutes per bed position (which is ap-
proximately 15–20 cm, depending on the scanner
manufacturer) the PET images represent an average
over the period of data acquisition, whereas as
the CT represents a snapshot from a fraction of
a second. Misregistration of tumors on the PET
compared with the CT portion of PET/CT is com-
mon because of respiratory motion, specifically at
46 Heron et al
the base of the lungs and in the upper abdomen
(liver). In breast cancer, differences in the breath-
ing cycle can cause misregistration, specifically of
tumor lesions in the breast, chest wall, and internal
mammary lymph node chain. Different filling of
the renal collecting system and the urinary bladder
are also frequently observed because of the sequen-
tial imaging of CT and PET. Finally, patient move-
ment may occur between the PET and the CT
portion, specifically in the head and neck region.
Heron and colleagues [42] have used a thermoplas-
tic mask for radiation treatment planning with
FDG-PET/CT of head and beck cancers, and suc-
cessfully immobilized patients to improve the
precise coregistration between PET and CT. Laser-
assisted localization for treatment planning is often
used for current CT-based radiation treatment plan-
ning, and might also be necessary for PET/CT radia-
tion treatment planning.
Radiosurgery treatment planning and
positron emission tomography/computed
tomography
There are several techniques being developed to
extend radiosurgery from the brain to the body.
The CyberKnife (Accuray, Sunnyvale, California) sys-
tem [48] for example, consists of a lightweight linear
accelerator mounted on a robotic arm with 1-mm
spatial accuracy. Real-time radiograph imaging track-
ing allows for correction of patient movement by
bringing the radiation beam into alignment with
the observed position of the treatment target. An
important advantage of this type of radiosurgery
is that each beam is delivered independently, with-
out a fixed isocenter. Because of the spatial preci-
sion with which radiation can be delivered, it is
feasible to administer a tumoricidal radiation dose
in a single outpatient treatment. Radiosurgery is
specifically helpful in spinal metastases from breast
and other cancers, but can potentially be used for
any tumor localization within the body [49].
Because of the high precision and the steep decline
in delivered tumor dose, extension beyond the ana-
tomical abnormality would lead to underdosing of
the tumor.
The major potential benefits of radiosurgery are
the short treatment time in an outpatient setting,
with rapid recovery and good symptomatic
response. Tumor target definition and dose plan-
ning are currently based on CT, however, which
frequently provides poor delineation of tumor
involvement, specifically in bone. When possible,
MRI is being used to more accurately define the
extension of tumor tissue; however, in previously
irradiated areas, MRI frequently demonstrates un-
specific increased signal intensity in the bone mar-
row, limiting the definition of viable tumor.
The potential advantage of FDG-PET and PET/CT
is in providing more accurate treatment planning
compared with CT and MRI alone. This could result
in improved local control rates after radiosurgery,
and reduce subsequent tumor recurrence rates.
There is only very little information so far describ-
ing the role of FDG-PET and PET/CT for CyberKnife
treatment planning, however. Prospective studies
are needed comparing volumetric dose planning
for target definition and radiosurgical treatment
planning using FDG-PET and PET/CT with cur-
rently used CT and MRI. For potential clinical ap-
plication, it is important to demonstrate that the
metabolic information is different from the ana-
tomical tumor delineation, and provides specific
additional information. It has also to be shown
that the ability to tailor radiation doses to a specific
metabolic activity, as opposed to using conven-
tional doses based upon histology alone, results
in improved local control. It is also important to
determine specific features of FDG-PET and PET/CT
that would predict local failure to standard radio-
surgical dose prescriptions.
Future perspectives
Preliminary data provide early proof of the princi-
ple that multimodality imaging can facilitate indi-
vidualized treatment by better delineation of target.
In theory, better delineation of the target should
mean better tumor coverage, and hence better local
control. Better target delineation will not help if
we are missing the target because it moves as a
result of physiological or other mechanical forces.
Future studies are needed to integrate respiratory
gating with PET/CT scanning to account for target
motion in radiation treatment planning and deliv-
ery. The use of 4D PET/CT can correct for respira-
tory motion artifacts seen in conventional PET/CT
imaging. It has potential to reduce smearing and
improve the accuracy in PET/CT coregistration.
The metabolically active disease seen by PET can
be treated with escalated dose to improve chances
of local control. Prospective trials are needed to
evaluate and validate the role of dose escalation
in this setting using IMRT-based technique. Besides,
the studies performed to date using PET/CT in ra-
diation planning have only shown that treatment
volume changes significantly with the use of func-
tional imaging. It has yet to be determined whether
the PET-defined target volume leads to improved
outcomes in terms of local control, survival, and
reduced toxicities.
PET/CT also has application in assessment of treat-
ment response. The utility of post-therapy FDG-PET

to monitor tumor response also has been evalu-
ated in several tumors, including lymphoma,
breast, cervix, and colorectal cancers [50–53]. There
are also encouraging results available for treatment
evaluation of radiation therapy. The concept of as-
sessment and prediction of treatment response by
sequential FDG-PET imaging needs to be validated
in larger studies to help in early identification of
treatment failure. This can lead to early intervention
with potential for salvage therapy, which might im-
prove outcome.
From the authors’ perspective, there is no doubt
that FDG-PET/CT will become an important part of
modern radiation treatment planning in the near
future; however, it should be emphasized that cur-
rently, FDG-PET can only be used as complemen-
tary imaging modality for detecting disease not
identified by CT. The routine use of FDG-PET or
PET/CT for radiation treatment planning requires a
thorough and careful evaluation in large prospec-
tive studies as it relates to the promise of improving
local control or increasing disease-free and over-
all survival.
Summary
PET-CT based imaging is a valuable tool, and one
of the most useful tools in the staging and restaging
of breast cancer, especially in patients who have
recurrent or locally advanced breast cancer. Its great-
est clinical applications are in detection and defini-
tion of extent of recurrent or metastatic disease, and
in monitoring response to therapy. The role of PET-
CT in radiation treatment planning for breast cancer
is still in the nascent stages, lacking controlled ran-
domized trials data; however, the potential to
improve radiation treatment planning by allowing
for the tailoring of comprehensive radiation portals,
particularly for locally advanced breast cancer,
makes this one of the most promising tools are we
enter the era of image-guided radiation therapy.
References
[1] Bucci MK, Bevan A, Roach 3rd M. Advances in
radiation therapy: conventional to 3D, to IMRT,
to 4D, and beyond. CA Cancer J Clin 2005;55(2):
117–34.
[2] Jansen EP, Dewit LG, van Herk M, et al. Target
volumes in radiotherapy for high-grade malig-
nant glioma of the brain. Radiother Oncol 2000;
56(2):151–6.
[3] Quon A, Gambhir SS. FDG-PET and beyond:
molecular breast cancer imaging. J Clin Oncol
2005;23(8):1664–73.
[4] Greenlee RT, Murray T, Bolden S. Cancer sta-
tistics, 2000. CA Cancer J Clin 2000;50(1):7–33.
FDG-PET and PET/CT Radiation Therapy Planning 47
[5] Recht A, Edge SB, Solin LJ, American Society of
Clinical Oncology. Postmastectomy radiother-
apy: clinical practice guidelines of the American
Society of Clinical Oncology. J Clin Oncol 2001;
19(5):1539–69.
[6] Jacobson JA, Danforth DN, Cowan KH, et al.
Ten-year results of a comparison of conservation
with mastectomy in the treatment of Stage I and
II breast cancer. N Engl J Med 1995;332(14):
907–11.
[7] van Dongen JA, Bartelink H, Fentiman IS, et al.
Randomized clinical trial to assess the value of
breast-conserving therapy in Stage I and II breast
cancer, EORTC 10801 trial. J Natl Cancer Inst
Monogr 1992;11:15–8.
[8] Sarrazin D, Le MG, Arriagada R, et al. Ten-year
results of a randomized trial comparing a con-
servative treatment to mastectomy in early breast
cancer. Radiother Oncol 1989;14(3):177–84.
[9] Blichert-Toft M, Rose C, Anderson JA, et al.
Danish randomized trial comparing breast con-
servation therapy with mastectomy. Six years of
life-table analysis. J Natl Cancer Inst Monogr
1992;11:19–25.
[10] Veronesi U, Banfi A, Salvadori B, et al. Breast
conservation is the treatment of choice in small
breast cancer. Long-term results of a randomized
trial. Eur J Cancer 1990;26(6):668–70.
[11] Fisher B, Anderson S, Redmond C, et al.
Reanalysis and results after 12 years of follow-
up in a randomized clinical trial comparing total
mastectomy with lumpectomy with or without
irradiation in the treatment of breast cancer.
N Engl J Med 1995;333(22):1456–61.
[12] Cuzick J, Stewart H, Rutqvist L, et al. Cause-
specific mortality in long-term survivors of breast
cancer who participated in trials of radiotherapy.
J Clin Oncol 1994;12(3):447–53.
[13] Gao X, Fisher S, Emami B. Risk of second pri-
mary cancer in the contralateral breast in women
treated for early stage breast cancer A population
based study. Int J Radiat Oncol Biol Phys 2003;
56:1038–45.
[14] Bhatnagar AK, Brandner E, Sonnik D, et al.
Intensity-modulated radiation therapy (IMRT)
reduces the dose to the contralateral breast when
compared to conventional tangential fields for
primary breast irradiation: Initial Report. Cancer J
2004;10(6):381–5.
[15] Schirrmeister H, Kuhn T, Guhlmann A. [F-18]-
2-deoxy-2-fluoro-D-glucose PET in the preopera-
tive staging of breast cancer: comparison with the
standard staging procedures. Eur J Nucl Med
2001;28:351–8.
[16] Adler LP, Crowe JP, Al-Kaisi NK. Evaluation
of breast masses and axillary lymph nodes with
[F-18]-2-deoxy-2-fluoro-D-glucose PET. Radiology
1993;187:743–50.
[17] Avril N, Rose CA, Schelling M. Breast imaging
with positron emission tomography and [fluo-
rine-18]-fluorodeoxyglucose: use and limitations.
J Clin Oncol 2000;18:3495–502.
48 Heron et al
[18] Adler LP, Faulhaber PF, Schnur KC. Axillary
lymph node metastases: screening with [F-18]-2-
deoxy-2-fluoro-D- glucose (FDG) PET. Radiology
1997;203:323–7.
[19] Avril N, Dose J, Janicke F. Assessment of axil-
lary lymph node involvement in breast cancer
patients with positron emission tomography
using radiolabeled 2-[fluorine-18]-fluoro-
2-deoxy-D-glucose. J Natl Cancer Inst 1996;88:
1204–9.
[20] Wahl RL, Siegel BA, Coleman RE. Prospective
multi-center study of axillary nodal staging with
FDG positron emission tomography in breast
cancer. J Clin Oncol 2004;22:277–85.
[21] Lovrics PJ, Chen V, Coates G. A prospective
evaluation of positron emission tomography
scanning, sentinel lymph node biopsy, and stan-
dard axillary dissection for axillary staging in
patients with early stage breast cancer. Ann Surg
Oncol 2004;11:846–53.
[22] Yang JH, Nam SJ, Lee TS. Comparison of
intraoperative frozen section analysis of sentinel
node with preoperative positron emission tomog-
raphy in the diagnosis of axillary lymph node
status in breast cancer patients. Jpn J Clin Oncol
2001;31:1–6.
[23] Wahl RL, Siegel BA, Coleman RE. Prospective
multi-center study of axillary nodal staging with
FDG positron emission tomography in breast
cancer. J Clin Oncol 2004;22:277–85.
[24] Eubank WB, Mankoff DA, Takasugi J. 18Fluoro-
deoxyglucose positron emission tomography to
detect mediastinal or internal mammary metas-
tases in breast cancer. J Clin Oncol 2001;19:
3516–23.
[25] van Oost FJ, van der Hoeven JJ, Hoekstra OS.
Staging in patients with locoregionally recurrent
breast cancer: current practice and prospects for
positron emission tomography. Eur J Cancer
2004;40:1545–53.
[26] Dose J, Bleckmann C, Bachmann S, et al.
Comparison of fluorodeoxyglucose positron
emission tomography and ‘‘conventional diag-
nostic procedures’’ for the detection of distant
metastases in breast cancer patients. Nucl Med
Commun 2002;23(9):857–64.
[27] Heinisch M, Gallowitsch HJ, Mikosch P, et al.
Comparison of FDG-PET and dynamic contrast-
enhanced MRI in the evaluation of suggestive
breast lesions. Breast 2003;12(1):17–22.
[28] Cook GJ, Houston S, Rubens R, et al. Detection
of bone metastases in breast cancer by 18FDG
PET: differing metabolic activity in osteoblastic
and osteolytic lesions. J Clin Oncol 1998;16:
3375–9.
[29] Yang SN, Liang JA, Lin FJ, et al. Comparing
whole body (18)F-2-deoxyglucose positron emis-
sion tomography and technetium-99m methy-
lene diphosphonate bone scan to detect bone
metastases in patients with breast cancer. J Can-
cer Res Clin Oncol 2002;128:325–8.
[30] Pierce LJ, Butler JB, Martel MK, et al. Postmas-
tectomy radiotherapy of the chest wall: dosi-
metric comparison of common techniques. Int J
Radiat Oncol Biol Phys 2002;52(5):1220–30.
[31] Cuzick J, Stewart H, Peto R, et al. Overview
of randomized trials of postoperative adjuvant
radiotherapy in breast cancer. Cancer Treat Rep
1987;71(1):15–29.
[32] Rietzel E, Chen GT, Choi NC, et al. Four-
dimensional image-based treatment planning:
target volume segmentation and dose calculation
in the presence of respiratory motion. Int J Radiat
Oncol Biol Phys 2005;61(5):1535–50.
[33] Lacour J, Le M, Caceres E, et al. Radical mas-
tectomy versus radical mastectomy plus inter-
nal mammary dissection. Ten year results of an
international cooperative trial in breast cancer.
Cancer 1983;51:1941–3.
[34] Veronesi U, Valagussa P. Inefficacy of internal
mammary node dissection in breast cancer sur-
gery. Cancer 1981;47:170–5.
[35] Lacour J, Le MG, Hill C. Is it useful to remove
internal mammary nodes in operable breast can-
cer? Eur J Surg Oncol 1987;13:309–14.
[36] Lohrisch C, Jackson J, Jones A, et al. Relationship
between tumor location and relapse in 6781
women with early invasive breast cancer. J Clin
Oncol 2000;18:2828–35.
[37] Overgaard M, Hansen PS, Overgaard J, et al. Post-
operative radiotherapy in high-risk premenopausal
women with breast cancer who receive adjuvant
chemotherapy. N Engl J Med 1997;337:949–55.
[38] Ragaz J, Jackson SM, Le N, et al. Adjuvant
radiotherapy and chemotherapy in node-positive
premenopausal women with breast cancer.
N Engl J Med 1977;377:956–62.
[39] Hathaway PB, Mankoff DA, Maravilla KR, et al.
Value of combined FDG PET and MR imaging
in the evaluation of suspected recurrent local-
regional breast cancer: preliminary experience.
Radiology 1999;210(3):807–14.
[40] Ahmad A, Barrington S, Maisey M, et al. Use of
positron emission tomography in evaluation of
brachial plexopathy in breast cancer patients. Br J
Cancer 1999;79(3–4):478–82.
[41] Esthappan J, Mutic S, Malyapa RS, et al. Treat-
ment planning guidelines regarding the use of
CT/PET-guided IMRT for cervical carcinoma with
positive paraaortic lymph nodes. Int J Radiat
Oncol Biol Phys 2004;58(4):1289–97.
[42] Heron DE, Andrade RS, Flickinger J, et al. Hybrid
PET-CT simulation for radiation treatment plan-
ning in head-and-neck cancers: a brief technical
report. Int J Radiat Oncol Biol Phys 2004;60(5):
1419–24.
[43] Calvo FA, Domper M, Matute R, et al. 18F-FDG
positron emission tomography staging and re-
staging in rectal cancer treated with preoperative
chemoradiation. Int J Radiat Oncol Biol Phys
2004;58(2):528–35.
[44] Erdi YE, Rosenzweig K, Erdi AK, et al. Radio-
therapy treatment planning for patients with
non-small cell lung cancer using positron emis-

sion tomography (PET). Radiother Oncol 2002;
62(1):51–60.
[45] Bradley J, Thorstad WL, Mutic S, et al. Impact of
FDG-PET on radiation therapy volume delinea-
tion in non-small-cell lung cancer. Int J Radiat
Oncol Biol Phys 2004;59(1):78–86.
[46] Ciernik IF, Dizendorf E, Baumert BG, et al.
Radiation treatment planning with an integrated
positron emission and computer tomography
(PET/CT): a feasibility study. Int J Radiat Oncol
Biol Phys 2003;57(3):853–63.
[47] Gorres GW, Steinert HC, von Schulthess GK.
PET and functional anatomic fusion imaging
in lung and breast cancers. Cancer J 2004;10(4):
251–61.
[48] Kuo JS, Yu C, Petrovich Z, et al. The CyberKnife
stereotactic radiosurgery system: description, in-
stallation, and an initial evaluation of use and
functionality. Neurosurgery 2003;53(5):1235–9.
[49] Rock JP, Ryu S, Yin FF, et al. The evolving role
FDG-PET and PET/CT Radiation Therapy Planning 49
of stereotactic radiosurgery and stereotactic ra-
diation therapy for patients with spine tumors.
J Neurooncol 2004;69(1–3):319–34.
[50] Grigsby P, Siegel BA, Dehdashti F, et al.
Posttherapy [18F] fluorodeoxyglucose positron
emission tomography in carcinoma of the cervix:
response and outcome. J Clin Oncol 2004;
22(11):2167–71.
[51] Lowe VJ, Boyd JH, Dunphy FR, et al. Surveillance
for recurrent head and neck cancer using posi-
tron emission tomography. J Clin Oncol 2000;
18(3):651–8.
[52] Kumar R, Xiu Y, Potenta S, et al. 18F-FDG PET for
evaluation of the treatment response in patients
with gastrointestinal tract lymphomas. J Nucl
Med 2004;45(11):1796–803.
[53] Larson SM, Schoder H, Yeung H. Positron
emission tomography/computerized tomogra-
phy functional imaging of esophageal and colo-
rectal cancer. Cancer J 2004;10(4):243–50.

Steroid Receptor Imaging in
Breast Cancer
Jean-Mathieu Beauregard, MD,
François Bénard, MD*
& Molecular biology of the estrogen and
progesterone receptors
& Transport and delivery of steroid hormones
in plasma
& Prognostic value of estrogen receptor and
progesterone receptor expression in
breast cancer
& Current treatment strategies for estrogen
receptor/progesterone receptor positive
breast cancers
Many breast cancers express high levels of estro-
gen receptors (ER) or progesterone receptors (PR),
which have both been linked with the pathogene-
sis of this disease. Estrogen acts as a stimulant on
the growth of ER-positive breast tumors, and the
reduction of this stimulatory effect remains the
key strategy in the treatment of ER- or PR-positive
breast cancers, either by reducing estrogen produc-
tion or by antagonizing the effects of estrogens on
the ER.
Hormone therapy has shown promise in the pre-
vention of breast cancer, and is routinely used as an
adjuvant treatment to complement surgery and
radiation therapy in the treatment of primary breast
cancer. About 50% to 60% of patients will respond
to hormone therapy [1]. Tamoxifen, a partial estrogen
antagonist, remains the most widely used pharmaco-
logical agent in routine clinical use, but aromatase
inhibitors and pure antiestrogens are increasingly
This work was supported by grant #015388 from the Canadian Breast Cancer Research Alliance.
Metabolic and Functional Imaging Center, Clinical Research Center, Centre Hospitalier Universitaire de
Sherbrooke, 3001 12th Avenue N., Sherbrooke, Québec, J1H 5N4 Canada
* Corresponding author.
E-mail address: francois.benard@usherbrooke.ca (F. Bénard).
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
51
POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 51–70
Éric Turcotte, MD,
& Role of estrogen receptor/progesterone
receptor expression imaging in the
management of breast cancer
& Radioligands for steroid receptor imaging
Estrogen receptors
Progesterone receptors
Clinical imaging studies with 18F-FES
& Summary
& References
being used as first- or second-line agents. Because
these drugs are relatively well-tolerated by patients,
with low toxicity compared with most chemotherapy
regimen, hormone therapy is recommended to all
postmenopausal women who have ER- or PR-positive
breast cancer.
Because ER and PR are overexpressed in 60% to
80% of breast cancers, these proteins were identi-
fied many years ago as excellent targets for imag-
ing breast cancer with radiopharmaceuticals. There
have been several efforts undertaken over the past
25 years to identify suitable agents for imaging
ER or PR expression in breast cancers, with the goal
of providing noninvasive characterization of breast
cancers to guide hormone therapy. Some highly
effective ligands have been identified and well-
characterized, but ER imaging is still not yet used as
a routine clinical procedure because of the limited
number of available clinical trials.
doi:10.1016/j.cpet.2005.09.006
52 Beauregard et al
Molecular biology of the estrogen and
progesterone receptors
ER are nuclear proteins, members of the nuclear
superfamily of transcription factors that mediate
cell response to estrogens (or antiestrogens) by
the regulation of the synthesis of specific RNAs
and proteins [2]. Estrogens regulate, via ER, the
development and function of the reproductive sys-
tem, and also affect many other organs. ER was first
identified in rat uterus in 1966 [3]. In 1986, the
human ER was cloned and sequenced from MCF-7
human cancer cells [4]. This ER is now known as
ERα, since the discovery in 1996 of a second ER
subtype, named ERβ [5].
ERα is 595 amino acids long (66 kDa), whereas
ERβ is 530 amino acids long. Other isoforms of
both ER subtypes have been described [6,7]. ERα
and ERβ have the same structure, composed of six
functional domains that are responsible for specific
functions. The A/B domain is the least conserved
between ERα and ERβ, and only ERα A/B domain
contains the activation function 1 (AF1). The C
domain, or DNA binding domain (DBD), is the
most conserved between the two subtypes, suggest-
ing they interact with the same genes. The D
domain is called ‘‘hinge region,’’ and contains the
nuclear localization signal. E/F domain (ligand bind-
ing domain, LBD) contains AF2, which is found in
both ER subtypes [2,6].
ER α and β have slightly different affinities for
their natural endogenous or exogenous ligands. Affi-
nity is often expressed as relative binding affinity
(RBA), which is 100 times the ratio between the
quantities of the reference ligand (eg, estradiol in
the case of ER) and the tested ligand needed to
displace 50% of the activity of radioactive reference
ligand (eg, tritiated estradiol) in a competitive bind-
ing assay on ER (either isolated receptors, cell ex-
tracts, or cell culture). The estradiol RBA is, by
default, 100 for both ER subtypes. The potency of
different ER ligands depends, in part, on their RBA.
The binding of a ligand to the ligand binding
domain (LBD) of ER causes dissociation of protec-
tion proteins (called heat shock proteins) from ER,
dimerization and conformational change of ER,
resulting in increased affinity of its DBD for the
estrogen response elements (ERE). ERE are short
DNA sequences located in the promoter region
on genes, near the transcriptional sequence, that
enhance the transcriptional potential of a gene
[2]. This is not sufficient to activate transcription,
which is the role of activation functions AF1 and
AF2. The activity of each AF is dependent on the
target gene promoter context, cellular environment,
and ligand [2,8]. ER DNA binding affinity and
transcriptional activation are also modulated by
the state of phosphorylation of ER, which is influ-
enced by ligand binding [2].
Ligands can be classified as pure agonists,
(eg, estradiol), pure antagonists (eg, fulvestrant)
or mixed agonist-antagonists, also called selective
ER modulators, or SERMs (eg, tamoxifen or raloxi-
fene). Pure agonist and antagonist ligands activate
or block (via conformational change in ER) both
AF1 and AF2, whereas SERMs have different effects
on each AF. The ability of a ligand to initiate or
block gene transcription in a specific cell type de-
pends on the dominance of ERα versus ERβ and
AF1 versus AF2. Initiation of transcription is much
more complex, because it involves the interaction
of many proteins that act as coregulators (general
transcription factors, coactivators, corepressors,
cointegrators, histone acetyltransferases, and his-
tone deacetylases), further refining this process
and making it even more cell-specific [8]. Acom-
plete review of these coregulators is beyond the
scope of this article. Moreover, other nonclassical
pathways of gene regulation by ER (at different
DNA sites and with other coactivators) and non-
genomic ER actions (involving membrane ER out-
side the nucleus) are studied, and these could provide
some explanations in drug-resistance mechanisms,
particularly in the case of selective ER modula-
tors [9].
Both ER subtypes are found in brain, cardiovas-
cular system, urogenital tract, and bone, in various
ratios. ERα predominates in the liver, whereas ERβ
is the main ER subtype in the colon [2,6]. In nor-
mal breast tissue, both ERα and ERβ are expressed,
but ERβ predominates. In premalignant lesions,
however, there is an increase in ERα, which pro-
motes cell proliferation under estrogen stimulation,
and a decrease in ERβ expression, which is asso-
ciated with lower proliferation and pathological
grade, and thought to downregulate ERα, acting
as a tumor suppressor [7,8,10]. Thus ERα is the
therapeutic target of choice in breast cancer. ERα
is routinely dosed in vitro by immunohistochem-
istry on breast cancer biopsy specimens for treat-
ment planning, as is discussed later. The clinical
value of ERβ, however, is not established [7,10].
PR are, like ER, members of the nuclear receptor
superfamily. Like ER, they are ligand-activated nu-
clear proteins that regulate expression of specific
subset of genes [8,11]. PR are found in the uterus
and many other organs, including the brain. One
of the main roles of progesterone is to prepare
the uterus for pregnancy. Two isoforms of PR are
known: Progesterone receptor A (PRA) and proges-
terone receptor B (PRB). They are encoded by the
same gene. PRB is a longer version of PRA contain-
ing 164 additional amino acids. Both subtypes are
normally found in the normal breast [8]. In malig-

