Lung Cancer, An Issue of PET Clinics - Fletcher.pdf
Lung Cancer, An Issue of PET Clinics - Fletcher.pdf
P O S I T R O N
E M I S S I O N
T O M O G R A P H Y
PET Clin 1 (2006) ix–x
ixPreface
James W. Fletcher, MD
Guest Editor
James W. Fletcher, MD
Division of Nuclear Medicine
PET Imaging Center
Department of Radiology
Indiana University School of Medicine
Indiana/Purdue University at Indianapolis University
Hospital, Room 0655
550 North University Boulevard
Indianapolis, IN 46234, USA
E-mail address:
jwfletch@iupui.eduPositron emission tomography (PET) is now an
important cancer-imaging tool, both for diagnosis
and staging, as well as offering prognostic in-
formation based on response. PET is the gold
standard in the evaluation of an indeterminate
solitary pulmonary nodule (SPN) or mass, where
PET has proved to be significantly more accurate
than computed tomography (CT). PET is signifi-
cantly more accurate than CT in the evaluation of
metastatic spread to locoregional lymph nodes, so
that invasive surgical staging may be omitted in
many patients with negative mediastinal PET
images. Invasive surgical staging remains mandatory
in patients with positive mediastinal PET images
because of the possibility of false-positive findings
owing to inflammatory nodes or granulomatous
disorders. PET is a useful adjunct to conventional
imaging in the search for metastatic spread. This
may be due to the finding of unexpected metastatic
lesions or the exclusion of malignancy in lesions
that are equivocal on standard imaging. However,
at this time, PET does not replace conventional
imaging.1556-8598/07/$ – see front matter ª 2007 Elsevier Inc. All right
pet.theclinics.comThe evaluation of SPNs was one of the first
Centers for Medicare and Medicaid Services
approved and widespread Fluorodeoxyglucose-
PET oncologic applications. Collectively, more has
been published on the use of FDG-PET in
evaluation of SPNs and staging of non–small-cell
lung cancer (NSCLC) than any other clinical PET
application. The evaluation of lung cancer probably
represents the most common application of FDG-
PET in most clinical departments.
Lung cancer is currently the leading cause of
cancer deaths among both men and women in the
United States, with statistical estimates of 169,400
new diagnoses and 154,900 cancer deaths in 2002.
In editing this issue of Positron Emission Tomogra-
phy Clinics on lung cancer, I have endeavored to
incorporate the perspectives of a number of special-
ists who are typically involved in the diagnosis,
treatment, and management of patients with lung
cancer. Contributions are presented from the
perspective of the pulmonary radiologist, the
interventional radiologist, the respiratory medical
oncologist, the radiation therapy oncologist, ands reserved. doi:10.1016/j.cpet.2007.01.001
Prefacexnuclear medicine/PET specialists. Two articles were
written by individuals who have special additional
expertise and experience in medical economics and
cost–benefit, cost-effectiveness analysis (CEA).
The article on the diagnosis of lung cancer from
the perspective of the pulmonary radiologist is
presented by Drs. Teague and Conces. This article is
a superb summary of the radiographic features of
lung cancer that is very well illustrated with
excellent examples of the radiographic features of
both benign and malignant nodules and masses.
The roles of transthoracic needle biopsy and
bronchoscopic biopsy are also discussed.
In their article on staging of lung cancer,
Drs. Wynants et al. provide an excellent overview of
lung cancer staging as well as detailed information
on the N-, T-, and M-factors. They also discuss the
impact of FDG-PET in the workup of lung cancer,
with emphasis on the valuable role of PET in
guidance for invasive procedures. An example of
the beneficial impact of PET on overall stage and
patient management is the substantial reduction in
the number of futile thoracotomies observed when
PET is incorporated into the workup.
Dr. Michael Mac Manus provides an entire article
devoted to the role of PET in radiation therapy
planning. He presents a detailed summary and
review of the role of imaging in radiation oncology,
with special emphasis on PET and PET/CT. The
results of the use of PET for selection of patients for
radical radiation therapy and the impact of PET on
patient outcomes are also discussed. It is clear that
PET is transforming the way that radiation oncol-
ogists approach the treatment of lung cancer. The
use of PET for selection of patients for radical
therapy alone has already improved the apparent
success rate with radiation therapy.
In the first article on the cost-effectiveness of
FDG-PET, Drs. Hoekstra et al. detail their experience
with the introduction of PET in the Netherlands,
together with the design of parallel cost-effective-
ness studies from their institution and other
hospitals in their region. Techniques and designs
for decision modeling and clinical value studies are
described in detail, along with the intrinsic prob-
lems of implementation. The authors define reason-able outcome measures for randomized clinical
trials (RCTs) that employ diagnostic imaging
procedures, ie, the extent to which appropriate
therapy is applied as a result of the diagnostic
intervention. The results of their PET in lung cancer
staging (PLUS) multicenter study are presented as
an example of a well-designed RCT with FDG-PET as
the imaging modality, where the use of PET resulted
in a 50% reduction in futile thoracotomies.