nant transformation, there is an alteration of the
PRA:PRB ratio, and both PRA and PRB dominance
have been reported [8,11,12]. Loss of PR expression
in breast cancer is associated with more aggressive
tumors and worse prognosis [8].
Transport and delivery of steroid hormones
in plasma
In women, progestins (progesterone and 17-hydroxy-
progesterone) and estrogens (estrone, estradiol-
17β, and estriol) are synthesized from cholesterol,
mainly by the ovaries, and during pregnancy by the
placenta. These hormones, like other steroids, are
delivered to target organs by the blood. Because
they are lipophilic, they enter the cell and reach
the nucleus by passive diffusion. In blood, they are
bound to plasma proteins, mainly albumin and sex
hormone-binding globulin (SHBG). Sex steroids
bind with low affinity to albumin. Nevertheless,
this nonspecific binding is significant because of
the high concentration of albumin in plasma.
Like albumin, SHBG is synthesized mostly by the
liver. This glycoprotein plays a role in the transport
of sex steroids in plasma by allowing solubiliza-
tion of these lipophilic hormones and by protecting
them against metabolic inactivation [13]. SHBG
might also regulate the bioavailability of sex steroids,
because only the free (unbound) fraction is generally
considered available for cellular uptake [14].
More recently, a cell-membrane SHBG receptor
has been discovered. This receptor is thought to par-
ticipate in the delivery of estrogens into cells. Two
mechanisms have been proposed: (1) the steroid-
SHBG complex binds to the SHBG receptor and
then the steroid enters the cell [15], or (2) SHBG
binds to its receptor and then the free estrogen
binds to the SHBG-receptor complex [16]. There-
after, either the estrogen diffuses passively inside the
cell to reach ER in the nucleus, as mentioned above,
or alternatively, the estrogen-SHBG-receptor com-
plex is internalized. It is still not clear how SHBG
influences the delivery of steroids to their receptors.
It is currently difficult to ascertain whether the affin-
ity of estrogens and progestins for SHBG influ-
ences their biodistribution in humans. SHBG is
not expressed in small rodents used for preclinical
biodistribution studies of radiolabeled estrogens.
As it will be discussed later, this limits the extrapo-
lation of results from animal studies to humans.
Prognostic value of estrogen receptor and
progesterone receptor expression in breast
cancer
The hormone receptor status is one of the few well-
established prognostic indicators in breast cancer,
Steroid Receptor Imaging in Breast Cancer 53
along with the nodal status, tumor size, patient age,
and pathologic grade [17]. Although other prog-
nostic indicators, such as p53, HER2/neu status,
angiogenesis, and proliferation indices have also
shown potential, the value of these parameters
has not been as clearly established. Breast tumors
that express ER grow more slowly, have a lower
thymidine-labeling index, and typically have a more
differentiated phenotype than ER-negative tumors
[18]. The prevalence of ER-positive tumors increases
with patient age [19,20], and is significantly higher
in postmenopausal women (among whom 80%
of patients have ER positive tumors) than pre-
menopausal women (among whom approximately
50% of patients have ER-positive tumors) [17]. The
short-term recurrence rate is also much lower in
ER-positive than negative breast tumors.
Conversely, the ER status does not correlate with
the absence of nodal and distant metastases, and
some studies have questioned whether the long-
term survival of women who have ER-positive
tumors is really lower than those who have ER-
negative cancers [17]. Thus the presence of estrogen
receptors does not correlate with the metastatic
potential of breast tumors, but rather with their
growth rate. The predictive value of ER, therefore,
more accurately reflects the short-term outcome
of these patients rather than the absolute disease-
free survival.
The real value of measuring the expression of ste-
roid hormone receptors in breast cancers, howo-
ever, lies in their ability to predict a successful
response to hormonal therapy. A meta-analysis
published in 1998 of the available evidence from
55 studies that included 37 000 women [21]
showed that 5 years of adjuvant therapy with ta-
moxifen reduced recurrences by 47% in women
who had ER-positive primary tumors. No signifi-
cant benefit could be proven in women who had
ER-negative tumors.
Approximately 60% of previously untreated pa-
tients who have ER-positive tumors will respond
to hormonal therapy, whereas a response is ob-
served in only 5% to 10% of patients who have ER-
negative tumors [17].
Current treatment strategies for estrogen
receptor/progesterone receptor positive
breast cancers
Surgery and radiation therapy play important
roles in breast cancer treatment. Many reports have
shown that radiotherapy can significantly reduce
the locoregional recurrence rate, but does not sig-
nificantly improve the long-term survival of the pa-
tient [22,23]. Systemic adjuvant therapy, such as
chemotherapy and hormonal therapy, has been
54 Beauregard et al
proven useful to reduce the rate of recurrence of
breast cancers.
When compared with hormone receptor-negative
tumors, hormone receptor-positive breast cancers
exhibit stronger clinical responses to hormonal treat-
ment. Adjuvant tamoxifen prolonged both disease-
free and overall survival in patients who had
ER-rich tumors, but had little benefit in patients
who had ER-negative neoplasms [21]. According
to Harvey and colleagues [24], patients who have
breast cancers containing as few as 1% to 10% of
cells staining for ER respond to hormone therapy.
Patients who are ER-negative but PR-positive can
also benefit from tamoxifen, but the role of the
progesterone status in selecting therapy remains
controversial. In 2003, Bardou and coworkers [25]
showed that the combined measurement of ER and
PR is superior to ER alone in predicting benefit
from adjuvant hormone therapy, because the pre-
sence of PR enhances the ER predictive value in
both pre- and postmenopausal patients.
Tamoxifen is accepted as the standard treatment
for patients whose tumors express the estrogen or
progesterone receptors [26]—regardless of age, meno-
pausal status, axillary node involvement, or tumor
size [21,27,28]—to increase recurrence-free survival
and overall survival irrespective of whether che-
motherapy has been given [21]. According to trials,
5 years of tamoxifen treatment appeared supe-
rior to 2 years. The beneficial effect of 5 years of
adjuvant tamoxifen was demonstrated in both
receptor-positive postmenopausal patients [29] and
premenopausal patients under 50 years of age [21].
Currently, there are no data justifying the use of ta-
moxifen for a longer period than 5 years [21,30,31],
but trials are still ongoing to address this question.
Tamoxifen combined with chemotherapy further re-
duces the risk of disease recurrence, particularly in
premenopausal women [21,30].
According to Colleoni and colleagues [32], cyclo-
phosphamide, methotrexate and 5 fluorouracil
(CMF) combination chemotherapy given concur-
rently (early, delayed, or both) with tamoxifen was
more effective than tamoxifen alone for patients
who had node-positive, endocrine-responsive breast
cancer, supporting late administration of chemo-
therapy even after initiation of tamoxifen. In con-
trast, sequential CMF and tamoxifen for patients
who had node-negative, endocrine-responsive breast
cancer was ineffective [32]. Thus all patients who
have hormone-receptor positive tumors receiving
adjuvant chemotherapy should also receive tamoxi-
fen, but the optimal timing between chemotherapy
and tamoxifen (during or after chemotherapy) is still
under investigation [21,30].
Based on results from multiple large randomized
trials, postmenopausal women who have hormone-
receptor–positive breast cancer can also benefit from
aromatase inhibitors as adjuvant therapy. Anas-
trozole and letrozole are nonsteroidal aromatase
inhibitors compared with exemestane, which is a
steroidal aromatase inactivator. The main effect of
aromatase inhibitors is to decrease the concentra-
tion of circulating estrogen levels in postmeno-
pausal women. Aromatase inhibitors should not
be employed as monotherapy in premenopausal
women because estrogen suppression increases
gonadotrophin release by feedback mechanism, in
return increasing the production of estrogen by the
ovaries. Both anastrozole and letrozole have been
shown to be equivalent or superior to tamoxifen in
a variety of clinical parameters, including the treat-
ment of metastatic breast cancer [33,34]. In a ran-
domized unblinded trial comparing anastrozole
and letrozole in patients who had metastatic breast
cancer [35], there was no significant difference
between the two agents, even if letrozole was
known to have a more pronounced effect on re-
ducing estrogen levels. Letrozole was associated
with a higher overall response rate, but this was
not statistically significant. The two agents did not
differ significantly in terms of clinical benefit rate
or overall survival. Both aromatase inhibitors were
associated with significantly fewer endometrial can-
cers and venous and arterial vascular events com-
pared with tamoxifen [36,37]. Although this was
not statistically significant, both anastrozole and
letrozole were associated with a higher incidence
of cardiovascular disease [37,38].
Results of the Arimidex, Tamoxifen, Alone or
in Combination (ATAC) trial after completion of
5 years’ adjuvant treatment for breast cancer have
been published recently [39], and propose anastro-
zole for 5 years to be the preferred initial adju-
vant treatment over tamoxifen for postmenopausal
women having localized, hormone-receptor–positive
breast cancer. This recommendation is based on a
significantly increase disease-free survival (hazard ra-
tio [HR] 0.87), greater time to recurrence (HR 0.79),
greater time to distant recurrence (HR 0.86), and
reduced distant metastases (HR 0.86). Overall sur-
vival rates were similar in the two groups, however,
despite a death reduction by 12% in the anastro-
zole group that was not statistically significant.
Aromatase inhibitors have not been evaluated in
the adjuvant setting in women whose tumors lack
hormone receptors. The Canadian MA-17 trial [38]
is so far the only study that could demonstrate a
survival benefit for node-positive disease by extend-
ing 5 years adjuvant therapy of tamoxifen with
letrozole in ER-positive postmenopausal women
[38,40,41]. Postmenopausal women concluding
2 to 3 years of tamoxifen therapy may consider
crossover to an aromatase inhibitor, and should

plan on a total of 5 years of adjuvant endocrine ther-
apy. The optimal timing of transition between
tamoxifen to the aromatase inhibitor and the opti-
mal duration of aromatase inhibition are still under
investigation. According to the American Society of
Clinical Oncology (ASCO) [42], there are no data
on the use of tamoxifen after an aromatase inhibitor
in the adjuvant setting. Patients intolerant to aroma-
tase inhibitors should receive tamoxifen as adjuvant
therapy, and women who have hormone-receptor–
negative tumors should not receive adjuvant endo-
crine therapy.
Fulvestrant is one of a new type of ER antagonists
that bind to estrogen receptor monomers, inhibit
receptor dimerization [43], disable activating func-
tion AF1 and AF2 [44], reduce translocation of
receptors to the nucleus, and accelerate degrada-
tion of the ER, ultimately resulting in a reduction
in cellular ER [45,46]. This results in pure anti-
estrogenic effects [47], and unlike tamoxifen, ful-
vestrant is devoid of any known agonist activity
[46]. Fulvestrant has a steroidal structure that com-
petitively binds to the ER with an affinity much
greater than that of tamoxifen (~100 times) [46].
Many trials are ongoing to evaluate the clinical effi-
cacy of fulvestrant after recurrence or progression
on a nonsteroidal aromatase inhibitor. Current re-
sults suggest that fulvestrant is at least as effective as
anastrozole [48]. Fulvestrant has been approved for
clinical use in the United States for the treatment of
ER-positive metastatic breast cancer that fails to
respond to tamoxifen. The efficacy of fulvestrant in
sequential endocrine therapy and in combination
with other agents appears promising, and active inves-
tigations are ongoing to explore the clinical potential
of this medication.
New markers to predict response to tamoxifen are
continuously being evaluated, and several investi-
gators have focused their attention on HER-2/neu
[49–53]. In breast cell lines and in model tumor
systems, overexpression of the HER-2/neu gene has
been associated with increased mitogenesis, malig-
nant transformation, increased cell motility, inva-
sion, and metastasis [54]. In human breast cancer,
amplification of the HER-2/neu gene is found in
15% to 30% of primary invasive tumors. For pa-
tients who have both early and advanced disease,
overexpression of HER-2/neu correlates with either
relative resistance or adverse outcome after treat-
ment with hormonal therapy [55,56]. The relation-
ship between HER-2/neu expression and response
to chemotherapy in breast cancers appears to de-
pend on the type of drug administered. With adju-
vant CMF, the majority of studies showed a
reduced benefit in HER-2/neu–positive compared
with negative patients [56–58]; however, patients
who have cancers overexpressing HER-2/neu are
Steroid Receptor Imaging in Breast Cancer 55
likely to derive benefit from treatment with CMF-
based regimes compared with no treatment. Several
studies suggest that overexpression of this gene may
predict enhanced sensitivity to anthracycline-based
regimens in the adjuvant setting [55–57]. Thus the
available data suggest that patients who have HER-2/
neu–positive cancers are more likely to respond to
anthracycline regimens than HER-2/neu–negative.
According to the ASCO, the use of HER-2/neu to
decide whether to prescribe endocrine therapy either
in the adjuvant or metastatic setting is not recom-
mended. HER-2/neu may identify patients who par-
ticularly benefit from anthracycline-based adjuvant
therapy, but levels of HER-2/neu should not be
used to exclude patients from this type of treatment
[42]. For postmenopausal women who have breast
cancer overexpressing HER-2/neu, higher response
rates have been reported for aromatase inhibitors
compared with tamoxifen [59,60]; however, the
role of HER-2/neu status in selecting optimal endo-
crine therapy remains controversial and needs fur-
ther investigations.
Role of estrogen receptor/progesterone
receptor expression imaging in the
management of breast cancer
The variability of ER/PR measurements across labo-
ratories using radioligand binding assays on tissue
extracts, the heterogeneity of tumor sampling [61],
and the limitations of the technique to provide
accurate results using very small samples were all
important factors that motivated the development
of nuclear imaging to characterize the ER status of
breast cancers. The use of immunohistochemistry
to assess the steroid receptor status of breast cancers
has overcome many of the limitations of radioli-
gand binding assays, and has found widespread
acceptance. Because the presence of estrogen and
progesterone receptors is routinely assessed at pre-
sentation from small specimens, there is little jus-
tification to perform an imaging procedure that
would duplicate this information. Because up to
20% of postmenopausal and 50% of premeno-
pausal women who have breast cancer have tumors
that do not exhibit ER-positivity, radioligand-based
imaging methods would miss an unacceptably high
number of breast cancers if used for screening or
diagnostic purposes.
Although the status of ER and PR receptors in
primary breast cancers is usually well established at
the time of diagnosis with the routine use of immu-
nohistochemistry, this is not always the case at the
time of recurrence. A biopsy of a recurrent lesion is
always recommended, because the ER/PR status of
the recurrence can then be documented, and the
56 Beauregard et al
Fig. 1. (A–F ) Chemical structures of some estrogen-binding radiopharmaceuticals used in clinical studies. The
structure of the endogenous hormone, 17-ß-estradiol, is shown in A.
status of other antigens with important prognostic
and therapeutic value, such as HER2/neu, can also
be obtained on the specimen. Not all sites of recur-
rences are easily amenable to biopsy, however, and
heterogeneity of ER/PR expression across metastatic
sites could have strong prognostic implications.
Furthermore, the mechanisms of hormone therapy
resistance are multiple and complex, with loss of
expression of ER being just one of multiple path-
ways by which malignant breast cancer cells
develop resistance. Only about 20% to 40% of pa-
tients who have a primary tumor that expressed
ER will have lost ER expression at the time of re-
currence [62–64]. With resistance acquired during
hormone therapy, 28% of patients will lose ER ex-
pression in their tumors [65]. With the availability
of second-line agents with proven efficacy even
when tamoxifen has failed, the accurate assessment
of the status of ER expression gains even more
importance. A diagnostic test that can provide func-
tional information about the status of ER or PR
across all tumor sites could have profound prog-
nostic implications and influence the decision to
treat individual patients with hormone therapy or
chemotherapy. Fig. 1 illustrates several ER-binding
radiopharmaceuticals that have been proposed for
this purpose.
Radioligands for steroid receptor imaging
Estrogen receptors
Iodinated ligands
16αα-Iodoestradiol Hochberg synthesized 16α-
[125I]iodoestradiol in 1979 [66]. This compound
showed specific ER-mediated uptake in vivo, in
rat uterus. It thus became one of the first poten-
tial single photon-emitting ER imaging agent, be-
cause it could be labeled with either 131I or 123I [see
Fig. 1B]. 16α-[123I]iodoestradiol (123I-E2) could
bind to ER-positive tissues in animals, including
7,12-dimethylbenz[a]anthracene (DMBA)-induced
mammary tumors [67]. In a series of 21 patients
who had breast masses [68], 123I-E2 was positive in
7 of 9 patients who had ER-positive malignant tu-
mors, and negative in all 12 ER-negative tumors (in-
cluding 7 benign tumors). Kenady and coworkers
[69] reported uptake in chest wall tumors and in-
flammatory breast cancer. In another study [70], in
42 patients, sensitivity of 123I-E2 for ER+ tumors
 Steroid Receptor Imaging in Breast Cancer 57

was 66%. For imaging purpose, 131I had less desir- Van de Wiele and coworkers [98] imaged nine pa-
able physical characteristics than 123I, which ac- tients who had primary breast cancer using [123I]
counted for disappointing results obtained with iodomethyl-N,N-dimethyltamoxifen (123ITX) and
131
I-E2 [71]. Rapid metabolism of I-E2 appears to SPECT. 123ITX lacked sensitivity—only four pa-
prevent its optimal localization in ER-positive tu- tients out of six who had ER+/PR+ tumors had
mors [71,72]. positive 123ITX scintigraphy. The three other pa-
tients (two who had ER+/PR-, and one who had
Z-MIVE Ali and colleagues [73] reported synthe-
ER-/PR- tumors) had negative 123ITX studies. Sen-
sis of (17α,20Z)-[125I]iodoestradiol, which showed
sitivity was also low for detecting nodal and dis-
higher ER-mediated uptake in target tissue than its
tant metastasis.
20E isomer. They also reported that the addition
of a 11β-methoxy substituent to these isomers,
Tc-99m ligands
producing 11β-methoxy-(17α,20E/Z)-[125I]iodovi-
Because of its wider availability, lower cost, and
nylestradiol (E- and Z-MIVE, see Fig. 1C), which
convenient physical characteristics, Tc-99m (140 keV,
increased their in vivo specific uptake [74]. Both
6 hours half-life), would be a very desirable single-
E- and Z-MIVE were labeled with I-123 [75]. Ribeiro-
photon emitting label for ER imaging agents.
Barras and coworkers [76] chose E-MIVE for the
Skaddan and colleagues [99] described the syn-
first clinical study, but only 2 of their 4 patients
thesis of four estradiol complexes, labeled with
who had high ER content primary breast cancers
Tc-99m at position 7α, along with their bio-
had positive E-MIVE uptake on single photon emis-
distribution in rodents. These compounds had
sion computed tomography (SPECT) imaging. Rijks
poor ER-mediated uptake in target organs, along
and colleagues [77,78] preferred Z-MIVE over E-MIVE
with high nontarget organs uptake. Others reported
on the basis of their animal experiments. In 25 pa-
on Tc-99m labeled estrogens, including tamoxi-
tients imaged with Z-MIVE SPECT, there was 90%
fen conjugates [100,101]. Again, low uterus uptake
agreement (9/10 patients, 1 false-positive) between
was observed.
SPECT and ER status for primary breast cancer,
Single-photon emitting ER ligands are advanta-
and 82% agreement (9/11 patients, 1 false-positive,
geous because they can be readily used in any nu-
1 false-negative) in metastatic/recurrent breast can-
clear medicine department, using both planar and
cer [79]. The Sherbrooke group [80,81] achieved
SPECT imaging protocols; however, iodine-123, the
similar results (80% agreement, 2 false-positive)
current label of choice, remains expensive. With the
using Z-MIVE scintimammography in 10 patients
increasing availability of positron emission tomog-
who had primary breast cancer. Bennink and co-
raphy (PET) scanners in many major cancer centers,
workers [82] reported a sensitivity of 94% for planar
PET is an attractive alternative to SPECT because of
scintigraphy and 100% for SPECT imaging in
its superior spatial resolution and sensitivity over
25 patients who had primary breast cancer (includ-
traditional nuclear medicine techniques. Not sur-
ing 10 patients from reference [79]). Thus Z-MIVE
prisingly, several ligands were developed for ER
appears to be more sensitive than E-MIVE for ER
imaging with positron emitters.
imaging. Recently, it has been demonstrated that
Z-MIVE can predict response to hormonal therapy.
C-11 ligands
Among 21 patients who had clear Z-MIVE uptake in
Carbon-11 can be used to label molecules without
their metastatic lesions, the 17 patients who had
introducing a halogen or other foreign atoms, mak-
complete blockade of uptake 1 month after starting
ing their structure closer, if not identical, to en-
tamoxifen had a longer progression-free survival than
dogenous ligands; however, the radiosynthesis is
the 4 others [83].
generally more difficult because the labeling reaction
Other estrogens labeled with single photon is easily contaminated with cold carbon, reducing
emitting halogens Ali and colleagues [84–88] and specific activity. Because specific activity generally
others [89,90] reported synthesis and biologi- decreases at the same rate as activity decays, this
cal activity study of many other iodine and bro- may further reduce the sensitivity for imaging satu-
mine radiolabeled steroidal estrogens. Of these, rable receptor systems, like ER. The short half-life
only 16α-[77Br]bromoestradiol has been studied of C-11 (20 minutes) also prevents the distribution
in humans, but the images suffered from the poor of this radionuclide to distant sites. Nevertheless,
imaging characteristics of bromine-77 [91,92]. Radio- several reports C-11 labeled estrogens have been
iodinated tamoxifen analogs have been synthesized published [102–104]. These tracers showed poor to
[93–98]. Imaging breast cancer in humans with [131I] moderate receptor-mediated target organ uptake and
iodotamoxifen was unsuccessful [93]. The high lipo- target/nontarget ratios in biodistribution studies in
philicity of this molecule resulted in high nonspe- rodents, however. Tamoxifen was also labeled with
cific uptake and reduced target-to-nontarget ratio. C-11, but specific activity was low [105].
58 Beauregard et al

F-18 ligands sulfate as precursor [111], and methods have

Fluorine-18 is a useful radioisotope for PET ER been published to automate this synthesis [112].
imaging: it has a convenient half-life (allowing Several groups now use this synthesis as a rou-
off-site radiosynthesis), and its presence on the tine procedure.
ligands is well-tolerated by the ER (particularly at
16β-fluoromoxestrol To improve the uptake char-
the 16α position). It is relatively easy to radio-
acteristics of FES, many substituted FES derivatives
fluorinate estrogens with specific activities over
have been synthesized. 11β-ethyl, 11β-methoxy and
1000 Ci/mmol.
17α-ethynyl substitutions were synthesized in both
16α- [113] and 16β-FES [114]. These derivatives have
16α-[18F]fluoroestradiol In 1984, Kiesewetter and
been studied in vitro (ER and SHBG RBA) and in
coworkers [106] published the preparation of
vivo (biodistribution in Sprague-Dawley female
16α-[18F]fluoroestradiol (18F-FES, see Fig. 1D),
rats). Among these compounds, 17α-ethynyl-11β-
along with other fluorine-18 labeled estrogens
methoxy-16β-fluoroestradiol (16β-fluoromoxestrol,
(16β-[18F]fluoroestradiol, 1-[18F]fluoropentestrol,
βFMOX) was the most promising ligand (see
1-[18F]fluorohexestrol). Their RBAs for ER were 80,
Fig. 1E), with a very good ER relative binding affinity
30, 129 ,and 127, respectively. All these compounds
(78 at 25°C), very low SHBG relative binding affinity
showed ER mediated uptake in vivo in immature
(0.037), high uterus uptake (18.26 ± 7.850% I.D./g),
female rats, manifested by uterus uptake that can
and high uterus/blood and uterus/muscle ratios
be blocked by excess estradiol co-injection. In this
(66.26 ± 9.38 and 24.31 ± 6.14, respectively)
biodistribution study, 18F-FES showed the high-
[114,115]. These encouraging results were attri-
est selectivity for ER in vivo, manifested by the high-
buted to the greater metabolic stability of βFMOX
est uterus/blood and uterus/nontarget ratios over
over 16α-FES in vivo, contributing to high specific up-
2 hours (39 ± 16 and 28 ± 4.8, respectively, 1 hour
take and lower background activity from metabolites.
postinjection of 50 μCi of 18F-FES). 18F-FES also
These improved characteristics over 16α-FES led
had the highest blood clearance [106]. Good 18F-FES
to the first clinical study with βFMOX in 12 patients
uptake in DMBA-induced mammary tumors was
who had breast masses [116]. Only 3 of these
shown, although this uptake did not correlate
patients had biopsy-confirmed ER-positive malig-
well with the ER content of these tumors [107].
nant tumors. Unfortunately, βFMOX-PET was nega-
Mintun and colleagues [108] published the re-
tive in these 3 cases. These disappointing results
sults of the first clinical study with 18F-FES in 1988.
were attributed to the very low affinity of βFMOX
Thirteen patients who had breast masses under-
for SHBG. This supports the hypothesis that SHBG
went 18F-FES PET before biopsy. Except for one
favors steroid binding to ER in vivo in humans, by
benign lesion and one mass that was out of field
protecting estrogens from metabolism [13] and
of view, 18F-FES PET showed elevated uptake in all
facilitating their uptake via a cell-surface SHBG
other breast masses proven to be breast carcinoma,
receptor [16]. The authors compared FES and
and in some axillary metastases as well [108]. Since
βFMOX metabolic rates in rat, primate and hu-
then, 18F-FES has been considered the ER imaging
man isolated females hepatocytes. Results were
agent of choice.
consistent with a protective effect of sex hormone-
Mankoff and coworkers [109] established the ra-
binding globulin (SHBG) [116].
diation dosimetry in 18F-FES PET studies of 49 pa-
tients. The effective dose equivalent was 0.022 mSv/ 4-fluoro-11β-methoxy-fluoroestradiol In an effort
MBq, which would give 4.8 mSv for a typical activ- to improve metabolic stability of FES, the authors’
ity of 220 MBq (6 mCi). The critical organ was the group described the synthesis of a new series of
liver (0.13 mGy/MBq). Thus the radiation dose is fluorine-18 labeled estrogens, the 2- and 4-fluoro
reasonable and well below generally accepted limits derivatives of 16α-18F-FES, with or without a 11β-
for research studies. methoxy group [117]. Biodistribution studies in
Metabolism studies of 18F-FES in both animals immature Sprague-Dawley female rats identified
[107] and humans [110] showed that 18F-FES is 4-fluoro-11β-methoxy-fluoroestradiol (4F-MFES) as
metabolized rapidly, leaving conjugated circulating the most promising of these potential ER imag-
labeled metabolites that do not bind to ER, increas- ing agents [118]. Despite a more than fourfold
ing background activity. This rapid metabolism reduction of 4F-MFES relative binding affinity for
is also manifested by a rapid blood clearance of ER when compared with 18F-FES (on both isolated
18
F-FES, leaving only 20% of unmetabolized ERα and MCF7 cells), 4F-MFES achieved the highest
18
F-FES 20 minutes after injection [110], thus po- uterus uptake (14.66 ± 3.22% ID/g) and highest
tentially limiting the maximum achievable uptake uterus/blood and uterus/nontarget ratios (136.32 ±
in ER-containing cells. The F-18 labeling method 22.00 and 47.65 ± 7.23, respectively) at 1 hour
of 18F-FES has been improved by using a cyclic postinjection [118]. The performance of 4F-MFES
 Table 1: Comparative characteristics of selected radiolabeled estrogens
RBA (estradiol = 100) Uterus uptakeb Ratios at 1 h
a d
ER SHBG 1h 1 h (+ E2 ) (uterus/blood) (uterus/nontargete) Reference
f
FES 80 – 4.67 ± 1.50 0.61 ± 0.12 39 ± 16 28 ± 4.8 [106]
FES 107.7 ± 26 4.5 ± 1 11.75 ± 3.20 0.70 ± 0.11 66.41 ± 20.44 36.10 ± 6.85 [118]
FES 54 9.47 – – – – [115]
βFMOX 78 0.037 18.26 ± 6.80 2.548 ± 0.593 66.27 ± 9.38 24.31 ± 6.14g [115]
4F-MFES 24.4 ± 5 <0.0017 14.66 ± 3.22 1.14 ± 0.29 136.32 ± 22.00 47.65 ± 7.23 [118]
Z-MIVE 32.0 ± 6.6 – 17.49 (15)c 1.87 (17)c 22.71 13.81 [74]

Abbreviations: CV, coefficient of variation; ID/g, percent injected dose per gram of tissue; SD, standard deviation.
a
at 25°C or room temperature.
b
% ID/g ± SD 1 hour after 50 μCi injection in immature female Sprague-Dawley rats.
c
% ID/g (% CV) 1 hour after 3 μCi injection in immature female Fischer rats.
d
coinjection of excess unlabeled estradiol.
e
nontarget organs: lung, spleen and muscle.
f
nontarget organs: oesophagus, lung, spleen and muscle.
g
nontarget organ: muscle.
Steroid Receptor Imaging in Breast Cancer
59
60 Beauregard et al

Fig. 2. (A–C ) Chemical structure of progesterone and other analogs.

relative to other ER-binding radiopharmaceuticals Fluorotamoxifen Because the ultimate goal of

is summarized in Table 1. Higher binding selec-
tivity, greater ability to cross membrane barriers,
and reduced interaction with metabolic enzymes
could all explain the superior localization properties
of 4F-MFES. Biodistribution studies in Balb/c mice
bearing MC7-L1 and MC4-L2 ER+ murine breast
cancer tumors also showed high uptake of 4F-MFES
in these tumors. This uptake was ER mediated be-
cause it could be blocked by coinjection of excess
estradiol [119].
4F-MFES has very low relative binding affinity for
SHBG (< 0.0017), like the other 11β-methoxy de-
rivatives in this series [118] and like, as discussed
above, βFMOX. Because the role of SHBG in trans-
port and metabolism of steroids is not completely
elucidated, we cannot predict how 4F-MFES will
behave in human. Accordingly, 4F-FMES remains
a promising but speculative ER imaging PET agent.
A Phase I clinical trial is currently underway in the
authors’ center with this agent.
imaging ER in breast cancer is to guide the choice
of therapy, it appears attractive to use radiolabeled
forms of SERM drugs used in breast cancer treat-
ment. To date, Tamoxifen remains the most widely
used antiestrogen drug for this purpose. The syn-
thesis of a fluorine-18 labeled derivative of ta-
moxifen, [18F]fluoromethyl-N,N-dimethyltamoxifen
(18FTX), was described by Yang and colleagues
[96]. Biodistribution studies of this compound in
rats showed some ER-mediated uptake in both
uterus and inoculated mammary tumors, although
there was only slight to moderate (less than 50%)
decrease in uptake when blocking with an excess
of cold estradiol. Also, the absolute uptake in the
target organs was rather low compared with the
very high uptake in liver, lung, and kidney [96]. A
clinical study was conducted in 10 patients who
had ER-positive breast tumors [120]. 18FTX PET was
able to detect 16 of the 20 lesions (in five of seven
patients) considered to be truly positive for breast

Fig. 3. Typical 18F-FDG (A) and 18F-FES (B) maximum-intensity projection images in a patient with recurrent breast
cancer and multiple bone metastases. Most of the disease sites are concordant between both tracers. Note the
high liver, biliary tract, and bowel caused by excretion of 18F-FES metabolites.
 Steroid Receptor Imaging in Breast Cancer 61

Fig. 4. Discordant 18F-FDG (A) and 18F-FES (B) findings in a patient with recurrent breast cancer in the left axilla.
Although the primary tumor was ER-positive, a biopsy of a left axillary lymph node confirmed that the recurrent
disease no longer expressed ER. The panels show coronal, sagittal, and transaxial images from left to right.

cancer. Equivocal lesions on 18FTX PET were in-
cluded in the positive results. 18FTX PET predicted
response to tamoxifen therapy after the scan in five
of six patients: three had positive lesions and a
good response, two had negative lesions and poor
response, and the other nonresponder had mixed
results on PET [120]. Similarly to what was ob-
served in the animal studies, markedly high up-
take in liver, lung, and heart was observed. This
nonspecific uptake, attributable to the high lipo-

Fig. 5. 18F-FDG PET maximum-intensity projection images in a patient with diffuse bone metastasis (A). The 18F-FES
PET images were negative in this patient (B). Note the uptake of both 18F-FDG and 18F-FES in the apex of the
left lung, which is the site of radiation pneumonitis.
62 Beauregard et al

compounds have been synthesized for ER imaging

[90,121,122]. The authors’ group recently reported
the synthesis three fluorine-18 labeled analogs of
fulvestrant, a pure ER antagonist [123]; however,
the fluorine substitution at position 16α reduced
considerably the RBA (16α-fluoro-fulvestrant = 1.01;
fulvestrant = 89; estradiol = 100). Likewise, low ER-
mediated uptake was observed in immature female
rats [123].

Fig. 6. 18F-FES PET study (anterior projection image) of
a patient. The dose of 18F-FES was injected in a port-a-
cath, with retention of the tracer in the tubing.
philicity of 18FTX and its liver metabolism, contrib-
uted to reduce its sensitivity, particularly for tho-
racic lesions.
Other fluorinated PET ER imaging agents Many
other flurine-18 labeled steroidal and nonsteroidal
Progesterone receptors
21-[18F]fluoro-16α-ethyl-norprogesterone ([18F]FENP)
was the first progestin labeled with fluorine-18
[124], and is the most studied positron emitting
PR imaging agent to date [Fig. 2]. Its RBA value of
6000 (progesterone = 100), and its highly selective
uptake in rat uterus were encouraging. It is the only
PR radiopharmaceutical studied in humans to date
[125]. This study included eight patients who had
primary breast carcinoma. [18F]FENP PET was only
50% accurate in identifying PR-positive primary
tumors (3/6 patients). Rapid metabolism, mani-
fested by high bone (defluorination) and hepatic
uptake, contributed to these disappointing results.
A few other fluorine-18 labeled progestins have
been synthesized and studied in small animals
[126]. Radioiodinated PR imaging agents have

Fig. 7. Patient with recurrent breast cancer manifested by serum tumor marker CA-125 elevation. The 18F-FDG PET
images showed subtle bone metastases (A – coronal and sagittal images; left), that are much more numerous
and obvious on the 18F-FES images (A – right). During aromatase inhibitor therapy, the 18F-FDG-PET study became
almost completely negative (B, left), while extensive residual disease was still apparent of 18F-FES PET (B, right).
 Steroid Receptor Imaging in Breast Cancer 63

been proposed by Ali and coworkers [127] and Mankoff and coworkers [133] showed that
18
others [128]. Tc-99m labeled progestins are re- F-FES uptake could be heterogeneous across
viewed in reference [129]. tumor sites in a given patient, with a coefficient
of variation on the order of 30%. They also showed
18
that 18F-FES uptake tended to be higher in post-
Clinical imaging studies with F-FES menopausal patients who had low serum estradiol
As discussed above, Mintun and colleagues [108] and SHBG levels, and in patients who received
showed that 18F-FES uptake could detect ER-positive prior hormonal therapy. Recently, the same group
breast tumors in a series of patients who had rela- showed in a study conducted on 46 patients who
tively large primary tumors. They observed a good had ER-positive primary and recurrent breast can-
correlation (r = 0.96) between ER concentration cers [134] that 18F-FES uptake could predict re-
and the uptake of 18F-FES. These results were sponse to hormonal therapy (mostly aromatase
further supported in a subsequent study of 16 pa- inhibitors) depending on the HER2/neu status of
tients who had recurrent or metastatic breast the tumors. 18F-FES uptake was predictive of re-
carcinoma eligible to antiestrogen therapy [130]. sponse in HER2/neu-negative patients, but could not
18
F-FES uptake was seen in 14 of 16 patients and predict the outcome when HER2/neu was positive.
53 of 57 known lesions. Interestingly, three of In practical terms, 18F-FES is usually prepared in
the four false-negative lesions occurred in 2 pa- a saline solution containing from 5% to 10% of
tients injected with 18F-FES of low specific activity ethanol. 18F-FES can also be prepared in a lipid
(278 and 230 Ci/mmol). This study showed that emulsion without significant alterations in the bio-
18
F-FES could also assess ER expression in nodal
and distant metastases. Dehdashti and coworkers
[131] reported a good 88% agreement between
estrogen receptor status determined by 18F-FES imag-
ing and in-vitro analysis of biopsy samples. Their
study included 32 women who had primary breast
lesions and 21 patients who had recurrent breast
cancer. In primary breast tumors, they observed an
agreement of 82%. None of the benign breast
lesions incorporated 18F-FES. In recurrent or meta-
static sites, the rate of agreement between in vitro
results and 18F-FES-PET was 94%. They observed
intrapatient discordance (some lesions incorpo-
rating 18F-FES whereas others did not) in 2 out of
13 18F-FES positive patients in whom multiple sites
were evaluated, illustrating possible ER expression het-
erogeneity across metastatic sites. No correlation was
observed visually or quantitatively between 18F-FES
and 18F-Fluorodeoxyglucose (18F-FDG) uptake.
Mortimer and colleagues [132] reported a study
with combined 18F-FDG and 18F-FES performed
before and after treatment with tamoxifen to detect
hormone-induced changes in tumor metabolism
(metabolic flare), changes in available levels of
estrogen receptor (ER), and to correlate if those find-
ings would predict hormonally responsive breast
cancer. The 18F-FES uptake values were higher on
average for responders than nonresponders, but
there was considerable overlap between baseline
18
F-FES uptake values of responders and nonrespon-
ders. A successful response to tamoxifen was asso- Fig. 8. Patient recurrent metastatic breast cancer in
ciated with ER blockage confirmed by 18F-FES and a the mediastinum and bones. The 18F-FDG PET study
metabolic ‘‘flare’’ response corresponding to an (A) was obtained before (left) and after two months
of aromatase inhibitors (right), with a moderate re-
increased 18F-FDG uptake 7 to 10 days after the
duction in the metabolic activity of the lesions. The
initiation of tamoxifen. Results of PET were predic-
decrease in 18F-FES uptake (B) was also significant,
tive of responsiveness to tamoxifen therapy in pa- although more disease sites were still apparent on
tients who had advanced ER-positive breast cancer. the 18F-FES study after 2 months of therapy.
64 Beauregard et al
distribution of this compound, but there was a
trend for slightly higher tumor uptake with the
ethanol formulation in animal studies [119]. The
typical injected activity is 6 mCi (220 MBq).
18
F-FES uptake reaches a plateau between 60 to
90 minutes after injection, and remains stable for
up to 2 hours. 18F-FES is rapidly metabolized from
the plasma by the liver, and there is intense liver
and biliary excretion that interferes with the evalua-
tion of lesions in the abdomen and pelvis. 18F-FES
is also partly excreted by the kidneys, and the
bladder is usually apparent in the images. Apart
from these two excretory pathways, there is little
background activity in other organs. The brain
takes up 18F-FES early after injection, presumably
by blood-flow–related diffusion, and washes out
progressively afterwards, with minimal residual
uptake on delayed images. The respective biodistri-
bution of 18F-FES relative to 18F-FDG is illustrated
in Fig. 3, which shows several sites of recurrent
breast cancer that express estrogen receptors. Con-
Fig. 9. Cross-sectional images of a patient with an isolated site of recurrent disease in the sternum, obtained on
a PET/CT scanner. The recurrence in the sternum is readily identified on the 18F-FDG PET study (A). The presence
of estrogen receptors in that lesion is also obvious on the 18F-FES PET study obtained the following week (B).
(C ) shows the fused 18F-FES and CT studies.
versely, loss of ER expression is demonstrated
by a complete lack of 18F-FES uptake at disease
sites identified on 18F-FDG PET images [Fig. 4].
Although the uptake of 18F-FES is generally highly
specific, some inflammatory lesions such as radia-
tion pneumonitis can accumulate low amounts of
this radiopharmaceutical [Fig. 5]. Injections into
implanted central venous lines (such as a port-
a-cath; Smiths Medical Canada Ltd., Markham, On-
tario) should be avoided, because this lipophilic
compound tends to stick to tubing and can inter-
fere with lesion visualization in the thorax [Fig. 6].
In most cases, the distribution of 18F-FES will cor-
respond to sites of disease that are highlighted on
18
F-FDG PET imaging [see Fig. 3], but the extent of
disease is sometimes much more pronounced on
18
F-FES images [Fig. 7A]. In general, the response
to aromatase inhibitors is much more apparent on
18
F-FDG PET studies [see Fig. 7; Fig. 8], and despite
an almost complete resolution of 18F-FDG uptake,
residual tumors can still be depicted under ther-

apy by 18F-FES imaging [see Fig. 7B]. Tamoxifen
typically blocks 18F-FES uptake by ER; in this case,
18
F-FES is useful in demonstrating the successful
blockage of the receptors by ER antagonists [83,132].
With the availability of PET/CT, the interpreta-
tion of 18F-FDG and 18F-FES PET studies is greatly
facilitated to identify the precise sites of disease
involvement [Fig. 9]. The CT component could
improve the accuracy of these tests by identify-
ing some small pulmonary metastases and FES- or
FDG-negative lesions that are occasionally unde-
tected by PET imaging.
Thus there is a wide spectrum of 18F-FES uptake
across tumor sites, and heterogeneity can be ob-
served within a single patient. The current data
suggest that 18F-FES is sensitive to detect ER-posi-
tive tumor sites, and complements the information
provided by 18F-FDG to assess or predict response
to therapy. Various studies are conducted at the
atuhors’ center and at least three other institutions
to assess whether 18F-FES imaging has clinical value
to select patients who have recurrent breast cancer
and who can benefit from second-line hormone
therapy by aromatase inhibitors or pure antagonists
such as fulvestrant. With sufficient follow-up infor-
mation, the precise role of ER imaging agents in
this patient population will likely become clear.
Summary
ER expression measured by competitive binding
assays or immunohistochemistry has a proven util-
ity in identifying a subset of patients who have
slower growing tumors and who will benefit from
hormonal therapy, either with tamoxifen or second-
line agents such as aromatase inhibitors or pure
antagonists. Several promising ligands that bind
selectively to the estrogen receptors are available
for use in clinical research for PET and SPECT
imaging. The search for a suitable PR ligand is
still ongoing, but PR imaging might provide a bet-
ter estimate of tumor response to hormonal ther-
apy than ER imaging. Although the number of
published clinical studies remains limited, the
available data suggest that ER imaging, as an
adjunct to 18F-FDG PET imaging, might be useful
to predict the response to second-line hormonal
therapy in recurrent or metastatic breast cancer.
References
[1] Fisher B, Osborne K, Margolese R, et al.
Neoplasms of the breast. In: Holland J, Blast
R, Morton D, et al, editors. Cancer Medicine.
4th edition, vol 1. Baltimore (MD): Williams &
Wilkins; 1997. p. 2349–429.
[2] Ray S, Rastogi R, Kumar A. Current status of
Steroid Receptor Imaging in Breast Cancer 65
estrogen receptors. In: Jucker E, editor. Progress
in drug research, vol 59. Basel (Switzerland):
Birkhäuser Verlag; 2002. p. 201–32.
[3] Toft D, Gorski J. A receptor molecule for es-
trogens: isolation from the rat uterus and pre-
liminary characterization. Proc Natl Acad Sci
USA 1966;55(6):1574–81.
[4] Greene GL, Gilna P, Waterfield M, et al.
Sequence and expression of human estrogen
receptor complementary DNA. Science 1986;
231(4742):1150–4.
[5] Mosselman S, Polman J, Dijkema R. ER beta:
identification and characterization of a novel hu-
man estrogen receptor. FEBS Lett 1996;392(1):
49–53.
[6] Pearce ST, Jordan VC. The biological role of
estrogen receptors alpha and beta in cancer.
Crit Rev Oncol Hematol 2004;50(1):3–22.
[7] Speirs V, Carder PJ, Lane S, et al. Oestrogen
receptor [beta]: what it means for patients with
breast cancer. Lancet Oncol 2004;5(3):174–81.
[8] Clarke RB, Anderson E, Howell A. Steroid re-
ceptors in human breast cancer. Trends Endo-
crinol Metab 2004;15(7):316–23.
[9] Osborne CK, Schiff R. Estrogen-receptor biol-
ogy: continuing progress and therapeutic im-
plications. J Clin Oncol 2005;23(8):1616–22.
[10] Balfe PJ, McCann AH, Welch HM, et al. Estrogen
receptor beta and breast cancer. Eur J Surg
Oncol 2004;30(10):1043–50.
[11] Keen JC, Davidson NE. The biology of breast
carcinoma. Cancer 2003;97(3):825–33.
[12] Mote PA, Bartow S, Tran N, et al. Loss of
co-ordinate expression of progesterone recep-
tors A and B is an early event in breast carci-
nogenesis. Breast Cancer Res Treat 2002;72(2):
163–72.
[13] Tait JF, Tait SA. The effect of plasma protein
binding on the metabolism of steroid hor-
mones. J Endocrinol 1991;131(3):339–57.
[14] Siiteri PK, Murai JT, Hammond GL, et al. The
serum transport of steroid hormones. Recent
Prog Horm Res 1982;38:457–510.
[15] Strel’chyonok OA, Avvakumov GV. Specific
steroid-binding glycoproteins of human blood
plasma: novel data on their structure and func-
tion. J Steroid Biochem 1990;35(5):519–34.
[16] Hryb DJ, Khan MS, Romas NA, et al. The
control of the interaction of sex hormone-
binding globulin with its receptor by steroid
hormones. J Biol Chem 1990;265(11):6048–54.
[17] Osborne CK. Steroid hormone receptors in
breast cancer management. Breast Cancer Res
Treat 1998;51(3):227–38.
[18] Henderson IC, Patek AJ. The relationship
between prognostic and predictive factors in
the management of breast cancer. Breast Cancer
Res Treat 1998;52(1–3):261–88.
[19] Yasui Y, Potter JD. The shape of age-incidence
curves of female breast cancer by hormone-
receptor status. Cancer Causes Control 1999;
10(5):431–7.
66 Beauregard et al
[20] Anderson WF, Chatterjee N, Ershler WB, et al.
Estrogen receptor breast cancer phenotypes in
the Surveillance, Epidemiology, and End Results
database. Breast Cancer Res Treat 2002;76(1):
27–36.
[21] Early Breast Cancer Trialists’ Collaborative
Group. Tamoxifen for early breast cancer: an
overview of the randomised trials. Lancet 1998;
351(9114):1451–67.
[22] Overgaard M. Radiotherapy as part of a multi-
disciplinary treatment strategy in early breast
cancer. Eur J Cancer 2001;37(Suppl 7):S33–43.
[23] Olson JE, Neuberg D, Pandya KJ, et al. The role
of radiotherapy in the management of operable
locally advanced breast carcinoma: results of a
randomized trial by the Eastern Cooperative
Oncology Group. Cancer 1997;79(6):1138–49.
[24] Harvey JM, Clark GM, Osborne CK, et al.
Estrogen receptor status by immunohistochem-
istry is superior to the ligand-binding assay for
predicting response to adjuvant endocrine
therapy in breast cancer. J Clin Oncol 1999;
17(5):1474–81.
[25] Bardou VJ, Arpino G, Elledge RM, et al. Pro-
gesterone receptor status significantly improves
outcome prediction over estrogen receptor sta-
tus alone for adjuvant endocrine therapy in
two large breast cancer databases. J Clin Oncol
2003;21(10):1973–9.
[26] Early Breast Cancer Trialists’ Collaborative
Group. Systemic treatment of early breast
cancer by hormonal, cytotoxic, or immune
therapy. 133 randomised trials involving
31,000 recurrences and 24,000 deaths among
75,000 women. Lancet 1992;339(8785):71–85.
[27] Goldhirsch A, Glick JH, Gelber RD, et al. Inter-
national consensus panel on the treatment of
primary breast cancer. V: Update 1998. Recent
Results Cancer Res 1998;152:481–97.
[28] Abrams JS. Adjuvant therapy for breast cancer—
results from the USA consensus conference.
Breast Cancer 2001;8(4):298–304.
[29] Ferno M, Stal O, Baldetorp B, et al. Results of
two or five years of adjuvant tamoxifen corre-
lated to steroid receptor and S-phase levels.
South Sweden Breast Cancer Group, and South-
East Sweden Breast Cancer Group. Breast Can-
cer Res Treat 2000;59(1):69–76.
[30] National Institutes of Health Consensus Devel-
opment Conference statement: adjuvant ther-
apy for breast cancer, November 1–3, 2000.
J Natl Cancer Inst Monogr 2001;(30):5–15.
[31] Fisher B, Dignam J, Bryant J, et al. Five versus
more than five years of tamoxifen therapy for
breast cancer patients with negative lymph
nodes and estrogen receptor-positive tumors.
J Natl Cancer Inst 1996;88(21):1529–42.
[32] Colleoni M, Li S, Gelber RD, et al. Timing of
CMF chemotherapy in combination with ta-
moxifen in postmenopausal women with breast
cancer: role of endocrine responsiveness of the
tumor. Ann Oncol 2005;16(5):716–25.
[33] Bonneterre J, Buzdar A, Nabholtz JM, et al. Anas-
trozole is superior to tamoxifen as first-line
therapy in hormone receptor positive advanced
breast carcinoma. Cancer 2001;92(9):2247–58.
[34] Nabholtz JM, Bonneterre J, Buzdar A, et al.
Anastrozole (Arimidex) versus tamoxifen as
first-line therapy for advanced breast cancer
in postmenopausal women: survival analysis
and updated safety results. Eur J Cancer 2003;
39(12):1684–9.
[35] Geisler J, Haynes B, Anker G, et al. Influence
of letrozole and anastrozole on total body aro-
matization and plasma estrogen levels in post-
menopausal breast cancer patients evaluated in
a randomized, cross-over study. J Clin Oncol
2002;20(3):751–7.
[36] Coombes RC, Hall E, Gibson LJ, et al. A ran-
domized trial of exemestane after two to three
years of tamoxifen therapy in postmenopausal
women with primary breast cancer. N Engl J
Med 2004;350(11):1081–92.
[37] Baum M, Buzdar A, Cuzick J, et al. Anastrozole
alone or in combination with tamoxifen ver-
sus tamoxifen alone for adjuvant treatment of
postmenopausal women with early-stage breast
cancer: results of the ATAC (Arimidex, Tamoxi-
fen Alone or in Combination) trial efficacy and
safety update analyses. Cancer 2003;98(9):
1802–10.
[38] Goss PE, Ingle JN, Martino S, et al. A random-
ized trial of letrozole in postmenopausal
women after five years of tamoxifen therapy
for early-stage breast cancer. N Engl J Med 2003;
349(19):1793–802.
[39] Howell A, Cuzick J, Baum M, et al. Results of the
ATAC (Arimidex, Tamoxifen, Alone or in Com-
bination) trial after completion of 5 years’
adjuvant treatment for breast cancer. Obstet
Gynecol Surv 2005;60(5):307–9.
[40] Kahlert S, von Koch F, Ditsch N, et al. [Adjuvant
treatment of breast cancer in 2004. A review of
the 8th consensus conference of St. Gallen and
new study results]. Gynakol Geburtshilfliche
Rundsch 2004;44(2):102–12.
[41] Simpson D, Curran MP, Perry CM. Letrozole:
a review of its use in postmenopausal women
with breast cancer. Drugs 2004;64(11):1213–30.
[42] Bast Jr RC, Ravdin P, Hayes DF, et al. 2000
update of recommendations for the use of tu-
mor markers in breast and colorectal cancer:
clinical practice guidelines of the American So-
ciety of Clinical Oncology. J Clin Oncol 2001;
19(6):1865–78.
[43] Parker MG. Action of ‘‘pure’’ antiestrogens in
inhibiting estrogen receptor action. Breast Can-
cer Res Treat 1993;26(2):131–7.
[44] Webb P, Nguyen P, Kushner PJ. Differential
SERM effects on corepressor binding dictate
ERalpha activity in vivo. J Biol Chem 2003;
278(9):6912–20.
[45] Pink JJ, Jordan VC. Models of estrogen recep-
tor regulation by estrogens and antiestrogens in