A more specific example of cost-effectiveness
studies in imaging is presented in an article on
the cost-effectiveness of PET for characterizing SPNs
by Dr. Michael Gould. This article introduces
the topic of CEA with a very clear and understand-
able discussion of CEA compared with other types of
economic evaluations. The author also provides the
reader with a concise guide to interpreting the results
of a CEA. The results of several CEA studies on FDG-
PET in SPN are reviewed, with comments and
critiques regarding strengths and weaknesses of
study design. As the author points out, payers will
likely require evidence of cost-effectiveness when
making future coverage decisions as health care
costs continue to increase in this country.
An important treatise on the accuracy of dual-
modality FDG-PET/CT imaging in the staging of
NSCLC compared with FDG-PET alone, as well as
FDG-PET and CT read side by side, is presented
in the article on anato-metabolic imaging by
Drs. Freudenberg et al. These authors discuss the
value and limitations of software versus hardware
imaging fusion in FDG-PET/CT. The results of
several recent large studies indicate a definite
advantage for hardware dual-modality imaging
fusion, inasmuch as the combined modality is able
to detect significantly more lesions and has a larger
impact on change/modification of patient manage-
ment. As a bonus, the article provides the reader
with an excellent summary of features designed to
optimize FDG-PET/CT imaging protocols.
Many thanks are extended to all of the authors
who contributed to this issue of Positron Emission
Tomography Clinics for their diligence and support.
A variation on the ancient Chinese proverb/curse
is offered to all: ‘‘May we continue to live in
interesting times.’’
P O S I T R O N
E M I S S I O N
T O M O G R A P H Y
PET Clin 1 (2006) 289–300
289Diagnosis of Lung Cancer:
Perspective of a Pulmonary
Radiologist
Shawn D. Teague, MD*, Dewey J. Conces, Jr, MD
- Clinical
- Radiography
- Location
- Number
- Size
- Calcification
- Dynamic enhancement
- Attenuation
- Morphology
- Margin
- Cavitation
- Growth rate
- Bayesian analysis
- Follow-up
- Tissue diagnosis
- ReferencesThe solitary pulmonary nodule (SPN), a common
incidental finding on routine radiographic imaging,
is defined as a focal, oval, or round area of increased
density in the lung that is less than 3 cm in diame-
ter. In fact, there are an estimated 150,000 SPNs de-
tected annually in the United States [1], and a single
pulmonary nodule is detected on up to 0.2% of all
chest radiographs [2]. The majority of SPNs are due
to benign diseases such as hamartomas or granulo-
mas [3,4]. A significant number, however, are ma-
lignant (up to 40% of single pulmonary nodules)
[5]. Because the care of lung cancer is dependent
on diagnosis at an early stage, it is important to
evaluate nodules as expeditiously as possible. The
evaluation of an SPN involves the use of a variety
of modalities, because up to 25% to 39% of malig-
nant nodules are inaccurately classified as benign
after radiologic assessment of size, margin, contour,
and internal characteristics [6]. In this article, the
authors address the assessment of SPN radiograph-
ically for benign or malignant characteristics.1556-8598/07/$ – see front matter ª 2007 Elsevier Inc. All righ
pet.theclinics.comClinical
Patient history plays a significant role in the evalu-
ation of SPNs. A patient under 30 years old has
a very low risk of an SPN being malignant unless
the patient has a known extrathoracic malignancy.
As the age of the patient increases, the likelihood
of malignancy tends to increase.
There are other risk factors that significantly in-
crease the likelihood of malignancy, such as a his-
tory of smoking, asbestos exposure, coal mining
occupation, fibrosis, and known extrathoracic ma-
lignancy. In a patient who has a history of smoking,
the incidence of lung cancer no longer increases
once the patient ceases to smoke; however, the inci-
dence is never equal to that of an individual who
has never smoked. Presenting symptoms can be
an important factor from a clinical standpoint in
determining the likelihood of malignancy. In a pa-
tient who has suspected infection, it is reasonable
to pursue a less aggressive course and obtainDepartment of Radiology, Indiana University School of Medicine, 550 North University Boulevard, Room
0279, University Hospital, Indianapolis, IN 46250, USA
* Corresponding author.