breast cancer cell lines. Cancer Res 1996;56(10):
2321–30.
[46] Howell A, Osborne CK, Morris C, et al. ICI
182,780 (Faslodex): Development of a novel,
pure antiestrogen. Cancer 2000;89(4):817–25.
[47] Wakeling AE, Bowler J. Ici 182,780, a new anti-
oestrogen with clinical potential. J Steroid Bio-
chem Mol Biol 1992;43(1–3):173–7.
[48] Mauriac L, Pippen JE, Quaresma Albano J, et al.
Fulvestrant (Faslodex) versus anastrozole for the
second-line treatment of advanced breast cancer
in subgroups of postmenopausal women with
visceral and non-visceral metastases: combined
results from two multicentre trials. Eur J Cancer
2003;39(9):1228–33.
[49] Menard S, Fortis S, Castiglioni F, et al. HER2 as
a prognostic factor in breast cancer. Oncology
2001;61(Suppl 2):67–72.
[50] Duffy MJ. Predictive markers in breast and
other cancers: a review. Clin Chem 2005;
51(3):494–503.
[51] Borg A, Baldetorp B, Ferno M, et al. ERBB2 am-
plification is associated with tamoxifen resis-
tance in steroid-receptor positive breast cancer.
Cancer Lett 1994;81(2):137–44.
[52] Arpino G, Green SJ, Allred DC, et al. HER-2
amplification, HER-1 expression, and tamoxifen
response in estrogen receptor-positive meta-
static breast cancer: a southwest oncology group
study. Clin Cancer Res 2004;10(17):5670–6.
[53] Agrup M, Stal O, Olsen K, et al. C-erbB-2 over-
expression and survival in early onset breast
cancer. Breast Cancer Res Treat 2000;63(1):23–9.
[54] Olayioye MA, Neve RM, Lane HA, et al. The
ErbB signaling network: receptor heterodimer-
ization in development and cancer. EMBO J
2000;19(13):3159–67.
[55] Piccart MJ, Di Leo A, Hamilton A. HER2. a ’pre-
dictive factor’ ready to use in the daily manage-
ment of breast cancer patients? Eur J Cancer
2000;36(14):1755–61.
[56] Yamauchi H, Stearns V, Hayes DF. When is a
tumor marker ready for prime time? A case
study of c-erbB-2 as a predictive factor in breast
cancer. J Clin Oncol 2001;19(8):2334–56.
[57] Winston JS, Ramanaryanan J, Levine E. HER-2/
neu evaluation in breast cancer are we there yet?
Am J Clin Pathol 2004;121(Suppl):S33–49.
[58] Piccart MJ, Awada A. State-of-the-art chemother-
apy for advanced breast cancer. Semin Oncol
2000;27(5, Suppl 9):3–12.
[59] Ellis MJ, Coop A, Singh B, et al. Letrozole is more
effective neoadjuvant endocrine therapy than
tamoxifen for ErbB-1- and/or ErbB-2-positive,
estrogen receptor-positive primary breast cancer:
evidence from a phase III randomized trial.
J Clin Oncol 2001;19(18):3808–16.
[60] Ellis MJ, Coop A, Singh B, et al. Letrozole in-
hibits tumor proliferation more effectively than
tamoxifen independent of HER1/2 expression
status. Cancer Res 2003;63(19):6523–31.
[61] Hull D, Clark G, Osborne C, et al. Multiple es-
Steroid Receptor Imaging in Breast Cancer 67
trogen receptor assays in human breast cancer.
Cancer Res 1983;43:413–6.
[62] Castagnetta L, Traina A, DiCarlo A, et al.
Heterogeneity of soluble and nuclear oestrogen
receptor status of involved nodes in relation to
primary breast cancer. Eur J Cancer Clin Oncol
1987;23:31–5.
[63] Johnston S, Saccani-Jotti G, Smith E, et al. Changes
in estrogen receptor, progesteron receptor, and
pS2 expression in tamoxifen-resistant human
breast cancer. Cancer Res 1995;55:3331–8.
[64] Encarnacion C, Ciocca D, McGuire W, et al.
Measurement of steroid hormone receptors in
breast cancer patients on tamoxifen. Breast
Cancer Res Treat 1993;26:237–46.
[65] Johnston S, Haynes B, Smith I, et al. Acquired
tamoxifen resistance in human breast cancer
and reduced intra-tumoral drug concentration.
Lancet 1993;342:1521–2.
[66] Hochberg RB. Iodine-125–labeled estradiol: a
gamma-emitting analog of estradiol that binds
to the estrogen receptor. Science 1979;205(4411):
1138–40.
[67] Pavlik EJ, Nelson K, Gallion HH, et al. Charac-
terization of high specific activity [16 alpha-
123I]Iodo-17 beta-estradiol as an estrogen
receptor-specific radioligand capable of imaging
estrogen receptor-positive tumors. Cancer Res
1990;50(24):7799–805.
[68] Schober O, Scheidhauer K, Jackisch C, et al.
Breast cancer imaging with radioiodinated
oestradiol. Lancet 1990;335(8704):1522.
[69] Kenady DE, Pavlik EJ, Nelson K, et al. Images
of estrogen-receptor-positive breast tumors
produced by estradiol labeled with iodine I
123 at 16 alpha. Arch Surg 1993;128(12):
1373–81.
[70] Scheidhauer K, Muller S, Smolarz K, et al. Tumor-
Szintigraphie mit 123J-markiertem Ostradiol
beim Mammakarzinom–Rezeptorszintigraphie
[Tumor scintigraphy using 123I-labeled estradiol
in breast cancer-receptor scintigraphy]. Nucl Med
(Stuttg) 1991;30(3):84–99 [in German].
[71] Symes EK, Coulson WF, Ralphs DN. The
synthesis of 16 alpha-[131I]iodo-oestradiol
and evaluation of its use as a radiotracer for
oestrogen receptor positive breast tumours.
J Steroid Biochem 1985;22(2):155–60.
[72] Gatley SJ, Shaughnessy WJ, Inhorn L, et al.
Studies with 17 beta(16 alpha-[125i]iodo)-
estradiol, an estrogen receptor-binding radio-
pharmaceutical, in rats bearing mammary
tumors. J Nucl Med 1981;22(5):459–64.
[73] Ali H, Rousseau J, Ghaffari MA, et al. Synthesis,
receptor binding, and tissue distribution of
(17 alpha,20E)- and (17 alpha,20Z)-21-[125I]
iodo-19-norpregna-1,3,5(10), 20-tetraene-3,
17-diol. J Med Chem 1988;31(10):1946–50.
[74] Ali H, Rousseau J, Ghaffari MA, et al. Synthesis,
receptor binding, and tissue distribution of 7
alpha- and 11 beta-substituted (17 alpha,20E)-
and (17 alpha,20Z)-21-[125I]iodo-19-norpreg-
68 Beauregard et al
na-1,3,5(10),20-tetraene-3,17-dio ls. J Med
Chem 1991;34(2):854–60.
[75] Foulon C, Guilloteau D, Baulieu JL, et al. Es-
trogen receptor imaging with 17 alpha-[123I]
iodovinyl-11 beta-methoxyestradiol (MIVE2)–
Part I. Radiotracer preparation and charac-
terization. Int J Rad Appl Instrum B 1992;
19(3):257–61.
[76] Ribeiro-Barras MJ, Foulon C, Baulieu JL, et al.
Estrogen receptor imaging with 17 alpha-[123I]
iodovinyl-11 beta-methoxyestradiol (MIVE2)—
Part II. Preliminary results in patients with breast
carcinoma. Int J Rad Appl Instrum B 1992;
19(3):263–7.
[77] Rijks LJ, Boer GJ, Endert E, et al. The stereo-
isomers of 17alpha-[123I]iodovinyloestradiol
and its 11beta-methoxy derivative evaluated
for their oestrogen receptor binding in human
MCF-7 cells and rat uterus, and their distribu-
tion in immature rats. Eur J Nucl Med 1996;
23(3):295–307.
[78] Rijks LJ, Boer GJ, Endert E, et al. The Z-isomer
of 11 beta-methoxy-17 alpha-[123I]iodovinyl-
estradiol is a promising radioligand for estrogen
receptor imaging in human breast cancer. Nucl
Med Biol 1997;24(1):65–75.
[79] Rijks LJ, Bakker PJ, van Tienhoven G, et al.
Imaging of estrogen receptors in primary and
metastatic breast cancer patients with iodine-
123-labeled Z-MIVE. J Clin Oncol 1997;15(7):
2536–45.
[80] Nachar O, Rousseau JA, Lefebvre B, et al. Bio-
distribution, dosimetry and metabolism of
11beta-methoxy-(17alpha,20E/Z)-[123I]iodo-
vinylestradiol in healthy women and breast can-
cer patients. J Nucl Med 1999;40(10):1728–36.
[81] Nachar O, Rousseau JA, Ouellet R, et al.
Scintimammography with 11beta-methoxy-
(17alpha,20Z)-[123I]iodovinylestradiol: a com-
plementary role to 99mTc-methoxyisobutyl
isonitrile in the characterization of breast tu-
mors. J Nucl Med 2000;41(8):1324–31.
[82] Bennink ML, Leuba SH, Leno GH, et al.
Unfolding individual nucleosomes by stretch-
ing single chromatin fibers with optical twee-
zers. Nat Struct Biol 2001;8(7):606–10.
[83] Bennink RJ, van Tienhoven G, Rijks LJ, et al. In
vivo prediction of response to antiestrogen treat-
ment in estrogen receptor-positive breast cancer.
J Nucl Med 2004;45(1):1–7.
[84] Ali H, Rousseau J, van Lier JE. Synthesis of
A-ring fluorinated derivatives of (17 alpha,20E/Z)-
[125I]iodovinylestradiols: effect on receptor
binding and receptor-mediated target tissue up-
take. J Med Chem 1993;36(21):3061–72.
[85] Ali H, Rousseau J, van Lier JE. 7 alpha-Methyl-
and 11 beta-ethoxy-substitution of [125I]-16
alpha-iodoestradiol: effect on estrogen receptor-
mediated target tissue uptake. J Med Chem
1993;36(2):264–71.
[86] Ali H, Rousseau J, Gantchev TG, et al. 2- and
4-fluorinated 16 alpha(-)[125I]iodoestradiol de-
rivatives: synthesis and effect on estrogen recep-
tor binding and receptor-mediated target tissue
uptake. J Med Chem 1993;36(26):4255–63.
[87] Ali H, Rousseau J, van Lier JE. Synthesis, recep-
tor binding and biodistribution of the gem-
21-chloro-21-iodovinylestradiol derivatives. J
Steroid Biochem Mol Biol 1993;46(5):613–22.
[88] Ali H, Rousseau J, Paquette B, et al. Synthe-
sis and biological properties of 7alpha-cyano
derivatives of the (17alpha,20E/Z)-[125I]iodo-
vinyl- and 16alpha-[125I]iodo-estradiols. Steroids
2003;68(14):1189–200.
[89] Rijks LJ, van den Bos JC, van Doremalen PA, et
al. Synthesis, estrogen receptor binding, and
tissue distribution of a new iodovinylestradiol
derivative (17alpha,20E)-21-[123I]Iodo-
11beta-nitrato-19-norp regna-1,3,5 (10),20-tet-
raene-3,17-diol (E-[123I]NIVE). Nucl Med Biol
1998;25(4):411–21.
[90] Cummins CH. Radiolabeled steroidal estrogens
in cancer research. Steroids 1993;58(6):245–59.
[91] McElvany KD, Katzenellenbogen JA, Shafer KE,
et al. 16 alpha-[77Br]bromoestradiol: dosimetry
and preliminary clinical studies. J Nucl Med
1982;23(5):425–30.
[92] McElvany KD, Carlson KE, Welch MJ, et al. In
vivo comparison of 16 alpha[77Br]bromoestra-
diol-17 beta and 16 alpha-[125I]iodoestradiol-
17 beta. J Nucl Med 1982;23(5):420–4.
[93] Strickland LA, Ponce YZ, Hunter DH, et al.
Amino and iodotamoxifens: synthesis, estrogen
receptor affinity and biodistribution. Drug Des
Deliv 1990;6(3):195–212.
[94] Maaroufi Y, Quivy J, Trivedi S, et al.
4-iodotamoxifen aziridine, a new affinity label-
ing agent for the rapid detection of estrogen
receptor isoforms. J Steroid Biochem Mol Biol
1998;67(2):95–104.
[95] Hanson RN, Seitz DE. Tissue distribution of the
radiolabeled antiestrogen [125I]iodotamoxifen.
Int J Nucl Med Biol 1982;9(2):105–7.
[96] Yang DJ, Li C, Kuang LR, et al. Imaging,
biodistribution and therapy potential of haloge-
nated tamoxifen analogues. Life Sci 1994;55(1):
53–67.
[97] Van de Wiele C, De Vos F, De Sutter J, et al.
Biodistribution and dosimetry of (iodine-123)-
iodomethyl-N, N-diethyltamoxifen, an (anti)oestro-
gen receptor radioligand. Eur J Nucl Med 1999;
26(10):1259–64.
[98] Van de Wiele C, Cocquyt V, VandenBroecke R,
et al. Iodine-labeled tamoxifen uptake in pri-
mary human breast carcinoma. J Nucl Med 2001;
42(12):1818–20.
[99] Skaddan MB, Wust FR, Jonson S, et al. Radio-
chemical synthesis and tissue distribution of
Tc-99m-labeled 7alpha-substituted estradiol
complexes. Nucl Med Biol 2000;27(3):269–78.
[100] Unak P, Enginar H, Biber FZ, et al. A correlative
study between 99mTc-ESTCPTA and 99mTc-MIBI
in rats. Appl Radiat Isot 2002;57(5):733–42.
[101] Hunter DH, Luyt LG. Single isomer technetium-

99m tamoxifen conjugates. Bioconjug Chem
2000;11(2):175–81.
[102] Ouellet R, Rousseau J, Brasseur N, et al. Syn-
thesis, receptor binding, and target-tissue up-
take of carbon-11 labeled carbamate derivatives
of estradiol and hexestrol. J Med Chem 1984;
27(4):509–13.
[103] Dence CS, Napolitano E, Katzenellenbogen JA,
et al. Carbon-11-labeled estrogens as potential
imaging agents for breast tumors. Nucl Med
Biol 1996;23(4):491–6.
[104] Ali H, Rousseau J, Diksic M, et al. Synthesis,
receptor binding and target-tissue uptake of
carbon-11-labeled carbamate derivatives of es-
tradiol and hexestrol. Int J Rad Appl Instrum B
1992;19(2):175–82.
[105] Ram S, Spicer LD. Direct Incorporation of
[C-11] Carbon Dioxide for Labeling Bioactive
Molecules - an Application to [C-11] Labeled
Tamoxifen. J Labelled Comp Radiopharm 1989;
27(6):661–8.
[106] Kiesewetter D, Kilbourn M, Landvatter S, et al.
Preparation of four fluorine-18-labeled estro-
gens and their selective uptake in target tissues
of immature rats. J Nucl Med 1984;25(11):
1212–21.
[107] Mathias CJ, Welch MJ, Katzenellenbogen JA,
et al. Characterization of the uptake of 16 alpha-
([18F]fluoro)-17 beta-estradiol in DMBA-
induced mammary tumors. Int J Rad Appl
Instrum B 1987;14(1):15–25.
[108] Mintun MA, Welch MJ, Siegel BA, et al. Breast
cancer: PET imaging of estrogen receptors. Ra-
diology 1988;169(1):45–8.
[109] Mankoff DA, Peterson LM, Tewson TJ, et al.
[18F]fluoroestradiol radiation dosimetry in
human PET studies. J Nucl Med 2001;42(4):
679–84.
[110] Mankoff DA, Tewson TJ, Eary JF. Analysis of
blood clearance and labeled metabolites for the
estrogen receptor tracer [F-18]-16 alpha-fluo-
roestradiol (FES). Nucl Med Biol 1997;24(4):
341–8.
[111] Lim JL, Zheng L, Berridge MS, et al. The use of
3-methoxymethyl-16 beta, 17 beta-epiestriol-O-
cyclic sulfone as the precursor in the synthesis
of F-18 16 alpha-fluoroestradiol. Nucl Med Biol
1996;23(7):911–5.
[112] Romer J, Fuchtner F, Steinbach J, et al. Auto-
mated production of 16alpha-[18F]fluoroestra-
diol for breast cancer imaging. Nucl Med Biol
1999;26(4):473–9.
[113] Pomper MG, VanBrocklin H, Thieme AM, et al.
11 beta-methoxy-, 11 beta-ethyl- and 17 alpha-
ethynyl-substituted 16 alpha-fluoroestradiols:
receptor-based imaging agents with enhanced
uptake efficiency and selectivity. J Med Chem
1990;33(12):3143–55.
[114] VanBrocklin HF, Carlson KE, Katzenellenbogen
JA, et al. 16 beta-([18F]fluoro)estrogens: sys-
tematic investigation of a new series of fluorine-
18-labeled estrogens as potential imaging
Steroid Receptor Imaging in Breast Cancer 69
agents for estrogen-receptor-positive breast tu-
mors. J Med Chem 1993;36(11):1619–29.
[115] VanBrocklin HF, Rocque PA, Lee HV, et al. 16
beta-[18F]fluoromoxestrol: a potent, metaboli-
cally stable positron emission tomography imag-
ing agent for estrogen receptor positive human
breast tumors. Life Sci 1993;53(10):811–9.
[116] Jonson SD, Bonasera TA, Dehdashti F, et al.
Comparative breast tumor imaging and com-
parative in vitro metabolism of 16alpha-[18F]
fluoroestradiol-17beta and 16beta-[18F]fluoro-
moxestrol in isolated hepatocytes. Nucl Med
Biol 1999;26(1):123–30.
[117] Seimbille Y, Bénard F, van Lier JE. Synthesis of
16alpha-fluoro ICI 182,780 derivatives: power-
ful antiestrogens to image estrogen receptor
densities in breast cancer by positron emission
tomography. J Chem Soc [Perkin 1] 2002;20:
2275–81.
[118] Seimbille Y, Rousseau J, Benard F, et al. 18F-
labeled difluoroestradiols: preparation and pre-
clinical evaluation as estrogen receptor-binding
radiopharmaceuticals. Steroids 2002;67(9):
765–75.
[119] Bénard F, Aliaga A, Ahmed N, et al. Biodis-
tribution of fluorinated estradiol derivatives
in ER + tumor-bearing mice: impact of sub-
stituents, formulation and specific activity
[abstract]. J Nucl Med 2004;45(5):329P.
[120] Inoue T, Kim EE, Wallace S, et al. Posi-
tron emission tomography using [18F]fluoro-
tamoxifen to evaluate therapeutic responses
in patients with breast cancer: preliminary
study. Cancer Biother Radiopharm 1996;11(4):
235–45.
[121] Landvatter SW, Kiesewetter DO, Kilbourn MR,
et al. (2R*, 3S*)-1-[18F]fluoro-2,3-bis(4-hydro-
xyphenyl)pentane [(18F]fluoronor-hexestrol), a
positron-emitting estrogen that shows highly-
selective, receptor-mediated uptake by target
tissues in vivo. Life Sci 1983;33(19):1933–8.
[122] Hostetler ED, Jonson SD, Welch PJ, et al.
Synthesis of 2-[F-18]fluoroestradiol, a potential
diagnostic imaging agent for breast cancer:
Strategies to achieve nucleophilic substitution
of an electron-rich aromatic ring with [F-18]F-.
J Org Chem 1999;64(1):178–85.
[123] Seimbille Y, Benard F, Rousseau J, et al. Impact
on estrogen receptor binding and target tissue
uptake of [18F]fluorine substitution at the 16
[alpha]-position of fulvestrant (faslodex; ICI
182,780). Nucl Med Biol 2004;31(6):691–8.
[124] Pomper MG, Katzenellenbogen JA, Welch MJ,
et al. 21-[18F]fluoro-16 alpha-ethyl-19-norpro-
gesterone: synthesis and target tissue selective
uptake of a progestin receptor based radio-
tracer for positron emission tomography.
J Med Chem 1988;31(7):1360–3.
[125] Dehdashti F, McGuire AH, Van Brocklin HF,
et al. Assessment of 21-[18F]fluoro-16 alpha-
ethyl-19-norprogesterone as a positron-emitting
radiopharmaceutical for the detection of pro-
70 Beauregard et al
gestin receptors in human breast carcinomas.
J Nucl Med 1991;32(8):1532–7.
[126] Jonson SD, Welch MJ. PET imaging of breast
cancer with fluorine-18 radiolabeled estro-
gens and progestins. Q J Nucl Med 1998;42(1):
8–17.
[127] Ali H, Rousseau J, Ahmed N, et al. Synthesis
of the 7alpha-cyano-(17alpha,20E/Z)-[125I]
iodovinyl-19-nortestosterones: potential radio-
ligands for androgen and progesterone recep-
tors. Steroids 2003;68(14):1163–71.
[128] Rijks LJ, van den Bos JC, van Doremalen PA,
et al. New iodinated progestins as potential
ligands for progesterone receptor imaging in
breast cancer. Part 2: In vivo pharmacological
characterization. Nucl Med Biol 1998;25(8):
791–8.
[129] Katzenellenbogen JA. Designing steroid receptor-
based radiotracers to image breast and prostate
tumors. J Nucl Med 1995;36(Suppl 6):8S–13S.
[130] McGuire AH, Dehdashti F, Siegel BA, et al.
Positron tomographic assessment of 16 alpha-
[18F] fluoro-17 beta-estradiol uptake in meta-
static breast carcinoma. J Nucl Med 1991;32(8):
1526–31.
[131] Dehdashti F, Mortimer JE, Siegel BA, et al. Posi-
tron tomographic assessment of estrogen recep-
tors in breast cancer: comparison with FDG-PET
and in vitro receptor assays. J Nucl Med 1995;
36(10):1766–74.
[132] Mortimer JE, Dehdashti F, Siegel BA, et al.
Metabolic flare: indicator of hormone respon-
siveness in advanced breast cancer. J Clin Oncol
2001;19(11):2797–803.
[133] Mankoff DA, Peterson LM, Petra PH, et al.
Factors affecting the level and heterogeneity
of uptake of [18F]Fluoroestradiol in patients
with estrogen receptor positive breast cancer
[abstract]. J Nucl Med 2002;43(5):286P.
[134] Linden HM, Stekhova S, Link JM, et al. HER2
expression and uptake of 18F-Fluoroestradiol
(FES) predict response of breast cancer to hor-
monal therapy [abstract]. J Nucl Med 2004;
45(5):85P.

Changes in Glucose Metabolism
and Blood Flow Following
Chemotherapy for Breast Cancer
David A. Mankoff, MD, PhD*,
& Positron emission tomography
methodology
18
& F-fluordeoxyglucose positron emission
tomography and neoadjuvant
chemotherapy
& 18F-fluordeoxyglucose positron emission
tomography and systemic therapy of
metastatic disease
One of the important advances in breast cancer
treatment over the past 2 decades is the recognition
that breast cancer is a systemic disease necessitating
systemic therapy, even when apparently localized
[1]. Although a variety of options for breast cancer
systemic therapy have emerged, cytotoxic chemo-
therapy remains an essential and commonly used
form of treatment. Most breast cancer chemo-
therapy occurs in the adjuvant setting, following
definitive surgery, designed to eradicate systemic
micrometastases [2]. Imaging currently has a small
role in these patients, and is reserved for sur-
veillance or work-up of symptoms suggesting pos-
sible recurrence; however, a substantial fraction
of women present with more advanced cancer, at
Stage III or Stage IV, when resection may not be
feasible nor appropriate. Furthermore, there is an
increasing trend toward presurgical or neoadjuvant
chemotherapy, even for resectable disease [3]. For
all these reasons, the systemic treatment of women
who have one or more macroscopic tumor sites is
an important public health problem. There are
This work was supported by NIH Grants R01CA72064, P01CA42045, R01CA90771, and S10RR17229; by the
Avon Cancer Crusader Fund; and by the British Columbia Cancer Fund.
Division of Nuclear Medicine, Department of Radiology, Box 356113, Room NN203, University of Washington,
1959 NE Pacific Street, Seattle, WA 98195, USA
* Corresponding author.
E-mail address: dam@u.washington.edu (D.A. Mankoff).
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
71
POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 71–81
Lisa K. Dunnwald, BS, CNMT
& Other positron emission tomography
radiopharmaceuticals
& Summary
& References
40,000 women per year dying of breast cancer in
the United States, and their mortality is largely
related to progressive metastatic disease [4]. The
ability to monitor response in patients who have
macroscopic breast cancer treated by systemic ther-
apy is therefore an important task for which posi-
tron emission tomography (PET) is playing an
emerging role. This article focuses on this applica-
tion of PET to breast cancer.
Many solid tumors respond poorly to systemic
therapy; however, breast cancer is one of the more
chemotherapy-sensitive solid tumors [5]. Women
who have locally advanced or metastatic breast can-
cer can have prolonged remissions [6–8]. Those
that have failed first-line chemotherapy still have
a number of reasonable choices for second-line
therapy with substantial response rates [5,9]; how-
ever, the ability to evaluate chemotherapy response,
and thereby to make appropriate adjustments in
systemic therapy, is limited. For primary breast tu-
mors, anatomic imaging and sized-based response
evaluation can substantially over- or underestimate
doi:10.1016/j.cpet.2005.09.001
72 Mankoff & Dunnwald
response [6,10,11]. For certain sites of metastasis,
for example bone metastasis, changes in imaging
for conventional methods such as CT and bone
scintigraphy can significantly lag response or pro-
gression [12]. Furthermore, it takes several weeks
to months to evaluate efficacy using changes in
tumor size to assess response. For some therapies
that are potentially cytostatic, such as hormonal
therapy, it can be impossible to discern tumor
response from slow disease progression when rely-
ing on anatomically based measures of response.
These are tasks to which a quantitative, functional
imaging method such as PET is ideally suited. Better
methods of predicting and evaluating response are
critical to the individualized therapy of patients
who have breast cancer, especially as advances in
basic biology give rise to a wider variety of thera-
peutic options [2,9].
This article reviews the PET methodology used for
breast cancer response assessment, with an empha-
sis on quantitative methods. This is followed by a
review of results to date for neoadjuvant chemo-
therapy and therapy of metastatic breast cancer.
Preliminary studies with tracers other than 18F-
fluordeoxyglucose (FDG) are reviewed, followed
by a summary and discussion of future directions.
Positron emission tomography methodology
Most of the studies using PET to monitor breast
cancer chemotherapy response have used 18F-
fluordeoxyglucose (FDG) [13]. Most applications
have used fairly standard approaches, with patients
fasting at least 4 to 6 hours before imaging, and with
imaging times typically 45 to 60 minutes after injec-
tion, and attenuation-corrected imaging on whole-
body PET tomographs. Some studies [14,15] have
also performed dynamic imaging over a particular
tumor site, typically the primary breast tumor,
allowing for more rigorous quantitative analysis.
Because quantitative analysis of serial FDG PET
studies is key in application to response monitor-
ing, some aspects of quantitation are highlighted
here. More detailed discussion of quantitative
methods can be found in other studies [16–19].
The most commonly used measured of FDG uptake
in published studies of breast cancer response is
most frequently known as the standardized up-
take value (SUV), defined by the following for-
mula [17]:
Ct
SUV ¼
ID=wt
In this formula, Ct bar is the average tissue
uptake during the course of the static imaging
period (μCi/g or μCi/mL), ID is the injected dose
(mCi) and wt is the patient weight (kg). This mea-
sure is simple to implement and clinically feasi-
ble, but it is an approximation of more complex
FDG kinetics, and therefore has a number of limi-
tations, which are well-described in the litera-
ture [20]. Many of these problems arise from the
fact that simple uptake measures include the FDG
that is present as unphosphorylated FDG in addi-
tion to phosphorylated FDG in the trapped com-
partment [17,19]. Only the phosphorylated FDG
carries specific information on glucose metabo-
lism. For tissues with high metabolic rate of FDG
(MRFDG), most of the label in the tumor region-
of-interest late after injection will be in the form of
FDG-6-phosphate, and therefore SUV provides a
reasonable measure for quantifying FDG uptake.
For less metabolically active tumors, a greater frac-
tion of the label is nonphosphorylated FDG, and
SUV measures FDG metabolism less well. This
has implications for tracking response to ther-
apy in which tumors move from high glucose
metabolism pretreatment to lower glucose me-
tabolism post-treatment, and may account for
some of the differences in results between stud-
ies using SUV versus those using more rigorous
methods. An important consideration is that SUV
changes quickly in metabolically active tumors
[21,22]; therefore maintaining a constant time be-
tween injection and imaging is important in com-
paring serial studies.
More rigorous quantitative analysis of dynamic
FDG PET images is upon a model for FDG
kinetics [Fig. 1], based upon the original work
of Sokoloff and colleagues [23], Reivich and col-
leagues [24], and Phelps and colleagues [25] for
brain imaging. This model depicts the rate con-
stants for FDG as tracer moves from the blood to
tissue, where transport is largely reversible (K1 and
k2), and as it is phosphorylated to form FDG-
6-phosphate (k3) and trapped, with allowance
for a small amount of dephosphorylation in
some tissues (k4). In this model, initial transport
and retention in the reversible compartment are
nonspecific, in that FDG is not yet committed to
the glucose metabolic pathway and may be influ-
enced by factors that are unrelated to glucose
metabolism. These factors include blood flow, cap-
illary permeability, and the size of the local inter-
stitial fluid pool [17]. Glucose metabolism is more
specifically indicated by the FDG trapped as FDG-
6-phosphate, and the most relevant measure of
glucose metabolism that can be obtained from
FDG PET is the flux of FDG from the blood into
the trapped compartment—the phosphorylated
state. This measure is typically described in units
of μmoles/min/g of tissue and is related to the

Blood Tissue
K1
Exchangeable
Blood Tissue FDG
k2
FDG
Fig. 1. Compartmental model of FDG kinetics.
FDG compartmental model rate constants by the
formula below:
K1 k 3
MRFDG ¼ ½Glucose
¼ ½Glucose
Ki
k2 þ k3
In this formula, [Glucose] is the plasma glucose
concentration (μmole/mL). The combination of
rate constants in the fraction term, which describes
a rate constant for flux from the blood to the
phosphorylated state, is often termed the flux con-
stant, Ki (mL/min/g). The equation differs from
the classic work of Reivich and coworkers [24]
and Phelps and coworkers [25] for brain imaging,
in that there is no attempt to relate FDG kinetics
directly to glucose kinetics. Instead, the term
MRFDG is used to indicate glucose flux estimated
by FDG PET, and to reflect the fact that there are
differences between FDG and glucose metabolism,
especially in tumors [26]. Glucose metabolic rate or
MRFDG estimated from full compartmental mod-
eling has been reported in several studies of neoad-
juvant chemotherapy of breast cancer [14,27,28].
An alternate and slightly simplified approach is
the graphic method devised by Patlak and Blasberg
[29] and Gjedde [30]. This method requires
dynamic tissue uptake curves, but transforms the
problem of estimating the flux constant to linear
regression, or in other words, a line fit. In this
formulation, a plot of the ratio of tissue activity
to blood activity versus normalized time, which
takes the blood clearance curve into account, ap-
proaches a straight line with slope equal to the flux
constant, Ki, under the condition that there is no
loss from the trapped compartment (ie, k4 = 0).
Graphic analysis is a simple and reasonably robust
approach for quantifying FDG uptake kinetics. The
graphic approach does not yield all of the kinetic
insight that more rigorous quantitative analysis
provides; however, it is simple and robust. Graphic
analysis has been applied in some of the published
literature on breast cancer neoadjuvant chemother-
apy [15,31].
PET & Breast Cancer Therapy 73
Total Activity in PET Image
k3
Trapped
k4 FDG-
6P
18
F-fluordeoxyglucose positron emission
tomography and neoadjuvant chemotherapy
Presurgical or neoadjuvant therapy is increasingly
used in patients presenting with nonmetastatic
breast cancer. The majority of studies performed
evaluating PET to measure breast cancer response
to systemic therapy have been done in this set-
ting. Neoadjuvant chemotherapy was originally
applied to locally advanced breast cancer (LABC),
defined as primary tumors that are greater than
5 cm, fixed axillary lymph nodes, or skin/chest
wall invasion [32]. More recently, neoadjuvant che-
motherapy has been applied to smaller, resectable
breast cancers [3], often with the goal of shrinking
the tumor and offering the possibility of breast-
conserving surgery instead of mastectomy [33].
Neoadjuvant chemotherapy regimens are most
commonly anthracylcine-based (doxorubicin or
epirubicin), and in recent trends they are combined
sequentially with a taxane [2,34]. Other chemo-
therapy regimens have been successfully applied.
Neoadjuvant hormonal therapy using tamoxifen or
aromatase inhibitors has been tested, but with
slower response rates and less complete response
compared with chemotherapy [35].
Studies have shown that the extent of residual
breast and axillary tumor post-neoadjuvant che-
motherapy is prognostic of disease-free survival
(DFS) and overall survival (OS) [6–8,36,37]. Pa-
tients who have a pathologic complete response
(CR)—defined as disappearance of the primary tu-
mor on histopathologic analysis of the breast speci-
men from definitive post-therapy surgery–have
better disease-free survival than those patients
who have residual tumor [6,7,37]. This has there-
fore become an important endpoint in evaluating
neoadjuvant therapy regimens. Earlier investiga-
tions used a macroscopic complete response (mCR)
as the endpoint, with mCR defined as the absence
of gross disease post-therapy, even if scattered mi-
croscopic disease remained [6]. More recent studies
have used the microscopic resolution of all invasive
74 Mankoff & Dunnwald
Pre-Therapy Chemotherapy
8-12 wks
Baseline
Early Therapy Mid-Therapy
Fig. 2. Timeline for neoadjuvant treatment of breast cancer showing times of tumor assessment by FDG PET in
published studies.
cancer (pCR) as the preferred endpoint, following
the lead of recent trials of neoadjuvant therapy of
resectable breast cancer [37]. The different defini-
tion of complete response can cause confusion in
comparing studies, and should be considered in
evaluating studies of imaging to measure response
to neoadjuvant therapy. The number of residual
lymph node metastases post-therapy is also prog-
nostic. The studies of McCready and colleagues [8]
showed that patients who had three or fewer lymph
nodes post-therapy were significantly more likely to
remain disease-free than those patients who had
four or more positive lymph nodes post-therapy.
A number of investigators have tested the appli-
cation of FDG PET to evaluate breast cancer re-
sponse to neoadjuvant chemotherapy. In fact,
some of the earliest studies testing FDG PET to
measure chemotherapy response were performed
in this clinical setting [14,38]. In these studies
of presurgical breast cancer treatment, FDG PET
was applied at two or more time points along
the course of chemotherapy [Fig. 2]: pretherapy,
early in the course of therapy (after a single cycle
of chemotherapy), midtherapy, and post-therapy
(presurgery). These time points are discussed sepa-
rately below.
The pretherapy scan serves as an important base-
line in measuring changes in the extent and level
of FDG uptake post-therapy; however, it also pro-
vides information on the extent of disease that may
be important in determining postsurgical treat-
ment, especially regional radiotherapy. PET pro-
vides information on the presence or absence of
axillary lymph node metastasis. Because sentinel
lymph node biopsy can be misleading post-therapy
if there was macroscopic disease pretherapy [39],
this can be very helpful in directing the approach
to surgical staging or lymph node dissection. In
addition, PET accurately depicted the extent of
macroscopic axillary disease (reviewed in [40])
[Fig. 3]. This can be especially important in defin-
ing radiotherapy fields post-therapy when there has
been substantial resolution of macroscopic bulky
disease by systemic treatment. Finally, FDG PET can
identify spread to extra-axillary nodes. Bellon and
Surgery
16 - 24 wks
Post-Therapy
coworkers [41] found evidence of internal mam-
mary (IM) nodal involvement in approximately
20% of patients who had locally advanced breast
cancer, and the presence of pretherapy FDG up-
take in IM lymph nodes predicted regional or sys-
temic failure.
The majority of studies of FDG PET for neoadju-
vant chemotherapy have measured FDG uptake
midtherapy as a measure of response, and tested
the ability of midtherapy measurements of FDG
uptake to predict post-therapy pathologic response
[14,15,42–44] [Fig. 4]. Results are summarized in
Table 1. Almost all studies show a significant dif-
ference in midtherapy FDG uptake for responders
versus nonresponders. Echoing the early studies of
Wahl and colleagues [14], all studies have shown
that in patients who go on to have a good response
to neoadjuvant chemotherapy, primary tumor FDG
uptake declines by approximately 50% or more
from pretherapy levels. Studies have shown that
nonresponding tumors also had a modest decline
in FDG uptake. This was as high as a 40% decline
in nonresponders [15,43], but typically is less than
50%. It therefore appears that a 50% or greater
decline in FDG uptake midtherapy is a marker of
breast cancer response to neoadjuvant chemother-
apy. So far, studies have not established the prog-
nostic significance of this finding, although the
studies of Mankoff and coworkers [15] show a
Fig. 3. Coronal FDG PET images from a patient with
multifocal left breast cancer (thin arrows) and exten-
sive axillary disease (thick arrow), occupying all three
axillary levels.
 PET & Breast Cancer Therapy 75

Fig. 4. Axial FDG PET images through the thorax and breasts in a patient with inflammatory breast cancer treated
with neoadjuvant chemotherapy and imaged after 2 and 4 months of treatment. The pretherapy scan shows a
large tumor with uptake in the skin consistent with dermal invasion of the cancer. Abnormalities progressively
decline over the course of treatment. The patient was found to have no residual invasive disease, and only
noninvasive disease ductal carcinoma in situ (DCIS) at surgery post-therapy. The effect of marrow-stimulating
cytokines granulocyte colony stimulating factor (GCSF), given as part of the chemotherapy regimen, is seen on the
2- and 4-month scans as increased vertebral marrow uptake (dashed arrows).

surements [14,43,44] [Fig. 5]. These exciting results

trend for patients who had lower midtherapy FDG
suggest that FDG PET may be an early predictor of
uptake, measured as MRFDG, to have improved
chemotherapy response and merit further study.
disease-free survival. Demonstration that serial
Some studies have looked at post-therapy FDG
FDG PET predicts survival in the neoadjuvant set-
PET as a measure of the extent of residual disease
ting is an important area of future investigation that
[Table 3] [42,45,46]. This information can help
could establish FDG PET as a quantitative surrogate
direct post-therapy definitive surgery and possibly
in future drug therapy trials.
adjuvant radiotherapy. Studies have shown that
A smaller number of studies have looked at the
although residual FDG uptake post-therapy is a
change in FDG uptake early in the course of ther-
good indicator of residual macroscopic disease,
apy as a predictor of pathologic response. Wahl
the absence of FDG uptake post-therapy is not a
and colleagues [14], Smith and colleagues [43],
reliable indicator of a complete pathologic re-
and Schelling and colleagues [44] all measured
sponse. This is especially true for axillary nodal
FDG uptake after the first cycle of every-3-weeks
metastases, in which the sensitivity of post-therapy
anthracycline-based chemotherapy [Table 2]. All
FDG for residual nodal disease appears to be low
three of these studies showed that early FDG PET
[42,45]. This is likely due to several factors, includ-
predicted subsequent response to chemotherapy,
ing the observation that chemotherapy lowers
in some cases with greater separation between re-
FDG uptake even in patients who have a modest
sponders and nonresponders than midtherapy mea-
response, and that incompletely treated lymph
nodes often harbor small-volume disease for
Table 1: FDG PET results for midtherapy which PET sensitivity is limited even in the absence
evaluation of treatment [47]. These results limit the utility of
Reference N Therapy Results post-therapy FDG PET to determine the extent of
residual disease in locally advanced breast cancer. A
Wahl 11 AC R: −48% SUV possible role for FDG PET in this setting, however,
et al [14] NR: −19% SUV was highlighted in a preliminary study by Chen
Bassa 15 FAC All: −51% SUV
and coworkers [48], who used FDG PET in
et al [42]
Schelling 24 EC or mCR: −46% SUV conjunction with breast MRI. In this study, FDG
et al [44] ET not mCR: −8% SUV PET was complementary to the detailed anatomic
Smith 30 CVAP mCR: −86% SUV information provided by MRI on the extent of re-
et al [43] not mCR: −40% SUV sidual disease. FDG PET added specificity to the
Mankoff 35 FAC mCR: −65%
et al [15] or AC MRFDG
PR: −49% Table 2: FDG PET results for early therapy
MRFDG evaluation
NR: −40% Reference N Therapy Results
MRFDG
Wahl et al 11 AC R: −22% SUV
Abbreviations: AC, epirubicin/cyclophosphamide; CVAP, [14] NR: no change
epirubicin/cyclophosphamide/vincristine/prednisolone; Schelling 24 EC or mCR: −54% SUV
EC, epirubicin/cyclophosphamide; ET, epirubicin/pacli- et al [44] ET not mCR: −19% SUV
taxel; FAC, 5-fluorouracil/doxorubicin/cyclophosphamide; Smith 30 CVAP mCR: −77% SUV
mCR, macroscopic complete response; NR, no response;
et al [43] not mCR: +1% SUV
PR, partial response; R, response.
76 Mankoff & Dunnwald

A 140%
B 0.12
120% 0.1
% Change Mean

100%
0.08
80% mCR

DUR
GRD
SUV

0.06
60% MRD Other
0.04
40%

20% 0.02

0% 0
0 1 2 3 0 1 2 3 4 5
Cycle # Cycle #
Fig. 5. Time course of FDG uptake taken from [44] (A) and [43] (B). Both plots show an early decline in tumor
uptake after the first course of chemotherapy for minimal residual disease (MRD) versus gross residual disease
(GRD) (A) and for mCR versus other response (B). There is no further separation of the different response curves at
the midtherapy time point. In the Smith et al data [43] (B), there is less separation of mCR versus other at
midtherapy versus early therapy, due to a decline in uptake in patients with other than mCR at midtherapy not
seen early. DUR, differential uptake ratio (equivalent to the SUV).