E-mail address: sdteague@iupui.edu (S.D. Teague).ts reserved. doi:10.1016/j.cpet.2006.09.004
Teague & Conces290short-term follow-up radiographic imaging to
document improvement in the nodule. In certain
extrathoracic primary malignancies, the risk of an
SPN being a metastases is much greater (such as
melanoma, sarcoma, or testicular carcinoma) com-
pared with other malignancies (such as head
and neck squamous cell carcinoma) in which an
SPN is much more likely to be a primary lung
malignancy [7].
Radiography
The first imaging usually performed is the plain
chest radiograph. An SPN, however, is seldom
noted on a chest radiograph before it reaches
9 mm in diameter [8]. The false-negative rate for
primary lung cancer on routine radiographic exam-
inations has been reported to be 30% to 40% [9].
When evaluating plain radiographs, it is important
to obtain prior examinations for comparison. A
multicenter trial from the Mayo Clinic indicated
that nearly 90% of peripheral lung cancers could
be seen retrospectively on prior radiographic exam-
ination that were reported as negative at the time of
initial interpretation [10]. Missed or overlooked
lung nodules tend to occur in the lung apices, peri-
hilar regions, and behind the heart and diaphragm
on the frontal view. On the lateral view, lesions are
typically missed when located posteriorly over the
spine or anteriorly over the heart [11,12].
Some lesions can be determined as benign on
a plain radiograph; however, most lesions will re-
quire further evaluation. If there is a question that
the density seen on a chest radiograph represents
a real nodule, chest fluoroscopy can be performed.
Fluoroscopy can be used to determine if the nodule
is real and not just an overlap of other soft tissues
resulting in a ‘‘pseudonodule.’’ Up to 20% of sus-
pected nodules are actually mimics of other entities
such as rib fractures, skin lesions, or combined
areas of soft tissue attenuation [1]. Low kilovolts
peak chest radiographs can be obtained to deter-
mine if calcification is present.
If these two examinations are not performed or
are nonconclusive, the next diagnostic imaging
test to consider is a thoracic CT examination. This
examination cannot only define the morphologic
characteristics of the nodule, but excels at identifica-
tion of calcification within the nodule. CT is 10 to
20 times more sensitive for calcification when com-
pared with traditional tomography, and 22% to
36% of nodules not obviously calcified on tomo-
grams are shown to contain calcification on CT
examination [13–15]. CT also allows quantitative
assessment of calcification present in the nodule
[15–18]. If the calcification is visible on thin-section CT images, the Hounsfield measurement
will usually have values of 400 H or greater. How-
ever, calcium can be present and not grossly visible.
Therefore, if the measurement is 200 H or greater,
then the nodule can be declared as containing
calcification [19]. CT can also further characterize
the features of the nodule, including nodule mar-
gin, tissue composition, number and size of lesions,
and growth rate (when serial examinations are
available). However, nodules can also be missed
on CT scans. Swensen and colleagues [20] reported
that of nodules that were missed on previous
screening CT examinations, 62% were less than
4 mm in diameter, 37% were between 4 mm and
7 mm, and 2% were 8 mm to 20 mm in diameter.
Ko and colleagues [21] reported that ground glass
nodules were less frequently detected (65%)
compared with solid nodules (83%) and there
was a decreased sensitivity for detecting central
compared with peripheral nodules (61% versus
80%). Finally, they also reported nodules adjacent
to the pleura were better detected (84%) versus
those not in contact with the pleural surface (75%).
Thinner sections, multiplanar reformats (espe-
cially MIPs), and cine viewing have all been shown
to improve nodule detection [22–25]. CAD systems
are being developed to help improve the nodule
detection on both plain radiography and CT exam-
inations. The efficiency of CAD systems has been
reported to be between 38% and 95% [21].
The CT evaluation of a patient who has proven
or suspected SPN should start with an examination
without use of intravenous contrast through the
entire chest. This serves to locate the nodule for
further evaluation with thin sections, and also
to document other abnormalities that include
effusions, lymphadenopathy, or additional nod-
ules. If there is only an SPN, then thin-section
(1–3 mm) CT examination is suggested to fully
characterize an SPN. Unless characteristics such
as central calcification or fat are noted, the nodule
will still remain indeterminate. In this situation,
the nodule can be further evaluated with dynamic
contrast-enhanced CT scan if it meets the appropri-
ate criteria, or followed up over time with serial
CT scans for growth assessment. Those nodules
with characteristics suggesting malignancy can be
further evaluated with a radiographic or surgical
biopsy procedure.
With multidetector CT scanners, the thoracic
scans can be performed with isotropic images retro-
spectively reconstructed from the raw data of rou-
tine helical scans provided the thinnest detector
configuration is used for the initial routine scan.
This somewhat eliminates the need to obtain addi-
tional thin sections specifically through the nodule
in question. These thin sections are also useful for
...