MRI examinations and was accurate in identifying and therefore FDG PET may be particularly help-
resistant cancer. These encouraging results warrant ful in this clinical setting. Unfortunately, there
further multimodality studies in the setting of neo- have been relatively few studies of Stage IV breast
adjuvant chemotherapy. cancer response monitoring using FDG PET
[50,51], complicated by small patient numbers, a
variety of disease presentations, and a number
18F-fluordeoxyglucose positron emission of different types of therapy including chemo-
tomography and systemic therapy of therapy, hormonal therapy, and biologic therapy
(eg, trastuzumab).
metastatic disease
Gennari and colleagues [50] studied 13 patients
Unlike many other solid tumors, breast cancer is who had metastatic breast cancer undergoing che-
often responsive to systemic therapy even at ad- motherapy of Stage IV breast cancer using one of
vanced stages [49]. Although Stage IV (metastatic) two different epirubicin-based regimens. The study
breast cancer is rarely ‘‘cured,’’ patients who have authors found that after a single cycle of che-
Stage IV disease often have indolent disease, and motherapy (day 8), those patients who went on
there is a wide and expanding range of systemic to a clinical response had an average 25% decline
therapy options. These patients may therefore in FDG SUV, whereas nonresponders had no
have prolonged survival with good quality of life. change. By the completion of therapy, responders
Evaluation of metastatic breast cancer response had a greater than 80% average decline in FDG
by conventional imaging, looking for change in SUV, similar to the neoadjuvant setting, whereas
tumor size [12], can be particularly challenging, nonresponders had essentially no change. These
preliminary results suggest that, as in neoadjuvant
chemotherapy, objective response is accompanied
Table 3: FDG PET results for post-therapy by a large decline in FDG uptake (50%–75% or
Results (for more) after at least 2 or 3 months of treatment, and
Reference N Therapy residual disease) by a smaller decline early in the course of che-
Bassa 11 FAC Primary: Sens 75% motherapy. Further studies are needed to better
et al [42] Axilla: Sens 42%, define endpoints for a variety of treatment regi-
Spec 100% mens in the metastatic setting.
Burcombe 9 FEC Primary: Sens 0% (0/9) Bone metastases present a particularly troubling
et al [45] Axilla: Sens 0% (0/3) site of disease for evaluating response. Although
Kim et al 50 ACT Primary: bone scan and MRI are highly accurate in detecting
[46] or XT SUV: CR, −97%; PR, breast cancer bone metastasis, they are ineffective
−88%; NR, −56%; for measuring response [52,53]. Changes in bone
Sens, 86%; Spec, 83% metastasis appearance on bone scan and MRI can
(−79% threshold)
lag metastasis response, and lesions may appear to
Abbreviations: Sens, sensitivity; Spec, specificity; XT, ca- worsen early in the course of successful treatment
pecitabine/docetaxol. [54,55]. Given the possibility of prolonged sur-

vival, the wide array of therapeutic choices, and the
inadequacy of current conventional imaging for
evaluating bone metastasis response, better tools
for evaluating response are clearly needed. Taking
advantage of the fact that breast cancer bone metas-
tases appear to indicate the tumor itself, rather than
the bony reaction to the tumor [56,57], Stafford
and colleagues [51] tested the use of serial FDG PET
to measure breast cancer bone metastasis response.
They found that the change in FDG PET correlated
with clinical response, based upon a combination
of conventional imaging, clinical symptoms, and
tumor markers. Furthermore, there was a quantita-
tive correlation between the fractional change in
FDG SUV and fractional change in tumor markers,
two measurements often used to assess bone metas-
tasis response [53]. Preliminary follow-up of these
results suggests that the change in FDG uptake in
treated bone metastases is not only an indicator of
response, but also appears to predict progression-
free survival [58], in that patients who have a larger
decline in FDG SUV over the course of therapy have
a longer time to progression. These encouraging
preliminary results merit further prospective study.
Some caution must be taken in interpreting the
results of serial FDG PET studies in evaluating the
response of metastatic breast cancer to systemic
therapy other than chemotherapy. For example,
Mortimer and coworkers [59] found that FDG up-
take increased early in the course of successful
tamoxifen therapy. A ‘‘flare’’ of disease in the first
7 to 10 days after starting tamoxifen accurately
predicted good response to subsequent treatment,
likely a result of the early agonist effect of tamoxi-
fen in hormone-sensitive breast cancer. First-line
hormonal therapy is more commonly given, at
Fig. 6. Response to aromatase inhibitors. Coronal FDG
PET images of a patient with recurrent breast cancer
presenting as isolated sternal disease (arrows). FDG
PET shows a substantial decline in uptake after only
6 weeks of treatment. The patient went on to have a
prolonged response to treatment with minimal resi-
dual disease.
PET & Breast Cancer Therapy 77
present, in the form of aromatase inhibitors, which
do not have an agonist effect and therefore likely
do not have an associated flare. The authors have
observed substantial declines in FDG uptake as
little as 6 weeks after initiation of aromatase in-
hibitors for metastatic breast cancer in our pre-
liminary experience [Fig. 6].
Other positron emission tomography
radiopharmaceuticals
Several studies have used PET radiopharmaceuticals
other than FDG to examine other aspects of breast
cancer biology in patients undergoing systemic
therapy. Based on the work of Beaney and col-
leagues [60] and Wilson and colleagues [61], sev-
eral studies have used 15O-water to measure blood
flow in patients who had LABC treated by neo-
adjuvant chemotherapy. Simultaneous studies of
blood flow and metabolism using 15O-water and
FDG PET have yielded some interesting results.
Zasadny and coworkers [28] found some correla-
tion between blood flow and metabolism, but bet-
ter correlation between blood flow and early FDG
delivery (FDG K1). Tseng and coworkers [27] found
similar results; in addition, their data suggested that
blood flow and early delivery of glucose were not
the rate-limiting step in the LABC glucose metabolic
rate in most cases. Interestingly, Mankoff and co-
workers [31] found that a metabolism-blood flow
mismatch, in the form of aberrantly high FDG
metabolism relative to blood flow, was an indicator
of poorer response to neoadjuvant chemotherapy
and poorer prognosis.
In serial 15O-water PET studies over the course of
neoadjuvant chemotherapy of locally advanced
breast cancer, Mankoff and colleagues [15] found
that the change in tumor blood flow midtherapy
was an excellent predictor of response and of
patient outcome, better than FDG uptake in both
regards. Patients who responded to neoadjuvant
chemotherapy had an average decline in tumor
blood flow, whereas nonresponders had an aver-
age increase in blood flow. Furthermore, residual
blood flow midtherapy predicted both disease-free
and overall survival; lower midtherapy blood flow
was associated with better disease-free and over-
all survival. This echoes results with other imag-
ing modalities strongly influenced by tumor blood
flow, such as contrast-enhanced MRI [62–65],
Doppler ultrasound [66,67], and technetium-99m
(99mTc)-sestamibi [68–72]. A comparison of changes
in blood flow and FDG metabolism over the course
of therapy by Tseng and coworkers [27] suggested
that the pattern of glucose metabolism changed.
Blood flow and metabolism were better matched
after chemotherapy, with a ratio closer to the nor-
78 Mankoff & Dunnwald
mal breast than the untreated tumor, even in non-
responders. This provides an interesting insight into
the biology of chemotherapy-treated tumors, and
may help explain the decline in FDG uptake seen
after chemotherapy, even in nonresponders.
PET can be used to measure the pharmacologic
effect of drugs directly. Dehdashti and colleagues
[73] and Mortimer and colleagues [59] performed
serial 18F-fluoroestradiol (FES) PET studies to docu-
ment tamoxifen blockade of the estrogen receptor
(ER). These studies showed that high pretherapy
FES uptake, as an indicator of estrogen receptor
expression, and a decline in FES uptake, as an in-
dicator of estrogen receptor blockade, were pre-
dictive of response to primary tamoxifen therapy.
The studies set forth an important paradigm for the
use of PET imaging in early drug trials, and this
merits further study in a variety of clinical settings
and for a variety of treatments.
Finally, some studies have investigated the use of
other PET radiopharmaceuticals to measure breast
cancer response. Energy metabolism fuels tumor
growth, but it also fuels a variety of other biologic
processes, including cellular repair. Therefore, it is
likely that radiopharmaceuticals measuring pro-
cesses more specific to cellular growth may be
advantageous for response evaluation. In early
and pioneering studies, Huovinen and coworkers
[74] and Jansson and coworkers [75] measured
breast cancer response to chemotherapy using la-
beled amino acids as a measure of cellular bio-
synthesis with promising initial results, In more
recent studies paralleling work with cellular prolif-
eration tracers in other tumors [76], Pio and col-
leagues [77] measured early breast cancer response
using the thymidine analog, 18F-fluorothymidine
(FLT) with promising early results. Imaging cellular
death or apoptosis as an early measure of therapeu-
tic efficacy also holds promise, and has been tested
in preliminary studies of 99mTc-annexin V [78].
18
F-labeled versions of annexin have been tested
[79–81], and may offer advantages for quantitative
serial imaging.
Summary
The evaluation of breast cancer response to sys-
temic therapy is a clinical task to which quantitative
PET imaging is ideally suited. FDG PET already
plays a role in response evaluation for neoadjuvant
therapy of primary breast cancer and systemic ther-
apy of metastatic breast cancer. Breast cancer is one
of the few cancers for which FDG PET is approved
by US health care insurances for therapy response
evaluation. Increasing use of FDG PET to measure
breast cancer response is likely, especially in appli-
cations such as bone-dominant breast cancer, in
which conventional imaging is inadequate for mea-
suring response. Importantly, future use will likely
include use of PET radiopharmaceuticals beyond
FDG to help guide and monitor breast cancer sys-
temic therapy by helping to identify therapeutic
targets, and by determining response early in the
course of therapy.
References
[1] Fisher B, Wickerham DL, Redmond C. Recent
developments in the use of systemic adjuvant
therapy for the treatment of breast cancer. Semin
Oncol 1992;19(3):263–77.
[2] Thomas E, Hortobagyi GN. New paradigms in
adjuvant systemic therapy of breast cancer.
Endocr Relat Cancer 2003;10(1):75–89.
[3] Bonadonna G, Valagussa P, Zucali R, et al.
Primary chemotherapy in surgically resectable
breast cancer. CA Cancer J Clin 1995;45:227–43.
[4] Jemal A, Murray T, Samuels A, et al. Cancer Sta-
tistics, 2003. CA Cancer J Clin 2003;53(1):5–26.
[5] Hortobagyi GN. Progress in systemic chemother-
apy of primary breast cancer: an overview. J Natl
Cancer Inst Monogr 2001;30:72–9.
[6] Feldman LD, Hortobagyi GN, Buzdar AU, et al.
Pathological assessment of response to induction
chemotherapy in breast cancer. Cancer Res
1986;46:2578–81.
[7] Machiavelli MR, Romero AO, Perez JE, et al.
Prognostic significance of pathological response
of primary tumor and metastatic axillary lymph
nodes after neoadjuvant chemotherapy for lo-
cally advanced breast carcinoma. Cancer J Sci Am
1998;4(2):125–31.
[8] McCready DR, Hortobagyi GN, Kau SW, et al.
The prognostic significance of lymph node
metastases after preoperative chemotherapy for
locally advanced breast cancer. Arch Surg 1989;
124:21–5.
[9] Kaklamani V, O’Regan RM. New targeted therapies
in breast cancer. Semin Oncol 2004;31(2)(Suppl 4):
20–5.
[10] Moscovic EC, Mansi JL, King DM, et al. Mam-
mography in the assessment of response to
medical treatment of large primary breast tumor.
Clin Radiol 1993;47:339–44.
[11] Segel MC, Paulus DD, Hortobagyi GN. Advanced
primary breast cancer: assessment at mammog-
raphy of response to chemotherapy. Radiology
1988;169:49–54.
[12] Husband JE. Monitoring tumor response. Eur
Radiol 1996;6:775–85.
[13] Hoekstra CJ, Paglianiti I, Hoekstra OS, et al.
Monitoring response to therapy in cancer using
[18F]-2-fluoro-2-deoxy-D-glucose and positron
emission tomography: an overview of different
analytical methods. Eur J Nucl Med 2000;27:
731–43.
[14] Wahl RL, Zasadny K, Helvie M, et al. Metabolic
monitoring of breast cancer chemohormono-

therapy using positron emission tomography:
initial evaluation. J Clin Oncol 1993;11:2101–11.
[15] Mankoff DA, Dunnwald LK, Gralow JR, et al.
Changes in blood flow and metabolism in lo-
cally advanced breast cancer treated with neoad-
juvant chemotherapy. J Nucl Med 2003;44(11):
1806–14.
[16] Graham MM, Peterson LM, Hayward RM. Com-
parison of simplified quantitative analyses of
FDG uptake. Nucl Med Biol 2000;27(7):647–55.
[17] Huang S-C. Anatomy of SUV. Nucl Med Biol
2000;27:643–6.
[18] Lammertsma AA. Measurement of tumor re-
sponse using [18F]-2-fluoro-2-deoxy-D-glucose
and positron-emission tomography. J Clin Phar-
macol 2001;41:104–6.
[19] Mankoff DA, Muzi M, Krohn KA. Quantita-
tive positron emission tomography imaging to
measure tumor response to therapy: what is
the best method? Mol Imaging Biol 2003;5(5):
281–5.
[20] Keyes JWJ. SUV: standard uptake or silly useless
value? J Nucl Med 1995;36:1836–9.
[21] Beaulieu S, Kinahan P, Tseng J, et al. SUV varies
with time after injection in (18)F-FDG PET of
breast cancer: characterization and method to
adjust for time differences. J Nucl Med 2003;
44(7):1044–50.
[22] Thie JA, Hubner KF, Smith GT. Optimizing imag-
ing time for improved performance in oncology
PET studies. Mol Imaging Biol 2002;4(3):238–44.
[23] Sokoloff L, Reivich M, Kennedy C, et al. The
[14C]deoxyglucose method for the measurement
of local cerebral glucose utilization: theory,
procedure, and normal values in the conscious
and anesthetized albino rat. J Neurochem 1977;
28:897–916.
[24] Reivich M, Alavi A, Wolf A, et al. Glucose
metabolic rate kinetic model parameter determi-
nation in humans: the lumped constant and rate
constants for [18F]fluorodeoxyglucose and [11C]
deoxyglucose. J Cereb Blood Flow Metab 1985;5:
179–92.
[25] Phelps M, Huang S, Hoffman E. Tomographic
measurement of local cerebral glucose metabolic
rate in humans with (18F)2-fluoro-2-deoxy-D-
glucose: validation of method. Ann Neurol 1979;
6(5):371–88.
[26] Spence A, Muzi M, Graham M, et al. Glucose
metabolism in human malignant gliomas mea-
sured quantitatively with PET, 1-[C-11]glucose
and FDG: analysis of the FDG lumped constant.
J Nucl Med 1998;39:440–8.
[27] Tseng J, Dunnwald LK, Schubert EK, et al. 18F-
FDG kinetics in locally advanced breast can-
cer: correlation with tumor blood flow and
changes in response to neoadjuvant chemother-
apy. J Nucl Med 2004;45(11):1829–37.
[28] Zasadny KR, Tatsumi M, Wahl RL. FDG metabo-
lism and uptake versus blood flow in women
with untreated primary breast cancers. Eur J Nucl
Med Mol Imaging 2003;30(2):274–80.
PET & Breast Cancer Therapy 79
[29] Patlak CS, Blasberg RG. Graphical evaluation of
blood-to-brain transfer constants from multiple-
time uptake data. Generalizations. J Cereb Blood
Flow Metab 1985;5(4):584–90.
[30] Gjedde A. Calculation of cerebral glucose phos-
phorylation from brain uptake of glucose ana-
logs in vivo: a re-examination. Brain Res 1982;
257(2):237–74.
[31] Mankoff DA, Dunnwald LK, Gralow JR, et al.
Blood flow and metabolism in locally advanced
breast cancer: relationship to response to ther-
apy. J Nucl Med 2002;43(4):500–9.
[32] Beahrs OH. Staging of cancer. CA Cancer J Clin
1991;41(2):121–5.
[33] Schwartz GF, Birchansky CA, Komarnicky LT,
et al. Induction chemotherapy followed by breast
conservation for locally advanced carcinoma of
the breast. Cancer 1994;73:362–9.
[34] Schwartz GF, Hortobagyi GN. Proceedings of the
Consensus Conference on Neoadjuvant Che-
motherapy in Carcinoma of the Breast, April
26–28, 2003, Philadelphia, Pennsylvania. Cancer
2004;100(12):2512–32.
[35] Wong ZW, Ellis MJ. Neoadjuvant endocrine
therapy for breast cancer: an overlooked option?
Oncology (Williston Park) 2004;18(4):411–20.
[36] Valagussa P, Zambetti M, Bonadonna G, et al.
Prognostic factors in locally advanced nonin-
flammatory breast cancer. Long-term results
following primary chemotherapy. Breast Ca Res
Treat 1990;15:137–47.
[37] Wolmark N, Wang J, Mamounas E, et al.
Preoperative chemotherapy in patients with
operable breast cancer: nine-year results from
National Surgical Adjuvant Breast and Bowel
Project B-18. J Natl Cancer Inst Monogr 2001;
30:96–102.
[38] Minn H, Soini I. [18F]fluorodeoxyglucose scintig-
raphy in diagnosis and follow up of treatment
in advanced breast cancer. Eur J Nucl Med 1989;
15:61–6.
[39] Nason KS, Anderson BO, Byrd DR, et al.
Increased false negative sentinel node biopsy
rates after preoperative chemotherapy for inva-
sive breast carcinoma. Cancer 2000;89(11):
2187–94.
[40] Eubank WB, Mankoff DA. Current and future
uses of positron emission tomography in breast
cancer imaging. Semin Nucl Med 2004;34(3):
224–40.
[41] Bellon JR, Livingston RB, Eubank WB, et al.
Evaluation of the internal mammary lymph
nodes by FDG-PET in locally advanced breast
cancer (LABC). Am J Clin Oncol 2004;27(4):
407–10.
[42] Bassa P, Kim EE, Inoue T, et al. Evaluation of
preoperative chemotherapy using PET with
fluorine-18-fluorodeoxyglucose in breast cancer.
J Nucl Med 1996;37:931–8.
[43] Smith IC, Welch AE, Hutcheon AW, et al.
Positron emission tomography using [18F]-
fluorodeoxy-D-glucose to predict the pathologic
80 Mankoff & Dunnwald
response of breast cancer to primary chemother-
apy. J Clin Oncol 2000;18:1676–88.
[44] Schelling M, Avril N, Nahrig J, et al. Positron
emission tomography using [18F] fluorodeoxy-
glucose for monitoring primary chemotherapy in
breast cancer. J Clin Oncol 2000;18:1689–95.
[45] Burcombe RJ, Makris A, Pittam M, et al. Evalua-
tion of good clinical response to neoadjuvant
chemotherapy in primary breast cancer using
[18F]-fluorodeoxyglucose positron emission to-
mography. Eur J Cancer 2002;38(3):375–9.
[46] Kim SJ, Kim SK, Lee ES, et al. Predictive value of
[18F]FDG PET for pathological response of breast
cancer to neoadjuvant chemotherapy. Ann Oncol
2004;15(9):1352–7.
[47] Barranger E, Grahek D, Antoine M, et al. Evalua-
tion of fluorodeoxyglucose positron emission
tomography in the detection of axillary lymph
node metastases in patients with early-stage breast
cancer. Ann Surg Oncol 2003;10(6):622–7.
[48] Chen X, Moore MO, Lehman CD, et al.
Combined use of MRI and PET to monitor
response and assess residual disease for locally
advanced breast cancer treated with neoadju-
vant chemotherapy. Acad Radiol 2004;11(10):
1115–24.
[49] Seidman A. Introduction. Single-agent or com-
bination chemotherapy in metastatic breast can-
cer. Oncology (Williston Park) 2003;17(Suppl 14):
9–14.
[50] Gennari A, Donati S, Salvadori B, et al. Role of
2-[18F]-fluorodeoxyglucose (FDG) positron emis-
sion tomography (PET) in the early assessment
of response to chemotherapy in metastatic breast
cancer patients. Clin Breast Cancer 2000;1(2):
156–61.
[51] Stafford SE, Gralow JR, Schubert EK, et al. Use of
serial FDG PET to measure the response of bone-
dominant breast cancer to therapy. Acad Radiol
2002;9(8):913–21.
[52] Nielsen OS, Munro AJ, Tannock IF. Bone
metastases: pathophysiology and management
policy. J Clin Oncol 1991;9(3):509–24.
[53] Bast Jr RC, Ravdin P, Hayes DF, et al. 2000 up-
date of recommendations for the use of tumor
markers in breast and colorectal cancer: clini-
cal practice guidelines of the American Society
of Clinical Oncology. J Clin Oncol 2001;19(6):
1865–78.
[54] Vogel CL, Schoenfelder J, Shemano I, et al. Wors-
ening bone scan in the evaluation of antitumor
response during hormonal therapy of breast
cancer. J Clin Oncol 1995;13(5):1123–8.
[55] Schneider JA, Divgi CR, Scott AM, et al. Flare on
bone scintigraphy following Taxol chemotherapy
for metastatic breast cancer. J Nucl Med 1994;
35(11):1748–52.
[56] Mundy GR. Mechanisms of bone metastasis.
Cancer 1997;80(Suppl):1546–56.
[57] Cook GJ, Houston S, Rubens R, et al. Detection
of bone metastases in breast cancer by 18FDG
PET: differing metabolic activity in osteoblastic
and osteolytic lesions. J Clin Oncol 1998;16(10):
3375–9.
[58] Tam SL, Gralow JR, Livingston RB, et al.
Serial FDG-PET to monitor treatment of bone-
dominant metastatic breast cancer predicts time
to progression (TTP). Proceedings of ASCO 2005;
23:36s.
[59] Mortimer JE, Dehdashti F, Siegel BA, et al.
Metabolic flare: indicator of hormone respon-
siveness in advanced breast cancer. J Clin Oncol
2001;19(11):2797–803.
[60] Beaney R, Jones T, Lammertsma A, et al. Positron
emission tomography for in-vivo measurement
of regional blood flow, oxygen utilisation, and
blood volume in patients with breast carcinoma.
Lancet 1984;1(8369):131–4.
[61] Wilson CBJH, Lammertsma AA, McKenzie CG,
et al. Measurements of blood flow and exchang-
ing water space in breast tumors using positron
emission tomography: a rapid and non-invasive
dynamic method. Cancer Res 1992;52:1592–7.
[62] Abraham DC, Jones RC, Jones SE, et al. Evalua-
tion of neoadjuvant chemotherapeutic response
of locally advanced breast cancer by magnetic
resonance imaging. Cancer 1996;78:91–100.
[63] Esserman L, Kaplan E, Partridge S, et al. MRI
phenotype is associated with response to doxo-
rubicin and cyclophosphamide neoadjuvant che-
motherapy in stage III breast cancer. Ann Surg
Oncol 2001;8(6):549–59.
[64] Gilles R, Guinebretiere JM, Toussaint C, et al. Lo-
cally advanced breast cancer: contrast-enhanced
subtraction MR imaging of response to pre-
operative chemotherapy. Radiology 1994;191:
633–8.
[65] Knopp MV, Brix G, Junkerman HJ, et al. MR
mammography with pharmacokinetic mapping
for monitoring breast cancer treatment during
neoadjuvant chemotherapy. Magn Reson Imag-
ing Clin N Am 1994;2(4):633–58.
[66] Kedar RP, Cosgrove DO, Smith IE, et al. Breast
carcinoma: measurement of tumor response to
primary medical therapy with color flow Dopp-
ler imaging. Radiology 1994;190:825–30.
[67] Peters-Engl C, Frank W, Leodolter S, et al. Tumor
flow in malignant breast tumors measured by
Doppler ultrasound: an independent predictor
of survival. Breast Cancer Res Treat 1999;54(1):
65–71.
[68] Maini CL, Tofani A, Sciuto R, et al. Technetium-
99m-MIBI scintigraphy in the assessment of
neoadjuvant chemotherapy in breast carcinoma.
J Nucl Med 1997;38:1546–50.
[69] Mankoff DA, Dunnwald LK, Gralow JR, et al.
Monitoring the response of patients with locally
advanced breast carcinoma to neoadjuvant che-
motherapy using [technetium-99m]-sestamibi
scintimammography. Cancer 1999;85:2410–23.
[70] Tiling R, Kessler M, Untch M, et al. Breast cancer:
monitoring response to neoadjuvant chemother-
apy using Tc-99m sestamibi scintimammogra-
phy. Onkologie 2003;26(1):27–31.

[71] Mankoff DA, Dunnwald LK, Gralow JR, et al. [Tc-
99m]-sestamibi uptake and washout in locally
advanced breast cancer are correlated with tumor
blood flow. Nucl Med Biol 2002;29(7):719–27.
[72] Dunnwald LK, Gralow JR, Ellis GK, et al.
Residual tumor uptake of [99mTc]-sestamibi after
neoadjuvant chemotherapy for locally advanced
breast carcinoma predicts survival. Cancer 2005;
103(4):680–8.
[73] Dehdashti F, Flanagan FL, Mortimer JE, et al.
Positron emission tomographic assessment of
‘‘metabolic flare’’ to predict response of meta-
bolic breast cancer to antiestrogen therapy. Eur J
Nucl Med 1999;26:51–6.
[74] Huovinen R, Leskinen-Kallio S, Någren K, et al.
Carbon-11-methionine and PET in evaluation of
treatment response of breast cancer. Br J Cancer
1993;67:787–91.
[75] Jansson T, Westlin J, Ahlstrom H, et al. Positron
emission tomography studies in patients with
locally advanced and/or metastatic breast cancer:
a method for early therapy evaluation? J Clin
Oncol 1995;13(6):1470–7.
PET & Breast Cancer Therapy 81
[76] Shields AF, Mankoff DA, Link JM, et al. Carbon-
11-thymidine and FDG to measure therapy
response. J Nucl Med 1998;39(10):1757–62.
[77] Pio BS, Park CK, Satyamurthy N, et al. PET with
fluoro-L-thmyidine allows early prediction of
breast cancer response to chemotherapy. J Nucl
Med 2003;44(5):76P.
[78] Blankenberg F, Ohtsuki K, Strauss HW. Dying a
thousand deaths. Radionuclide imaging of apop-
tosis. Q J Nucl Med 1999;43(2):170–6.
[79] Grierson JR, Yagle KJ, Eary JF, et al. Production
of [F-18]fluoroannexin for imaging apoptosis
with PET. Bioconjug Chem 2004;15(2):373–9.
[80] Yagle KJ, Eary JF, Tait JF, et al. Evaluation of
18
F-Annexin V as a PET imaging agent in an
animal model of apoptosis. J Nucl Med 2005;
46(4):658–66.
[81] Zijlstra S, Gunawan J, Burchert W. Synthesis
and evaluation of a 18F-labelled recombinant
annexin-V derivative, for identification and quan-
tification of apoptotic cells with PET. Appl Radiat
Isot 2003;58(2):201–7.

Histopathologic and Metabolic
Criteria for Assessment of
Treatment Response in
Breast Cancer
a,*
Stefanie Sassen, MD , Falko Fend,
& Neoadjuvant treatment of breast cancer
& Assessment of treatment response by
anatomical imaging
& Assessment of treatment response by
histopathology
Predictive markers of pathologic
response
Limitations and future perspectives of
histopathologic response criteria
& Prediction of response to chemotherapy by
F-18 fluorodeoxyglucose-positron
emission tomography
Most breast cancers develop from the terminal
ducto-lobular unit, which comprises the glandular
component of normal breast parenchyma. There-
fore, virtually all breast cancers are adenocarcino-
mas. Invasive ductal carcinoma is the most common
histological type (70%–80%; Fig. 1), followed by
invasive lobular carcinoma (~10%; Fig. 2). A com-
bination of lobular and ductal features is not
uncommon [1]. The remaining cases fall into a
variety of histological types that are generally asso-
ciated with less aggressive biologic behavior and a
better prognosis. These include mucinous (mucoid
or colloid) carcinoma, consisting of neoplastic cells
floating in pools of extravasated mucin; and med-
ullary carcinoma, which is characterized by large
pleomorphic cells, a prominent intratumoral lym-
a
Department of Pathology, Technical University Munich, Ismaningerstrasse 22, 81675 Munich, Germany
b
Department of Nuclear Medicine, Barts and the London School of Medicine, Queen Mary, University of
London, West Smithfield (QEII), London, EC1A 7BE, UK
* Corresponding author.
E-mail address: stefanie.sassen@lrz.tum.de (S. Sassen).
1556-8598/06/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
pet.theclinics.com
83
POSITRON
EMISSION
TOMOGRAPHY
PET Clin 1 (2006) 83–94
a b
MD , Norbert Avril, MD
& Limitation of F-18
fluorodeoxyglucose-positron emission
tomography for prediction of treatment
response
& Comparison between histopathologic and
metabolic response criteria
& References
phoplasmacytic infiltrate, and frequent lack of estro-
gen receptor expression [1].
Grading of breast cancer is usually performed
according to Bloom and Richardson, modified by
Elston and Ellis [2]. It takes into account the fre-
quency of mitotic cells (rate of cell division), the
architecture, namely tubule formation (percentage
of cancer composed of tubular structures), and nu-
clear pleomorphism (variation in size and shape
of nuclei). Each of these features is assigned a score
ranging from 1 to 3 (3 indicating higher prolifera-
tive activity, nuclear atypia, and less differentiated
architecture). After adding the scores, breast can-
cer is graded as well-differentiated Grade I (score 3,
4 or 5), moderately-differentiated Grade II (score 6
or 7), or poorly-differentiated Grade III (score 8
doi:10.1016/j.cpet.2005.09.007
84 Sassen et al
Fig. 1. Invasive ductal carcinoma, moderately differ-
entiated (grade 2), with features of glandular dif-
ferentiation, fibrosis, and diffuse infiltration of the
surrounding fat.
or 9). The histologic grade is of prognostic rele-
vance, especially in smaller-sized tumors that show
a strong correlation between higher grade and
shorter disease-free and overall survival [3].
The prognostic factors in locally advanced breast
cancer are similar to those for earlier tumor stages,
with lymph node status and tumor size having the
strongest effect on survival. The prognostic informa-
tion of the hormone receptor status is less apparent,
although earlier studies found a better prognosis for
patients who had estrogen or progesterone recep-
tor positive breast cancer [4,5]. Independent of its
rather complex prognostic relevance, the assessment
of hormone receptor status, nowadays usually by
immunohistochemistry, is very useful to predict re-
sponse to hormonal treatment. Human epidermal
growth factor receptor 2 (HER-2/ErbB2), a transmem-
brane protein receptor, is a key component in regu-
lating cell growth. Alterations in the HER-2 gene with
subsequent overexpression of HER-2/neu protein
affect approximately 25% to 30% of breast cancer
patients, and are associated with poor clinical out-
come [6,7]. Other extensively studied prognostic
markers, including measurement of cell prolifera-
tion and mutations in p53, have not consistently
emerged as independent predictors of survival [8,9].
Neoadjuvant treatment of breast cancer
The treatment of newly diagnosed breast cancer
is primarily determined by the extent of disease,
and generally consists of surgery—either breast-
conserving surgery, usually combined with post-
operative radiation therapy, or total mastectomy.
Axillary lymph node dissection has now been re-
placed by sentinel lymph node biopsy in many
patients who have clinically negative axillary find-
ings. In addition, adjuvant systemic therapy is fre-
quently recommended after surgery to reduce the
risk for tumor recurrence, especially in patients who
have positive lymph nodes, tumors of more than
1 cm in size, or other unfavorable prognostic fea-
tures. Doxorubicin and cyclophosphamide, often
combined with taxanes, is the most common adju-
vant therapy [10,11]. Anti-estrogen treatment with
tamoxifen or other estrogen antagonists is often ap-
plied to postmenopausal patients who have estro-
gen receptor-positive tumors.
Neoadjuvant (preoperative) chemotherapy is in-
creasingly used for managing patients who have
locally advanced breast cancer. Disease-free and
overall survival and loco-regional control are simi-
lar to conventional adjuvant chemotherapy; how-
ever, neoadjuvant chemotherapy reduces the tumor
volume, which enables more patients to be treated
with breast-conserving surgery [12]. In addition,
this approach frequently produces better cosmetic
results and allows for observation of response
to chemotherapy. Approximately 70% of patients
demonstrate clinical response to neoadjuvant che-
motherapy, but only 20% to 40% have partial or
complete pathologic response [12–14]. The assess-
ment of response to neoadjuvant chemotherapy
provides important prognostic information, and
patients who had complete pathological response
were found to have significantly longer disease-free
and overall survival than nonresponders [15,16].
An important aspect of neoadjuvant treatment is
that response reflects an individual in-vivo assay of
tumor chemosensitivity, which ultimately deter-
mines survival [17].
Monitoring response to treatment is highly rele-
vant in the current clinical setting. Chemotherapy
for breast cancer is still associated with significant
side effects, such as nausea, vomiting, alopecia, vari-
ous grades of hematological toxicity with the risk
of subsequent (life-threatening) infections, and po-
tential neurotoxicity (eg, taxanes) and cardiotoxicity
Fig. 2. Invasive lobular carcinoma, moderately differ-
entiated (grade 2), with a typical Indian-file pattern
and concentrically surrounding non-tumor–involved
ducts (arrow).

(eg, anthracyclines). Early identification of tumor
resistance allows one to avoid ineffective therapies
and unnecessary side effects, and to change to
potentially more effective treatments. The rate of
complete pathologic response after primary che-
motherapy varies between 3% and 27% [13–16,
18–21]. The largest series of neoadjuvant chemo-
therapy [14], with 608 breast cancer patients, found
a complete pathologic response rate of 20%. Of
note, in most cases, no definite correlation was
found between clinical response and pathologic re-
sponse. Only about one third of patients with a
complete clinical response also had a complete
pathologic response [14,18]. Conversely, a com-
plete pathologic response can be observed in some
patients who have partial or no clinical response,
mainly because of the persistence of collagenous
scar tissue despite disappearance of tumor cells.
Assessment of treatment response by
anatomical imaging
The ultimate goal of breast cancer treatment is to
achieve cure, or at least a complete remission. The
current standard end point for assessing response
to therapy in solid tumors is by measuring the
change in tumor size [22]. Anatomical imaging,
predominantly CT, MRI, and ultrasound (US) is
used to obtain unidimensional or bidimensional
measurements of reference tumor lesions from
pretreatment scans relative to follow-up; however,
there are several limitations of anatomic imaging-
based assessment of treatment response. According
to the World Health Organization criteria, a tumor
is classified as responding when the product of two
perpendicular diameters has decreased by at least
50%. In cases that have multiple lesions, the sum-
mation of the products should decrease by more
than 50% [23]. More recently, the response evalua-
tion criteria in solid tumors (RECIST) define tumor
response as a decrease of the maximum tumor di-
ameter by at least 30% [22]. For a spherical tumor,
a decrease of its diameter by 30% corresponds to a
decrease of the product of two perpendicular di-
ameters by 50%. Therefore, the RECIST criteria
have been primarily designed to simplify and stan-
dardize the tumor size measurement, and have not
changed the threshold values for definition of a
tumor response. Despite the widely accepted prac-
tice of using these criteria, it is important to note
that a 30% (or 50%) decrease in tumor size is a
more or less arbitrary convention, which is not
based on results of outcome studies. In fact, the
correlation between radiographic tumor response
and patient outcome is frequently weak. Therefore,
tumor response as defined by anatomical imaging
techniques is frequently not accepted as a surrogate
Histopathologic and Metabolic Response Criteria 85
end point for the evaluation of new anticancer
drugs [24]. Furthermore, dissolving and shrinkage
of a tumor mass is the final step in a complex cas-
cade of cellular and subcellular changes after initia-
tion of treatment. Frequently, several cycles of
chemotherapy or other treatment modalities need
to be applied before treatment response can be as-
sessed by current anatomical imaging. Even the eval-
uation of response after completion of treatment
can pose a challenge. If a residual mass is present,
it is difficult to differentiate viable tumor tissue from
post-treatment changes such as scarring and fibrosis.
Assessment of treatment response by
histopathology
Histopathologic tumor regression is the current
gold standard for assessment of treatment response
in several types of tumors. Most criteria have been
established for response evaluation of surgical
specimens after neoadjuvant chemotherapy. Re-
sponse criteria were initially developed for osteosar-
coma, with six response grades describing the range
of histopathological treatment changes based on
the percentage of residual viable tumor [25].
Similar histopathologic response criteria have been
defined for gastric cancer [26], esophageal cancer
[27], pancreatic cancer [28], and non-small–cell lung
cancer [29]. Generally, the surgical specimens are
analyzed for residual viable tumor relative to the
macroscopically identifiable tumor bed. Two thresh-
olds have been identified to provide prognostic in-
formation following neoadjuvant chemotherapy:
response defined by less than 10% viable tumor,
and minimal or no response defined by more than
50% residual tumor.
For breast cancer, histopathological response to
neoadjuvant chemotherapy is based on the pres-
ence or absence of residual tumor and on regressive
changes within the remaining tumor tissue [20,30].
Regressive changes are observed in approximately
50% of breast cancers after chemotherapy [14,31].
Regression is commonly defined as replacement of
invasive tumor by fibrous or hyaline stroma, along
with a reduction in the relative number of tumor
cells [Figs. 3, 4]. Regressive changes also include
cellular changes, such as cytoplasmic vacuolization
of tumor cells [Fig. 5] [32,33], an increase in cell
size, and the presence of giant tumor cells [Fig. 6]
[14,20,34,35]. Regression in the tumoral stroma is
characterized by lymphoplasmacytic or histiocytic
infiltrates, iron-storing macrophages, or calcifica-
tions [14,36]. Common changes within the normal
breast tissue after chemotherapy include lobular
atrophy, sclerosis of ductal and ductolobular base-
ment membranes, and mild epithelial atypia [34].
86 Sassen et al

Fig. 3. Highly pleomorphic tumor cells scattered within

Fig. 5. Cytoplasmic vacuolization of tumor cells (arrows).
extensive hyaline fibrosis.
compared with the whole patient group [37]. A
Chevallier and colleagues [30] and Sinn and co-
significantly longer overall survival in breast cancer
workers [20] defined various response grades fol-
patients who had pathologic complete response
lowing neoadjuvant chemotherapy of breast cancer
after neoadjuvant chemotherapy has also been re-
using opposite scaling directions [Table1]. Patho-
ported by others [13,14,16,19,38].
logic complete response after neoadjuvant che-
Honkoop and colleagues [38,39] suggested a
motherapy is generally defined by the absence of
modified classification for pathologic response. In
residual invasive tumor. Consequently, patients
42 breast cancer patients, they found no difference
who have residual ductal carcinoma in situ are
in disease-free and overall survival for patients who
still classified as pathologic complete response in
had scattered microscopic foci of tumor cells and
most studies [see Fig. 4]. The largest study of
no macroscopically visible residual tumor, com-
neoadjuvant chemotherapy in breast cancer, the
pared with patients who had complete pathologic
National Surgical Adjuvant Breast and Bowel Pro-
response. These two groups were combined in a re-
ject (NSABP) B-18 trial [37], showed a significant
sponse category ‘‘minimal residual disease’’ (MRD),
correlation between primary tumor response and
and all other patients were classified as ‘‘gross re-
patient outcome. At 9 years of follow-up, the over-
sidual disease’’ (GRD). Patients who had MRD after
all survival rate for patients who had a pathologic
chemotherapy had a 2-year overall survival rate of
complete response was 85%, compared with 73%
94%, compared with 68% for patients who had
in patients who had residual invasive tumor [37].
GRD. The respective rates for disease-free survival
The respective rates for disease-free survival were
were 80% and 45% [38].
75% and 58%. After adjustment for other prognos-
The presence of axillary lymph node metastasis is
tic factors, patients who had a pathologic complete
generally associated with a poor prognosis. The
response had a 50% reduction in risk of death
lymph node status remains an important inde-
pendent prognostic factor, even after neoadjuvant
chemotherapy. Disease-free and overall survival
correlated significantly with the number of resid-

Fig. 4. Residual intraductal carcinoma within fibrotic

scar tissue (arrow). Intraductal residual tumor is com-
monly classified as complete pathologic response
(response Grade III [20].). Fig. 6. The appearance of giant tumor cells (arrow).

Table 1: Histopathologic response classifications
for breast cancer
Chevallier Sinn
Response et al [30] et al [20]
No residual tumor Grade I Grade IV
No residual invasion Grade II Grade III
tumor
Minimal residual — Grade II
tumor with
extensive fibrosis
Residual tumor Grade III Grade I
with fibrosis and
regressive changes
No apparent Grade IV Grade 0
histopathological
changes
ual tumor-involved axillary lymph nodes [14,31,
38,40]. In 135 patients who had locally advanced
breast cancer, patients who had tumor-free axillary
lymph nodes after neoadjuvant chemotherapy had
an overall 5-year survival rate of 71%, compared
with 54% in patients who had one to nine tumor-
involved lymph nodes [31]. The overall 5-year sur-
vival rate was only 35% in patients who had 10 or
more positive lymph nodes after neoadjuvant che-
motherapy [31]. The NSABP B-18 trial also noted a
significantly longer overall and disease-free survival
for patients who had negative axillary lymph nodes
compared with either one to three or more than
three tumor-involved lymph nodes after neoadju-
vant chemotherapy [14]. There was a significant
correlation between the pathologic response of the
primary tumors and the axillary lymph nodes
[14,16,31]. Kuerer and coworkers [16] studied the
histopathologic response of lymph nodes in 356 pa-
tients, and noted residual tumor involvement of
axillary lymph nodes in 28% of patients who
achieved a complete pathologic response in their
primary tumor, compared with 74% of patients
who did not have a complete pathologic response.
Predictive markers of pathologic response
A complete pathologic response of the primary
tumor and of axillary lymph nodes after che-
motherapy is an important predictive factor for
survival. Pathologic response was consistently cor-
related with smaller tumor size at initial diagnosis
[16,18,31] and a lack of estrogen-receptor expres-
sion [16,41,42]. In several studies, including the
NSABP B-18 trial [14,16,43], a high nuclear grade
at initial diagnosis was predictive of a subsequent
pathologic response to chemotherapy. Mutations
in the p53 gene did not predict response to che-
motherapy in breast cancer [44–46]. Results regard-
ing the role of HER-2/neu are still conflicting
Histopathologic and Metabolic Response Criteria 87
[42,47]. A measurable increase in apoptosis was
observed within 1 to 2 days after initiation of che-
motherapy in breast cancer, but this does not allow
consistently predicting response to chemother-
apy so far [43,48]. Currently, there are no reliable
markers available to predict treatment response for
individual patients based on the pretreatment
breast cancer specimens [11].
In-vitro chemosensitivity testing in breast cancer
was introduced over 2 decades ago [49]; however,
the need for adequate tissue samples and low yield
of tumor cells from breast biopsies has limited
the implementation in clinical routine use [50]. It
remains uncertain if chemosensitivity of cancer
cells in two-dimensional (2D) culture assays is
comparable to the chemoresponsiveness in vivo,
which is also determined by local tissue-specific
factors, the three-dimensional (3D) microenviron-
ment, and individual metabolism and clearance
of the chemotherapeutic agents. Using an extreme
drug resistance assay, Mehta and colleagues [51]
demonstrated in a study of 103 newly diagnosed
breast cancer patients that extreme drug resistance
scores were significantly associated with time to
tumor progression and overall survival. Newer as-
says are able to test smaller tumors.
Limitations and future perspectives of
histopathologic response criteria
It is important to note that regressive changes such
as stromal hyalinization and fibrosis, necrosis, or
reactive inflammation may be a priori present in
breast cancer specimens. The NSABP B-18 trial [14]
noted a higher frequency of cytoplasmic vacuoliza-
tion in tumor cells, the presence of giant tumor
cells, and stromal hyalinization in patients who
had received preoperative chemotherapy. Most of
these histologic changes have a moderate-to-high
specificity (48%–99%), but relatively low sensitivity
(7%–67%), when compared with nontreated tumor
specimens. Thus, there is no single morphologic
marker that identifies treatment-related changes
with high accuracy in the majority of patients [14].
In contrast to complete pathological response,
the presence of chemotherapy-induced regressive
changes was not related to better patient outcome
[14,52]. The clinical relevance and potential prog-
nostic information of histomorphologic and cyto-
pathic changes after chemotherapy still remains to
be defined. The histopathologic assessment of sur-
gical specimens may also be affected by sampling
error, depending on the technique used to examine
the initial tumor bed. Currently, pathologic response
is evaluated after several cycles of chemotherapy
(three to six cycles) or several months of endocrine
therapy; however, this approach does not address
the early identification of nonresponders, which is
88 Sassen et al
of critical importance because these patients may
benefit from alternative treatment regimens [53].
Recently, various molecular markers, including
large-scale gene expression profiling, and the pres-
ence of tumor cells in the blood have been sug-
gested to be helpful in predicting response to
treatment [54,55]. Gene expression profiling with
microarrays allows simultaneous assessment of
thousands of genes potentially involved in the sen-
sitivity or resistance of cancer cells to chemotherapy
[56]. In breast cancer, recent studies identified gene
expression profiles with independent prognos-
tic information in early disease stages [57,58]. A
70-gene profile predicted the outcome in young
patients who had Stage I or II breast cancer more
accurately than standard clinical and histological
criteria [58]. A poor-prognosis gene profile was
associated with a fivefold increased relative risk of
distant metastases as compared with patients who
had a good-prognosis profile. Recent studies also
focused on the potential use of gene expression
profiles for prediction of response to neoadjuvant
chemotherapy. A gene expression signature com-
prising of 92 genes predicted clinical response to
a taxane-based neoadjuvant chemotherapy with
positive and negative predictive values of 92%
and 83%, respectively [59]. Another group also
identified a gene expression signature comprising
of 74 genes that predicted pathologic response with
an accuracy of 78% and a positive predictive value
of 100% [60]. Principal limitations of this promis-
ing approach currently include the need for ade-
quate tissue samples, tumor heterogeneity, and
host factors such as drug delivery and metabolism
that may not be reflected by gene expression pro-
files of tumor cells.
Prediction of response to chemotherapy by
F-18 fluorodeoxyglucose-positron emission
tomography
The uptake mechanism and biochemical pathway
of the glucose analog [F-18] fluorodeoxyglucose
(FDG) has been extensively studied in vitro and
in vivo. The transport of the radiotracer through
the cell membrane via glucose transport proteins
(GLUT) and subsequent intracellular phosphoryla-
tion by hexokinase (HK) have been identified as
key steps for subsequent tissue accumulation [61].
Because FDG-6-phosphate is not a suitable substrate
for glucose-6-phosphate isomerase, and the enzyme
levels of glucose-6-phosphatase are generally low in
tumors, the intracellular accumulation of FDG-6-
phosphate reflects tissue glucose metabolism.
Attenuation-corrected PET images provide quan-
titative information about the tracer concentra-
tion in tissue, which is particularly important for
monitoring therapeutic response. Frequently, stan-
dardized uptake values (SUV) are being calculated,
providing a semiquantitative measure of FDG ac-
cumulation in tissue by normalizing the tissue
radioactivity concentration measured with PET to
injected dose and patient’s body weight [62]. Pre-
vious studies have demonstrated that SUV provide
highly reproducible parameters of tumor glucose
use [63,64]. Weber and coworkers [64] studied
16 patients twice within 10 days with FDG-PET
while they were receiving no therapy. None of vari-
ous semiquantitative parameters of FDG uptake
showed a significant increase or decrease between
the two PET scans. The study authors found that
changes of a parameter of more than 20% are out-
side the 95% range for spontaneous fluctuations,
and can be considered to reflect true changes in
tumor glucose metabolism. At least two sequential
FDG-PET scans are currently necessary for predic-
tion of treatment response: one PET scan before
treatment serving as baseline, and one PET scan
after initiation of chemotherapy (eg, after the first
or second cycle). Comparing the percentage change
of FDG uptake in tumors provides specific infor-
mation about the response to a given treatment.
This characteristic behavior of tumor metabolism is
based on the observation that the majority of
changes in FDG uptake occur early after initiation
of treatment. In successfully treated non-Hodgkin’s
lymphoma for example, the FDG uptake decreased
by 60% to 67% from baseline to day 7 [65]. Simi-
lar changes were also observed in breast cancer
[66,67].
In 1993, Wahl and colleagues [68] studied
11 women who had newly diagnosed, locally ad-
vanced primary breast cancers and who were un-
dergoing chemo-hormonotherapy with sequential
FDG-PET imaging. Patients had a baseline and four
follow-up PET scans during the first three cycles of
treatment, and final tumor response was deter-
mined by histopathology after nine cycles of treat-
ment. The FDG uptake in 8 patients who had
partial or complete pathologic responses decreased
promptly with treatment. On day 8 of therapy,
mean tumor FDG uptake was 78% of baseline,
and decreased further to 68% at day 21, 60% at
day 42, and 52% at day 63. In contrast, 3 patients
who had nonresponding tumors did not show a
significant decrease in FDG uptake. In 30 breast
cancer patients who received eight doses of primary
chemotherapy, the mean reduction in FDG uptake
after the first cycle of chemotherapy was signifi-
cantly higher in lesions with partial, complete mac-
roscopic, or complete microscopic response [67].
After a single cycle of chemotherapy, PET predicted
complete pathologic response with a sensitivity of
 Histopathologic and Metabolic Response Criteria 89

90% and specificity of 74%. Schelling and cowork-
ers [66] compared results from PET imaging with
pathological response using distinct histopatho-
logical criteria, namely MRD and GRD, previously
identified to provide prognostic information [38].
FDG uptake in breast cancer after the first and
second cycle of chemotherapy was compared with
baseline PET. Patients classified as responders
by histopathology had a significantly more pro-
nounced decrease in FDG uptake than nonre-
sponding patients. Therefore, responding and
nonresponding tumors could be differentiated by
PET as early as after the first cycle of chemotherapy.
By a threshold defined as an SUV decrease below
55% compared with the baseline, all responders
were correctly identified after the first cycle (sensi-
tivity 100%, specificity 85%). The accuracy in pre-
dicting histopathological response was 88% after
the first cycle of therapy and 91% after the second.
Other studies also addressed the role of FDG-PET
at later time points during and after chemotherapy
[Fig. 7] [69–71].
It is important to note that a transient increase
in glucose use has been found in responding tu-
mors treated with hormonotherapy. In a first series,
Dehdashti and colleagues [72] performed FDG-PET
in 11 women before and approximately 1 week
after initiation of tamoxifen therapy. In all patients,
clinical and radiological follow-up was performed
with a median interval of 12 months. Seven pa-
tients responded, and all showed an initial increase
of FDG uptake 1 week after therapy. This ‘‘meta-
bolic flare’’ effect is a recognized side effect of anti-
estrogen therapy, which is clinically characterized
by pain and erythema in soft-tissue lesions and
increased pain in osseous metastases. An expla-
nation for this metabolic flare effect is that anti-
estrogen therapy first has an agonist effect before
the antagonist effect overrules. This agonist effect
occurs within the first 1 to 2 weeks after the begin-

Fig. 7. Sixty-eight-year-old breast cancer patient, postmastectomy of the left breast. Baseline FDG-PET/CT before
chemotherapy (A) demonstrates recurrence in the right axilla. FDG-PET/CT after chemotherapy (B) shows a
decrease in the size of axillary lymph nodes, but mild increased FDG uptake indicates viable residual tumor.
90 Sassen et al
ning of a treatment, and is usually followed by
disease remission. Recently, the same group con-
firmed their findings in a larger series including
40 patients [73]. In responders, the tumor FDG up-
take increased after tamoxifen by 28.4%, with only
5 of these patients having evidence of a clinical flare
reaction. In nonresponders, there was no signifi-
cant change in tumor FDG uptake from baseline.
There are no reports so far about metabolic flare
in more aggressive chemotherapeutic regimens.
There are significant differences between the neo-
adjuvant setting and the treatment of metastatic
breast cancer. Generally, more aggressive chemo-
therapeutic regimens are used in the neoadjuvant
setting compared with a large variety of treatment
options available for metastatic disease. Modern
treatment regimens are developing toward individual-
ized treatment tailored to the need of each patient,
emphasizing the importance of methods for predic-
tion of treatment response. There is little informa-
tion available on the clinical utility of sequential
FDG-PET imaging in patients who have metastatic
disease. A decrease in FDG uptake was observed 6
to 13 days after the first cycle of chemotherapy in 8
out of 12 patients with metastatic disease who had
responded to chemotherapy [74]. Gennari and co-
workers [75] also observed a rapid and significant
decrease in tumor glucose metabolism after the first
course of chemotherapy in 6 out of 9 responding
patients, and no significant decrease in nonrespond-
ing patients. A close correlation was found between
changes in FDG uptake and the overall clinical
assessment of response in breast cancer patients
who had bone metastases [76].
Dose Schwarz and colleagues [77] recently con-
firmed previous reports on the predictive informa-
tion of early changes in glucose metabolism after
initiation of chemotherapy in patients who had
metastatic breast cancer. Changes in tumor FDG
uptake were statistically significantly different be-
tween responding and nonresponding metastatic
lesions after the first and second cycle of chemo-
therapy. When compared with the baseline FDG-
PET, the FDG uptake in responding metastatic
lesions decreased to 72 ± 21% after the first cycle,
and to 54 ± 16% after the second cycle of che-
motherapy. In contrast, the FDG uptake in metas-
tases not responding to chemotherapy declined
only to 94 ± 19% after the first cycle, and to 79 ±
9% after the second cycle of chemotherapy. FDG-
PET whole-body imaging correctly predicted the
response in all patients as early as after the first
cycle (3 weeks) of chemotherapy. FDG-PET was
more accurate than conventional imaging proce-
dures obtained after the third cycle (9 weeks) of
chemotherapy, which misclassified three patients
(27%) falsely as responders. A specific strength of
FDG-PET appears to be the identification of non-
responders who are characterized by virtually no
change in FDG uptake after initiation of che-
motherapy. The overall survival in patients who
did not respond according to FDG-PET was 8.8 ±
6.7 months, compared with 19.2 ± 13.6 months in
responding patients. An important finding of the
study from Dose Schwarz and coworkers [77] is
that by using a PET-based strategy, 14 cycles of
chemotherapy could have been avoided in 5 non-
responding patients, and they could have earlier
received second-line chemotherapy.
Limitation of F-18
fluorodeoxyglucose-positron emission
tomography for prediction of treatment
response
Although numerous studies indicate that SUV may
be used to determine quantitative changes in tumor
glucose use over time, it is important to note that
adherence to a strict PET protocol is required to
minimize the variation between FDG-PET studies.
The radioactivity in the syringe should be measured
before and after injection, and corrected for radio-
active decay to accurately determine the dose of
FDG injected. Extravasation of the radiotracer dur-
ing injection will influence quantification of PET
imaging. The radioactivity concentration in the
body measured by PET is affected by the time
interval between tracer administration and PET
imaging. Many institutions use a 60-minute uptake
period in which the patient is resting before ac-
quiring the PET scan. The uptake period needs to
be standardized, and should vary less than 5 min-
utes between baseline and follow-up PET. Beaulieu
and colleagues [78] recently described a method to
adjust for time differences. There is an inverse rela-
tionship between FDG uptake and blood glucose
level that has to taken into account for the compari-
son of serial PET scans. The patient’s blood glucose
levels should be comparable between the PET
scans, and ideally less than 150 mg/dl at the time
of tracer injection. The specific method for calcula-
tion of SUV, for example, normalization to body
weight or the lean body mass, is less important for
the purpose of therapy monitoring because intra-
individual comparisons of SUV are being made.
When FDG-PET is used for early assessment of ther-
apy response, significant changes in body weight
influencing the results of the SUV calculation are
unlikely. In contrast, normalization to body surface
area or lean body mass is preferable for interindi-
vidual comparison of SUV, especially if the study
population includes obese patients [62].
The SUV measurements are also affected by the
size of a lesion. Partial volume effects cause signifi-

cant underestimation of isotope concentration in
structures smaller than two times the scanner reso-
lution at full-width, half-maximum (FWHM). Ap-
propriate recovery coefficients, which represent the
ratio of apparent to true isotope concentration in
the structure of interest, can be used to correct for
partial volume effects. SUV are obtained by placing
a region of interest (ROI) on the PET images. Of
note, the size of the ROI affects the radioactivity
measurements—the larger the ROI, the lower the
mean SUV. In the clinical setting, the maximum
SUV within the ROI, which represents the highest
radioactivity concentration in one voxel within the
tumor, is frequently being used for comparison be-
tween PET studies. Of note, quantitative measure-
ments with PET require appropriate and regularly
quality control, including scanner cross calibration.
In the Unites States, health insurance providers
reimburse monitoring treatment response in breast
cancer, as an adjunct to standard imaging modali-
ties for women who have locally advanced and
metastatic breast cancer, when a change in ther-
apy is anticipated. Specific criteria to determine re-
sponse by FDG-PET have yet to be established,
however. The threshold to differentiate between
responders and nonresponders based on changes
in FDG uptake is of critical importance. Chemo-
therapy should not be discontinued in respond-
ing patients, even at the cost of not identifying
all nonresponding patients. Weber and coworkers
[64,79] recently proposed to define metabolic re-
sponse as an SUV decrease larger than two times
the standard deviation (20%) of spontaneous
changes in FDG uptake. In breast cancer, a decrease
in FDG uptake of more than 20% compared with
baseline correctly identified 5 out of 7 (71.4%)
nonresponding lesions, and 12 out of 14 (85.7%)
responding lesions after the first cycle of che-
motherapy [77]. The proposed approach is also
supported by a recent study in non-small–cell
lung cancer, in which this criterion was prospec-
tively applied to 57 patients who had advanced
stage of disease [79]. Only 1 out of 27 metabolic
nonresponders achieved a partial response after
completion of chemotherapy, which resulted in a
negative predictive value of 96%; however, this
strategy needs to be validated in a separate prospec-
tive study for metastatic breast cancer patients.
Comparison between histopathologic and
metabolic response criteria
Important differences between histopathology and
FDG-PET imaging for assessment of treatment
response in breast cancer should be considered.
Histopathology is based on morphologic features
Histopathologic and Metabolic Response Criteria 91
of tissue specimens, and therapeutic changes are
assessed in vitro on a microscopic, cellular level.
In contrast, FDG-PET imaging provides informa-
tion about tumor metabolism in vivo, although
on a macroscopic level. Whereas histopathological
response assessment requires invasive procedures
to obtain surgical specimens and is routinely only
performed after completion of therapy, PET imag-
ing may provide early response evaluation. FDG-
PET imaging allows monitoring of lesions in soft
tissue, lymph nodes, the liver, the lungs, and in the
bone in one imaging procedure, in contrast to
histopathology. Histopathology is affected by the
experience of the pathologist, with some unavoid-
able observer variability. PET provides quantitative
measurements that are less observer-dependent. On
the other hand, histopathology is widely available,
whereas PET imaging is generally less accessible,
although it is becoming standard of care in major
hospitals throughout the United States. Histopa-
thology is currently the gold standard for assess-
ment of treatment response, with promising new
developments in genetic and proteomic profiling.
Sequential PET imaging, although promising, needs
to be validated and its criteria standardized in
appropriately sized prospective clinical trials before
it can and should be used in the clinical setting. The
metabolic information from PET requires thorough
evaluation regarding differences in disease-free and
overall survival of metabolic responders versus
nonresponders, as well as evaluation of the cost-
effectiveness of this approach.
References
[1] Rosai J, editor. Rosai and Ackermann’s surgical
pathology. 9th edition. Philadelphia: Mosby,
Elsevier; 2004. p. 1787–827.
[2] Elston CW, Ellis IO. Pathological prognostic fac-
tors in breast cancer. I. The value of histological
grade in breast cancer: experience from a large
study with long-term follow-up. Histopathology
1991;19:403–10.
[3] Genestie C, Zafrani B, Asselain B, et al. Compari-
son of the prognostic value of Scarff-Bloom-
Richardson and Nottingham histological grades
in a series of 825 cases of breast cancer: major
importance of the mitotic count as a component
of both grading systems. Anticancer Res 1998;18:
571–6.
[4] Aamdal S, Bormer O, Jorgensen O, et al. Estrogen
receptors and long-term prognosis in breast
cancer. Cancer 1984;53:2525–9.
[5] Butler JA, Bretsky S, Menendez-Botet C, et al. Es-
trogen receptor protein of breast cancer as a pre-
dictor of recurrence. Cancer 1985;55:1178–81.
[6] Slamon DJ, Godolphin W, Jones LA, et al. Stud-
ies of the HER-2/neu proto-oncogene in human
92 Sassen et al
breast and ovarian cancer. Science 1989;244:
707–12.
[7] Menard S, Fortis S, Castiglioni F, et al. HER2 as
a prognostic factor in breast cancer. Oncology
2001;61(Suppl 2):67–72.
[8] Barnes DM, Dublin EA, Fisher CJ, et al. Im-
munohistochemical detection of p53 protein in
mammary carcinoma: an important new in-
dependent indicator of prognosis? Hum Pathol
1993;24:469–76.
[9] Rosen PP, Lesser ML, Arroyo CD, et al. p53 in
node-negative breast carcinoma: an immuno-
histochemical study of epidemiologic risk factors,
histologic features, and prognosis. J Clin Oncol
1995;13:821–30.
[10] Cocconi G. Chemotherapy with and without an-
thracycline. J Clin Oncol 2000;18:1392–3.
[11] Hamilton A, Hortobagyi G. Chemotherapy: what
progress in the last 5 years? J Clin Oncol 2005;
23:1760–75.
[12] van der Hage JA, van de Velde CJ, Julien JP, et al.
Preoperative chemotherapy in primary opera-
ble breast cancer: results from the European
Organization for Research and Treatment of Can-
cer trial 10902. J Clin Oncol 2001;19:4224–37.
[13] Bonadonna G, Veronesi U, Brambilla C, et al.
Primary chemotherapy to avoid mastectomy in
tumors with diameters of three centimeters or
more. J Natl Cancer Inst 1990;82:1539–45.
[14] Fisher ER, Wang J, Bryant J, et al. Pathobiology
of preoperative chemotherapy: findings from the
National Surgical Adjuvant Breast and Bowel
(NSABP) protocol B-18. Cancer 2002;95:681–95.
[15] Feldman LD, Hortobagyi GN, Buzdar AU, et al.
Pathological assessment of response to induction
chemotherapy in breast cancer. Cancer Res 1986;
46:2578–81.
[16] Kuerer HM, Newman LA, Smith TL, et al. Clinical
course of breast cancer patients with complete
pathologic primary tumor and axillary lymph
node response to doxorubicin-based neoadjuvant
chemotherapy. J Clin Oncol 1999;17:460–9.
[17] Fisher B, Bryant J, Wolmark N, et al. Effect of
preoperative chemotherapy on the outcome of
women with operable breast cancer. J Clin Oncol
1998;16:2672–85.
[18] Chollet P, Amat S, Cure H, et al. Prognostic sig-
nificance of a complete pathological response
after induction chemotherapy in operable breast
cancer. Br J Cancer 2002;86:1041–6.
[19] Machiavelli MR, Romero AO, Perez JE, et al.
Prognostic significance of pathological response
of primary tumor and metastatic axillary lymph
nodes after neoadjuvant chemotherapy for lo-
cally advanced breast carcinoma. Cancer J Sci Am
1998;4:125–31.
[20] Sinn HP, Schmid H, Junkermann H, et al. [His-
tologic regression of breast cancer after pri-
mary (neoadjuvant) chemotherapy]. Geburtshilfe
Frauenheilkd 1994;54:552–8 [in German].
[21] Wang J, Buchholz TA, Middleton LP, et al. As-
sessment of histologic features and expression of
biomarkers in predicting pathologic response to
anthracycline-based neoadjuvant chemotherapy
in patients with breast carcinoma. Cancer 2002;
94:3107–14.
[22] Therasse P, Arbuck SG, Eisenhauer EA, et al. New
guidelines to evaluate the response to treatment
in solid tumors. European Organization for Re-
search and Treatment of Cancer, National Cancer
Institute of the United States, National Cancer
Institute of Canada. J Natl Cancer Inst 2000;92:
205–16.
[23] Miller AB, Hoogstraten B, Staquet M, et al. Re-
porting results of cancer treatment. Cancer 1981;
47:207–14.
[24] Buyse M, Thirion P, Carlson RW, et al. Relation
between tumour response to first-line chemo-
therapy and survival in advanced colorectal can-
cer: a meta-analysis. Meta-Analysis Group in
Cancer. Lancet 2000;356:373–8.
[25] Salzer-Kuntschik M, Brand G, Delling G. [Deter-
mination of the degree of morphological regres-
sion following chemotherapy in malignant bone
tumors]. Pathologe 1983;4:135–41 [in German].
[26] Becker K, Mueller JD, Schulmacher C, et al.
Histomorphology and grading of regression in
gastric carcinoma treated with neoadjuvant che-
motherapy. Cancer 2003;98:1521–30.
[27] Mandard AM, Dalibard F, Mandard JC, et al.
Pathologic assessment of tumor regression after
preoperative chemoradiotherapy of esophageal
carcinoma. Clinicopathologic correlations. Can-
cer 1994;73:2680–6.
[28] Evans DB, Rich TA, Byrd DR, et al. Preoperative
chemoradiation and pancreaticoduodenectomy
for adenocarcinoma of the pancreas. Arch Surg
1992;127:1335–9.
[29] Junker K, Thomas M, Schulmann K, et al.
[Regression grading of neoadjuvant non-small-
cell lung carcinoma treatment]. Pathologe 1997;
18:131–40 [in German].
[30] Chevallier B, Roche H, Olivier JP, et al. Inflam-
matory breast cancer. Pilot study of intensive
induction chemotherapy (FEC-HD) results in a
high histologic response rate. Am J Clin Oncol
1993;16:223–8.
[31] Gajdos C, Tartter PI, Estabrook A, et al. Relation-
ship of clinical and pathologic response to neo-
adjuvant chemotherapy and outcome of locally
advanced breast cancer. J Surg Oncol 2002;80:
4–11.
[32] Aktepe F, Kapucuoglu N, Pak I. The effects of
chemotherapy on breast cancer tissue in locally
advanced breast cancer. Histopathology 1996;29:
63–7.
[33] Seno R, Sparano JA, Fineberg SA. Gross and
histologic features of locally advanced breast
cancer after neoadjuvant chemotherapy. Anat
Pathol 1998;3:169–80.
[34] Kennedy S, Merino MJ, Swain SM, et al. The ef-
fects of hormonal and chemotherapy on tumoral
and nonneoplastic breast tissue. Hum Pathol
1990;21:192–8.

[35] Rasbridge SA, Gillett CE, Seymour AM, et al. The
effects of chemotherapy on morphology, cellular
proliferation, apoptosis and oncoprotein expres-
sion in primary breast carcinoma. Br J Cancer
1994;70:335–41.
[36] Moreno A, Escobedo A, Benito E, et al. Patho-
logic changes related to CMF primary chemo-
therapy in breast cancer. Pathological evaluation
of response predicts clinical outcome. Breast
Cancer Res Treat 2002;75:119–25.
[37] Wolmark N, Wang J, Mamounas E, et al. Pre-
operative chemotherapy in patients with op-
erable breast cancer: nine-year results from
National Surgical Adjuvant Breast and Bowel
Project B-18. J Natl Cancer Inst Monogr 2001;
30:96–102.
[38] Honkoop AH, van Diest PJ, de Jong JS, et al.
Prognostic role of clinical, pathological and
biological characteristics in patients with locally
advanced breast cancer. Br J Cancer 1998;77:
621–6.
[39] Honkoop AH, Pinedo HM, De Jong JS, et al.
Effects of chemotherapy on pathologic and bio-
logic characteristics of locally advanced breast
cancer. Am J Clin Pathol 1997;107:211–8.
[40] Pierga JY, Mouret E, Dieras V, et al. Prognostic
value of persistent node involvement after neo-
adjuvant chemotherapy in patients with opera-
ble breast cancer. Br J Cancer 2000;83:1480–7.
[41] Buzdar AU, Singletary SE, Theriault RL, et al.
Prospective evaluation of paclitaxel versus com-
bination chemotherapy with fluorouracil, doxo-
rubicin, and cyclophosphamide as neoadjuvant
therapy in patients with operable breast cancer.
J Clin Oncol 1999;17:3412–7.
[42] Colleoni M, Minchella I, Mazzarol G, et al. Re-
sponse to primary chemotherapy in breast can-
cer patients with tumors not expressing estrogen
and progesterone receptors. Ann Oncol 2000;11:
1057–9.
[43] Stearns V, Singh B, Tsangaris T, et al. A pro-
spective randomized pilot study to evaluate pre-
dictors of response in serial core biopsies to
single agent neoadjuvant doxorubicin or pacli-
taxel for patients with locally advanced breast
cancer. Clin Cancer Res 2003;9:124–33.
[44] MacGrogan G, Mauriac L, Durand M, et al.
Primary chemotherapy in breast invasive carci-
noma: predictive value of the immunohisto-
chemical detection of hormonal receptors, p53,
c-erbB-2, MiB1, pS2 and GST pi. Br J Cancer
1996;74:1458–65.
[45] Niskanen E, Blomqvist C, Franssila K, et al. Pre-
dictive value of c-erbB-2, p53, cathepsin-D and
histology of the primary tumour in metastatic
breast cancer. Br J Cancer 1997;76:917–22.
[46] Rozan S, Vincent-Salomon A, Zafrani B, et al. No
significant predictive value of c-erbB-2 or p53
expression regarding sensitivity to primary che-
motherapy or radiotherapy in breast cancer. Int J
Cancer 1998;79:27–33.
[47] Cleator S, Parton M, Dowsett M. The biology of
Histopathologic and Metabolic Response Criteria 93
neoadjuvant chemotherapy for breast cancer.
Endocr Relat Cancer 2002;9:183–95.
[48] Buchholz TA, Davis DW, McConkey DJ, et al.
Chemotherapy-induced apoptosis and Bcl-2 lev-
els correlate with breast cancer response to che-
motherapy. Cancer J 2003;9:33–41.
[49] Oude Elferink RF, Goldschmidt HM, Majoor HJ,
et al. Rapid assay for the in vitro chemosensitivity
testing of human breast tumours. Control of
assay conditions and quality assurance. Pharm
Weekbl Sci 1989;11:155–60.
[50] Krasna L, Netikova I, Chaloupkova A, et al. As-
sessment of in vitro drug resistance of human
breast cancer cells subcultured from biopsy speci-
mens. Anticancer Res 2003;23:2593–9.
[51] Mehta RS, Bornstein R, Yu IR, et al. Breast cancer
survival and in vitro tumor response in the ex-
treme drug resistance assay. Breast Cancer Res
Treat 2001;66:225–37.
[52] Ziegler LD, Connelly JH, Frye D, et al. Lack of
correlation between histologic findings and re-
sponse to chemotherapy in metastatic breast
cancer. Cancer 1991;68:628–33.
[53] Thomas E, Holmes FA, Smith TL, et al. The use of
alternate, non-cross-resistant adjuvant chemo-
therapy on the basis of pathologic response to
a neoadjuvant doxorubicin-based regimen in
women with operable breast cancer: long-term
results from a prospective randomized trial.
J Clin Oncol 2004;22:2294–302.
[54] Cristofanilli M, Budd GT, Ellis MJ, et al. Cir-
culating tumor cells, disease progression, and
survival in metastatic breast cancer. N Engl J
Med 2004;351:781–91.
[55] Shimizu D, Ishikawa T, Ichikawa Y, et al. Cur-
rent progress in the prediction of chemosensi-
tivity for breast cancer. Breast Cancer 2004;11:
42–8.
[56] Perou CM, Sorlie T, Eisen MB, et al. Molecular
portraits of human breast tumours. Nature 2000;
406:747–52.
[57] Sorlie T, Perou CM, Tibshirani R, et al. Gene
expression patterns of breast carcinomas distin-
guish tumor subclasses with clinical implications.
Proc Natl Acad Sci USA 2001;98:10869–74.
[58] van’t Veer LJ, Dai H, van de Vijver MJ, et al. Gene
expression profiling predicts clinical outcome of
breast cancer. Nature 2002;415:530–6.
[59] Chang JC, Wooten EC, Tsimelzon A, et al. Gene
expression profiling for the prediction of thera-
peutic response to docetaxel in patients with
breast cancer. Lancet 2003;362:362–9.
[60] Ayers M, Symmans WF, Stec J, et al. Gene ex-
pression profiles predict complete pathologic
response to neoadjuvant paclitaxel and fluoroura-
cil, doxorubicin, and cyclophosphamide chemo-
therapy in breast cancer. J Clin Oncol 2004;22:
2284–93.
[61] Avril N. GLUT1 expression in tissue and (18)F-
FDG uptake. J Nucl Med 2004;45:930–2.
[62] Zasadny KR, Wahl RL. Standardized uptake val-
ues of normal tissues at PET with 2-[fluorine-18]-
94 Sassen et al
fluoro-2-deoxy-D-glucose: variations with body
weight and a method for correction. Radiology
1993;189:847–50.
[63] Minn H, Zasadny KR, Quint LE, et al. Lung can-
cer: reproducibility of quantitative measurements
for evaluating 2-[F-18]-fluoro-2-deoxy-D-glucose
uptake at PET. Radiology 1995;196:167–73.
[64] Weber WA, Ziegler SI, Thodtmann R, et al. Re-
producibility of metabolic measurements in ma-
lignant tumors using FDG PET. J Nucl Med 1999;
40:1771–7.
[65] Romer W, Hanauske AR, Ziegler S, et al. Positron
emission tomography in non-Hodgkin’s lym-
phoma: assessment of chemotherapy with fluoro-
deoxyglucose. Blood 1998;91:4464–71.
[66] Schelling M, Avril N, Nahrig J, et al. Positron
emission tomography using [(18)F]fluorodeoxy-
glucose for monitoring primary chemotherapy in
breast cancer. J Clin Oncol 2000;18:1689–95.
[67] Smith IC, Welch AE, Hutcheon AW, et al.
Positron emission tomography using [(18)F]-
fluorodeoxy-D-glucose to predict the pathologic
response of breast cancer to primary chemother-
apy. J Clin Oncol 2000;18:1676–88.
[68] Wahl RL, Zasadny K, Helvie M, et al. Metabolic
monitoring of breast cancer chemohormono-
therapy using positron emission tomography:
initial evaluation. J Clin Oncol 1993;11:2101–11.
[69] Mankoff DA, Dunnwald LK, Gralow JR, et al.
Blood flow and metabolism in locally advanced
breast cancer: relationship to response to ther-
apy. J Nucl Med 2002;43:500–9.
[70] Mankoff DA, Dunnwald LK, Gralow JR, et al.
Changes in blood flow and metabolism in lo-
cally advanced breast cancer treated with neo-
adjuvant chemotherapy. J Nucl Med 2003;44:
1806–14.
[71] Tseng J, Dunnwald LK, Schubert EK, et al. 18F-
FDG kinetics in locally advanced breast cancer:
correlation with tumor blood flow and changes
in response to neoadjuvant chemotherapy.
J Nucl Med 2004;45:1829–37.
[72] Dehdashti F, Flanagan FL, Mortimer JE, et al.
Positron emission tomographic assessment of
‘‘metabolic flare’’ to predict response of meta-
static breast cancer to antiestrogen therapy. Eur J
Nucl Med 1999;26:51–6.
[73] Mortimer JE, Dehdashti F, Siegel BA, et al.
Metabolic flare: indicator of hormone respon-
siveness in advanced breast cancer. J Clin Oncol
2001;19:2797–803.
[74] Jansson T, Westlin JE, Ahlstrom H, et al. Positron
emission tomography studies in patients with
locally advanced and/or metastatic breast cancer:
a method for early therapy evaluation? J Clin
Oncol 1995;13:1470–7.
[75] Gennari A, Donati S, Salvadori B, et al. Role
of 2-[18F]-fluorodeoxyglucose (FDG) positron
emission tomography (PET) in the early assess-
ment of response to chemotherapy in metastatic
breast cancer patients. Clin Breast Cancer 2000;
1:156–61 [discussion: 162–3].
[76] Stafford SE, Gralow JR, Schubert EK, et al. Use of
serial FDG PET to measure the response of bone-
dominant breast cancer to therapy. Acad Radiol
2002;9:913–21.
[77] Dose Schwarz J, Bader M, Jenicke L, et al. Early
prediction of response to chemotherapy in meta-
static breast cancer using sequential 18F-FDG-PET
imaging. J Nucl Med 2005;46(7):1144–50.
[78] Beaulieu S, Kinahan P, Tseng J, et al. SUV varies
with time after injection in (18)F-FDG PET of
breast cancer: characterization and method to
adjust for time differences. J Nucl Med 2003;44:
1044–50.
[79] Weber WA, Petersen V, Schmidt B, et al. Positron
emission tomography in non-small-cell lung
cancer: prediction of response to chemotherapy
by quantitative assessment of glucose use. J Clin
Oncol 2003;21:2651–7.